{
  "best_global_step": 2600,
  "best_metric": 0.1722700148820877,
  "best_model_checkpoint": "saves/ia3/gemma-3-1b-it/train_copa_1745950321/checkpoint-2600",
  "epoch": 444.44444444444446,
  "eval_steps": 200,
  "global_step": 40000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05555555555555555,
      "grad_norm": 16.86914825439453,
      "learning_rate": 4.999999876629946e-05,
      "loss": 7.3595,
      "num_input_tokens_seen": 1376,
      "step": 5
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": 18.08570671081543,
      "learning_rate": 4.999999375439123e-05,
      "loss": 7.6691,
      "num_input_tokens_seen": 2800,
      "step": 10
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 18.84496307373047,
      "learning_rate": 4.9999984887169785e-05,
      "loss": 7.9391,
      "num_input_tokens_seen": 4240,
      "step": 15
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 18.026498794555664,
      "learning_rate": 4.9999972164636506e-05,
      "loss": 7.7538,
      "num_input_tokens_seen": 5664,
      "step": 20
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 16.30668830871582,
      "learning_rate": 4.999995558679334e-05,
      "loss": 7.1566,
      "num_input_tokens_seen": 7072,
      "step": 25
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 12.80604076385498,
      "learning_rate": 4.999993515364287e-05,
      "loss": 6.8696,
      "num_input_tokens_seen": 8480,
      "step": 30
    },
    {
      "epoch": 0.3888888888888889,
      "grad_norm": 20.748817443847656,
      "learning_rate": 4.999991086518822e-05,
      "loss": 7.8009,
      "num_input_tokens_seen": 9840,
      "step": 35
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 24.94146156311035,
      "learning_rate": 4.999988272143315e-05,
      "loss": 8.2798,
      "num_input_tokens_seen": 11200,
      "step": 40
    },
    {
      "epoch": 0.5,
      "grad_norm": 17.470874786376953,
      "learning_rate": 4.999985072238199e-05,
      "loss": 7.4146,
      "num_input_tokens_seen": 12640,
      "step": 45
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 11.26298999786377,
      "learning_rate": 4.999981486803969e-05,
      "loss": 6.8518,
      "num_input_tokens_seen": 14032,
      "step": 50
    },
    {
      "epoch": 0.6111111111111112,
      "grad_norm": 13.12817096710205,
      "learning_rate": 4.999977515841176e-05,
      "loss": 5.7519,
      "num_input_tokens_seen": 15472,
      "step": 55
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 11.03712272644043,
      "learning_rate": 4.9999731593504344e-05,
      "loss": 5.3544,
      "num_input_tokens_seen": 16896,
      "step": 60
    },
    {
      "epoch": 0.7222222222222222,
      "grad_norm": 13.830143928527832,
      "learning_rate": 4.999968417332415e-05,
      "loss": 4.1454,
      "num_input_tokens_seen": 18336,
      "step": 65
    },
    {
      "epoch": 0.7777777777777778,
      "grad_norm": 18.25334358215332,
      "learning_rate": 4.999963289787848e-05,
      "loss": 3.0636,
      "num_input_tokens_seen": 19760,
      "step": 70
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 19.46759796142578,
      "learning_rate": 4.999957776717526e-05,
      "loss": 3.6125,
      "num_input_tokens_seen": 21136,
      "step": 75
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 10.358040809631348,
      "learning_rate": 4.9999518781222984e-05,
      "loss": 3.0044,
      "num_input_tokens_seen": 22544,
      "step": 80
    },
    {
      "epoch": 0.9444444444444444,
      "grad_norm": 16.970083236694336,
      "learning_rate": 4.9999455940030746e-05,
      "loss": 2.0468,
      "num_input_tokens_seen": 23920,
      "step": 85
    },
    {
      "epoch": 1.0,
      "grad_norm": 10.724055290222168,
      "learning_rate": 4.999938924360824e-05,
      "loss": 2.4001,
      "num_input_tokens_seen": 25312,
      "step": 90
    },
    {
      "epoch": 1.0555555555555556,
      "grad_norm": 12.355313301086426,
      "learning_rate": 4.999931869196575e-05,
      "loss": 1.9364,
      "num_input_tokens_seen": 26720,
      "step": 95
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 7.455521106719971,
      "learning_rate": 4.999924428511416e-05,
      "loss": 1.6966,
      "num_input_tokens_seen": 28096,
      "step": 100
    },
    {
      "epoch": 1.1666666666666667,
      "grad_norm": 5.097175598144531,
      "learning_rate": 4.999916602306494e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 29488,
      "step": 105
    },
    {
      "epoch": 1.2222222222222223,
      "grad_norm": 10.398807525634766,
      "learning_rate": 4.999908390583016e-05,
      "loss": 1.8339,
      "num_input_tokens_seen": 30880,
      "step": 110
    },
    {
      "epoch": 1.2777777777777777,
      "grad_norm": 4.991722106933594,
      "learning_rate": 4.999899793342247e-05,
      "loss": 1.6853,
      "num_input_tokens_seen": 32256,
      "step": 115
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 4.353153228759766,
      "learning_rate": 4.999890810585516e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 33616,
      "step": 120
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 2.387917995452881,
      "learning_rate": 4.999881442314206e-05,
      "loss": 0.8846,
      "num_input_tokens_seen": 35040,
      "step": 125
    },
    {
      "epoch": 1.4444444444444444,
      "grad_norm": 8.633458137512207,
      "learning_rate": 4.9998716885297617e-05,
      "loss": 0.8034,
      "num_input_tokens_seen": 36480,
      "step": 130
    },
    {
      "epoch": 1.5,
      "grad_norm": 9.863308906555176,
      "learning_rate": 4.999861549233688e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 37888,
      "step": 135
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 7.095940589904785,
      "learning_rate": 4.999851024427548e-05,
      "loss": 0.8779,
      "num_input_tokens_seen": 39296,
      "step": 140
    },
    {
      "epoch": 1.6111111111111112,
      "grad_norm": 9.27439022064209,
      "learning_rate": 4.999840114112965e-05,
      "loss": 0.7859,
      "num_input_tokens_seen": 40704,
      "step": 145
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 4.413433074951172,
      "learning_rate": 4.999828818291621e-05,
      "loss": 0.6751,
      "num_input_tokens_seen": 42112,
      "step": 150
    },
    {
      "epoch": 1.7222222222222223,
      "grad_norm": 0.8119032979011536,
      "learning_rate": 4.999817136965259e-05,
      "loss": 0.6771,
      "num_input_tokens_seen": 43520,
      "step": 155
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 7.583204746246338,
      "learning_rate": 4.9998050701356794e-05,
      "loss": 0.8089,
      "num_input_tokens_seen": 44912,
      "step": 160
    },
    {
      "epoch": 1.8333333333333335,
      "grad_norm": 9.491842269897461,
      "learning_rate": 4.999792617804744e-05,
      "loss": 0.4834,
      "num_input_tokens_seen": 46272,
      "step": 165
    },
    {
      "epoch": 1.8888888888888888,
      "grad_norm": 1.9000917673110962,
      "learning_rate": 4.9997797799743724e-05,
      "loss": 0.524,
      "num_input_tokens_seen": 47664,
      "step": 170
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 17.943260192871094,
      "learning_rate": 4.999766556646545e-05,
      "loss": 0.6669,
      "num_input_tokens_seen": 49072,
      "step": 175
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.0800528526306152,
      "learning_rate": 4.9997529478232996e-05,
      "loss": 0.5133,
      "num_input_tokens_seen": 50464,
      "step": 180
    },
    {
      "epoch": 2.0555555555555554,
      "grad_norm": 2.6266374588012695,
      "learning_rate": 4.9997389535067365e-05,
      "loss": 0.7405,
      "num_input_tokens_seen": 51904,
      "step": 185
    },
    {
      "epoch": 2.111111111111111,
      "grad_norm": 6.269982814788818,
      "learning_rate": 4.999724573699012e-05,
      "loss": 0.5696,
      "num_input_tokens_seen": 53280,
      "step": 190
    },
    {
      "epoch": 2.1666666666666665,
      "grad_norm": 3.7333648204803467,
      "learning_rate": 4.9997098084023457e-05,
      "loss": 0.5297,
      "num_input_tokens_seen": 54720,
      "step": 195
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 1.568212628364563,
      "learning_rate": 4.999694657619013e-05,
      "loss": 0.5214,
      "num_input_tokens_seen": 56176,
      "step": 200
    },
    {
      "epoch": 2.2222222222222223,
      "eval_loss": 0.5373854041099548,
      "eval_runtime": 1.2303,
      "eval_samples_per_second": 32.513,
      "eval_steps_per_second": 16.257,
      "num_input_tokens_seen": 56176,
      "step": 200
    },
    {
      "epoch": 2.2777777777777777,
      "grad_norm": 0.9515780806541443,
      "learning_rate": 4.999679121351352e-05,
      "loss": 0.5265,
      "num_input_tokens_seen": 57584,
      "step": 205
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 7.1056671142578125,
      "learning_rate": 4.9996631996017565e-05,
      "loss": 0.4839,
      "num_input_tokens_seen": 58944,
      "step": 210
    },
    {
      "epoch": 2.388888888888889,
      "grad_norm": 2.663908004760742,
      "learning_rate": 4.9996468923726835e-05,
      "loss": 0.4941,
      "num_input_tokens_seen": 60352,
      "step": 215
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 10.243932723999023,
      "learning_rate": 4.999630199666647e-05,
      "loss": 0.5928,
      "num_input_tokens_seen": 61728,
      "step": 220
    },
    {
      "epoch": 2.5,
      "grad_norm": 4.802247047424316,
      "learning_rate": 4.999613121486222e-05,
      "loss": 0.5933,
      "num_input_tokens_seen": 63136,
      "step": 225
    },
    {
      "epoch": 2.5555555555555554,
      "grad_norm": 2.6775033473968506,
      "learning_rate": 4.999595657834041e-05,
      "loss": 0.5869,
      "num_input_tokens_seen": 64528,
      "step": 230
    },
    {
      "epoch": 2.611111111111111,
      "grad_norm": 3.1681978702545166,
      "learning_rate": 4.999577808712798e-05,
      "loss": 0.4105,
      "num_input_tokens_seen": 65936,
      "step": 235
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 17.032817840576172,
      "learning_rate": 4.999559574125244e-05,
      "loss": 0.4468,
      "num_input_tokens_seen": 67312,
      "step": 240
    },
    {
      "epoch": 2.7222222222222223,
      "grad_norm": 2.72995924949646,
      "learning_rate": 4.9995409540741934e-05,
      "loss": 0.5751,
      "num_input_tokens_seen": 68768,
      "step": 245
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 1.0495034456253052,
      "learning_rate": 4.999521948562516e-05,
      "loss": 0.3427,
      "num_input_tokens_seen": 70144,
      "step": 250
    },
    {
      "epoch": 2.8333333333333335,
      "grad_norm": 5.613031387329102,
      "learning_rate": 4.999502557593143e-05,
      "loss": 0.6076,
      "num_input_tokens_seen": 71568,
      "step": 255
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 6.207342147827148,
      "learning_rate": 4.999482781169066e-05,
      "loss": 0.3981,
      "num_input_tokens_seen": 72976,
      "step": 260
    },
    {
      "epoch": 2.9444444444444446,
      "grad_norm": 5.402806282043457,
      "learning_rate": 4.9994626192933324e-05,
      "loss": 0.5428,
      "num_input_tokens_seen": 74368,
      "step": 265
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.7881920337677,
      "learning_rate": 4.999442071969054e-05,
      "loss": 0.4567,
      "num_input_tokens_seen": 75760,
      "step": 270
    },
    {
      "epoch": 3.0555555555555554,
      "grad_norm": 1.4912617206573486,
      "learning_rate": 4.999421139199397e-05,
      "loss": 0.5965,
      "num_input_tokens_seen": 77168,
      "step": 275
    },
    {
      "epoch": 3.111111111111111,
      "grad_norm": 1.2257336378097534,
      "learning_rate": 4.999399820987592e-05,
      "loss": 0.5092,
      "num_input_tokens_seen": 78624,
      "step": 280
    },
    {
      "epoch": 3.1666666666666665,
      "grad_norm": 2.91635799407959,
      "learning_rate": 4.999378117336924e-05,
      "loss": 0.3152,
      "num_input_tokens_seen": 80016,
      "step": 285
    },
    {
      "epoch": 3.2222222222222223,
      "grad_norm": 4.086484432220459,
      "learning_rate": 4.9993560282507415e-05,
      "loss": 0.4362,
      "num_input_tokens_seen": 81424,
      "step": 290
    },
    {
      "epoch": 3.2777777777777777,
      "grad_norm": 2.279388189315796,
      "learning_rate": 4.9993335537324495e-05,
      "loss": 0.5072,
      "num_input_tokens_seen": 82784,
      "step": 295
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 2.8126630783081055,
      "learning_rate": 4.999310693785516e-05,
      "loss": 0.3551,
      "num_input_tokens_seen": 84208,
      "step": 300
    },
    {
      "epoch": 3.388888888888889,
      "grad_norm": 7.199731826782227,
      "learning_rate": 4.9992874484134653e-05,
      "loss": 0.4938,
      "num_input_tokens_seen": 85632,
      "step": 305
    },
    {
      "epoch": 3.4444444444444446,
      "grad_norm": 7.512269496917725,
      "learning_rate": 4.999263817619882e-05,
      "loss": 0.3235,
      "num_input_tokens_seen": 87040,
      "step": 310
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.7775722146034241,
      "learning_rate": 4.9992398014084105e-05,
      "loss": 0.3059,
      "num_input_tokens_seen": 88432,
      "step": 315
    },
    {
      "epoch": 3.5555555555555554,
      "grad_norm": 7.950046539306641,
      "learning_rate": 4.999215399782754e-05,
      "loss": 0.3622,
      "num_input_tokens_seen": 89824,
      "step": 320
    },
    {
      "epoch": 3.611111111111111,
      "grad_norm": 2.5228848457336426,
      "learning_rate": 4.999190612746675e-05,
      "loss": 0.3005,
      "num_input_tokens_seen": 91216,
      "step": 325
    },
    {
      "epoch": 3.6666666666666665,
      "grad_norm": 2.658169984817505,
      "learning_rate": 4.999165440303998e-05,
      "loss": 0.5193,
      "num_input_tokens_seen": 92640,
      "step": 330
    },
    {
      "epoch": 3.7222222222222223,
      "grad_norm": 1.3305848836898804,
      "learning_rate": 4.999139882458603e-05,
      "loss": 0.3433,
      "num_input_tokens_seen": 94032,
      "step": 335
    },
    {
      "epoch": 3.7777777777777777,
      "grad_norm": 2.717548131942749,
      "learning_rate": 4.9991139392144314e-05,
      "loss": 0.3881,
      "num_input_tokens_seen": 95424,
      "step": 340
    },
    {
      "epoch": 3.8333333333333335,
      "grad_norm": 5.798102855682373,
      "learning_rate": 4.999087610575485e-05,
      "loss": 0.3243,
      "num_input_tokens_seen": 96816,
      "step": 345
    },
    {
      "epoch": 3.888888888888889,
      "grad_norm": 2.9053900241851807,
      "learning_rate": 4.999060896545824e-05,
      "loss": 0.3575,
      "num_input_tokens_seen": 98192,
      "step": 350
    },
    {
      "epoch": 3.9444444444444446,
      "grad_norm": 2.090167760848999,
      "learning_rate": 4.999033797129568e-05,
      "loss": 0.4584,
      "num_input_tokens_seen": 99584,
      "step": 355
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.109574794769287,
      "learning_rate": 4.999006312330894e-05,
      "loss": 0.2425,
      "num_input_tokens_seen": 100960,
      "step": 360
    },
    {
      "epoch": 4.055555555555555,
      "grad_norm": 4.20232629776001,
      "learning_rate": 4.998978442154043e-05,
      "loss": 0.3382,
      "num_input_tokens_seen": 102400,
      "step": 365
    },
    {
      "epoch": 4.111111111111111,
      "grad_norm": 0.620210587978363,
      "learning_rate": 4.9989501866033125e-05,
      "loss": 0.4006,
      "num_input_tokens_seen": 103792,
      "step": 370
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 1.6776630878448486,
      "learning_rate": 4.998921545683059e-05,
      "loss": 0.2626,
      "num_input_tokens_seen": 105136,
      "step": 375
    },
    {
      "epoch": 4.222222222222222,
      "grad_norm": 2.580178737640381,
      "learning_rate": 4.9988925193976996e-05,
      "loss": 0.234,
      "num_input_tokens_seen": 106528,
      "step": 380
    },
    {
      "epoch": 4.277777777777778,
      "grad_norm": 10.14448356628418,
      "learning_rate": 4.998863107751711e-05,
      "loss": 0.4278,
      "num_input_tokens_seen": 107936,
      "step": 385
    },
    {
      "epoch": 4.333333333333333,
      "grad_norm": 2.283970832824707,
      "learning_rate": 4.998833310749629e-05,
      "loss": 0.3548,
      "num_input_tokens_seen": 109296,
      "step": 390
    },
    {
      "epoch": 4.388888888888889,
      "grad_norm": 2.2505338191986084,
      "learning_rate": 4.998803128396047e-05,
      "loss": 0.2431,
      "num_input_tokens_seen": 110688,
      "step": 395
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 5.188375473022461,
      "learning_rate": 4.9987725606956215e-05,
      "loss": 0.4327,
      "num_input_tokens_seen": 112064,
      "step": 400
    },
    {
      "epoch": 4.444444444444445,
      "eval_loss": 0.3699286878108978,
      "eval_runtime": 1.2368,
      "eval_samples_per_second": 32.341,
      "eval_steps_per_second": 16.17,
      "num_input_tokens_seen": 112064,
      "step": 400
    },
    {
      "epoch": 4.5,
      "grad_norm": 6.9356184005737305,
      "learning_rate": 4.998741607653066e-05,
      "loss": 0.4252,
      "num_input_tokens_seen": 113472,
      "step": 405
    },
    {
      "epoch": 4.555555555555555,
      "grad_norm": 5.830610275268555,
      "learning_rate": 4.9987102692731523e-05,
      "loss": 0.2961,
      "num_input_tokens_seen": 114880,
      "step": 410
    },
    {
      "epoch": 4.611111111111111,
      "grad_norm": 2.02085018157959,
      "learning_rate": 4.9986785455607157e-05,
      "loss": 0.2748,
      "num_input_tokens_seen": 116224,
      "step": 415
    },
    {
      "epoch": 4.666666666666667,
      "grad_norm": 1.771722674369812,
      "learning_rate": 4.9986464365206456e-05,
      "loss": 0.3734,
      "num_input_tokens_seen": 117680,
      "step": 420
    },
    {
      "epoch": 4.722222222222222,
      "grad_norm": 4.021349906921387,
      "learning_rate": 4.9986139421578956e-05,
      "loss": 0.394,
      "num_input_tokens_seen": 119072,
      "step": 425
    },
    {
      "epoch": 4.777777777777778,
      "grad_norm": 3.8249731063842773,
      "learning_rate": 4.998581062477477e-05,
      "loss": 0.2937,
      "num_input_tokens_seen": 120496,
      "step": 430
    },
    {
      "epoch": 4.833333333333333,
      "grad_norm": 1.7053364515304565,
      "learning_rate": 4.998547797484458e-05,
      "loss": 0.2795,
      "num_input_tokens_seen": 121936,
      "step": 435
    },
    {
      "epoch": 4.888888888888889,
      "grad_norm": 2.370164632797241,
      "learning_rate": 4.9985141471839706e-05,
      "loss": 0.2765,
      "num_input_tokens_seen": 123344,
      "step": 440
    },
    {
      "epoch": 4.944444444444445,
      "grad_norm": 6.871527671813965,
      "learning_rate": 4.998480111581203e-05,
      "loss": 0.2962,
      "num_input_tokens_seen": 124736,
      "step": 445
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.1808879375457764,
      "learning_rate": 4.998445690681405e-05,
      "loss": 0.2691,
      "num_input_tokens_seen": 126144,
      "step": 450
    },
    {
      "epoch": 5.055555555555555,
      "grad_norm": 3.1696889400482178,
      "learning_rate": 4.9984108844898834e-05,
      "loss": 0.2669,
      "num_input_tokens_seen": 127520,
      "step": 455
    },
    {
      "epoch": 5.111111111111111,
      "grad_norm": 3.9747207164764404,
      "learning_rate": 4.9983756930120076e-05,
      "loss": 0.2635,
      "num_input_tokens_seen": 128944,
      "step": 460
    },
    {
      "epoch": 5.166666666666667,
      "grad_norm": 6.3094258308410645,
      "learning_rate": 4.9983401162532025e-05,
      "loss": 0.4758,
      "num_input_tokens_seen": 130336,
      "step": 465
    },
    {
      "epoch": 5.222222222222222,
      "grad_norm": 4.34121036529541,
      "learning_rate": 4.998304154218955e-05,
      "loss": 0.278,
      "num_input_tokens_seen": 131696,
      "step": 470
    },
    {
      "epoch": 5.277777777777778,
      "grad_norm": 1.7400267124176025,
      "learning_rate": 4.998267806914812e-05,
      "loss": 0.2415,
      "num_input_tokens_seen": 133088,
      "step": 475
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 1.64749014377594,
      "learning_rate": 4.998231074346378e-05,
      "loss": 0.2592,
      "num_input_tokens_seen": 134464,
      "step": 480
    },
    {
      "epoch": 5.388888888888889,
      "grad_norm": 1.4608403444290161,
      "learning_rate": 4.998193956519317e-05,
      "loss": 0.2455,
      "num_input_tokens_seen": 135856,
      "step": 485
    },
    {
      "epoch": 5.444444444444445,
      "grad_norm": 2.1841394901275635,
      "learning_rate": 4.9981564534393545e-05,
      "loss": 0.2477,
      "num_input_tokens_seen": 137248,
      "step": 490
    },
    {
      "epoch": 5.5,
      "grad_norm": 0.8324402570724487,
      "learning_rate": 4.998118565112272e-05,
      "loss": 0.2152,
      "num_input_tokens_seen": 138592,
      "step": 495
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 2.2707161903381348,
      "learning_rate": 4.998080291543914e-05,
      "loss": 0.2087,
      "num_input_tokens_seen": 140016,
      "step": 500
    },
    {
      "epoch": 5.611111111111111,
      "grad_norm": 1.9172308444976807,
      "learning_rate": 4.9980416327401826e-05,
      "loss": 0.2837,
      "num_input_tokens_seen": 141424,
      "step": 505
    },
    {
      "epoch": 5.666666666666667,
      "grad_norm": 2.4913368225097656,
      "learning_rate": 4.998002588707038e-05,
      "loss": 0.231,
      "num_input_tokens_seen": 142864,
      "step": 510
    },
    {
      "epoch": 5.722222222222222,
      "grad_norm": 1.3811036348342896,
      "learning_rate": 4.997963159450503e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 144320,
      "step": 515
    },
    {
      "epoch": 5.777777777777778,
      "grad_norm": 1.6968319416046143,
      "learning_rate": 4.9979233449766575e-05,
      "loss": 0.3067,
      "num_input_tokens_seen": 145712,
      "step": 520
    },
    {
      "epoch": 5.833333333333333,
      "grad_norm": 3.4956307411193848,
      "learning_rate": 4.997883145291641e-05,
      "loss": 0.2392,
      "num_input_tokens_seen": 147088,
      "step": 525
    },
    {
      "epoch": 5.888888888888889,
      "grad_norm": 3.190186023712158,
      "learning_rate": 4.9978425604016536e-05,
      "loss": 0.3151,
      "num_input_tokens_seen": 148464,
      "step": 530
    },
    {
      "epoch": 5.944444444444445,
      "grad_norm": 3.3422956466674805,
      "learning_rate": 4.9978015903129536e-05,
      "loss": 0.4079,
      "num_input_tokens_seen": 149840,
      "step": 535
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.9827764630317688,
      "learning_rate": 4.997760235031859e-05,
      "loss": 0.2627,
      "num_input_tokens_seen": 151248,
      "step": 540
    },
    {
      "epoch": 6.055555555555555,
      "grad_norm": 0.8102995157241821,
      "learning_rate": 4.9977184945647473e-05,
      "loss": 0.2154,
      "num_input_tokens_seen": 152624,
      "step": 545
    },
    {
      "epoch": 6.111111111111111,
      "grad_norm": 4.396462917327881,
      "learning_rate": 4.997676368918055e-05,
      "loss": 0.2857,
      "num_input_tokens_seen": 154032,
      "step": 550
    },
    {
      "epoch": 6.166666666666667,
      "grad_norm": 3.1466708183288574,
      "learning_rate": 4.9976338580982794e-05,
      "loss": 0.2887,
      "num_input_tokens_seen": 155456,
      "step": 555
    },
    {
      "epoch": 6.222222222222222,
      "grad_norm": 1.2655282020568848,
      "learning_rate": 4.9975909621119755e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 156864,
      "step": 560
    },
    {
      "epoch": 6.277777777777778,
      "grad_norm": 1.849257469177246,
      "learning_rate": 4.997547680965758e-05,
      "loss": 0.2138,
      "num_input_tokens_seen": 158304,
      "step": 565
    },
    {
      "epoch": 6.333333333333333,
      "grad_norm": 1.7839317321777344,
      "learning_rate": 4.997504014666302e-05,
      "loss": 0.2988,
      "num_input_tokens_seen": 159712,
      "step": 570
    },
    {
      "epoch": 6.388888888888889,
      "grad_norm": 4.711520671844482,
      "learning_rate": 4.997459963220342e-05,
      "loss": 0.2773,
      "num_input_tokens_seen": 161088,
      "step": 575
    },
    {
      "epoch": 6.444444444444445,
      "grad_norm": 0.9007275700569153,
      "learning_rate": 4.997415526634671e-05,
      "loss": 0.2357,
      "num_input_tokens_seen": 162464,
      "step": 580
    },
    {
      "epoch": 6.5,
      "grad_norm": 1.4829423427581787,
      "learning_rate": 4.99737070491614e-05,
      "loss": 0.1955,
      "num_input_tokens_seen": 163872,
      "step": 585
    },
    {
      "epoch": 6.555555555555555,
      "grad_norm": 2.076313018798828,
      "learning_rate": 4.997325498071663e-05,
      "loss": 0.2596,
      "num_input_tokens_seen": 165280,
      "step": 590
    },
    {
      "epoch": 6.611111111111111,
      "grad_norm": 0.7604356408119202,
      "learning_rate": 4.997279906108211e-05,
      "loss": 0.2026,
      "num_input_tokens_seen": 166704,
      "step": 595
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 1.109295129776001,
      "learning_rate": 4.9972339290328155e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 168112,
      "step": 600
    },
    {
      "epoch": 6.666666666666667,
      "eval_loss": 0.25527411699295044,
      "eval_runtime": 1.2428,
      "eval_samples_per_second": 32.184,
      "eval_steps_per_second": 16.092,
      "num_input_tokens_seen": 168112,
      "step": 600
    },
    {
      "epoch": 6.722222222222222,
      "grad_norm": 4.168839454650879,
      "learning_rate": 4.9971875668525646e-05,
      "loss": 0.3097,
      "num_input_tokens_seen": 169504,
      "step": 605
    },
    {
      "epoch": 6.777777777777778,
      "grad_norm": 1.7936726808547974,
      "learning_rate": 4.997140819574609e-05,
      "loss": 0.2271,
      "num_input_tokens_seen": 170880,
      "step": 610
    },
    {
      "epoch": 6.833333333333333,
      "grad_norm": 1.5015357732772827,
      "learning_rate": 4.997093687206159e-05,
      "loss": 0.2497,
      "num_input_tokens_seen": 172256,
      "step": 615
    },
    {
      "epoch": 6.888888888888889,
      "grad_norm": 1.4440075159072876,
      "learning_rate": 4.997046169754482e-05,
      "loss": 0.2117,
      "num_input_tokens_seen": 173664,
      "step": 620
    },
    {
      "epoch": 6.944444444444445,
      "grad_norm": 1.0461044311523438,
      "learning_rate": 4.996998267226905e-05,
      "loss": 0.2277,
      "num_input_tokens_seen": 175024,
      "step": 625
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.6692721843719482,
      "learning_rate": 4.996949979630817e-05,
      "loss": 0.2175,
      "num_input_tokens_seen": 176400,
      "step": 630
    },
    {
      "epoch": 7.055555555555555,
      "grad_norm": 0.45171448588371277,
      "learning_rate": 4.996901306973663e-05,
      "loss": 0.1663,
      "num_input_tokens_seen": 177792,
      "step": 635
    },
    {
      "epoch": 7.111111111111111,
      "grad_norm": 0.6437914967536926,
      "learning_rate": 4.996852249262949e-05,
      "loss": 0.1691,
      "num_input_tokens_seen": 179168,
      "step": 640
    },
    {
      "epoch": 7.166666666666667,
      "grad_norm": 2.6709277629852295,
      "learning_rate": 4.996802806506241e-05,
      "loss": 0.2486,
      "num_input_tokens_seen": 180544,
      "step": 645
    },
    {
      "epoch": 7.222222222222222,
      "grad_norm": 0.7791646122932434,
      "learning_rate": 4.996752978711164e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 181984,
      "step": 650
    },
    {
      "epoch": 7.277777777777778,
      "grad_norm": 2.13019061088562,
      "learning_rate": 4.996702765885401e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 183392,
      "step": 655
    },
    {
      "epoch": 7.333333333333333,
      "grad_norm": 2.5034077167510986,
      "learning_rate": 4.9966521680366964e-05,
      "loss": 0.2545,
      "num_input_tokens_seen": 184784,
      "step": 660
    },
    {
      "epoch": 7.388888888888889,
      "grad_norm": 0.8435221314430237,
      "learning_rate": 4.9966011851728524e-05,
      "loss": 0.2067,
      "num_input_tokens_seen": 186192,
      "step": 665
    },
    {
      "epoch": 7.444444444444445,
      "grad_norm": 1.5187093019485474,
      "learning_rate": 4.996549817301731e-05,
      "loss": 0.1434,
      "num_input_tokens_seen": 187568,
      "step": 670
    },
    {
      "epoch": 7.5,
      "grad_norm": 2.0387685298919678,
      "learning_rate": 4.9964980644312544e-05,
      "loss": 0.2672,
      "num_input_tokens_seen": 188928,
      "step": 675
    },
    {
      "epoch": 7.555555555555555,
      "grad_norm": 0.5301392674446106,
      "learning_rate": 4.996445926569403e-05,
      "loss": 0.2194,
      "num_input_tokens_seen": 190336,
      "step": 680
    },
    {
      "epoch": 7.611111111111111,
      "grad_norm": 1.0442380905151367,
      "learning_rate": 4.996393403724218e-05,
      "loss": 0.264,
      "num_input_tokens_seen": 191712,
      "step": 685
    },
    {
      "epoch": 7.666666666666667,
      "grad_norm": 1.4284602403640747,
      "learning_rate": 4.9963404959037985e-05,
      "loss": 0.195,
      "num_input_tokens_seen": 193056,
      "step": 690
    },
    {
      "epoch": 7.722222222222222,
      "grad_norm": 1.648135781288147,
      "learning_rate": 4.996287203116303e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 194496,
      "step": 695
    },
    {
      "epoch": 7.777777777777778,
      "grad_norm": 4.0916829109191895,
      "learning_rate": 4.996233525369951e-05,
      "loss": 0.2443,
      "num_input_tokens_seen": 195936,
      "step": 700
    },
    {
      "epoch": 7.833333333333333,
      "grad_norm": 1.8835772275924683,
      "learning_rate": 4.99617946267302e-05,
      "loss": 0.2807,
      "num_input_tokens_seen": 197360,
      "step": 705
    },
    {
      "epoch": 7.888888888888889,
      "grad_norm": 0.8576043248176575,
      "learning_rate": 4.996125015033846e-05,
      "loss": 0.2317,
      "num_input_tokens_seen": 198784,
      "step": 710
    },
    {
      "epoch": 7.944444444444445,
      "grad_norm": 1.0027538537979126,
      "learning_rate": 4.996070182460827e-05,
      "loss": 0.2256,
      "num_input_tokens_seen": 200192,
      "step": 715
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.4442576169967651,
      "learning_rate": 4.996014964962418e-05,
      "loss": 0.209,
      "num_input_tokens_seen": 201616,
      "step": 720
    },
    {
      "epoch": 8.055555555555555,
      "grad_norm": 0.7380709648132324,
      "learning_rate": 4.9959593625471344e-05,
      "loss": 0.2141,
      "num_input_tokens_seen": 203024,
      "step": 725
    },
    {
      "epoch": 8.11111111111111,
      "grad_norm": 0.5705077052116394,
      "learning_rate": 4.995903375223552e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 204400,
      "step": 730
    },
    {
      "epoch": 8.166666666666666,
      "grad_norm": 1.734371304512024,
      "learning_rate": 4.995847003000302e-05,
      "loss": 0.2054,
      "num_input_tokens_seen": 205824,
      "step": 735
    },
    {
      "epoch": 8.222222222222221,
      "grad_norm": 0.7726757526397705,
      "learning_rate": 4.9957902458860804e-05,
      "loss": 0.252,
      "num_input_tokens_seen": 207264,
      "step": 740
    },
    {
      "epoch": 8.277777777777779,
      "grad_norm": 0.689985990524292,
      "learning_rate": 4.995733103889639e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 208656,
      "step": 745
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 1.1886894702911377,
      "learning_rate": 4.99567557701979e-05,
      "loss": 0.1965,
      "num_input_tokens_seen": 210064,
      "step": 750
    },
    {
      "epoch": 8.38888888888889,
      "grad_norm": 1.0946857929229736,
      "learning_rate": 4.995617665285403e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 211440,
      "step": 755
    },
    {
      "epoch": 8.444444444444445,
      "grad_norm": 0.9379662871360779,
      "learning_rate": 4.99555936869541e-05,
      "loss": 0.3155,
      "num_input_tokens_seen": 212848,
      "step": 760
    },
    {
      "epoch": 8.5,
      "grad_norm": 1.0027194023132324,
      "learning_rate": 4.995500687258803e-05,
      "loss": 0.1787,
      "num_input_tokens_seen": 214272,
      "step": 765
    },
    {
      "epoch": 8.555555555555555,
      "grad_norm": 2.024844169616699,
      "learning_rate": 4.995441620984628e-05,
      "loss": 0.2159,
      "num_input_tokens_seen": 215696,
      "step": 770
    },
    {
      "epoch": 8.61111111111111,
      "grad_norm": 0.7903777360916138,
      "learning_rate": 4.995382169881996e-05,
      "loss": 0.1966,
      "num_input_tokens_seen": 217088,
      "step": 775
    },
    {
      "epoch": 8.666666666666666,
      "grad_norm": 0.5296844244003296,
      "learning_rate": 4.9953223339600755e-05,
      "loss": 0.1817,
      "num_input_tokens_seen": 218496,
      "step": 780
    },
    {
      "epoch": 8.722222222222221,
      "grad_norm": 0.8257622122764587,
      "learning_rate": 4.995262113228091e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 219840,
      "step": 785
    },
    {
      "epoch": 8.777777777777779,
      "grad_norm": 0.8792858123779297,
      "learning_rate": 4.995201507695332e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 221264,
      "step": 790
    },
    {
      "epoch": 8.833333333333334,
      "grad_norm": 0.5362371802330017,
      "learning_rate": 4.995140517371144e-05,
      "loss": 0.2136,
      "num_input_tokens_seen": 222656,
      "step": 795
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 1.6798100471496582,
      "learning_rate": 4.995079142264932e-05,
      "loss": 0.1897,
      "num_input_tokens_seen": 224048,
      "step": 800
    },
    {
      "epoch": 8.88888888888889,
      "eval_loss": 0.21392026543617249,
      "eval_runtime": 1.239,
      "eval_samples_per_second": 32.284,
      "eval_steps_per_second": 16.142,
      "num_input_tokens_seen": 224048,
      "step": 800
    },
    {
      "epoch": 8.944444444444445,
      "grad_norm": 1.0414284467697144,
      "learning_rate": 4.995017382386162e-05,
      "loss": 0.2092,
      "num_input_tokens_seen": 225440,
      "step": 805
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.7658756971359253,
      "learning_rate": 4.994955237744356e-05,
      "loss": 0.1809,
      "num_input_tokens_seen": 226816,
      "step": 810
    },
    {
      "epoch": 9.055555555555555,
      "grad_norm": 2.1816484928131104,
      "learning_rate": 4.994892708349101e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 228176,
      "step": 815
    },
    {
      "epoch": 9.11111111111111,
      "grad_norm": 0.5286153554916382,
      "learning_rate": 4.994829794210035e-05,
      "loss": 0.1992,
      "num_input_tokens_seen": 229584,
      "step": 820
    },
    {
      "epoch": 9.166666666666666,
      "grad_norm": 1.5833561420440674,
      "learning_rate": 4.994766495336864e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 230944,
      "step": 825
    },
    {
      "epoch": 9.222222222222221,
      "grad_norm": 0.8390882015228271,
      "learning_rate": 4.994702811739348e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 232368,
      "step": 830
    },
    {
      "epoch": 9.277777777777779,
      "grad_norm": 0.7819733619689941,
      "learning_rate": 4.994638743427308e-05,
      "loss": 0.1896,
      "num_input_tokens_seen": 233792,
      "step": 835
    },
    {
      "epoch": 9.333333333333334,
      "grad_norm": 0.5129129886627197,
      "learning_rate": 4.994574290410624e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 235216,
      "step": 840
    },
    {
      "epoch": 9.38888888888889,
      "grad_norm": 0.9194492697715759,
      "learning_rate": 4.9945094526992364e-05,
      "loss": 0.2115,
      "num_input_tokens_seen": 236576,
      "step": 845
    },
    {
      "epoch": 9.444444444444445,
      "grad_norm": 1.6027228832244873,
      "learning_rate": 4.994444230303142e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 237952,
      "step": 850
    },
    {
      "epoch": 9.5,
      "grad_norm": 1.7122023105621338,
      "learning_rate": 4.994378623232402e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 239360,
      "step": 855
    },
    {
      "epoch": 9.555555555555555,
      "grad_norm": 1.3446722030639648,
      "learning_rate": 4.99431263149713e-05,
      "loss": 0.2049,
      "num_input_tokens_seen": 240784,
      "step": 860
    },
    {
      "epoch": 9.61111111111111,
      "grad_norm": 2.316411018371582,
      "learning_rate": 4.9942462551075056e-05,
      "loss": 0.1274,
      "num_input_tokens_seen": 242208,
      "step": 865
    },
    {
      "epoch": 9.666666666666666,
      "grad_norm": 3.08766770362854,
      "learning_rate": 4.994179494073764e-05,
      "loss": 0.1978,
      "num_input_tokens_seen": 243584,
      "step": 870
    },
    {
      "epoch": 9.722222222222221,
      "grad_norm": 2.7254252433776855,
      "learning_rate": 4.9941123484062e-05,
      "loss": 0.2778,
      "num_input_tokens_seen": 244960,
      "step": 875
    },
    {
      "epoch": 9.777777777777779,
      "grad_norm": 1.426692008972168,
      "learning_rate": 4.99404481811517e-05,
      "loss": 0.2453,
      "num_input_tokens_seen": 246384,
      "step": 880
    },
    {
      "epoch": 9.833333333333334,
      "grad_norm": 1.437313437461853,
      "learning_rate": 4.9939769032110864e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 247840,
      "step": 885
    },
    {
      "epoch": 9.88888888888889,
      "grad_norm": 0.6410755515098572,
      "learning_rate": 4.993908603704423e-05,
      "loss": 0.2403,
      "num_input_tokens_seen": 249216,
      "step": 890
    },
    {
      "epoch": 9.944444444444445,
      "grad_norm": 1.0482217073440552,
      "learning_rate": 4.9938399196057126e-05,
      "loss": 0.206,
      "num_input_tokens_seen": 250592,
      "step": 895
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.0709227323532104,
      "learning_rate": 4.993770850925547e-05,
      "loss": 0.2406,
      "num_input_tokens_seen": 251968,
      "step": 900
    },
    {
      "epoch": 10.055555555555555,
      "grad_norm": 0.6985483765602112,
      "learning_rate": 4.993701397674577e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 253392,
      "step": 905
    },
    {
      "epoch": 10.11111111111111,
      "grad_norm": 1.6414226293563843,
      "learning_rate": 4.993631559863515e-05,
      "loss": 0.1932,
      "num_input_tokens_seen": 254752,
      "step": 910
    },
    {
      "epoch": 10.166666666666666,
      "grad_norm": 1.459583044052124,
      "learning_rate": 4.9935613375031283e-05,
      "loss": 0.2033,
      "num_input_tokens_seen": 256176,
      "step": 915
    },
    {
      "epoch": 10.222222222222221,
      "grad_norm": 0.8507170081138611,
      "learning_rate": 4.993490730604248e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 257584,
      "step": 920
    },
    {
      "epoch": 10.277777777777779,
      "grad_norm": 1.6489183902740479,
      "learning_rate": 4.993419739177761e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 259008,
      "step": 925
    },
    {
      "epoch": 10.333333333333334,
      "grad_norm": 0.56942218542099,
      "learning_rate": 4.9933483632346164e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 260416,
      "step": 930
    },
    {
      "epoch": 10.38888888888889,
      "grad_norm": 0.8923459649085999,
      "learning_rate": 4.993276602785821e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 261872,
      "step": 935
    },
    {
      "epoch": 10.444444444444445,
      "grad_norm": 1.960074782371521,
      "learning_rate": 4.993204457842441e-05,
      "loss": 0.193,
      "num_input_tokens_seen": 263232,
      "step": 940
    },
    {
      "epoch": 10.5,
      "grad_norm": 0.8318308591842651,
      "learning_rate": 4.993131928415602e-05,
      "loss": 0.2259,
      "num_input_tokens_seen": 264656,
      "step": 945
    },
    {
      "epoch": 10.555555555555555,
      "grad_norm": 0.6224443912506104,
      "learning_rate": 4.993059014516489e-05,
      "loss": 0.1636,
      "num_input_tokens_seen": 266016,
      "step": 950
    },
    {
      "epoch": 10.61111111111111,
      "grad_norm": 1.5698004961013794,
      "learning_rate": 4.9929857161563464e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 267408,
      "step": 955
    },
    {
      "epoch": 10.666666666666666,
      "grad_norm": 0.7479705214500427,
      "learning_rate": 4.992912033346477e-05,
      "loss": 0.1877,
      "num_input_tokens_seen": 268800,
      "step": 960
    },
    {
      "epoch": 10.722222222222221,
      "grad_norm": 0.7814568281173706,
      "learning_rate": 4.992837966098245e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 270208,
      "step": 965
    },
    {
      "epoch": 10.777777777777779,
      "grad_norm": 0.7119309306144714,
      "learning_rate": 4.992763514423071e-05,
      "loss": 0.2067,
      "num_input_tokens_seen": 271552,
      "step": 970
    },
    {
      "epoch": 10.833333333333334,
      "grad_norm": 1.4959992170333862,
      "learning_rate": 4.992688678332437e-05,
      "loss": 0.1964,
      "num_input_tokens_seen": 272928,
      "step": 975
    },
    {
      "epoch": 10.88888888888889,
      "grad_norm": 3.8195817470550537,
      "learning_rate": 4.992613457837884e-05,
      "loss": 0.1935,
      "num_input_tokens_seen": 274352,
      "step": 980
    },
    {
      "epoch": 10.944444444444445,
      "grad_norm": 1.4165693521499634,
      "learning_rate": 4.992537852951011e-05,
      "loss": 0.2192,
      "num_input_tokens_seen": 275760,
      "step": 985
    },
    {
      "epoch": 11.0,
      "grad_norm": 3.2431254386901855,
      "learning_rate": 4.9924618636834785e-05,
      "loss": 0.239,
      "num_input_tokens_seen": 277152,
      "step": 990
    },
    {
      "epoch": 11.055555555555555,
      "grad_norm": 0.925532341003418,
      "learning_rate": 4.9923854900470046e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 278512,
      "step": 995
    },
    {
      "epoch": 11.11111111111111,
      "grad_norm": 0.4565618932247162,
      "learning_rate": 4.992308732053367e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 279904,
      "step": 1000
    },
    {
      "epoch": 11.11111111111111,
      "eval_loss": 0.20405562222003937,
      "eval_runtime": 1.2372,
      "eval_samples_per_second": 32.331,
      "eval_steps_per_second": 16.166,
      "num_input_tokens_seen": 279904,
      "step": 1000
    },
    {
      "epoch": 11.166666666666666,
      "grad_norm": 0.7674880027770996,
      "learning_rate": 4.992231589714402e-05,
      "loss": 0.2029,
      "num_input_tokens_seen": 281360,
      "step": 1005
    },
    {
      "epoch": 11.222222222222221,
      "grad_norm": 1.206112027168274,
      "learning_rate": 4.992154063042007e-05,
      "loss": 0.1728,
      "num_input_tokens_seen": 282752,
      "step": 1010
    },
    {
      "epoch": 11.277777777777779,
      "grad_norm": 1.6875271797180176,
      "learning_rate": 4.992076152048136e-05,
      "loss": 0.2096,
      "num_input_tokens_seen": 284160,
      "step": 1015
    },
    {
      "epoch": 11.333333333333334,
      "grad_norm": 1.882201075553894,
      "learning_rate": 4.991997856744807e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 285536,
      "step": 1020
    },
    {
      "epoch": 11.38888888888889,
      "grad_norm": 0.35457760095596313,
      "learning_rate": 4.9919191771440905e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 286896,
      "step": 1025
    },
    {
      "epoch": 11.444444444444445,
      "grad_norm": 2.4337880611419678,
      "learning_rate": 4.991840113258122e-05,
      "loss": 0.2105,
      "num_input_tokens_seen": 288256,
      "step": 1030
    },
    {
      "epoch": 11.5,
      "grad_norm": 0.7331763505935669,
      "learning_rate": 4.9917606650990933e-05,
      "loss": 0.1643,
      "num_input_tokens_seen": 289680,
      "step": 1035
    },
    {
      "epoch": 11.555555555555555,
      "grad_norm": 0.5503110885620117,
      "learning_rate": 4.9916808326792566e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 291040,
      "step": 1040
    },
    {
      "epoch": 11.61111111111111,
      "grad_norm": 1.7860690355300903,
      "learning_rate": 4.9916006160109235e-05,
      "loss": 0.2461,
      "num_input_tokens_seen": 292400,
      "step": 1045
    },
    {
      "epoch": 11.666666666666666,
      "grad_norm": 2.3030123710632324,
      "learning_rate": 4.991520015106464e-05,
      "loss": 0.2037,
      "num_input_tokens_seen": 293824,
      "step": 1050
    },
    {
      "epoch": 11.722222222222221,
      "grad_norm": 0.8643040060997009,
      "learning_rate": 4.991439029978308e-05,
      "loss": 0.2059,
      "num_input_tokens_seen": 295232,
      "step": 1055
    },
    {
      "epoch": 11.777777777777779,
      "grad_norm": 1.5813279151916504,
      "learning_rate": 4.9913576606389434e-05,
      "loss": 0.1652,
      "num_input_tokens_seen": 296640,
      "step": 1060
    },
    {
      "epoch": 11.833333333333334,
      "grad_norm": 0.5652819275856018,
      "learning_rate": 4.991275907100919e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 298048,
      "step": 1065
    },
    {
      "epoch": 11.88888888888889,
      "grad_norm": 0.7392434477806091,
      "learning_rate": 4.9911937693768434e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 299456,
      "step": 1070
    },
    {
      "epoch": 11.944444444444445,
      "grad_norm": 0.8316829204559326,
      "learning_rate": 4.991111247479382e-05,
      "loss": 0.2104,
      "num_input_tokens_seen": 300848,
      "step": 1075
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.6481261253356934,
      "learning_rate": 4.9910283414212605e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 302224,
      "step": 1080
    },
    {
      "epoch": 12.055555555555555,
      "grad_norm": 0.8392964005470276,
      "learning_rate": 4.990945051215265e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 303648,
      "step": 1085
    },
    {
      "epoch": 12.11111111111111,
      "grad_norm": 1.3891217708587646,
      "learning_rate": 4.99086137687424e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 305024,
      "step": 1090
    },
    {
      "epoch": 12.166666666666666,
      "grad_norm": 0.45470935106277466,
      "learning_rate": 4.9907773184110874e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 306448,
      "step": 1095
    },
    {
      "epoch": 12.222222222222221,
      "grad_norm": 1.4731347560882568,
      "learning_rate": 4.9906928758387715e-05,
      "loss": 0.2182,
      "num_input_tokens_seen": 307856,
      "step": 1100
    },
    {
      "epoch": 12.277777777777779,
      "grad_norm": 0.9751033186912537,
      "learning_rate": 4.9906080491703146e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 309280,
      "step": 1105
    },
    {
      "epoch": 12.333333333333334,
      "grad_norm": 1.0742043256759644,
      "learning_rate": 4.990522838418797e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 310704,
      "step": 1110
    },
    {
      "epoch": 12.38888888888889,
      "grad_norm": 1.0589925050735474,
      "learning_rate": 4.9904372435973604e-05,
      "loss": 0.1875,
      "num_input_tokens_seen": 312080,
      "step": 1115
    },
    {
      "epoch": 12.444444444444445,
      "grad_norm": 1.0174062252044678,
      "learning_rate": 4.990351264719203e-05,
      "loss": 0.2226,
      "num_input_tokens_seen": 313520,
      "step": 1120
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.981279730796814,
      "learning_rate": 4.990264901797586e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 314912,
      "step": 1125
    },
    {
      "epoch": 12.555555555555555,
      "grad_norm": 2.6535048484802246,
      "learning_rate": 4.990178154845826e-05,
      "loss": 0.1722,
      "num_input_tokens_seen": 316336,
      "step": 1130
    },
    {
      "epoch": 12.61111111111111,
      "grad_norm": 0.7514998912811279,
      "learning_rate": 4.9900910238773014e-05,
      "loss": 0.1434,
      "num_input_tokens_seen": 317744,
      "step": 1135
    },
    {
      "epoch": 12.666666666666666,
      "grad_norm": 0.5858056545257568,
      "learning_rate": 4.990003508905448e-05,
      "loss": 0.215,
      "num_input_tokens_seen": 319136,
      "step": 1140
    },
    {
      "epoch": 12.722222222222221,
      "grad_norm": 0.5759261846542358,
      "learning_rate": 4.989915609943763e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 320528,
      "step": 1145
    },
    {
      "epoch": 12.777777777777779,
      "grad_norm": 1.8025212287902832,
      "learning_rate": 4.9898273270058e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 321936,
      "step": 1150
    },
    {
      "epoch": 12.833333333333334,
      "grad_norm": 0.7287235260009766,
      "learning_rate": 4.989738660105174e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 323344,
      "step": 1155
    },
    {
      "epoch": 12.88888888888889,
      "grad_norm": 0.4928489029407501,
      "learning_rate": 4.989649609255559e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 324784,
      "step": 1160
    },
    {
      "epoch": 12.944444444444445,
      "grad_norm": 1.1161869764328003,
      "learning_rate": 4.989560174470687e-05,
      "loss": 0.2332,
      "num_input_tokens_seen": 326208,
      "step": 1165
    },
    {
      "epoch": 13.0,
      "grad_norm": 2.8376364707946777,
      "learning_rate": 4.989470355764351e-05,
      "loss": 0.1817,
      "num_input_tokens_seen": 327600,
      "step": 1170
    },
    {
      "epoch": 13.055555555555555,
      "grad_norm": 0.3444328010082245,
      "learning_rate": 4.9893801531504e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 329024,
      "step": 1175
    },
    {
      "epoch": 13.11111111111111,
      "grad_norm": 0.9473903775215149,
      "learning_rate": 4.9892895666427475e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 330400,
      "step": 1180
    },
    {
      "epoch": 13.166666666666666,
      "grad_norm": 0.8591474890708923,
      "learning_rate": 4.9891985962553606e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 331808,
      "step": 1185
    },
    {
      "epoch": 13.222222222222221,
      "grad_norm": 0.9329907894134521,
      "learning_rate": 4.989107242002269e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 333200,
      "step": 1190
    },
    {
      "epoch": 13.277777777777779,
      "grad_norm": 0.2622354328632355,
      "learning_rate": 4.989015503897561e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 334640,
      "step": 1195
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 1.1023985147476196,
      "learning_rate": 4.988923381955383e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 336032,
      "step": 1200
    },
    {
      "epoch": 13.333333333333334,
      "eval_loss": 0.18691900372505188,
      "eval_runtime": 1.235,
      "eval_samples_per_second": 32.389,
      "eval_steps_per_second": 16.195,
      "num_input_tokens_seen": 336032,
      "step": 1200
    },
    {
      "epoch": 13.38888888888889,
      "grad_norm": 0.6724897623062134,
      "learning_rate": 4.988830876189942e-05,
      "loss": 0.1722,
      "num_input_tokens_seen": 337440,
      "step": 1205
    },
    {
      "epoch": 13.444444444444445,
      "grad_norm": 0.9162903428077698,
      "learning_rate": 4.988737986615503e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 338832,
      "step": 1210
    },
    {
      "epoch": 13.5,
      "grad_norm": 1.0843533277511597,
      "learning_rate": 4.988644713246391e-05,
      "loss": 0.1762,
      "num_input_tokens_seen": 340224,
      "step": 1215
    },
    {
      "epoch": 13.555555555555555,
      "grad_norm": 1.4023430347442627,
      "learning_rate": 4.988551056096991e-05,
      "loss": 0.19,
      "num_input_tokens_seen": 341616,
      "step": 1220
    },
    {
      "epoch": 13.61111111111111,
      "grad_norm": 0.6041615605354309,
      "learning_rate": 4.988457015181743e-05,
      "loss": 0.1911,
      "num_input_tokens_seen": 342992,
      "step": 1225
    },
    {
      "epoch": 13.666666666666666,
      "grad_norm": 0.4652657210826874,
      "learning_rate": 4.988362590515153e-05,
      "loss": 0.2368,
      "num_input_tokens_seen": 344384,
      "step": 1230
    },
    {
      "epoch": 13.722222222222221,
      "grad_norm": 2.1760923862457275,
      "learning_rate": 4.9882677821117805e-05,
      "loss": 0.2104,
      "num_input_tokens_seen": 345824,
      "step": 1235
    },
    {
      "epoch": 13.777777777777779,
      "grad_norm": 2.450000524520874,
      "learning_rate": 4.988172589986246e-05,
      "loss": 0.1781,
      "num_input_tokens_seen": 347248,
      "step": 1240
    },
    {
      "epoch": 13.833333333333334,
      "grad_norm": 1.0895534753799438,
      "learning_rate": 4.9880770141532304e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 348672,
      "step": 1245
    },
    {
      "epoch": 13.88888888888889,
      "grad_norm": 0.731010377407074,
      "learning_rate": 4.987981054627472e-05,
      "loss": 0.1899,
      "num_input_tokens_seen": 350032,
      "step": 1250
    },
    {
      "epoch": 13.944444444444445,
      "grad_norm": 0.5511336326599121,
      "learning_rate": 4.987884711423769e-05,
      "loss": 0.2094,
      "num_input_tokens_seen": 351392,
      "step": 1255
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.7629051804542542,
      "learning_rate": 4.9877879845569784e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 352784,
      "step": 1260
    },
    {
      "epoch": 14.055555555555555,
      "grad_norm": 0.42648282647132874,
      "learning_rate": 4.9876908740420175e-05,
      "loss": 0.2093,
      "num_input_tokens_seen": 354176,
      "step": 1265
    },
    {
      "epoch": 14.11111111111111,
      "grad_norm": 0.406075119972229,
      "learning_rate": 4.987593379893861e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 355568,
      "step": 1270
    },
    {
      "epoch": 14.166666666666666,
      "grad_norm": 1.3424891233444214,
      "learning_rate": 4.987495502127545e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 356960,
      "step": 1275
    },
    {
      "epoch": 14.222222222222221,
      "grad_norm": 1.1744002103805542,
      "learning_rate": 4.987397240758162e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 358368,
      "step": 1280
    },
    {
      "epoch": 14.277777777777779,
      "grad_norm": 0.5298190712928772,
      "learning_rate": 4.9872985958008664e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 359760,
      "step": 1285
    },
    {
      "epoch": 14.333333333333334,
      "grad_norm": 0.7154019474983215,
      "learning_rate": 4.987199567270871e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 361200,
      "step": 1290
    },
    {
      "epoch": 14.38888888888889,
      "grad_norm": 1.1485415697097778,
      "learning_rate": 4.9871001551834444e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 362656,
      "step": 1295
    },
    {
      "epoch": 14.444444444444445,
      "grad_norm": 1.2515934705734253,
      "learning_rate": 4.98700035955392e-05,
      "loss": 0.1663,
      "num_input_tokens_seen": 364032,
      "step": 1300
    },
    {
      "epoch": 14.5,
      "grad_norm": 0.7106055021286011,
      "learning_rate": 4.986900180397686e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 365408,
      "step": 1305
    },
    {
      "epoch": 14.555555555555555,
      "grad_norm": 1.1808903217315674,
      "learning_rate": 4.9867996177301926e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 366784,
      "step": 1310
    },
    {
      "epoch": 14.61111111111111,
      "grad_norm": 1.1130658388137817,
      "learning_rate": 4.9866986715669464e-05,
      "loss": 0.1943,
      "num_input_tokens_seen": 368176,
      "step": 1315
    },
    {
      "epoch": 14.666666666666666,
      "grad_norm": 0.4224929213523865,
      "learning_rate": 4.9865973419235155e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 369552,
      "step": 1320
    },
    {
      "epoch": 14.722222222222221,
      "grad_norm": 1.204710841178894,
      "learning_rate": 4.986495628815526e-05,
      "loss": 0.1887,
      "num_input_tokens_seen": 370944,
      "step": 1325
    },
    {
      "epoch": 14.777777777777779,
      "grad_norm": 1.183781623840332,
      "learning_rate": 4.986393532258663e-05,
      "loss": 0.212,
      "num_input_tokens_seen": 372368,
      "step": 1330
    },
    {
      "epoch": 14.833333333333334,
      "grad_norm": 0.8928014039993286,
      "learning_rate": 4.986291052268671e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 373760,
      "step": 1335
    },
    {
      "epoch": 14.88888888888889,
      "grad_norm": 1.1414697170257568,
      "learning_rate": 4.986188188861355e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 375120,
      "step": 1340
    },
    {
      "epoch": 14.944444444444445,
      "grad_norm": 1.2513052225112915,
      "learning_rate": 4.9860849420525766e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 376528,
      "step": 1345
    },
    {
      "epoch": 15.0,
      "grad_norm": 2.298698902130127,
      "learning_rate": 4.9859813118582575e-05,
      "loss": 0.1994,
      "num_input_tokens_seen": 377952,
      "step": 1350
    },
    {
      "epoch": 15.055555555555555,
      "grad_norm": 0.8822804689407349,
      "learning_rate": 4.98587729829438e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 379328,
      "step": 1355
    },
    {
      "epoch": 15.11111111111111,
      "grad_norm": 1.5971630811691284,
      "learning_rate": 4.985772901376983e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 380720,
      "step": 1360
    },
    {
      "epoch": 15.166666666666666,
      "grad_norm": 1.0133291482925415,
      "learning_rate": 4.9856681211221666e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 382096,
      "step": 1365
    },
    {
      "epoch": 15.222222222222221,
      "grad_norm": 1.1113307476043701,
      "learning_rate": 4.985562957546089e-05,
      "loss": 0.168,
      "num_input_tokens_seen": 383488,
      "step": 1370
    },
    {
      "epoch": 15.277777777777779,
      "grad_norm": 1.2665283679962158,
      "learning_rate": 4.9854574106649686e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 384848,
      "step": 1375
    },
    {
      "epoch": 15.333333333333334,
      "grad_norm": 0.34602221846580505,
      "learning_rate": 4.985351480495081e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 386256,
      "step": 1380
    },
    {
      "epoch": 15.38888888888889,
      "grad_norm": 0.6908441185951233,
      "learning_rate": 4.985245167052762e-05,
      "loss": 0.1794,
      "num_input_tokens_seen": 387696,
      "step": 1385
    },
    {
      "epoch": 15.444444444444445,
      "grad_norm": 1.6676795482635498,
      "learning_rate": 4.9851384703544066e-05,
      "loss": 0.235,
      "num_input_tokens_seen": 389072,
      "step": 1390
    },
    {
      "epoch": 15.5,
      "grad_norm": 1.1230659484863281,
      "learning_rate": 4.985031390416469e-05,
      "loss": 0.185,
      "num_input_tokens_seen": 390464,
      "step": 1395
    },
    {
      "epoch": 15.555555555555555,
      "grad_norm": 1.3729908466339111,
      "learning_rate": 4.984923927255461e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 391904,
      "step": 1400
    },
    {
      "epoch": 15.555555555555555,
      "eval_loss": 0.18082234263420105,
      "eval_runtime": 1.2355,
      "eval_samples_per_second": 32.376,
      "eval_steps_per_second": 16.188,
      "num_input_tokens_seen": 391904,
      "step": 1400
    },
    {
      "epoch": 15.61111111111111,
      "grad_norm": 0.4575163722038269,
      "learning_rate": 4.984816080887958e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 393344,
      "step": 1405
    },
    {
      "epoch": 15.666666666666666,
      "grad_norm": 0.6559146642684937,
      "learning_rate": 4.9847078513305875e-05,
      "loss": 0.1914,
      "num_input_tokens_seen": 394784,
      "step": 1410
    },
    {
      "epoch": 15.722222222222221,
      "grad_norm": 0.8845818042755127,
      "learning_rate": 4.984599238600043e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 396176,
      "step": 1415
    },
    {
      "epoch": 15.777777777777779,
      "grad_norm": 1.14346182346344,
      "learning_rate": 4.9844902427130716e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 397552,
      "step": 1420
    },
    {
      "epoch": 15.833333333333334,
      "grad_norm": 0.8397122025489807,
      "learning_rate": 4.984380863686482e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 398960,
      "step": 1425
    },
    {
      "epoch": 15.88888888888889,
      "grad_norm": 0.8849000930786133,
      "learning_rate": 4.984271101537143e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 400384,
      "step": 1430
    },
    {
      "epoch": 15.944444444444445,
      "grad_norm": 0.7772799730300903,
      "learning_rate": 4.9841609562819816e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 401776,
      "step": 1435
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.434482604265213,
      "learning_rate": 4.984050427937983e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 403168,
      "step": 1440
    },
    {
      "epoch": 16.055555555555557,
      "grad_norm": 0.6334783434867859,
      "learning_rate": 4.983939516522191e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 404592,
      "step": 1445
    },
    {
      "epoch": 16.11111111111111,
      "grad_norm": 0.4115030765533447,
      "learning_rate": 4.983828222051711e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 405984,
      "step": 1450
    },
    {
      "epoch": 16.166666666666668,
      "grad_norm": 0.9357030391693115,
      "learning_rate": 4.983716544543705e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 407392,
      "step": 1455
    },
    {
      "epoch": 16.22222222222222,
      "grad_norm": 1.2161204814910889,
      "learning_rate": 4.983604484015395e-05,
      "loss": 0.1927,
      "num_input_tokens_seen": 408768,
      "step": 1460
    },
    {
      "epoch": 16.27777777777778,
      "grad_norm": 0.5247777700424194,
      "learning_rate": 4.983492040484064e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 410144,
      "step": 1465
    },
    {
      "epoch": 16.333333333333332,
      "grad_norm": 1.9491363763809204,
      "learning_rate": 4.98337921396705e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 411568,
      "step": 1470
    },
    {
      "epoch": 16.38888888888889,
      "grad_norm": 0.4658815860748291,
      "learning_rate": 4.983266004481753e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 412976,
      "step": 1475
    },
    {
      "epoch": 16.444444444444443,
      "grad_norm": 0.48442408442497253,
      "learning_rate": 4.9831524120456316e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 414368,
      "step": 1480
    },
    {
      "epoch": 16.5,
      "grad_norm": 1.9384971857070923,
      "learning_rate": 4.9830384366762026e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 415760,
      "step": 1485
    },
    {
      "epoch": 16.555555555555557,
      "grad_norm": 0.7000146508216858,
      "learning_rate": 4.9829240783910436e-05,
      "loss": 0.1763,
      "num_input_tokens_seen": 417184,
      "step": 1490
    },
    {
      "epoch": 16.61111111111111,
      "grad_norm": 1.015797734260559,
      "learning_rate": 4.982809337207789e-05,
      "loss": 0.1822,
      "num_input_tokens_seen": 418544,
      "step": 1495
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 0.3452983796596527,
      "learning_rate": 4.9826942131441337e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 419952,
      "step": 1500
    },
    {
      "epoch": 16.72222222222222,
      "grad_norm": 1.984300136566162,
      "learning_rate": 4.9825787062178315e-05,
      "loss": 0.1984,
      "num_input_tokens_seen": 421392,
      "step": 1505
    },
    {
      "epoch": 16.77777777777778,
      "grad_norm": 0.531822144985199,
      "learning_rate": 4.9824628164466945e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 422784,
      "step": 1510
    },
    {
      "epoch": 16.833333333333332,
      "grad_norm": 0.5932440161705017,
      "learning_rate": 4.982346543848595e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 424192,
      "step": 1515
    },
    {
      "epoch": 16.88888888888889,
      "grad_norm": 1.0701526403427124,
      "learning_rate": 4.9822298884414626e-05,
      "loss": 0.1762,
      "num_input_tokens_seen": 425568,
      "step": 1520
    },
    {
      "epoch": 16.944444444444443,
      "grad_norm": 0.28415748476982117,
      "learning_rate": 4.982112850243288e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 426960,
      "step": 1525
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.47374215722084045,
      "learning_rate": 4.98199542927212e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 428384,
      "step": 1530
    },
    {
      "epoch": 17.055555555555557,
      "grad_norm": 0.9635041356086731,
      "learning_rate": 4.981877625546066e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 429776,
      "step": 1535
    },
    {
      "epoch": 17.11111111111111,
      "grad_norm": 0.6363587975502014,
      "learning_rate": 4.981759439083293e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 431248,
      "step": 1540
    },
    {
      "epoch": 17.166666666666668,
      "grad_norm": 1.1263530254364014,
      "learning_rate": 4.981640869902027e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 432656,
      "step": 1545
    },
    {
      "epoch": 17.22222222222222,
      "grad_norm": 0.680877149105072,
      "learning_rate": 4.9815219180205517e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 434048,
      "step": 1550
    },
    {
      "epoch": 17.27777777777778,
      "grad_norm": 0.5235930681228638,
      "learning_rate": 4.9814025834572126e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 435488,
      "step": 1555
    },
    {
      "epoch": 17.333333333333332,
      "grad_norm": 0.29449033737182617,
      "learning_rate": 4.981282866230411e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 436880,
      "step": 1560
    },
    {
      "epoch": 17.38888888888889,
      "grad_norm": 1.2929610013961792,
      "learning_rate": 4.981162766358611e-05,
      "loss": 0.2009,
      "num_input_tokens_seen": 438256,
      "step": 1565
    },
    {
      "epoch": 17.444444444444443,
      "grad_norm": 0.5355544686317444,
      "learning_rate": 4.9810422838603316e-05,
      "loss": 0.1909,
      "num_input_tokens_seen": 439664,
      "step": 1570
    },
    {
      "epoch": 17.5,
      "grad_norm": 0.6386119723320007,
      "learning_rate": 4.9809214187541533e-05,
      "loss": 0.1812,
      "num_input_tokens_seen": 441088,
      "step": 1575
    },
    {
      "epoch": 17.555555555555557,
      "grad_norm": 0.30592676997184753,
      "learning_rate": 4.980800171058715e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 442496,
      "step": 1580
    },
    {
      "epoch": 17.61111111111111,
      "grad_norm": 1.4244132041931152,
      "learning_rate": 4.980678540792715e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 443872,
      "step": 1585
    },
    {
      "epoch": 17.666666666666668,
      "grad_norm": 1.3800442218780518,
      "learning_rate": 4.980556527974909e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 445312,
      "step": 1590
    },
    {
      "epoch": 17.72222222222222,
      "grad_norm": 0.30796366930007935,
      "learning_rate": 4.980434132624114e-05,
      "loss": 0.1789,
      "num_input_tokens_seen": 446736,
      "step": 1595
    },
    {
      "epoch": 17.77777777777778,
      "grad_norm": 1.023626685142517,
      "learning_rate": 4.980311354759205e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 448112,
      "step": 1600
    },
    {
      "epoch": 17.77777777777778,
      "eval_loss": 0.17904119193553925,
      "eval_runtime": 1.2395,
      "eval_samples_per_second": 32.271,
      "eval_steps_per_second": 16.135,
      "num_input_tokens_seen": 448112,
      "step": 1600
    },
    {
      "epoch": 17.833333333333332,
      "grad_norm": 0.709891676902771,
      "learning_rate": 4.980188194399116e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 449504,
      "step": 1605
    },
    {
      "epoch": 17.88888888888889,
      "grad_norm": 1.3170000314712524,
      "learning_rate": 4.9800646515628384e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 450896,
      "step": 1610
    },
    {
      "epoch": 17.944444444444443,
      "grad_norm": 1.4679826498031616,
      "learning_rate": 4.979940726269426e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 452256,
      "step": 1615
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.4703870117664337,
      "learning_rate": 4.979816418537988e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 453648,
      "step": 1620
    },
    {
      "epoch": 18.055555555555557,
      "grad_norm": 1.263370394706726,
      "learning_rate": 4.979691728387696e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 455024,
      "step": 1625
    },
    {
      "epoch": 18.11111111111111,
      "grad_norm": 0.9282490611076355,
      "learning_rate": 4.979566655837776e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 456384,
      "step": 1630
    },
    {
      "epoch": 18.166666666666668,
      "grad_norm": 0.7424452900886536,
      "learning_rate": 4.9794412009075184e-05,
      "loss": 0.1739,
      "num_input_tokens_seen": 457824,
      "step": 1635
    },
    {
      "epoch": 18.22222222222222,
      "grad_norm": 0.6640689969062805,
      "learning_rate": 4.979315363616269e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 459136,
      "step": 1640
    },
    {
      "epoch": 18.27777777777778,
      "grad_norm": 0.31681588292121887,
      "learning_rate": 4.979189143983434e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 460544,
      "step": 1645
    },
    {
      "epoch": 18.333333333333332,
      "grad_norm": 0.6364949345588684,
      "learning_rate": 4.979062542028478e-05,
      "loss": 0.1994,
      "num_input_tokens_seen": 461968,
      "step": 1650
    },
    {
      "epoch": 18.38888888888889,
      "grad_norm": 0.427234947681427,
      "learning_rate": 4.978935557770923e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 463376,
      "step": 1655
    },
    {
      "epoch": 18.444444444444443,
      "grad_norm": 0.33037468791007996,
      "learning_rate": 4.978808191230353e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 464768,
      "step": 1660
    },
    {
      "epoch": 18.5,
      "grad_norm": 1.1001328229904175,
      "learning_rate": 4.9786804424264085e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 466144,
      "step": 1665
    },
    {
      "epoch": 18.555555555555557,
      "grad_norm": 1.1451537609100342,
      "learning_rate": 4.978552311378792e-05,
      "loss": 0.2208,
      "num_input_tokens_seen": 467504,
      "step": 1670
    },
    {
      "epoch": 18.61111111111111,
      "grad_norm": 0.2321372926235199,
      "learning_rate": 4.978423798107261e-05,
      "loss": 0.1316,
      "num_input_tokens_seen": 468912,
      "step": 1675
    },
    {
      "epoch": 18.666666666666668,
      "grad_norm": 0.4103918969631195,
      "learning_rate": 4.978294902631635e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 470304,
      "step": 1680
    },
    {
      "epoch": 18.72222222222222,
      "grad_norm": 0.5348076224327087,
      "learning_rate": 4.9781656249717914e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 471712,
      "step": 1685
    },
    {
      "epoch": 18.77777777777778,
      "grad_norm": 1.7878485918045044,
      "learning_rate": 4.9780359651476645e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 473072,
      "step": 1690
    },
    {
      "epoch": 18.833333333333332,
      "grad_norm": 0.48370397090911865,
      "learning_rate": 4.977905923179251e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 474512,
      "step": 1695
    },
    {
      "epoch": 18.88888888888889,
      "grad_norm": 0.4051567614078522,
      "learning_rate": 4.977775499086606e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 475936,
      "step": 1700
    },
    {
      "epoch": 18.944444444444443,
      "grad_norm": 0.6083509922027588,
      "learning_rate": 4.97764469288984e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 477392,
      "step": 1705
    },
    {
      "epoch": 19.0,
      "grad_norm": 1.1760424375534058,
      "learning_rate": 4.977513504609127e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 478768,
      "step": 1710
    },
    {
      "epoch": 19.055555555555557,
      "grad_norm": 0.837444007396698,
      "learning_rate": 4.9773819342646965e-05,
      "loss": 0.1709,
      "num_input_tokens_seen": 480176,
      "step": 1715
    },
    {
      "epoch": 19.11111111111111,
      "grad_norm": 1.0464037656784058,
      "learning_rate": 4.97724998187684e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 481552,
      "step": 1720
    },
    {
      "epoch": 19.166666666666668,
      "grad_norm": 0.4518974721431732,
      "learning_rate": 4.9771176474659045e-05,
      "loss": 0.1617,
      "num_input_tokens_seen": 482928,
      "step": 1725
    },
    {
      "epoch": 19.22222222222222,
      "grad_norm": 0.2580694258213043,
      "learning_rate": 4.976984931052299e-05,
      "loss": 0.1986,
      "num_input_tokens_seen": 484320,
      "step": 1730
    },
    {
      "epoch": 19.27777777777778,
      "grad_norm": 1.30778169631958,
      "learning_rate": 4.976851832656489e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 485712,
      "step": 1735
    },
    {
      "epoch": 19.333333333333332,
      "grad_norm": 0.3528192341327667,
      "learning_rate": 4.9767183522990004e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 487168,
      "step": 1740
    },
    {
      "epoch": 19.38888888888889,
      "grad_norm": 0.3838421106338501,
      "learning_rate": 4.9765844900004176e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 488528,
      "step": 1745
    },
    {
      "epoch": 19.444444444444443,
      "grad_norm": 1.8655658960342407,
      "learning_rate": 4.9764502457813834e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 489920,
      "step": 1750
    },
    {
      "epoch": 19.5,
      "grad_norm": 0.4097006320953369,
      "learning_rate": 4.9763156196626005e-05,
      "loss": 0.2036,
      "num_input_tokens_seen": 491328,
      "step": 1755
    },
    {
      "epoch": 19.555555555555557,
      "grad_norm": 0.8221561312675476,
      "learning_rate": 4.97618061166483e-05,
      "loss": 0.1814,
      "num_input_tokens_seen": 492704,
      "step": 1760
    },
    {
      "epoch": 19.61111111111111,
      "grad_norm": 0.44096699357032776,
      "learning_rate": 4.9760452218088915e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 494096,
      "step": 1765
    },
    {
      "epoch": 19.666666666666668,
      "grad_norm": 1.3051478862762451,
      "learning_rate": 4.975909450115663e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 495504,
      "step": 1770
    },
    {
      "epoch": 19.72222222222222,
      "grad_norm": 0.6437891125679016,
      "learning_rate": 4.975773296606084e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 496944,
      "step": 1775
    },
    {
      "epoch": 19.77777777777778,
      "grad_norm": 0.49528566002845764,
      "learning_rate": 4.97563676130115e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 498336,
      "step": 1780
    },
    {
      "epoch": 19.833333333333332,
      "grad_norm": 0.8525067567825317,
      "learning_rate": 4.9754998442219166e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 499776,
      "step": 1785
    },
    {
      "epoch": 19.88888888888889,
      "grad_norm": 0.4116416871547699,
      "learning_rate": 4.9753625453894984e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 501120,
      "step": 1790
    },
    {
      "epoch": 19.944444444444443,
      "grad_norm": 0.7633364200592041,
      "learning_rate": 4.975224864825068e-05,
      "loss": 0.1959,
      "num_input_tokens_seen": 502480,
      "step": 1795
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.6458736658096313,
      "learning_rate": 4.9750868025498576e-05,
      "loss": 0.1728,
      "num_input_tokens_seen": 503920,
      "step": 1800
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.1772594004869461,
      "eval_runtime": 1.2301,
      "eval_samples_per_second": 32.517,
      "eval_steps_per_second": 16.259,
      "num_input_tokens_seen": 503920,
      "step": 1800
    },
    {
      "epoch": 20.055555555555557,
      "grad_norm": 0.4867231845855713,
      "learning_rate": 4.974948358585158e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 505296,
      "step": 1805
    },
    {
      "epoch": 20.11111111111111,
      "grad_norm": 0.4236871600151062,
      "learning_rate": 4.9748095329523205e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 506688,
      "step": 1810
    },
    {
      "epoch": 20.166666666666668,
      "grad_norm": 0.8124524354934692,
      "learning_rate": 4.974670325672752e-05,
      "loss": 0.1806,
      "num_input_tokens_seen": 508112,
      "step": 1815
    },
    {
      "epoch": 20.22222222222222,
      "grad_norm": 0.49758198857307434,
      "learning_rate": 4.974530736767921e-05,
      "loss": 0.1935,
      "num_input_tokens_seen": 509488,
      "step": 1820
    },
    {
      "epoch": 20.27777777777778,
      "grad_norm": 1.6452497243881226,
      "learning_rate": 4.9743907662593524e-05,
      "loss": 0.1973,
      "num_input_tokens_seen": 510928,
      "step": 1825
    },
    {
      "epoch": 20.333333333333332,
      "grad_norm": 0.4660825729370117,
      "learning_rate": 4.974250414168633e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 512352,
      "step": 1830
    },
    {
      "epoch": 20.38888888888889,
      "grad_norm": 0.34138473868370056,
      "learning_rate": 4.974109680517407e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 513792,
      "step": 1835
    },
    {
      "epoch": 20.444444444444443,
      "grad_norm": 1.0403227806091309,
      "learning_rate": 4.973968565327376e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 515232,
      "step": 1840
    },
    {
      "epoch": 20.5,
      "grad_norm": 0.9274554252624512,
      "learning_rate": 4.973827068620303e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 516608,
      "step": 1845
    },
    {
      "epoch": 20.555555555555557,
      "grad_norm": 0.8772063255310059,
      "learning_rate": 4.973685190418008e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 517984,
      "step": 1850
    },
    {
      "epoch": 20.61111111111111,
      "grad_norm": 0.6366112232208252,
      "learning_rate": 4.97354293074237e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 519392,
      "step": 1855
    },
    {
      "epoch": 20.666666666666668,
      "grad_norm": 0.7338708639144897,
      "learning_rate": 4.9734002896153276e-05,
      "loss": 0.1989,
      "num_input_tokens_seen": 520768,
      "step": 1860
    },
    {
      "epoch": 20.72222222222222,
      "grad_norm": 0.9496849179267883,
      "learning_rate": 4.973257267058877e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 522192,
      "step": 1865
    },
    {
      "epoch": 20.77777777777778,
      "grad_norm": 1.1969400644302368,
      "learning_rate": 4.973113863095076e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 523616,
      "step": 1870
    },
    {
      "epoch": 20.833333333333332,
      "grad_norm": 0.9239012598991394,
      "learning_rate": 4.9729700777460384e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 525008,
      "step": 1875
    },
    {
      "epoch": 20.88888888888889,
      "grad_norm": 0.37232983112335205,
      "learning_rate": 4.972825911033937e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 526384,
      "step": 1880
    },
    {
      "epoch": 20.944444444444443,
      "grad_norm": 0.872170627117157,
      "learning_rate": 4.9726813629810056e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 527808,
      "step": 1885
    },
    {
      "epoch": 21.0,
      "grad_norm": 1.223137378692627,
      "learning_rate": 4.9725364336095326e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 529216,
      "step": 1890
    },
    {
      "epoch": 21.055555555555557,
      "grad_norm": 0.5099790692329407,
      "learning_rate": 4.972391122941871e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 530624,
      "step": 1895
    },
    {
      "epoch": 21.11111111111111,
      "grad_norm": 1.7254536151885986,
      "learning_rate": 4.972245431000428e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 532032,
      "step": 1900
    },
    {
      "epoch": 21.166666666666668,
      "grad_norm": 0.7739202380180359,
      "learning_rate": 4.972099357807671e-05,
      "loss": 0.1663,
      "num_input_tokens_seen": 533408,
      "step": 1905
    },
    {
      "epoch": 21.22222222222222,
      "grad_norm": 1.2649383544921875,
      "learning_rate": 4.971952903386127e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 534816,
      "step": 1910
    },
    {
      "epoch": 21.27777777777778,
      "grad_norm": 0.5280109643936157,
      "learning_rate": 4.971806067758381e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 536208,
      "step": 1915
    },
    {
      "epoch": 21.333333333333332,
      "grad_norm": 0.4728420674800873,
      "learning_rate": 4.971658850947076e-05,
      "loss": 0.1299,
      "num_input_tokens_seen": 537632,
      "step": 1920
    },
    {
      "epoch": 21.38888888888889,
      "grad_norm": 0.9929671287536621,
      "learning_rate": 4.9715112529749165e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 539040,
      "step": 1925
    },
    {
      "epoch": 21.444444444444443,
      "grad_norm": 0.44252896308898926,
      "learning_rate": 4.9713632738646624e-05,
      "loss": 0.1815,
      "num_input_tokens_seen": 540464,
      "step": 1930
    },
    {
      "epoch": 21.5,
      "grad_norm": 0.5587652325630188,
      "learning_rate": 4.971214913639134e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 541872,
      "step": 1935
    },
    {
      "epoch": 21.555555555555557,
      "grad_norm": 0.5052659511566162,
      "learning_rate": 4.9710661723212104e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 543264,
      "step": 1940
    },
    {
      "epoch": 21.61111111111111,
      "grad_norm": 0.5894092321395874,
      "learning_rate": 4.9709170499338295e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 544688,
      "step": 1945
    },
    {
      "epoch": 21.666666666666668,
      "grad_norm": 0.3685329556465149,
      "learning_rate": 4.9707675464999895e-05,
      "loss": 0.198,
      "num_input_tokens_seen": 546096,
      "step": 1950
    },
    {
      "epoch": 21.72222222222222,
      "grad_norm": 0.7492983937263489,
      "learning_rate": 4.970617662042743e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 547472,
      "step": 1955
    },
    {
      "epoch": 21.77777777777778,
      "grad_norm": 1.0163826942443848,
      "learning_rate": 4.970467396585206e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 548912,
      "step": 1960
    },
    {
      "epoch": 21.833333333333332,
      "grad_norm": 1.0986512899398804,
      "learning_rate": 4.97031675015055e-05,
      "loss": 0.1387,
      "num_input_tokens_seen": 550304,
      "step": 1965
    },
    {
      "epoch": 21.88888888888889,
      "grad_norm": 0.8267799019813538,
      "learning_rate": 4.9701657227620075e-05,
      "loss": 0.1263,
      "num_input_tokens_seen": 551664,
      "step": 1970
    },
    {
      "epoch": 21.944444444444443,
      "grad_norm": 1.3908714056015015,
      "learning_rate": 4.9700143144428685e-05,
      "loss": 0.1652,
      "num_input_tokens_seen": 553040,
      "step": 1975
    },
    {
      "epoch": 22.0,
      "grad_norm": 1.417788028717041,
      "learning_rate": 4.969862525216482e-05,
      "loss": 0.2138,
      "num_input_tokens_seen": 554400,
      "step": 1980
    },
    {
      "epoch": 22.055555555555557,
      "grad_norm": 0.6534412503242493,
      "learning_rate": 4.9697103551062556e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 555856,
      "step": 1985
    },
    {
      "epoch": 22.11111111111111,
      "grad_norm": 0.8015350699424744,
      "learning_rate": 4.9695578041356565e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 557232,
      "step": 1990
    },
    {
      "epoch": 22.166666666666668,
      "grad_norm": 0.3808780908584595,
      "learning_rate": 4.969404872328209e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 558640,
      "step": 1995
    },
    {
      "epoch": 22.22222222222222,
      "grad_norm": 1.084282636642456,
      "learning_rate": 4.969251559707498e-05,
      "loss": 0.1878,
      "num_input_tokens_seen": 560016,
      "step": 2000
    },
    {
      "epoch": 22.22222222222222,
      "eval_loss": 0.1755533665418625,
      "eval_runtime": 1.2392,
      "eval_samples_per_second": 32.278,
      "eval_steps_per_second": 16.139,
      "num_input_tokens_seen": 560016,
      "step": 2000
    },
    {
      "epoch": 22.27777777777778,
      "grad_norm": 0.6856106519699097,
      "learning_rate": 4.9690978662971674e-05,
      "loss": 0.1072,
      "num_input_tokens_seen": 561408,
      "step": 2005
    },
    {
      "epoch": 22.333333333333332,
      "grad_norm": 0.3182367980480194,
      "learning_rate": 4.968943792120916e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 562784,
      "step": 2010
    },
    {
      "epoch": 22.38888888888889,
      "grad_norm": 0.48531755805015564,
      "learning_rate": 4.9687893372025046e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 564224,
      "step": 2015
    },
    {
      "epoch": 22.444444444444443,
      "grad_norm": 2.1046535968780518,
      "learning_rate": 4.9686345015657535e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 565616,
      "step": 2020
    },
    {
      "epoch": 22.5,
      "grad_norm": 1.0263229608535767,
      "learning_rate": 4.968479285234538e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 566992,
      "step": 2025
    },
    {
      "epoch": 22.555555555555557,
      "grad_norm": 0.9265908598899841,
      "learning_rate": 4.9683236882327974e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 568384,
      "step": 2030
    },
    {
      "epoch": 22.61111111111111,
      "grad_norm": 1.306644320487976,
      "learning_rate": 4.968167710584526e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 569792,
      "step": 2035
    },
    {
      "epoch": 22.666666666666668,
      "grad_norm": 0.5203812718391418,
      "learning_rate": 4.968011352313775e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 571216,
      "step": 2040
    },
    {
      "epoch": 22.72222222222222,
      "grad_norm": 0.6089593768119812,
      "learning_rate": 4.967854613444659e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 572640,
      "step": 2045
    },
    {
      "epoch": 22.77777777777778,
      "grad_norm": 1.361174464225769,
      "learning_rate": 4.967697494001349e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 573968,
      "step": 2050
    },
    {
      "epoch": 22.833333333333332,
      "grad_norm": 1.374223232269287,
      "learning_rate": 4.9675399940080736e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 575360,
      "step": 2055
    },
    {
      "epoch": 22.88888888888889,
      "grad_norm": 0.5722189545631409,
      "learning_rate": 4.9673821134891226e-05,
      "loss": 0.1534,
      "num_input_tokens_seen": 576752,
      "step": 2060
    },
    {
      "epoch": 22.944444444444443,
      "grad_norm": 1.0351024866104126,
      "learning_rate": 4.967223852468842e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 578192,
      "step": 2065
    },
    {
      "epoch": 23.0,
      "grad_norm": 1.396262526512146,
      "learning_rate": 4.967065210971639e-05,
      "loss": 0.1832,
      "num_input_tokens_seen": 579584,
      "step": 2070
    },
    {
      "epoch": 23.055555555555557,
      "grad_norm": 1.1280466318130493,
      "learning_rate": 4.966906189021977e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 580992,
      "step": 2075
    },
    {
      "epoch": 23.11111111111111,
      "grad_norm": 0.6658402681350708,
      "learning_rate": 4.966746786644379e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 582368,
      "step": 2080
    },
    {
      "epoch": 23.166666666666668,
      "grad_norm": 0.6221481561660767,
      "learning_rate": 4.966587003863429e-05,
      "loss": 0.1034,
      "num_input_tokens_seen": 583760,
      "step": 2085
    },
    {
      "epoch": 23.22222222222222,
      "grad_norm": 0.3784397542476654,
      "learning_rate": 4.966426840703765e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 585184,
      "step": 2090
    },
    {
      "epoch": 23.27777777777778,
      "grad_norm": 1.6457993984222412,
      "learning_rate": 4.9662662971900875e-05,
      "loss": 0.1864,
      "num_input_tokens_seen": 586544,
      "step": 2095
    },
    {
      "epoch": 23.333333333333332,
      "grad_norm": 0.5695469975471497,
      "learning_rate": 4.9661053733471534e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 587952,
      "step": 2100
    },
    {
      "epoch": 23.38888888888889,
      "grad_norm": 0.9522337913513184,
      "learning_rate": 4.965944069199781e-05,
      "loss": 0.1806,
      "num_input_tokens_seen": 589376,
      "step": 2105
    },
    {
      "epoch": 23.444444444444443,
      "grad_norm": 0.8493772149085999,
      "learning_rate": 4.965782384772842e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 590832,
      "step": 2110
    },
    {
      "epoch": 23.5,
      "grad_norm": 0.4849632680416107,
      "learning_rate": 4.9656203200912734e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 592208,
      "step": 2115
    },
    {
      "epoch": 23.555555555555557,
      "grad_norm": 0.6985151767730713,
      "learning_rate": 4.965457875180067e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 593584,
      "step": 2120
    },
    {
      "epoch": 23.61111111111111,
      "grad_norm": 0.5956185460090637,
      "learning_rate": 4.9652950500642724e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 594928,
      "step": 2125
    },
    {
      "epoch": 23.666666666666668,
      "grad_norm": 0.5316794514656067,
      "learning_rate": 4.965131844769001e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 596304,
      "step": 2130
    },
    {
      "epoch": 23.72222222222222,
      "grad_norm": 0.7342842817306519,
      "learning_rate": 4.96496825931942e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 597712,
      "step": 2135
    },
    {
      "epoch": 23.77777777777778,
      "grad_norm": 0.7232881188392639,
      "learning_rate": 4.9648042937407566e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 599120,
      "step": 2140
    },
    {
      "epoch": 23.833333333333332,
      "grad_norm": 1.2872071266174316,
      "learning_rate": 4.964639948058297e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 600528,
      "step": 2145
    },
    {
      "epoch": 23.88888888888889,
      "grad_norm": 2.311302900314331,
      "learning_rate": 4.9644752222973846e-05,
      "loss": 0.1808,
      "num_input_tokens_seen": 601920,
      "step": 2150
    },
    {
      "epoch": 23.944444444444443,
      "grad_norm": 1.4072504043579102,
      "learning_rate": 4.964310116483422e-05,
      "loss": 0.1975,
      "num_input_tokens_seen": 603296,
      "step": 2155
    },
    {
      "epoch": 24.0,
      "grad_norm": 1.159556269645691,
      "learning_rate": 4.964144630641872e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 604720,
      "step": 2160
    },
    {
      "epoch": 24.055555555555557,
      "grad_norm": 1.494037389755249,
      "learning_rate": 4.9639787647982525e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 606112,
      "step": 2165
    },
    {
      "epoch": 24.11111111111111,
      "grad_norm": 0.6281684041023254,
      "learning_rate": 4.963812518978143e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 607488,
      "step": 2170
    },
    {
      "epoch": 24.166666666666668,
      "grad_norm": 0.9264286160469055,
      "learning_rate": 4.963645893207182e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 608912,
      "step": 2175
    },
    {
      "epoch": 24.22222222222222,
      "grad_norm": 0.38699766993522644,
      "learning_rate": 4.963478887511063e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 610320,
      "step": 2180
    },
    {
      "epoch": 24.27777777777778,
      "grad_norm": 0.5937960743904114,
      "learning_rate": 4.963311501915542e-05,
      "loss": 0.1562,
      "num_input_tokens_seen": 611760,
      "step": 2185
    },
    {
      "epoch": 24.333333333333332,
      "grad_norm": 0.32207226753234863,
      "learning_rate": 4.963143736446432e-05,
      "loss": 0.1914,
      "num_input_tokens_seen": 613120,
      "step": 2190
    },
    {
      "epoch": 24.38888888888889,
      "grad_norm": 1.3464572429656982,
      "learning_rate": 4.962975591129603e-05,
      "loss": 0.1144,
      "num_input_tokens_seen": 614544,
      "step": 2195
    },
    {
      "epoch": 24.444444444444443,
      "grad_norm": 1.2491759061813354,
      "learning_rate": 4.962807065990986e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 615952,
      "step": 2200
    },
    {
      "epoch": 24.444444444444443,
      "eval_loss": 0.17409680783748627,
      "eval_runtime": 1.2372,
      "eval_samples_per_second": 32.33,
      "eval_steps_per_second": 16.165,
      "num_input_tokens_seen": 615952,
      "step": 2200
    },
    {
      "epoch": 24.5,
      "grad_norm": 0.4176231026649475,
      "learning_rate": 4.9626381610565714e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 617360,
      "step": 2205
    },
    {
      "epoch": 24.555555555555557,
      "grad_norm": 0.32182252407073975,
      "learning_rate": 4.9624688763524043e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 618736,
      "step": 2210
    },
    {
      "epoch": 24.61111111111111,
      "grad_norm": 0.55516517162323,
      "learning_rate": 4.962299211904591e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 620128,
      "step": 2215
    },
    {
      "epoch": 24.666666666666668,
      "grad_norm": 0.5142238736152649,
      "learning_rate": 4.962129167739296e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 621488,
      "step": 2220
    },
    {
      "epoch": 24.72222222222222,
      "grad_norm": 0.5443081259727478,
      "learning_rate": 4.961958743882742e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 622880,
      "step": 2225
    },
    {
      "epoch": 24.77777777777778,
      "grad_norm": 1.736491322517395,
      "learning_rate": 4.961787940361211e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 624304,
      "step": 2230
    },
    {
      "epoch": 24.833333333333332,
      "grad_norm": 0.4860185384750366,
      "learning_rate": 4.961616757201043e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 625696,
      "step": 2235
    },
    {
      "epoch": 24.88888888888889,
      "grad_norm": 0.7403572201728821,
      "learning_rate": 4.961445194428637e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 627120,
      "step": 2240
    },
    {
      "epoch": 24.944444444444443,
      "grad_norm": 0.2989293038845062,
      "learning_rate": 4.9612732520704486e-05,
      "loss": 0.113,
      "num_input_tokens_seen": 628528,
      "step": 2245
    },
    {
      "epoch": 25.0,
      "grad_norm": 1.1656547784805298,
      "learning_rate": 4.961100930152994e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 630000,
      "step": 2250
    },
    {
      "epoch": 25.055555555555557,
      "grad_norm": 1.5994230508804321,
      "learning_rate": 4.960928228702849e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 631408,
      "step": 2255
    },
    {
      "epoch": 25.11111111111111,
      "grad_norm": 0.553300678730011,
      "learning_rate": 4.960755147746645e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 632800,
      "step": 2260
    },
    {
      "epoch": 25.166666666666668,
      "grad_norm": 1.613895058631897,
      "learning_rate": 4.9605816873110736e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 634240,
      "step": 2265
    },
    {
      "epoch": 25.22222222222222,
      "grad_norm": 0.33770695328712463,
      "learning_rate": 4.960407847422883e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 635632,
      "step": 2270
    },
    {
      "epoch": 25.27777777777778,
      "grad_norm": 1.2180849313735962,
      "learning_rate": 4.960233628108885e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 637008,
      "step": 2275
    },
    {
      "epoch": 25.333333333333332,
      "grad_norm": 1.220702886581421,
      "learning_rate": 4.960059029395942e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 638384,
      "step": 2280
    },
    {
      "epoch": 25.38888888888889,
      "grad_norm": 1.0822198390960693,
      "learning_rate": 4.959884051310983e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 639808,
      "step": 2285
    },
    {
      "epoch": 25.444444444444443,
      "grad_norm": 0.8714125752449036,
      "learning_rate": 4.959708693880991e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 641232,
      "step": 2290
    },
    {
      "epoch": 25.5,
      "grad_norm": 1.4673609733581543,
      "learning_rate": 4.9595329571330074e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 642608,
      "step": 2295
    },
    {
      "epoch": 25.555555555555557,
      "grad_norm": 0.7686823606491089,
      "learning_rate": 4.9593568410941326e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 644032,
      "step": 2300
    },
    {
      "epoch": 25.61111111111111,
      "grad_norm": 0.3965284526348114,
      "learning_rate": 4.959180345791528e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 645440,
      "step": 2305
    },
    {
      "epoch": 25.666666666666668,
      "grad_norm": 0.5500483512878418,
      "learning_rate": 4.9590034712524086e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 646816,
      "step": 2310
    },
    {
      "epoch": 25.72222222222222,
      "grad_norm": 0.6927191019058228,
      "learning_rate": 4.958826217504053e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 648240,
      "step": 2315
    },
    {
      "epoch": 25.77777777777778,
      "grad_norm": 0.4406355917453766,
      "learning_rate": 4.958648584573795e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 649632,
      "step": 2320
    },
    {
      "epoch": 25.833333333333332,
      "grad_norm": 0.41278380155563354,
      "learning_rate": 4.958470572489028e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 651040,
      "step": 2325
    },
    {
      "epoch": 25.88888888888889,
      "grad_norm": 1.4171417951583862,
      "learning_rate": 4.958292181277203e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 652432,
      "step": 2330
    },
    {
      "epoch": 25.944444444444443,
      "grad_norm": 0.8030608892440796,
      "learning_rate": 4.958113410965832e-05,
      "loss": 0.1164,
      "num_input_tokens_seen": 653808,
      "step": 2335
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.5085896849632263,
      "learning_rate": 4.957934261582481e-05,
      "loss": 0.1979,
      "num_input_tokens_seen": 655200,
      "step": 2340
    },
    {
      "epoch": 26.055555555555557,
      "grad_norm": 0.6095486879348755,
      "learning_rate": 4.95775473315478e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 656624,
      "step": 2345
    },
    {
      "epoch": 26.11111111111111,
      "grad_norm": 0.4362907409667969,
      "learning_rate": 4.9575748257104124e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 658048,
      "step": 2350
    },
    {
      "epoch": 26.166666666666668,
      "grad_norm": 1.5273178815841675,
      "learning_rate": 4.9573945392771224e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 659456,
      "step": 2355
    },
    {
      "epoch": 26.22222222222222,
      "grad_norm": 1.5962083339691162,
      "learning_rate": 4.9572138738827134e-05,
      "loss": 0.1762,
      "num_input_tokens_seen": 660864,
      "step": 2360
    },
    {
      "epoch": 26.27777777777778,
      "grad_norm": 0.3637585937976837,
      "learning_rate": 4.957032829555046e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 662224,
      "step": 2365
    },
    {
      "epoch": 26.333333333333332,
      "grad_norm": 0.9392313957214355,
      "learning_rate": 4.956851406322039e-05,
      "loss": 0.1835,
      "num_input_tokens_seen": 663680,
      "step": 2370
    },
    {
      "epoch": 26.38888888888889,
      "grad_norm": 1.183505892753601,
      "learning_rate": 4.9566696042116704e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 665104,
      "step": 2375
    },
    {
      "epoch": 26.444444444444443,
      "grad_norm": 0.599602460861206,
      "learning_rate": 4.9564874232519766e-05,
      "loss": 0.1157,
      "num_input_tokens_seen": 666496,
      "step": 2380
    },
    {
      "epoch": 26.5,
      "grad_norm": 0.44472813606262207,
      "learning_rate": 4.9563048634710516e-05,
      "loss": 0.1099,
      "num_input_tokens_seen": 667904,
      "step": 2385
    },
    {
      "epoch": 26.555555555555557,
      "grad_norm": 0.969132125377655,
      "learning_rate": 4.956121924897049e-05,
      "loss": 0.1313,
      "num_input_tokens_seen": 669280,
      "step": 2390
    },
    {
      "epoch": 26.61111111111111,
      "grad_norm": 1.2452738285064697,
      "learning_rate": 4.955938607558181e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 670720,
      "step": 2395
    },
    {
      "epoch": 26.666666666666668,
      "grad_norm": 1.598823070526123,
      "learning_rate": 4.955754911482715e-05,
      "loss": 0.1208,
      "num_input_tokens_seen": 672080,
      "step": 2400
    },
    {
      "epoch": 26.666666666666668,
      "eval_loss": 0.17653852701187134,
      "eval_runtime": 1.2368,
      "eval_samples_per_second": 32.342,
      "eval_steps_per_second": 16.171,
      "num_input_tokens_seen": 672080,
      "step": 2400
    },
    {
      "epoch": 26.72222222222222,
      "grad_norm": 1.521596074104309,
      "learning_rate": 4.9555708366989804e-05,
      "loss": 0.1578,
      "num_input_tokens_seen": 673440,
      "step": 2405
    },
    {
      "epoch": 26.77777777777778,
      "grad_norm": 0.8009665012359619,
      "learning_rate": 4.9553863832353655e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 674816,
      "step": 2410
    },
    {
      "epoch": 26.833333333333332,
      "grad_norm": 1.3025891780853271,
      "learning_rate": 4.955201551120313e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 676176,
      "step": 2415
    },
    {
      "epoch": 26.88888888888889,
      "grad_norm": 0.910458505153656,
      "learning_rate": 4.955016340382328e-05,
      "loss": 0.1806,
      "num_input_tokens_seen": 677552,
      "step": 2420
    },
    {
      "epoch": 26.944444444444443,
      "grad_norm": 1.5315111875534058,
      "learning_rate": 4.954830751049972e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 678960,
      "step": 2425
    },
    {
      "epoch": 27.0,
      "grad_norm": 1.3524807691574097,
      "learning_rate": 4.954644783151864e-05,
      "loss": 0.1274,
      "num_input_tokens_seen": 680432,
      "step": 2430
    },
    {
      "epoch": 27.055555555555557,
      "grad_norm": 0.5984790921211243,
      "learning_rate": 4.954458436716684e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 681808,
      "step": 2435
    },
    {
      "epoch": 27.11111111111111,
      "grad_norm": 1.6740776300430298,
      "learning_rate": 4.954271711773168e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 683216,
      "step": 2440
    },
    {
      "epoch": 27.166666666666668,
      "grad_norm": 0.5651077628135681,
      "learning_rate": 4.9540846083501115e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 684608,
      "step": 2445
    },
    {
      "epoch": 27.22222222222222,
      "grad_norm": 0.6258739233016968,
      "learning_rate": 4.953897126476369e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 686000,
      "step": 2450
    },
    {
      "epoch": 27.27777777777778,
      "grad_norm": 2.3720617294311523,
      "learning_rate": 4.9537092661808514e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 687408,
      "step": 2455
    },
    {
      "epoch": 27.333333333333332,
      "grad_norm": 1.1862773895263672,
      "learning_rate": 4.9535210274925306e-05,
      "loss": 0.105,
      "num_input_tokens_seen": 688800,
      "step": 2460
    },
    {
      "epoch": 27.38888888888889,
      "grad_norm": 2.9778709411621094,
      "learning_rate": 4.953332410440435e-05,
      "loss": 0.1829,
      "num_input_tokens_seen": 690208,
      "step": 2465
    },
    {
      "epoch": 27.444444444444443,
      "grad_norm": 2.341010808944702,
      "learning_rate": 4.9531434150536496e-05,
      "loss": 0.1636,
      "num_input_tokens_seen": 691568,
      "step": 2470
    },
    {
      "epoch": 27.5,
      "grad_norm": 0.9470853805541992,
      "learning_rate": 4.952954041361322e-05,
      "loss": 0.1123,
      "num_input_tokens_seen": 692976,
      "step": 2475
    },
    {
      "epoch": 27.555555555555557,
      "grad_norm": 0.9152123928070068,
      "learning_rate": 4.952764289392655e-05,
      "loss": 0.1005,
      "num_input_tokens_seen": 694336,
      "step": 2480
    },
    {
      "epoch": 27.61111111111111,
      "grad_norm": 1.3688881397247314,
      "learning_rate": 4.952574159176912e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 695744,
      "step": 2485
    },
    {
      "epoch": 27.666666666666668,
      "grad_norm": 0.41440221667289734,
      "learning_rate": 4.952383650743413e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 697152,
      "step": 2490
    },
    {
      "epoch": 27.72222222222222,
      "grad_norm": 0.547885537147522,
      "learning_rate": 4.952192764121536e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 698592,
      "step": 2495
    },
    {
      "epoch": 27.77777777777778,
      "grad_norm": 0.69452303647995,
      "learning_rate": 4.9520014993407185e-05,
      "loss": 0.1137,
      "num_input_tokens_seen": 700000,
      "step": 2500
    },
    {
      "epoch": 27.833333333333332,
      "grad_norm": 0.8713584542274475,
      "learning_rate": 4.951809856430456e-05,
      "loss": 0.183,
      "num_input_tokens_seen": 701424,
      "step": 2505
    },
    {
      "epoch": 27.88888888888889,
      "grad_norm": 0.867139458656311,
      "learning_rate": 4.951617835420303e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 702816,
      "step": 2510
    },
    {
      "epoch": 27.944444444444443,
      "grad_norm": 1.1480491161346436,
      "learning_rate": 4.951425436339869e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 704208,
      "step": 2515
    },
    {
      "epoch": 28.0,
      "grad_norm": 1.51802659034729,
      "learning_rate": 4.9512326592188274e-05,
      "loss": 0.1827,
      "num_input_tokens_seen": 705616,
      "step": 2520
    },
    {
      "epoch": 28.055555555555557,
      "grad_norm": 0.8370967507362366,
      "learning_rate": 4.9510395040869054e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 706992,
      "step": 2525
    },
    {
      "epoch": 28.11111111111111,
      "grad_norm": 1.5629960298538208,
      "learning_rate": 4.9508459709738905e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 708448,
      "step": 2530
    },
    {
      "epoch": 28.166666666666668,
      "grad_norm": 1.3116310834884644,
      "learning_rate": 4.950652059909627e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 709840,
      "step": 2535
    },
    {
      "epoch": 28.22222222222222,
      "grad_norm": 1.9878357648849487,
      "learning_rate": 4.95045777092402e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 711216,
      "step": 2540
    },
    {
      "epoch": 28.27777777777778,
      "grad_norm": 0.8635640144348145,
      "learning_rate": 4.950263104047031e-05,
      "loss": 0.1176,
      "num_input_tokens_seen": 712624,
      "step": 2545
    },
    {
      "epoch": 28.333333333333332,
      "grad_norm": 1.535151481628418,
      "learning_rate": 4.9500680593086775e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 714048,
      "step": 2550
    },
    {
      "epoch": 28.38888888888889,
      "grad_norm": 0.6365631222724915,
      "learning_rate": 4.94987263673904e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 715424,
      "step": 2555
    },
    {
      "epoch": 28.444444444444443,
      "grad_norm": 1.4065312147140503,
      "learning_rate": 4.949676836368256e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 716800,
      "step": 2560
    },
    {
      "epoch": 28.5,
      "grad_norm": 0.8330762386322021,
      "learning_rate": 4.949480658226518e-05,
      "loss": 0.1645,
      "num_input_tokens_seen": 718208,
      "step": 2565
    },
    {
      "epoch": 28.555555555555557,
      "grad_norm": 0.2378087043762207,
      "learning_rate": 4.949284102344082e-05,
      "loss": 0.1399,
      "num_input_tokens_seen": 719600,
      "step": 2570
    },
    {
      "epoch": 28.61111111111111,
      "grad_norm": 1.0021966695785522,
      "learning_rate": 4.9490871687512565e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 720976,
      "step": 2575
    },
    {
      "epoch": 28.666666666666668,
      "grad_norm": 0.728069543838501,
      "learning_rate": 4.948889857478413e-05,
      "loss": 0.1187,
      "num_input_tokens_seen": 722368,
      "step": 2580
    },
    {
      "epoch": 28.72222222222222,
      "grad_norm": 0.595953643321991,
      "learning_rate": 4.948692168555978e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 723808,
      "step": 2585
    },
    {
      "epoch": 28.77777777777778,
      "grad_norm": 0.6227155327796936,
      "learning_rate": 4.94849410201444e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 725216,
      "step": 2590
    },
    {
      "epoch": 28.833333333333332,
      "grad_norm": 0.3647812306880951,
      "learning_rate": 4.948295657884341e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 726624,
      "step": 2595
    },
    {
      "epoch": 28.88888888888889,
      "grad_norm": 1.3661571741104126,
      "learning_rate": 4.9480968361962835e-05,
      "loss": 0.1784,
      "num_input_tokens_seen": 728000,
      "step": 2600
    },
    {
      "epoch": 28.88888888888889,
      "eval_loss": 0.1722700148820877,
      "eval_runtime": 1.2549,
      "eval_samples_per_second": 31.874,
      "eval_steps_per_second": 15.937,
      "num_input_tokens_seen": 728000,
      "step": 2600
    },
    {
      "epoch": 28.944444444444443,
      "grad_norm": 1.9943803548812866,
      "learning_rate": 4.9478976369809305e-05,
      "loss": 0.1231,
      "num_input_tokens_seen": 729424,
      "step": 2605
    },
    {
      "epoch": 29.0,
      "grad_norm": 1.290898084640503,
      "learning_rate": 4.947698060268999e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 730816,
      "step": 2610
    },
    {
      "epoch": 29.055555555555557,
      "grad_norm": 0.43148481845855713,
      "learning_rate": 4.9474981060912665e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 732208,
      "step": 2615
    },
    {
      "epoch": 29.11111111111111,
      "grad_norm": 0.7145689725875854,
      "learning_rate": 4.94729777447857e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 733584,
      "step": 2620
    },
    {
      "epoch": 29.166666666666668,
      "grad_norm": 1.091159701347351,
      "learning_rate": 4.947097065461801e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 734992,
      "step": 2625
    },
    {
      "epoch": 29.22222222222222,
      "grad_norm": 0.5655441284179688,
      "learning_rate": 4.9468959790719125e-05,
      "loss": 0.1622,
      "num_input_tokens_seen": 736400,
      "step": 2630
    },
    {
      "epoch": 29.27777777777778,
      "grad_norm": 1.1347408294677734,
      "learning_rate": 4.9466945153399146e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 737776,
      "step": 2635
    },
    {
      "epoch": 29.333333333333332,
      "grad_norm": 0.7754456400871277,
      "learning_rate": 4.9464926742968755e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 739152,
      "step": 2640
    },
    {
      "epoch": 29.38888888888889,
      "grad_norm": 1.4254276752471924,
      "learning_rate": 4.946290455973921e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 740576,
      "step": 2645
    },
    {
      "epoch": 29.444444444444443,
      "grad_norm": 2.0206499099731445,
      "learning_rate": 4.9460878604022365e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 741968,
      "step": 2650
    },
    {
      "epoch": 29.5,
      "grad_norm": 0.9715413451194763,
      "learning_rate": 4.945884887613065e-05,
      "loss": 0.1365,
      "num_input_tokens_seen": 743328,
      "step": 2655
    },
    {
      "epoch": 29.555555555555557,
      "grad_norm": 1.1470414400100708,
      "learning_rate": 4.9456815376377055e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 744704,
      "step": 2660
    },
    {
      "epoch": 29.61111111111111,
      "grad_norm": 1.273575782775879,
      "learning_rate": 4.9454778105075195e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 746112,
      "step": 2665
    },
    {
      "epoch": 29.666666666666668,
      "grad_norm": 1.7525956630706787,
      "learning_rate": 4.945273706253924e-05,
      "loss": 0.1791,
      "num_input_tokens_seen": 747536,
      "step": 2670
    },
    {
      "epoch": 29.72222222222222,
      "grad_norm": 0.5751947164535522,
      "learning_rate": 4.9450692249083925e-05,
      "loss": 0.1339,
      "num_input_tokens_seen": 748944,
      "step": 2675
    },
    {
      "epoch": 29.77777777777778,
      "grad_norm": 0.536825954914093,
      "learning_rate": 4.9448643665024605e-05,
      "loss": 0.1065,
      "num_input_tokens_seen": 750368,
      "step": 2680
    },
    {
      "epoch": 29.833333333333332,
      "grad_norm": 1.586511492729187,
      "learning_rate": 4.944659131067719e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 751776,
      "step": 2685
    },
    {
      "epoch": 29.88888888888889,
      "grad_norm": 0.9577611684799194,
      "learning_rate": 4.944453518635818e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 753184,
      "step": 2690
    },
    {
      "epoch": 29.944444444444443,
      "grad_norm": 0.625480055809021,
      "learning_rate": 4.944247529238465e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 754544,
      "step": 2695
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.9112982153892517,
      "learning_rate": 4.944041162907427e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 755952,
      "step": 2700
    },
    {
      "epoch": 30.055555555555557,
      "grad_norm": 1.6375967264175415,
      "learning_rate": 4.943834419674529e-05,
      "loss": 0.1401,
      "num_input_tokens_seen": 757408,
      "step": 2705
    },
    {
      "epoch": 30.11111111111111,
      "grad_norm": 1.0467469692230225,
      "learning_rate": 4.9436272995716506e-05,
      "loss": 0.1694,
      "num_input_tokens_seen": 758816,
      "step": 2710
    },
    {
      "epoch": 30.166666666666668,
      "grad_norm": 0.9702094197273254,
      "learning_rate": 4.943419802630735e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 760224,
      "step": 2715
    },
    {
      "epoch": 30.22222222222222,
      "grad_norm": 0.7123199701309204,
      "learning_rate": 4.94321192888378e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 761616,
      "step": 2720
    },
    {
      "epoch": 30.27777777777778,
      "grad_norm": 0.5506933927536011,
      "learning_rate": 4.943003678362842e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 762992,
      "step": 2725
    },
    {
      "epoch": 30.333333333333332,
      "grad_norm": 0.5306255221366882,
      "learning_rate": 4.942795051100036e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 764384,
      "step": 2730
    },
    {
      "epoch": 30.38888888888889,
      "grad_norm": 0.3286009132862091,
      "learning_rate": 4.942586047127536e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 765760,
      "step": 2735
    },
    {
      "epoch": 30.444444444444443,
      "grad_norm": 0.7104036808013916,
      "learning_rate": 4.942376666477571e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 767168,
      "step": 2740
    },
    {
      "epoch": 30.5,
      "grad_norm": 1.043595790863037,
      "learning_rate": 4.9421669091824304e-05,
      "loss": 0.1643,
      "num_input_tokens_seen": 768560,
      "step": 2745
    },
    {
      "epoch": 30.555555555555557,
      "grad_norm": 0.9561747908592224,
      "learning_rate": 4.9419567752744634e-05,
      "loss": 0.1043,
      "num_input_tokens_seen": 769936,
      "step": 2750
    },
    {
      "epoch": 30.61111111111111,
      "grad_norm": 0.4862174689769745,
      "learning_rate": 4.941746264786074e-05,
      "loss": 0.168,
      "num_input_tokens_seen": 771360,
      "step": 2755
    },
    {
      "epoch": 30.666666666666668,
      "grad_norm": 0.6561657786369324,
      "learning_rate": 4.9415353777497254e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 772816,
      "step": 2760
    },
    {
      "epoch": 30.72222222222222,
      "grad_norm": 1.4369595050811768,
      "learning_rate": 4.9413241141979394e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 774208,
      "step": 2765
    },
    {
      "epoch": 30.77777777777778,
      "grad_norm": 1.0014433860778809,
      "learning_rate": 4.9411124741632956e-05,
      "loss": 0.1735,
      "num_input_tokens_seen": 775584,
      "step": 2770
    },
    {
      "epoch": 30.833333333333332,
      "grad_norm": 0.5889262557029724,
      "learning_rate": 4.940900457678431e-05,
      "loss": 0.1114,
      "num_input_tokens_seen": 776944,
      "step": 2775
    },
    {
      "epoch": 30.88888888888889,
      "grad_norm": 0.674523115158081,
      "learning_rate": 4.9406880647760425e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 778352,
      "step": 2780
    },
    {
      "epoch": 30.944444444444443,
      "grad_norm": 0.5828707218170166,
      "learning_rate": 4.9404752954888824e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 779744,
      "step": 2785
    },
    {
      "epoch": 31.0,
      "grad_norm": 0.6203059554100037,
      "learning_rate": 4.940262149849762e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 781088,
      "step": 2790
    },
    {
      "epoch": 31.055555555555557,
      "grad_norm": 1.3277438879013062,
      "learning_rate": 4.9400486278915526e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 782480,
      "step": 2795
    },
    {
      "epoch": 31.11111111111111,
      "grad_norm": 0.12956085801124573,
      "learning_rate": 4.939834729647181e-05,
      "loss": 0.0887,
      "num_input_tokens_seen": 783872,
      "step": 2800
    },
    {
      "epoch": 31.11111111111111,
      "eval_loss": 0.17663982510566711,
      "eval_runtime": 1.2469,
      "eval_samples_per_second": 32.08,
      "eval_steps_per_second": 16.04,
      "num_input_tokens_seen": 783872,
      "step": 2800
    },
    {
      "epoch": 31.166666666666668,
      "grad_norm": 1.368493914604187,
      "learning_rate": 4.9396204551496326e-05,
      "loss": 0.1209,
      "num_input_tokens_seen": 785264,
      "step": 2805
    },
    {
      "epoch": 31.22222222222222,
      "grad_norm": 1.9310373067855835,
      "learning_rate": 4.939405804431952e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 786704,
      "step": 2810
    },
    {
      "epoch": 31.27777777777778,
      "grad_norm": 0.740089476108551,
      "learning_rate": 4.9391907775272414e-05,
      "loss": 0.1132,
      "num_input_tokens_seen": 788064,
      "step": 2815
    },
    {
      "epoch": 31.333333333333332,
      "grad_norm": 0.9922474026679993,
      "learning_rate": 4.9389753744686604e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 789472,
      "step": 2820
    },
    {
      "epoch": 31.38888888888889,
      "grad_norm": 0.6507025957107544,
      "learning_rate": 4.938759595289426e-05,
      "loss": 0.1758,
      "num_input_tokens_seen": 790848,
      "step": 2825
    },
    {
      "epoch": 31.444444444444443,
      "grad_norm": 0.8400025963783264,
      "learning_rate": 4.938543440022815e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 792192,
      "step": 2830
    },
    {
      "epoch": 31.5,
      "grad_norm": 1.8651453256607056,
      "learning_rate": 4.938326908702161e-05,
      "loss": 0.2127,
      "num_input_tokens_seen": 793648,
      "step": 2835
    },
    {
      "epoch": 31.555555555555557,
      "grad_norm": 0.7733319401741028,
      "learning_rate": 4.9381100013608554e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 795056,
      "step": 2840
    },
    {
      "epoch": 31.61111111111111,
      "grad_norm": 1.5034613609313965,
      "learning_rate": 4.9378927180323485e-05,
      "loss": 0.1816,
      "num_input_tokens_seen": 796480,
      "step": 2845
    },
    {
      "epoch": 31.666666666666668,
      "grad_norm": 0.7863556742668152,
      "learning_rate": 4.937675058750148e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 797888,
      "step": 2850
    },
    {
      "epoch": 31.72222222222222,
      "grad_norm": 0.45489415526390076,
      "learning_rate": 4.937457023547819e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 799328,
      "step": 2855
    },
    {
      "epoch": 31.77777777777778,
      "grad_norm": 1.39766263961792,
      "learning_rate": 4.9372386124589876e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 800752,
      "step": 2860
    },
    {
      "epoch": 31.833333333333332,
      "grad_norm": 1.9824304580688477,
      "learning_rate": 4.937019825517333e-05,
      "loss": 0.1205,
      "num_input_tokens_seen": 802144,
      "step": 2865
    },
    {
      "epoch": 31.88888888888889,
      "grad_norm": 4.9405131340026855,
      "learning_rate": 4.9368006627565954e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 803504,
      "step": 2870
    },
    {
      "epoch": 31.944444444444443,
      "grad_norm": 0.5907888412475586,
      "learning_rate": 4.936581124210573e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 804944,
      "step": 2875
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.5741150975227356,
      "learning_rate": 4.9363612099131216e-05,
      "loss": 0.0966,
      "num_input_tokens_seen": 806352,
      "step": 2880
    },
    {
      "epoch": 32.05555555555556,
      "grad_norm": 0.7746472358703613,
      "learning_rate": 4.936140919898155e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 807760,
      "step": 2885
    },
    {
      "epoch": 32.111111111111114,
      "grad_norm": 1.119870662689209,
      "learning_rate": 4.9359202541996426e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 809184,
      "step": 2890
    },
    {
      "epoch": 32.166666666666664,
      "grad_norm": 0.5177024602890015,
      "learning_rate": 4.935699212851616e-05,
      "loss": 0.1072,
      "num_input_tokens_seen": 810576,
      "step": 2895
    },
    {
      "epoch": 32.22222222222222,
      "grad_norm": 1.009179711341858,
      "learning_rate": 4.935477795888162e-05,
      "loss": 0.1129,
      "num_input_tokens_seen": 812000,
      "step": 2900
    },
    {
      "epoch": 32.27777777777778,
      "grad_norm": 1.0807346105575562,
      "learning_rate": 4.935256003343426e-05,
      "loss": 0.1747,
      "num_input_tokens_seen": 813392,
      "step": 2905
    },
    {
      "epoch": 32.333333333333336,
      "grad_norm": 2.307401657104492,
      "learning_rate": 4.93503383525161e-05,
      "loss": 0.1041,
      "num_input_tokens_seen": 814800,
      "step": 2910
    },
    {
      "epoch": 32.388888888888886,
      "grad_norm": 2.185917854309082,
      "learning_rate": 4.934811291646977e-05,
      "loss": 0.1987,
      "num_input_tokens_seen": 816176,
      "step": 2915
    },
    {
      "epoch": 32.44444444444444,
      "grad_norm": 0.8642987012863159,
      "learning_rate": 4.934588372563845e-05,
      "loss": 0.1205,
      "num_input_tokens_seen": 817584,
      "step": 2920
    },
    {
      "epoch": 32.5,
      "grad_norm": 0.8956053853034973,
      "learning_rate": 4.93436507803659e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 818944,
      "step": 2925
    },
    {
      "epoch": 32.55555555555556,
      "grad_norm": 0.8802151083946228,
      "learning_rate": 4.934141408099649e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 820336,
      "step": 2930
    },
    {
      "epoch": 32.611111111111114,
      "grad_norm": 1.3656219244003296,
      "learning_rate": 4.9339173627875135e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 821728,
      "step": 2935
    },
    {
      "epoch": 32.666666666666664,
      "grad_norm": 1.15183424949646,
      "learning_rate": 4.9336929421347335e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 823136,
      "step": 2940
    },
    {
      "epoch": 32.72222222222222,
      "grad_norm": 0.6841804385185242,
      "learning_rate": 4.933468146175918e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 824528,
      "step": 2945
    },
    {
      "epoch": 32.77777777777778,
      "grad_norm": 2.0503807067871094,
      "learning_rate": 4.933242974945734e-05,
      "loss": 0.091,
      "num_input_tokens_seen": 825984,
      "step": 2950
    },
    {
      "epoch": 32.833333333333336,
      "grad_norm": 0.706894040107727,
      "learning_rate": 4.933017428478906e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 827360,
      "step": 2955
    },
    {
      "epoch": 32.888888888888886,
      "grad_norm": 0.7540112733840942,
      "learning_rate": 4.932791506810214e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 828752,
      "step": 2960
    },
    {
      "epoch": 32.94444444444444,
      "grad_norm": 0.37836313247680664,
      "learning_rate": 4.932565209974499e-05,
      "loss": 0.1304,
      "num_input_tokens_seen": 830160,
      "step": 2965
    },
    {
      "epoch": 33.0,
      "grad_norm": 0.6419324278831482,
      "learning_rate": 4.93233853800666e-05,
      "loss": 0.0969,
      "num_input_tokens_seen": 831520,
      "step": 2970
    },
    {
      "epoch": 33.05555555555556,
      "grad_norm": 0.8511049151420593,
      "learning_rate": 4.932111490941651e-05,
      "loss": 0.1042,
      "num_input_tokens_seen": 832880,
      "step": 2975
    },
    {
      "epoch": 33.111111111111114,
      "grad_norm": 0.9632062911987305,
      "learning_rate": 4.9318840688144876e-05,
      "loss": 0.1149,
      "num_input_tokens_seen": 834240,
      "step": 2980
    },
    {
      "epoch": 33.166666666666664,
      "grad_norm": 1.2681008577346802,
      "learning_rate": 4.9316562716602387e-05,
      "loss": 0.1076,
      "num_input_tokens_seen": 835616,
      "step": 2985
    },
    {
      "epoch": 33.22222222222222,
      "grad_norm": 0.5905352234840393,
      "learning_rate": 4.9314280995140346e-05,
      "loss": 0.1253,
      "num_input_tokens_seen": 837024,
      "step": 2990
    },
    {
      "epoch": 33.27777777777778,
      "grad_norm": 0.6009187698364258,
      "learning_rate": 4.931199552411063e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 838416,
      "step": 2995
    },
    {
      "epoch": 33.333333333333336,
      "grad_norm": 0.9227120876312256,
      "learning_rate": 4.930970630386568e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 839808,
      "step": 3000
    },
    {
      "epoch": 33.333333333333336,
      "eval_loss": 0.1776736080646515,
      "eval_runtime": 1.2361,
      "eval_samples_per_second": 32.36,
      "eval_steps_per_second": 16.18,
      "num_input_tokens_seen": 839808,
      "step": 3000
    },
    {
      "epoch": 33.388888888888886,
      "grad_norm": 1.0216569900512695,
      "learning_rate": 4.9307413334758524e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 841200,
      "step": 3005
    },
    {
      "epoch": 33.44444444444444,
      "grad_norm": 1.7926079034805298,
      "learning_rate": 4.930511661714276e-05,
      "loss": 0.1031,
      "num_input_tokens_seen": 842560,
      "step": 3010
    },
    {
      "epoch": 33.5,
      "grad_norm": 0.7680830955505371,
      "learning_rate": 4.9302816151372576e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 843984,
      "step": 3015
    },
    {
      "epoch": 33.55555555555556,
      "grad_norm": 0.6624910235404968,
      "learning_rate": 4.930051193780274e-05,
      "loss": 0.1276,
      "num_input_tokens_seen": 845392,
      "step": 3020
    },
    {
      "epoch": 33.611111111111114,
      "grad_norm": 0.9596295356750488,
      "learning_rate": 4.929820397678858e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 846832,
      "step": 3025
    },
    {
      "epoch": 33.666666666666664,
      "grad_norm": 1.698525071144104,
      "learning_rate": 4.9295892268686015e-05,
      "loss": 0.1835,
      "num_input_tokens_seen": 848224,
      "step": 3030
    },
    {
      "epoch": 33.72222222222222,
      "grad_norm": 0.565406858921051,
      "learning_rate": 4.9293576813851536e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 849616,
      "step": 3035
    },
    {
      "epoch": 33.77777777777778,
      "grad_norm": 0.4456782639026642,
      "learning_rate": 4.929125761264223e-05,
      "loss": 0.0876,
      "num_input_tokens_seen": 851040,
      "step": 3040
    },
    {
      "epoch": 33.833333333333336,
      "grad_norm": 0.4998404085636139,
      "learning_rate": 4.928893466541573e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 852480,
      "step": 3045
    },
    {
      "epoch": 33.888888888888886,
      "grad_norm": 0.9149336814880371,
      "learning_rate": 4.928660797253027e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 853872,
      "step": 3050
    },
    {
      "epoch": 33.94444444444444,
      "grad_norm": 1.3334101438522339,
      "learning_rate": 4.928427753434467e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 855280,
      "step": 3055
    },
    {
      "epoch": 34.0,
      "grad_norm": 2.043971300125122,
      "learning_rate": 4.9281943351218286e-05,
      "loss": 0.1057,
      "num_input_tokens_seen": 856688,
      "step": 3060
    },
    {
      "epoch": 34.05555555555556,
      "grad_norm": 1.8952771425247192,
      "learning_rate": 4.9279605423511095e-05,
      "loss": 0.1238,
      "num_input_tokens_seen": 858128,
      "step": 3065
    },
    {
      "epoch": 34.111111111111114,
      "grad_norm": 1.2542903423309326,
      "learning_rate": 4.927726375158363e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 859520,
      "step": 3070
    },
    {
      "epoch": 34.166666666666664,
      "grad_norm": 0.9127988219261169,
      "learning_rate": 4.9274918335797004e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 860928,
      "step": 3075
    },
    {
      "epoch": 34.22222222222222,
      "grad_norm": 1.0387693643569946,
      "learning_rate": 4.927256917651292e-05,
      "loss": 0.1116,
      "num_input_tokens_seen": 862320,
      "step": 3080
    },
    {
      "epoch": 34.27777777777778,
      "grad_norm": 1.336787223815918,
      "learning_rate": 4.927021627409364e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 863712,
      "step": 3085
    },
    {
      "epoch": 34.333333333333336,
      "grad_norm": 0.3736105263233185,
      "learning_rate": 4.9267859628902005e-05,
      "loss": 0.1033,
      "num_input_tokens_seen": 865104,
      "step": 3090
    },
    {
      "epoch": 34.388888888888886,
      "grad_norm": 0.9288823008537292,
      "learning_rate": 4.9265499241301454e-05,
      "loss": 0.0916,
      "num_input_tokens_seen": 866464,
      "step": 3095
    },
    {
      "epoch": 34.44444444444444,
      "grad_norm": 1.1889209747314453,
      "learning_rate": 4.926313511165598e-05,
      "loss": 0.1191,
      "num_input_tokens_seen": 867856,
      "step": 3100
    },
    {
      "epoch": 34.5,
      "grad_norm": 1.071210265159607,
      "learning_rate": 4.926076724033016e-05,
      "loss": 0.1789,
      "num_input_tokens_seen": 869216,
      "step": 3105
    },
    {
      "epoch": 34.55555555555556,
      "grad_norm": 1.5985215902328491,
      "learning_rate": 4.9258395627689146e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 870656,
      "step": 3110
    },
    {
      "epoch": 34.611111111111114,
      "grad_norm": 0.7508277297019958,
      "learning_rate": 4.925602027409868e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 872032,
      "step": 3115
    },
    {
      "epoch": 34.666666666666664,
      "grad_norm": 2.038886308670044,
      "learning_rate": 4.925364117992507e-05,
      "loss": 0.1666,
      "num_input_tokens_seen": 873440,
      "step": 3120
    },
    {
      "epoch": 34.72222222222222,
      "grad_norm": 1.0687428712844849,
      "learning_rate": 4.92512583455352e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 874864,
      "step": 3125
    },
    {
      "epoch": 34.77777777777778,
      "grad_norm": 0.42299580574035645,
      "learning_rate": 4.9248871771296536e-05,
      "loss": 0.0775,
      "num_input_tokens_seen": 876288,
      "step": 3130
    },
    {
      "epoch": 34.833333333333336,
      "grad_norm": 0.8802577257156372,
      "learning_rate": 4.924648145757711e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 877744,
      "step": 3135
    },
    {
      "epoch": 34.888888888888886,
      "grad_norm": 0.8778260946273804,
      "learning_rate": 4.924408740474554e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 879152,
      "step": 3140
    },
    {
      "epoch": 34.94444444444444,
      "grad_norm": 1.134452223777771,
      "learning_rate": 4.924168961317103e-05,
      "loss": 0.1428,
      "num_input_tokens_seen": 880544,
      "step": 3145
    },
    {
      "epoch": 35.0,
      "grad_norm": 2.24222993850708,
      "learning_rate": 4.9239288083223334e-05,
      "loss": 0.1935,
      "num_input_tokens_seen": 881984,
      "step": 3150
    },
    {
      "epoch": 35.05555555555556,
      "grad_norm": 1.657233715057373,
      "learning_rate": 4.9236882815272803e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 883392,
      "step": 3155
    },
    {
      "epoch": 35.111111111111114,
      "grad_norm": 1.810636043548584,
      "learning_rate": 4.9234473809690365e-05,
      "loss": 0.1029,
      "num_input_tokens_seen": 884848,
      "step": 3160
    },
    {
      "epoch": 35.166666666666664,
      "grad_norm": 2.072108745574951,
      "learning_rate": 4.923206106684752e-05,
      "loss": 0.1177,
      "num_input_tokens_seen": 886272,
      "step": 3165
    },
    {
      "epoch": 35.22222222222222,
      "grad_norm": 1.12697434425354,
      "learning_rate": 4.922964458711634e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 887680,
      "step": 3170
    },
    {
      "epoch": 35.27777777777778,
      "grad_norm": 0.8569563627243042,
      "learning_rate": 4.9227224370869474e-05,
      "loss": 0.125,
      "num_input_tokens_seen": 889056,
      "step": 3175
    },
    {
      "epoch": 35.333333333333336,
      "grad_norm": 1.8318535089492798,
      "learning_rate": 4.9224800418480155e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 890400,
      "step": 3180
    },
    {
      "epoch": 35.388888888888886,
      "grad_norm": 1.0961310863494873,
      "learning_rate": 4.9222372730322176e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 891792,
      "step": 3185
    },
    {
      "epoch": 35.44444444444444,
      "grad_norm": 0.5608189702033997,
      "learning_rate": 4.921994130676993e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 893248,
      "step": 3190
    },
    {
      "epoch": 35.5,
      "grad_norm": 1.2831895351409912,
      "learning_rate": 4.9217506148198366e-05,
      "loss": 0.1399,
      "num_input_tokens_seen": 894672,
      "step": 3195
    },
    {
      "epoch": 35.55555555555556,
      "grad_norm": 0.7256914973258972,
      "learning_rate": 4.921506725498302e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 896064,
      "step": 3200
    },
    {
      "epoch": 35.55555555555556,
      "eval_loss": 0.17774644494056702,
      "eval_runtime": 1.236,
      "eval_samples_per_second": 32.362,
      "eval_steps_per_second": 16.181,
      "num_input_tokens_seen": 896064,
      "step": 3200
    },
    {
      "epoch": 35.611111111111114,
      "grad_norm": 1.6363109350204468,
      "learning_rate": 4.9212624627499994e-05,
      "loss": 0.1238,
      "num_input_tokens_seen": 897456,
      "step": 3205
    },
    {
      "epoch": 35.666666666666664,
      "grad_norm": 1.8023390769958496,
      "learning_rate": 4.921017826612597e-05,
      "loss": 0.1236,
      "num_input_tokens_seen": 898832,
      "step": 3210
    },
    {
      "epoch": 35.72222222222222,
      "grad_norm": 0.3542575538158417,
      "learning_rate": 4.9207728171238223e-05,
      "loss": 0.066,
      "num_input_tokens_seen": 900192,
      "step": 3215
    },
    {
      "epoch": 35.77777777777778,
      "grad_norm": 1.0536742210388184,
      "learning_rate": 4.920527434321458e-05,
      "loss": 0.1126,
      "num_input_tokens_seen": 901536,
      "step": 3220
    },
    {
      "epoch": 35.833333333333336,
      "grad_norm": 0.49757152795791626,
      "learning_rate": 4.920281678243345e-05,
      "loss": 0.1155,
      "num_input_tokens_seen": 902960,
      "step": 3225
    },
    {
      "epoch": 35.888888888888886,
      "grad_norm": 0.4594939053058624,
      "learning_rate": 4.920035548927381e-05,
      "loss": 0.1123,
      "num_input_tokens_seen": 904384,
      "step": 3230
    },
    {
      "epoch": 35.94444444444444,
      "grad_norm": 0.5924867391586304,
      "learning_rate": 4.919789046411525e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 905792,
      "step": 3235
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.6375471353530884,
      "learning_rate": 4.919542170733787e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 907168,
      "step": 3240
    },
    {
      "epoch": 36.05555555555556,
      "grad_norm": 1.563642144203186,
      "learning_rate": 4.919294921932242e-05,
      "loss": 0.0943,
      "num_input_tokens_seen": 908544,
      "step": 3245
    },
    {
      "epoch": 36.111111111111114,
      "grad_norm": 1.2492492198944092,
      "learning_rate": 4.919047300045016e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 909952,
      "step": 3250
    },
    {
      "epoch": 36.166666666666664,
      "grad_norm": 1.426565408706665,
      "learning_rate": 4.918799305110299e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 911328,
      "step": 3255
    },
    {
      "epoch": 36.22222222222222,
      "grad_norm": 0.8016552329063416,
      "learning_rate": 4.918550937166331e-05,
      "loss": 0.1042,
      "num_input_tokens_seen": 912704,
      "step": 3260
    },
    {
      "epoch": 36.27777777777778,
      "grad_norm": 0.3468841016292572,
      "learning_rate": 4.918302196251415e-05,
      "loss": 0.1186,
      "num_input_tokens_seen": 914128,
      "step": 3265
    },
    {
      "epoch": 36.333333333333336,
      "grad_norm": 0.40067628026008606,
      "learning_rate": 4.91805308240391e-05,
      "loss": 0.1073,
      "num_input_tokens_seen": 915536,
      "step": 3270
    },
    {
      "epoch": 36.388888888888886,
      "grad_norm": 0.4732546806335449,
      "learning_rate": 4.9178035956622326e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 916944,
      "step": 3275
    },
    {
      "epoch": 36.44444444444444,
      "grad_norm": 2.6714587211608887,
      "learning_rate": 4.917553736064857e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 918368,
      "step": 3280
    },
    {
      "epoch": 36.5,
      "grad_norm": 1.4231961965560913,
      "learning_rate": 4.917303503650314e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 919792,
      "step": 3285
    },
    {
      "epoch": 36.55555555555556,
      "grad_norm": 0.920678973197937,
      "learning_rate": 4.917052898457194e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 921200,
      "step": 3290
    },
    {
      "epoch": 36.611111111111114,
      "grad_norm": 1.6975300312042236,
      "learning_rate": 4.916801920524141e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 922592,
      "step": 3295
    },
    {
      "epoch": 36.666666666666664,
      "grad_norm": 0.5349475145339966,
      "learning_rate": 4.916550569889862e-05,
      "loss": 0.0754,
      "num_input_tokens_seen": 924000,
      "step": 3300
    },
    {
      "epoch": 36.72222222222222,
      "grad_norm": 1.0051701068878174,
      "learning_rate": 4.916298846593116e-05,
      "loss": 0.1728,
      "num_input_tokens_seen": 925376,
      "step": 3305
    },
    {
      "epoch": 36.77777777777778,
      "grad_norm": 0.9057824611663818,
      "learning_rate": 4.916046750672722e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 926816,
      "step": 3310
    },
    {
      "epoch": 36.833333333333336,
      "grad_norm": 2.141697406768799,
      "learning_rate": 4.915794282167559e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 928176,
      "step": 3315
    },
    {
      "epoch": 36.888888888888886,
      "grad_norm": 1.030424952507019,
      "learning_rate": 4.915541441116558e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 929568,
      "step": 3320
    },
    {
      "epoch": 36.94444444444444,
      "grad_norm": 2.0296823978424072,
      "learning_rate": 4.915288227558711e-05,
      "loss": 0.0919,
      "num_input_tokens_seen": 930992,
      "step": 3325
    },
    {
      "epoch": 37.0,
      "grad_norm": 1.0230636596679688,
      "learning_rate": 4.915034641533066e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 932384,
      "step": 3330
    },
    {
      "epoch": 37.05555555555556,
      "grad_norm": 1.3585960865020752,
      "learning_rate": 4.914780683078731e-05,
      "loss": 0.1216,
      "num_input_tokens_seen": 933776,
      "step": 3335
    },
    {
      "epoch": 37.111111111111114,
      "grad_norm": 0.914517879486084,
      "learning_rate": 4.9145263522348695e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 935184,
      "step": 3340
    },
    {
      "epoch": 37.166666666666664,
      "grad_norm": 0.7652620673179626,
      "learning_rate": 4.9142716490407e-05,
      "loss": 0.1018,
      "num_input_tokens_seen": 936576,
      "step": 3345
    },
    {
      "epoch": 37.22222222222222,
      "grad_norm": 0.4598771333694458,
      "learning_rate": 4.914016573535504e-05,
      "loss": 0.1225,
      "num_input_tokens_seen": 938032,
      "step": 3350
    },
    {
      "epoch": 37.27777777777778,
      "grad_norm": 0.9155217409133911,
      "learning_rate": 4.9137611257586154e-05,
      "loss": 0.0991,
      "num_input_tokens_seen": 939456,
      "step": 3355
    },
    {
      "epoch": 37.333333333333336,
      "grad_norm": 0.6549411416053772,
      "learning_rate": 4.9135053057494274e-05,
      "loss": 0.1099,
      "num_input_tokens_seen": 940800,
      "step": 3360
    },
    {
      "epoch": 37.388888888888886,
      "grad_norm": 3.8623759746551514,
      "learning_rate": 4.913249113547392e-05,
      "loss": 0.1809,
      "num_input_tokens_seen": 942160,
      "step": 3365
    },
    {
      "epoch": 37.44444444444444,
      "grad_norm": 2.0697696208953857,
      "learning_rate": 4.912992549192016e-05,
      "loss": 0.1227,
      "num_input_tokens_seen": 943536,
      "step": 3370
    },
    {
      "epoch": 37.5,
      "grad_norm": 6.547144889831543,
      "learning_rate": 4.9127356127228665e-05,
      "loss": 0.1129,
      "num_input_tokens_seen": 944928,
      "step": 3375
    },
    {
      "epoch": 37.55555555555556,
      "grad_norm": 0.647189199924469,
      "learning_rate": 4.912478304179564e-05,
      "loss": 0.0708,
      "num_input_tokens_seen": 946336,
      "step": 3380
    },
    {
      "epoch": 37.611111111111114,
      "grad_norm": 0.43387606739997864,
      "learning_rate": 4.9122206236017896e-05,
      "loss": 0.1081,
      "num_input_tokens_seen": 947712,
      "step": 3385
    },
    {
      "epoch": 37.666666666666664,
      "grad_norm": 0.3431682884693146,
      "learning_rate": 4.911962571029282e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 949104,
      "step": 3390
    },
    {
      "epoch": 37.72222222222222,
      "grad_norm": 1.0622941255569458,
      "learning_rate": 4.9117041465018353e-05,
      "loss": 0.1022,
      "num_input_tokens_seen": 950496,
      "step": 3395
    },
    {
      "epoch": 37.77777777777778,
      "grad_norm": 0.5189652442932129,
      "learning_rate": 4.911445350059302e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 951888,
      "step": 3400
    },
    {
      "epoch": 37.77777777777778,
      "eval_loss": 0.18585829436779022,
      "eval_runtime": 1.2411,
      "eval_samples_per_second": 32.229,
      "eval_steps_per_second": 16.114,
      "num_input_tokens_seen": 951888,
      "step": 3400
    },
    {
      "epoch": 37.833333333333336,
      "grad_norm": 1.9749772548675537,
      "learning_rate": 4.9111861817415905e-05,
      "loss": 0.1514,
      "num_input_tokens_seen": 953264,
      "step": 3405
    },
    {
      "epoch": 37.888888888888886,
      "grad_norm": 1.5017238855361938,
      "learning_rate": 4.91092664158867e-05,
      "loss": 0.1154,
      "num_input_tokens_seen": 954656,
      "step": 3410
    },
    {
      "epoch": 37.94444444444444,
      "grad_norm": 1.8898310661315918,
      "learning_rate": 4.910666729640563e-05,
      "loss": 0.094,
      "num_input_tokens_seen": 956032,
      "step": 3415
    },
    {
      "epoch": 38.0,
      "grad_norm": 3.2630515098571777,
      "learning_rate": 4.910406445937353e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 957424,
      "step": 3420
    },
    {
      "epoch": 38.05555555555556,
      "grad_norm": 2.910778522491455,
      "learning_rate": 4.9101457905191774e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 958768,
      "step": 3425
    },
    {
      "epoch": 38.111111111111114,
      "grad_norm": 1.0486925840377808,
      "learning_rate": 4.909884763426233e-05,
      "loss": 0.1222,
      "num_input_tokens_seen": 960192,
      "step": 3430
    },
    {
      "epoch": 38.166666666666664,
      "grad_norm": 1.0855281352996826,
      "learning_rate": 4.9096233646987736e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 961600,
      "step": 3435
    },
    {
      "epoch": 38.22222222222222,
      "grad_norm": 1.0639597177505493,
      "learning_rate": 4.9093615943771104e-05,
      "loss": 0.0922,
      "num_input_tokens_seen": 963024,
      "step": 3440
    },
    {
      "epoch": 38.27777777777778,
      "grad_norm": 0.46118900179862976,
      "learning_rate": 4.909099452501611e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 964384,
      "step": 3445
    },
    {
      "epoch": 38.333333333333336,
      "grad_norm": 0.4168094992637634,
      "learning_rate": 4.908836939112702e-05,
      "loss": 0.1051,
      "num_input_tokens_seen": 965760,
      "step": 3450
    },
    {
      "epoch": 38.388888888888886,
      "grad_norm": 0.4072861969470978,
      "learning_rate": 4.908574054250865e-05,
      "loss": 0.0949,
      "num_input_tokens_seen": 967152,
      "step": 3455
    },
    {
      "epoch": 38.44444444444444,
      "grad_norm": 1.1234102249145508,
      "learning_rate": 4.9083107979566414e-05,
      "loss": 0.1701,
      "num_input_tokens_seen": 968528,
      "step": 3460
    },
    {
      "epoch": 38.5,
      "grad_norm": 1.5204277038574219,
      "learning_rate": 4.908047170270628e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 969968,
      "step": 3465
    },
    {
      "epoch": 38.55555555555556,
      "grad_norm": 0.458097368478775,
      "learning_rate": 4.9077831712334784e-05,
      "loss": 0.1111,
      "num_input_tokens_seen": 971376,
      "step": 3470
    },
    {
      "epoch": 38.611111111111114,
      "grad_norm": 1.305576205253601,
      "learning_rate": 4.907518800885907e-05,
      "loss": 0.1099,
      "num_input_tokens_seen": 972816,
      "step": 3475
    },
    {
      "epoch": 38.666666666666664,
      "grad_norm": 0.4904288649559021,
      "learning_rate": 4.907254059268681e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 974192,
      "step": 3480
    },
    {
      "epoch": 38.72222222222222,
      "grad_norm": 1.7514395713806152,
      "learning_rate": 4.906988946422628e-05,
      "loss": 0.1013,
      "num_input_tokens_seen": 975600,
      "step": 3485
    },
    {
      "epoch": 38.77777777777778,
      "grad_norm": 0.7222219705581665,
      "learning_rate": 4.9067234623886315e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 976992,
      "step": 3490
    },
    {
      "epoch": 38.833333333333336,
      "grad_norm": 1.2033828496932983,
      "learning_rate": 4.9064576072076316e-05,
      "loss": 0.0943,
      "num_input_tokens_seen": 978384,
      "step": 3495
    },
    {
      "epoch": 38.888888888888886,
      "grad_norm": 0.30113956332206726,
      "learning_rate": 4.906191380920628e-05,
      "loss": 0.1153,
      "num_input_tokens_seen": 979776,
      "step": 3500
    },
    {
      "epoch": 38.94444444444444,
      "grad_norm": 0.5119709372520447,
      "learning_rate": 4.905924783568675e-05,
      "loss": 0.1245,
      "num_input_tokens_seen": 981200,
      "step": 3505
    },
    {
      "epoch": 39.0,
      "grad_norm": 1.6427382230758667,
      "learning_rate": 4.905657815192886e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 982624,
      "step": 3510
    },
    {
      "epoch": 39.05555555555556,
      "grad_norm": 0.9240574240684509,
      "learning_rate": 4.90539047583443e-05,
      "loss": 0.1476,
      "num_input_tokens_seen": 984048,
      "step": 3515
    },
    {
      "epoch": 39.111111111111114,
      "grad_norm": 1.2464436292648315,
      "learning_rate": 4.905122765534534e-05,
      "loss": 0.0958,
      "num_input_tokens_seen": 985472,
      "step": 3520
    },
    {
      "epoch": 39.166666666666664,
      "grad_norm": 2.7247426509857178,
      "learning_rate": 4.9048546843344846e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 986880,
      "step": 3525
    },
    {
      "epoch": 39.22222222222222,
      "grad_norm": 1.0759532451629639,
      "learning_rate": 4.9045862322756206e-05,
      "loss": 0.1193,
      "num_input_tokens_seen": 988320,
      "step": 3530
    },
    {
      "epoch": 39.27777777777778,
      "grad_norm": 0.6545698642730713,
      "learning_rate": 4.904317409399342e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 989680,
      "step": 3535
    },
    {
      "epoch": 39.333333333333336,
      "grad_norm": 0.6208674311637878,
      "learning_rate": 4.904048215747104e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 991136,
      "step": 3540
    },
    {
      "epoch": 39.388888888888886,
      "grad_norm": 1.4778295755386353,
      "learning_rate": 4.90377865136042e-05,
      "loss": 0.1091,
      "num_input_tokens_seen": 992528,
      "step": 3545
    },
    {
      "epoch": 39.44444444444444,
      "grad_norm": 1.8273096084594727,
      "learning_rate": 4.90350871628086e-05,
      "loss": 0.0825,
      "num_input_tokens_seen": 993920,
      "step": 3550
    },
    {
      "epoch": 39.5,
      "grad_norm": 2.47780179977417,
      "learning_rate": 4.903238410550052e-05,
      "loss": 0.1149,
      "num_input_tokens_seen": 995312,
      "step": 3555
    },
    {
      "epoch": 39.55555555555556,
      "grad_norm": 0.3030542731285095,
      "learning_rate": 4.90296773420968e-05,
      "loss": 0.1208,
      "num_input_tokens_seen": 996656,
      "step": 3560
    },
    {
      "epoch": 39.611111111111114,
      "grad_norm": 0.46370938420295715,
      "learning_rate": 4.902696687301486e-05,
      "loss": 0.1108,
      "num_input_tokens_seen": 998064,
      "step": 3565
    },
    {
      "epoch": 39.666666666666664,
      "grad_norm": 0.8056023120880127,
      "learning_rate": 4.902425269867268e-05,
      "loss": 0.0901,
      "num_input_tokens_seen": 999504,
      "step": 3570
    },
    {
      "epoch": 39.72222222222222,
      "grad_norm": 1.2472426891326904,
      "learning_rate": 4.902153481948883e-05,
      "loss": 0.1231,
      "num_input_tokens_seen": 1000864,
      "step": 3575
    },
    {
      "epoch": 39.77777777777778,
      "grad_norm": 2.4401533603668213,
      "learning_rate": 4.901881323588244e-05,
      "loss": 0.1686,
      "num_input_tokens_seen": 1002272,
      "step": 3580
    },
    {
      "epoch": 39.833333333333336,
      "grad_norm": 2.5702426433563232,
      "learning_rate": 4.90160879482732e-05,
      "loss": 0.1205,
      "num_input_tokens_seen": 1003648,
      "step": 3585
    },
    {
      "epoch": 39.888888888888886,
      "grad_norm": 0.5803899765014648,
      "learning_rate": 4.9013358957081405e-05,
      "loss": 0.0708,
      "num_input_tokens_seen": 1005008,
      "step": 3590
    },
    {
      "epoch": 39.94444444444444,
      "grad_norm": 1.344221830368042,
      "learning_rate": 4.901062626272789e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 1006384,
      "step": 3595
    },
    {
      "epoch": 40.0,
      "grad_norm": 1.0325490236282349,
      "learning_rate": 4.900788986563406e-05,
      "loss": 0.1145,
      "num_input_tokens_seen": 1007760,
      "step": 3600
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.18119138479232788,
      "eval_runtime": 1.2317,
      "eval_samples_per_second": 32.476,
      "eval_steps_per_second": 16.238,
      "num_input_tokens_seen": 1007760,
      "step": 3600
    },
    {
      "epoch": 40.05555555555556,
      "grad_norm": 1.9649540185928345,
      "learning_rate": 4.9005149766221915e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 1009168,
      "step": 3605
    },
    {
      "epoch": 40.111111111111114,
      "grad_norm": 0.4299079477787018,
      "learning_rate": 4.9002405964914e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 1010576,
      "step": 3610
    },
    {
      "epoch": 40.166666666666664,
      "grad_norm": 1.7570360898971558,
      "learning_rate": 4.899965846213346e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 1011952,
      "step": 3615
    },
    {
      "epoch": 40.22222222222222,
      "grad_norm": 0.576318085193634,
      "learning_rate": 4.899690725830399e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 1013408,
      "step": 3620
    },
    {
      "epoch": 40.27777777777778,
      "grad_norm": 1.325811743736267,
      "learning_rate": 4.899415235384985e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 1014832,
      "step": 3625
    },
    {
      "epoch": 40.333333333333336,
      "grad_norm": 1.7588492631912231,
      "learning_rate": 4.899139374919589e-05,
      "loss": 0.0838,
      "num_input_tokens_seen": 1016256,
      "step": 3630
    },
    {
      "epoch": 40.388888888888886,
      "grad_norm": 0.504028856754303,
      "learning_rate": 4.898863144476752e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 1017632,
      "step": 3635
    },
    {
      "epoch": 40.44444444444444,
      "grad_norm": 2.148488998413086,
      "learning_rate": 4.898586544099072e-05,
      "loss": 0.0953,
      "num_input_tokens_seen": 1019008,
      "step": 3640
    },
    {
      "epoch": 40.5,
      "grad_norm": 0.7909969091415405,
      "learning_rate": 4.898309573829204e-05,
      "loss": 0.1022,
      "num_input_tokens_seen": 1020432,
      "step": 3645
    },
    {
      "epoch": 40.55555555555556,
      "grad_norm": 2.0010156631469727,
      "learning_rate": 4.898032233709862e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 1021776,
      "step": 3650
    },
    {
      "epoch": 40.611111111111114,
      "grad_norm": 1.289852499961853,
      "learning_rate": 4.8977545237838123e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 1023120,
      "step": 3655
    },
    {
      "epoch": 40.666666666666664,
      "grad_norm": 0.9158067107200623,
      "learning_rate": 4.8974764440938836e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 1024512,
      "step": 3660
    },
    {
      "epoch": 40.72222222222222,
      "grad_norm": 1.3535685539245605,
      "learning_rate": 4.897197994682959e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 1025920,
      "step": 3665
    },
    {
      "epoch": 40.77777777777778,
      "grad_norm": 0.30571427941322327,
      "learning_rate": 4.8969191755939786e-05,
      "loss": 0.1092,
      "num_input_tokens_seen": 1027360,
      "step": 3670
    },
    {
      "epoch": 40.833333333333336,
      "grad_norm": 0.736592710018158,
      "learning_rate": 4.8966399868699396e-05,
      "loss": 0.124,
      "num_input_tokens_seen": 1028752,
      "step": 3675
    },
    {
      "epoch": 40.888888888888886,
      "grad_norm": 0.5435478091239929,
      "learning_rate": 4.8963604285538965e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 1030128,
      "step": 3680
    },
    {
      "epoch": 40.94444444444444,
      "grad_norm": 1.6605803966522217,
      "learning_rate": 4.8960805006889604e-05,
      "loss": 0.0755,
      "num_input_tokens_seen": 1031584,
      "step": 3685
    },
    {
      "epoch": 41.0,
      "grad_norm": 0.9809513688087463,
      "learning_rate": 4.8958002033183004e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 1032960,
      "step": 3690
    },
    {
      "epoch": 41.05555555555556,
      "grad_norm": 1.4254568815231323,
      "learning_rate": 4.8955195364851414e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 1034416,
      "step": 3695
    },
    {
      "epoch": 41.111111111111114,
      "grad_norm": 1.245874285697937,
      "learning_rate": 4.895238500232766e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 1035824,
      "step": 3700
    },
    {
      "epoch": 41.166666666666664,
      "grad_norm": 2.573258876800537,
      "learning_rate": 4.8949570946045143e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 1037184,
      "step": 3705
    },
    {
      "epoch": 41.22222222222222,
      "grad_norm": 0.5077112317085266,
      "learning_rate": 4.89467531964378e-05,
      "loss": 0.0935,
      "num_input_tokens_seen": 1038544,
      "step": 3710
    },
    {
      "epoch": 41.27777777777778,
      "grad_norm": 0.35977599024772644,
      "learning_rate": 4.894393175394019e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 1039920,
      "step": 3715
    },
    {
      "epoch": 41.333333333333336,
      "grad_norm": 0.13396835327148438,
      "learning_rate": 4.8941106618987406e-05,
      "loss": 0.1429,
      "num_input_tokens_seen": 1041344,
      "step": 3720
    },
    {
      "epoch": 41.388888888888886,
      "grad_norm": 1.606570839881897,
      "learning_rate": 4.893827779201512e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 1042704,
      "step": 3725
    },
    {
      "epoch": 41.44444444444444,
      "grad_norm": 1.0771484375,
      "learning_rate": 4.893544527345957e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 1044080,
      "step": 3730
    },
    {
      "epoch": 41.5,
      "grad_norm": 1.6591973304748535,
      "learning_rate": 4.8932609063757563e-05,
      "loss": 0.1116,
      "num_input_tokens_seen": 1045520,
      "step": 3735
    },
    {
      "epoch": 41.55555555555556,
      "grad_norm": 0.9773233532905579,
      "learning_rate": 4.8929769163346484e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 1046944,
      "step": 3740
    },
    {
      "epoch": 41.611111111111114,
      "grad_norm": 1.5880424976348877,
      "learning_rate": 4.892692557266429e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 1048336,
      "step": 3745
    },
    {
      "epoch": 41.666666666666664,
      "grad_norm": 1.8189587593078613,
      "learning_rate": 4.8924078292149464e-05,
      "loss": 0.1075,
      "num_input_tokens_seen": 1049760,
      "step": 3750
    },
    {
      "epoch": 41.72222222222222,
      "grad_norm": 0.8393303751945496,
      "learning_rate": 4.892122732224114e-05,
      "loss": 0.069,
      "num_input_tokens_seen": 1051088,
      "step": 3755
    },
    {
      "epoch": 41.77777777777778,
      "grad_norm": 1.5729399919509888,
      "learning_rate": 4.8918372663378944e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 1052448,
      "step": 3760
    },
    {
      "epoch": 41.833333333333336,
      "grad_norm": 0.26502490043640137,
      "learning_rate": 4.89155143160031e-05,
      "loss": 0.1063,
      "num_input_tokens_seen": 1053840,
      "step": 3765
    },
    {
      "epoch": 41.888888888888886,
      "grad_norm": 0.851233959197998,
      "learning_rate": 4.891265228055441e-05,
      "loss": 0.101,
      "num_input_tokens_seen": 1055232,
      "step": 3770
    },
    {
      "epoch": 41.94444444444444,
      "grad_norm": 1.247749924659729,
      "learning_rate": 4.890978655747424e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 1056672,
      "step": 3775
    },
    {
      "epoch": 42.0,
      "grad_norm": 0.9136962890625,
      "learning_rate": 4.89069171472045e-05,
      "loss": 0.1079,
      "num_input_tokens_seen": 1058080,
      "step": 3780
    },
    {
      "epoch": 42.05555555555556,
      "grad_norm": 0.668990433216095,
      "learning_rate": 4.890404405018772e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 1059440,
      "step": 3785
    },
    {
      "epoch": 42.111111111111114,
      "grad_norm": 2.077491521835327,
      "learning_rate": 4.8901167266866934e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 1060864,
      "step": 3790
    },
    {
      "epoch": 42.166666666666664,
      "grad_norm": 1.1739898920059204,
      "learning_rate": 4.88982867976858e-05,
      "loss": 0.1224,
      "num_input_tokens_seen": 1062224,
      "step": 3795
    },
    {
      "epoch": 42.22222222222222,
      "grad_norm": 3.009368658065796,
      "learning_rate": 4.889540264308852e-05,
      "loss": 0.1049,
      "num_input_tokens_seen": 1063648,
      "step": 3800
    },
    {
      "epoch": 42.22222222222222,
      "eval_loss": 0.1894650161266327,
      "eval_runtime": 1.2313,
      "eval_samples_per_second": 32.485,
      "eval_steps_per_second": 16.243,
      "num_input_tokens_seen": 1063648,
      "step": 3800
    },
    {
      "epoch": 42.27777777777778,
      "grad_norm": 1.1623510122299194,
      "learning_rate": 4.889251480351986e-05,
      "loss": 0.1074,
      "num_input_tokens_seen": 1065024,
      "step": 3805
    },
    {
      "epoch": 42.333333333333336,
      "grad_norm": 0.42714715003967285,
      "learning_rate": 4.888962327942517e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 1066432,
      "step": 3810
    },
    {
      "epoch": 42.388888888888886,
      "grad_norm": 1.0990370512008667,
      "learning_rate": 4.8886728071250356e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 1067872,
      "step": 3815
    },
    {
      "epoch": 42.44444444444444,
      "grad_norm": 1.2414565086364746,
      "learning_rate": 4.8883829179441884e-05,
      "loss": 0.0928,
      "num_input_tokens_seen": 1069264,
      "step": 3820
    },
    {
      "epoch": 42.5,
      "grad_norm": 0.9634127020835876,
      "learning_rate": 4.888092660444682e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 1070688,
      "step": 3825
    },
    {
      "epoch": 42.55555555555556,
      "grad_norm": 1.5090807676315308,
      "learning_rate": 4.887802034671276e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 1072096,
      "step": 3830
    },
    {
      "epoch": 42.611111111111114,
      "grad_norm": 0.8738115429878235,
      "learning_rate": 4.88751104066879e-05,
      "loss": 0.1617,
      "num_input_tokens_seen": 1073456,
      "step": 3835
    },
    {
      "epoch": 42.666666666666664,
      "grad_norm": 0.9189335107803345,
      "learning_rate": 4.887219678482098e-05,
      "loss": 0.061,
      "num_input_tokens_seen": 1074864,
      "step": 3840
    },
    {
      "epoch": 42.72222222222222,
      "grad_norm": 1.7346930503845215,
      "learning_rate": 4.8869279481561316e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 1076256,
      "step": 3845
    },
    {
      "epoch": 42.77777777777778,
      "grad_norm": 0.3138115108013153,
      "learning_rate": 4.88663584973588e-05,
      "loss": 0.0911,
      "num_input_tokens_seen": 1077664,
      "step": 3850
    },
    {
      "epoch": 42.833333333333336,
      "grad_norm": 1.4403859376907349,
      "learning_rate": 4.8863433832663874e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 1079072,
      "step": 3855
    },
    {
      "epoch": 42.888888888888886,
      "grad_norm": 1.2047771215438843,
      "learning_rate": 4.886050548792757e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 1080480,
      "step": 3860
    },
    {
      "epoch": 42.94444444444444,
      "grad_norm": 2.6778934001922607,
      "learning_rate": 4.8857573463601465e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 1081888,
      "step": 3865
    },
    {
      "epoch": 43.0,
      "grad_norm": 1.0066698789596558,
      "learning_rate": 4.885463776013772e-05,
      "loss": 0.1044,
      "num_input_tokens_seen": 1083264,
      "step": 3870
    },
    {
      "epoch": 43.05555555555556,
      "grad_norm": 0.3944040536880493,
      "learning_rate": 4.8851698377989056e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 1084688,
      "step": 3875
    },
    {
      "epoch": 43.111111111111114,
      "grad_norm": 2.451805591583252,
      "learning_rate": 4.884875531760876e-05,
      "loss": 0.1029,
      "num_input_tokens_seen": 1086096,
      "step": 3880
    },
    {
      "epoch": 43.166666666666664,
      "grad_norm": 1.2799866199493408,
      "learning_rate": 4.88458085794507e-05,
      "loss": 0.1053,
      "num_input_tokens_seen": 1087456,
      "step": 3885
    },
    {
      "epoch": 43.22222222222222,
      "grad_norm": 1.0710240602493286,
      "learning_rate": 4.884285816396929e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 1088864,
      "step": 3890
    },
    {
      "epoch": 43.27777777777778,
      "grad_norm": 0.6450091004371643,
      "learning_rate": 4.8839904071619526e-05,
      "loss": 0.0522,
      "num_input_tokens_seen": 1090256,
      "step": 3895
    },
    {
      "epoch": 43.333333333333336,
      "grad_norm": 0.3330268859863281,
      "learning_rate": 4.8836946302856955e-05,
      "loss": 0.1222,
      "num_input_tokens_seen": 1091696,
      "step": 3900
    },
    {
      "epoch": 43.388888888888886,
      "grad_norm": 1.4091167449951172,
      "learning_rate": 4.8833984858137715e-05,
      "loss": 0.1807,
      "num_input_tokens_seen": 1093040,
      "step": 3905
    },
    {
      "epoch": 43.44444444444444,
      "grad_norm": 2.8666293621063232,
      "learning_rate": 4.8831019737918494e-05,
      "loss": 0.1848,
      "num_input_tokens_seen": 1094464,
      "step": 3910
    },
    {
      "epoch": 43.5,
      "grad_norm": 0.7046225666999817,
      "learning_rate": 4.882805094265655e-05,
      "loss": 0.079,
      "num_input_tokens_seen": 1095856,
      "step": 3915
    },
    {
      "epoch": 43.55555555555556,
      "grad_norm": 0.6780287623405457,
      "learning_rate": 4.8825078472809706e-05,
      "loss": 0.074,
      "num_input_tokens_seen": 1097264,
      "step": 3920
    },
    {
      "epoch": 43.611111111111114,
      "grad_norm": 1.0772883892059326,
      "learning_rate": 4.882210232883635e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 1098656,
      "step": 3925
    },
    {
      "epoch": 43.666666666666664,
      "grad_norm": 0.4247226417064667,
      "learning_rate": 4.881912251119546e-05,
      "loss": 0.0921,
      "num_input_tokens_seen": 1100096,
      "step": 3930
    },
    {
      "epoch": 43.72222222222222,
      "grad_norm": 0.9055408239364624,
      "learning_rate": 4.881613902034654e-05,
      "loss": 0.081,
      "num_input_tokens_seen": 1101568,
      "step": 3935
    },
    {
      "epoch": 43.77777777777778,
      "grad_norm": 0.5806609988212585,
      "learning_rate": 4.88131518567497e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 1102976,
      "step": 3940
    },
    {
      "epoch": 43.833333333333336,
      "grad_norm": 0.7213818430900574,
      "learning_rate": 4.881016102086558e-05,
      "loss": 0.0981,
      "num_input_tokens_seen": 1104352,
      "step": 3945
    },
    {
      "epoch": 43.888888888888886,
      "grad_norm": 0.7789344191551208,
      "learning_rate": 4.8807166513155425e-05,
      "loss": 0.1428,
      "num_input_tokens_seen": 1105776,
      "step": 3950
    },
    {
      "epoch": 43.94444444444444,
      "grad_norm": 0.7608314752578735,
      "learning_rate": 4.8804168334081004e-05,
      "loss": 0.0586,
      "num_input_tokens_seen": 1107168,
      "step": 3955
    },
    {
      "epoch": 44.0,
      "grad_norm": 0.21292871236801147,
      "learning_rate": 4.880116648410468e-05,
      "loss": 0.0952,
      "num_input_tokens_seen": 1108592,
      "step": 3960
    },
    {
      "epoch": 44.05555555555556,
      "grad_norm": 1.0074095726013184,
      "learning_rate": 4.879816096368939e-05,
      "loss": 0.104,
      "num_input_tokens_seen": 1110000,
      "step": 3965
    },
    {
      "epoch": 44.111111111111114,
      "grad_norm": 1.399000644683838,
      "learning_rate": 4.879515177329861e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 1111376,
      "step": 3970
    },
    {
      "epoch": 44.166666666666664,
      "grad_norm": 1.1195217370986938,
      "learning_rate": 4.8792138913396394e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 1112800,
      "step": 3975
    },
    {
      "epoch": 44.22222222222222,
      "grad_norm": 1.3405505418777466,
      "learning_rate": 4.8789122384447374e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 1114224,
      "step": 3980
    },
    {
      "epoch": 44.27777777777778,
      "grad_norm": 0.81708824634552,
      "learning_rate": 4.878610218691673e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 1115600,
      "step": 3985
    },
    {
      "epoch": 44.333333333333336,
      "grad_norm": 0.7316602468490601,
      "learning_rate": 4.87830783212702e-05,
      "loss": 0.0503,
      "num_input_tokens_seen": 1116976,
      "step": 3990
    },
    {
      "epoch": 44.388888888888886,
      "grad_norm": 1.1026273965835571,
      "learning_rate": 4.878005078797413e-05,
      "loss": 0.0701,
      "num_input_tokens_seen": 1118336,
      "step": 3995
    },
    {
      "epoch": 44.44444444444444,
      "grad_norm": 0.7572289109230042,
      "learning_rate": 4.877701958749539e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 1119744,
      "step": 4000
    },
    {
      "epoch": 44.44444444444444,
      "eval_loss": 0.19508138298988342,
      "eval_runtime": 1.235,
      "eval_samples_per_second": 32.388,
      "eval_steps_per_second": 16.194,
      "num_input_tokens_seen": 1119744,
      "step": 4000
    },
    {
      "epoch": 44.5,
      "grad_norm": 0.37366196513175964,
      "learning_rate": 4.877398472030142e-05,
      "loss": 0.078,
      "num_input_tokens_seen": 1121088,
      "step": 4005
    },
    {
      "epoch": 44.55555555555556,
      "grad_norm": 4.189672946929932,
      "learning_rate": 4.877094618686024e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 1122512,
      "step": 4010
    },
    {
      "epoch": 44.611111111111114,
      "grad_norm": 1.2933604717254639,
      "learning_rate": 4.876790398764045e-05,
      "loss": 0.1192,
      "num_input_tokens_seen": 1123936,
      "step": 4015
    },
    {
      "epoch": 44.666666666666664,
      "grad_norm": 1.8228843212127686,
      "learning_rate": 4.8764858123111167e-05,
      "loss": 0.1143,
      "num_input_tokens_seen": 1125312,
      "step": 4020
    },
    {
      "epoch": 44.72222222222222,
      "grad_norm": 0.4175952672958374,
      "learning_rate": 4.876180859374212e-05,
      "loss": 0.0793,
      "num_input_tokens_seen": 1126736,
      "step": 4025
    },
    {
      "epoch": 44.77777777777778,
      "grad_norm": 1.5941829681396484,
      "learning_rate": 4.875875540000357e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 1128176,
      "step": 4030
    },
    {
      "epoch": 44.833333333333336,
      "grad_norm": 2.5739498138427734,
      "learning_rate": 4.8755698542366376e-05,
      "loss": 0.1072,
      "num_input_tokens_seen": 1129568,
      "step": 4035
    },
    {
      "epoch": 44.888888888888886,
      "grad_norm": 2.2937722206115723,
      "learning_rate": 4.875263802130193e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 1131008,
      "step": 4040
    },
    {
      "epoch": 44.94444444444444,
      "grad_norm": 0.46574270725250244,
      "learning_rate": 4.8749573837282207e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 1132368,
      "step": 4045
    },
    {
      "epoch": 45.0,
      "grad_norm": 1.1356052160263062,
      "learning_rate": 4.874650599077974e-05,
      "loss": 0.0655,
      "num_input_tokens_seen": 1133744,
      "step": 4050
    },
    {
      "epoch": 45.05555555555556,
      "grad_norm": 0.6715356707572937,
      "learning_rate": 4.874343448226764e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 1135152,
      "step": 4055
    },
    {
      "epoch": 45.111111111111114,
      "grad_norm": 0.5907909870147705,
      "learning_rate": 4.874035931221955e-05,
      "loss": 0.0507,
      "num_input_tokens_seen": 1136560,
      "step": 4060
    },
    {
      "epoch": 45.166666666666664,
      "grad_norm": 0.700973391532898,
      "learning_rate": 4.8737280481109724e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 1137952,
      "step": 4065
    },
    {
      "epoch": 45.22222222222222,
      "grad_norm": 4.280134677886963,
      "learning_rate": 4.873419798941294e-05,
      "loss": 0.1277,
      "num_input_tokens_seen": 1139328,
      "step": 4070
    },
    {
      "epoch": 45.27777777777778,
      "grad_norm": 0.9603222608566284,
      "learning_rate": 4.873111183760458e-05,
      "loss": 0.087,
      "num_input_tokens_seen": 1140752,
      "step": 4075
    },
    {
      "epoch": 45.333333333333336,
      "grad_norm": 1.0625340938568115,
      "learning_rate": 4.8728022026160537e-05,
      "loss": 0.0999,
      "num_input_tokens_seen": 1142096,
      "step": 4080
    },
    {
      "epoch": 45.388888888888886,
      "grad_norm": 0.7345752716064453,
      "learning_rate": 4.872492855555732e-05,
      "loss": 0.0804,
      "num_input_tokens_seen": 1143440,
      "step": 4085
    },
    {
      "epoch": 45.44444444444444,
      "grad_norm": 3.4926605224609375,
      "learning_rate": 4.8721831426271956e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 1144832,
      "step": 4090
    },
    {
      "epoch": 45.5,
      "grad_norm": 0.8881168961524963,
      "learning_rate": 4.87187306387821e-05,
      "loss": 0.123,
      "num_input_tokens_seen": 1146240,
      "step": 4095
    },
    {
      "epoch": 45.55555555555556,
      "grad_norm": 0.6598724126815796,
      "learning_rate": 4.87156261935659e-05,
      "loss": 0.0552,
      "num_input_tokens_seen": 1147648,
      "step": 4100
    },
    {
      "epoch": 45.611111111111114,
      "grad_norm": 2.6644437313079834,
      "learning_rate": 4.871251809110211e-05,
      "loss": 0.1117,
      "num_input_tokens_seen": 1149040,
      "step": 4105
    },
    {
      "epoch": 45.666666666666664,
      "grad_norm": 0.1655622124671936,
      "learning_rate": 4.8709406331870044e-05,
      "loss": 0.089,
      "num_input_tokens_seen": 1150448,
      "step": 4110
    },
    {
      "epoch": 45.72222222222222,
      "grad_norm": 0.48918819427490234,
      "learning_rate": 4.8706290916349574e-05,
      "loss": 0.1063,
      "num_input_tokens_seen": 1151856,
      "step": 4115
    },
    {
      "epoch": 45.77777777777778,
      "grad_norm": 1.8537243604660034,
      "learning_rate": 4.8703171845021134e-05,
      "loss": 0.2263,
      "num_input_tokens_seen": 1153264,
      "step": 4120
    },
    {
      "epoch": 45.833333333333336,
      "grad_norm": 0.7973994016647339,
      "learning_rate": 4.870004911836572e-05,
      "loss": 0.1237,
      "num_input_tokens_seen": 1154656,
      "step": 4125
    },
    {
      "epoch": 45.888888888888886,
      "grad_norm": 0.8339221477508545,
      "learning_rate": 4.869692273686489e-05,
      "loss": 0.067,
      "num_input_tokens_seen": 1156096,
      "step": 4130
    },
    {
      "epoch": 45.94444444444444,
      "grad_norm": 0.9145405888557434,
      "learning_rate": 4.869379270100079e-05,
      "loss": 0.051,
      "num_input_tokens_seen": 1157488,
      "step": 4135
    },
    {
      "epoch": 46.0,
      "grad_norm": 0.4150006175041199,
      "learning_rate": 4.86906590112561e-05,
      "loss": 0.1202,
      "num_input_tokens_seen": 1158864,
      "step": 4140
    },
    {
      "epoch": 46.05555555555556,
      "grad_norm": 2.4293630123138428,
      "learning_rate": 4.8687521668114064e-05,
      "loss": 0.0783,
      "num_input_tokens_seen": 1160256,
      "step": 4145
    },
    {
      "epoch": 46.111111111111114,
      "grad_norm": 0.5616565942764282,
      "learning_rate": 4.868438067205853e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 1161616,
      "step": 4150
    },
    {
      "epoch": 46.166666666666664,
      "grad_norm": 1.1636528968811035,
      "learning_rate": 4.8681236023573844e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 1163040,
      "step": 4155
    },
    {
      "epoch": 46.22222222222222,
      "grad_norm": 1.2212045192718506,
      "learning_rate": 4.867808772314497e-05,
      "loss": 0.1059,
      "num_input_tokens_seen": 1164448,
      "step": 4160
    },
    {
      "epoch": 46.27777777777778,
      "grad_norm": 1.9382424354553223,
      "learning_rate": 4.867493577125741e-05,
      "loss": 0.0573,
      "num_input_tokens_seen": 1165888,
      "step": 4165
    },
    {
      "epoch": 46.333333333333336,
      "grad_norm": 1.5791140794754028,
      "learning_rate": 4.867178016839725e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 1167296,
      "step": 4170
    },
    {
      "epoch": 46.388888888888886,
      "grad_norm": 2.8361027240753174,
      "learning_rate": 4.8668620915051094e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 1168688,
      "step": 4175
    },
    {
      "epoch": 46.44444444444444,
      "grad_norm": 1.9388455152511597,
      "learning_rate": 4.866545801170616e-05,
      "loss": 0.1029,
      "num_input_tokens_seen": 1170064,
      "step": 4180
    },
    {
      "epoch": 46.5,
      "grad_norm": 1.014315128326416,
      "learning_rate": 4.86622914588502e-05,
      "loss": 0.1074,
      "num_input_tokens_seen": 1171472,
      "step": 4185
    },
    {
      "epoch": 46.55555555555556,
      "grad_norm": 2.1271286010742188,
      "learning_rate": 4.865912125697154e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 1172880,
      "step": 4190
    },
    {
      "epoch": 46.611111111111114,
      "grad_norm": 1.3846096992492676,
      "learning_rate": 4.865594740655907e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 1174288,
      "step": 4195
    },
    {
      "epoch": 46.666666666666664,
      "grad_norm": 0.45645177364349365,
      "learning_rate": 4.865276990810222e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 1175680,
      "step": 4200
    },
    {
      "epoch": 46.666666666666664,
      "eval_loss": 0.2037002146244049,
      "eval_runtime": 1.235,
      "eval_samples_per_second": 32.39,
      "eval_steps_per_second": 16.195,
      "num_input_tokens_seen": 1175680,
      "step": 4200
    },
    {
      "epoch": 46.72222222222222,
      "grad_norm": 0.35849279165267944,
      "learning_rate": 4.8649588762091016e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 1177056,
      "step": 4205
    },
    {
      "epoch": 46.77777777777778,
      "grad_norm": 3.136747360229492,
      "learning_rate": 4.8646403969016016e-05,
      "loss": 0.0909,
      "num_input_tokens_seen": 1178464,
      "step": 4210
    },
    {
      "epoch": 46.833333333333336,
      "grad_norm": 1.1086384057998657,
      "learning_rate": 4.864321552936838e-05,
      "loss": 0.0933,
      "num_input_tokens_seen": 1179872,
      "step": 4215
    },
    {
      "epoch": 46.888888888888886,
      "grad_norm": 0.4326907992362976,
      "learning_rate": 4.864002344363978e-05,
      "loss": 0.0505,
      "num_input_tokens_seen": 1181280,
      "step": 4220
    },
    {
      "epoch": 46.94444444444444,
      "grad_norm": 2.2608790397644043,
      "learning_rate": 4.863682771232248e-05,
      "loss": 0.0632,
      "num_input_tokens_seen": 1182704,
      "step": 4225
    },
    {
      "epoch": 47.0,
      "grad_norm": 0.4232003092765808,
      "learning_rate": 4.8633628335909324e-05,
      "loss": 0.0969,
      "num_input_tokens_seen": 1184112,
      "step": 4230
    },
    {
      "epoch": 47.05555555555556,
      "grad_norm": 1.340179681777954,
      "learning_rate": 4.8630425314893676e-05,
      "loss": 0.1155,
      "num_input_tokens_seen": 1185488,
      "step": 4235
    },
    {
      "epoch": 47.111111111111114,
      "grad_norm": 2.113987922668457,
      "learning_rate": 4.862721864976948e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 1186848,
      "step": 4240
    },
    {
      "epoch": 47.166666666666664,
      "grad_norm": 0.6828175783157349,
      "learning_rate": 4.862400834103125e-05,
      "loss": 0.0843,
      "num_input_tokens_seen": 1188240,
      "step": 4245
    },
    {
      "epoch": 47.22222222222222,
      "grad_norm": 1.9526026248931885,
      "learning_rate": 4.862079438917406e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 1189680,
      "step": 4250
    },
    {
      "epoch": 47.27777777777778,
      "grad_norm": 1.334497332572937,
      "learning_rate": 4.8617576794693536e-05,
      "loss": 0.0989,
      "num_input_tokens_seen": 1191072,
      "step": 4255
    },
    {
      "epoch": 47.333333333333336,
      "grad_norm": 0.29806530475616455,
      "learning_rate": 4.8614355558085875e-05,
      "loss": 0.0729,
      "num_input_tokens_seen": 1192480,
      "step": 4260
    },
    {
      "epoch": 47.388888888888886,
      "grad_norm": 1.978284478187561,
      "learning_rate": 4.861113067984783e-05,
      "loss": 0.107,
      "num_input_tokens_seen": 1193872,
      "step": 4265
    },
    {
      "epoch": 47.44444444444444,
      "grad_norm": 1.5610870122909546,
      "learning_rate": 4.860790216047671e-05,
      "loss": 0.1191,
      "num_input_tokens_seen": 1195296,
      "step": 4270
    },
    {
      "epoch": 47.5,
      "grad_norm": 1.5069422721862793,
      "learning_rate": 4.860467000047041e-05,
      "loss": 0.0824,
      "num_input_tokens_seen": 1196672,
      "step": 4275
    },
    {
      "epoch": 47.55555555555556,
      "grad_norm": 1.4576722383499146,
      "learning_rate": 4.860143420032737e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 1198064,
      "step": 4280
    },
    {
      "epoch": 47.611111111111114,
      "grad_norm": 2.8587868213653564,
      "learning_rate": 4.859819476054657e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 1199456,
      "step": 4285
    },
    {
      "epoch": 47.666666666666664,
      "grad_norm": 1.24491548538208,
      "learning_rate": 4.859495168162758e-05,
      "loss": 0.0642,
      "num_input_tokens_seen": 1200864,
      "step": 4290
    },
    {
      "epoch": 47.72222222222222,
      "grad_norm": 1.052321434020996,
      "learning_rate": 4.859170496407054e-05,
      "loss": 0.0763,
      "num_input_tokens_seen": 1202224,
      "step": 4295
    },
    {
      "epoch": 47.77777777777778,
      "grad_norm": 1.6354718208312988,
      "learning_rate": 4.8588454608376114e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 1203632,
      "step": 4300
    },
    {
      "epoch": 47.833333333333336,
      "grad_norm": 1.1339253187179565,
      "learning_rate": 4.8585200615045555e-05,
      "loss": 0.1035,
      "num_input_tokens_seen": 1205040,
      "step": 4305
    },
    {
      "epoch": 47.888888888888886,
      "grad_norm": 3.0631799697875977,
      "learning_rate": 4.8581942984580674e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 1206480,
      "step": 4310
    },
    {
      "epoch": 47.94444444444444,
      "grad_norm": 0.41661742329597473,
      "learning_rate": 4.857868171748384e-05,
      "loss": 0.0819,
      "num_input_tokens_seen": 1207952,
      "step": 4315
    },
    {
      "epoch": 48.0,
      "grad_norm": 1.7360998392105103,
      "learning_rate": 4.8575416814257976e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 1209328,
      "step": 4320
    },
    {
      "epoch": 48.05555555555556,
      "grad_norm": 0.3175080120563507,
      "learning_rate": 4.857214827540657e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 1210768,
      "step": 4325
    },
    {
      "epoch": 48.111111111111114,
      "grad_norm": 1.246970534324646,
      "learning_rate": 4.856887610143367e-05,
      "loss": 0.0859,
      "num_input_tokens_seen": 1212096,
      "step": 4330
    },
    {
      "epoch": 48.166666666666664,
      "grad_norm": 1.998848557472229,
      "learning_rate": 4.8565600292843896e-05,
      "loss": 0.098,
      "num_input_tokens_seen": 1213504,
      "step": 4335
    },
    {
      "epoch": 48.22222222222222,
      "grad_norm": 0.8064461350440979,
      "learning_rate": 4.856232085014241e-05,
      "loss": 0.077,
      "num_input_tokens_seen": 1214944,
      "step": 4340
    },
    {
      "epoch": 48.27777777777778,
      "grad_norm": 1.3713716268539429,
      "learning_rate": 4.855903777383495e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 1216336,
      "step": 4345
    },
    {
      "epoch": 48.333333333333336,
      "grad_norm": 0.996204674243927,
      "learning_rate": 4.85557510644278e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 1217760,
      "step": 4350
    },
    {
      "epoch": 48.388888888888886,
      "grad_norm": 1.689548134803772,
      "learning_rate": 4.855246072242782e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 1219136,
      "step": 4355
    },
    {
      "epoch": 48.44444444444444,
      "grad_norm": 1.5055071115493774,
      "learning_rate": 4.8549166748342414e-05,
      "loss": 0.1083,
      "num_input_tokens_seen": 1220512,
      "step": 4360
    },
    {
      "epoch": 48.5,
      "grad_norm": 0.7839334607124329,
      "learning_rate": 4.8545869142679556e-05,
      "loss": 0.08,
      "num_input_tokens_seen": 1221888,
      "step": 4365
    },
    {
      "epoch": 48.55555555555556,
      "grad_norm": 0.8191365003585815,
      "learning_rate": 4.8542567905947776e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 1223264,
      "step": 4370
    },
    {
      "epoch": 48.611111111111114,
      "grad_norm": 1.0465352535247803,
      "learning_rate": 4.853926303865618e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 1224640,
      "step": 4375
    },
    {
      "epoch": 48.666666666666664,
      "grad_norm": 1.1438379287719727,
      "learning_rate": 4.853595454131441e-05,
      "loss": 0.0428,
      "num_input_tokens_seen": 1226016,
      "step": 4380
    },
    {
      "epoch": 48.72222222222222,
      "grad_norm": 1.6011613607406616,
      "learning_rate": 4.8532642414432674e-05,
      "loss": 0.0651,
      "num_input_tokens_seen": 1227424,
      "step": 4385
    },
    {
      "epoch": 48.77777777777778,
      "grad_norm": 0.20648597180843353,
      "learning_rate": 4.8529326658521754e-05,
      "loss": 0.0831,
      "num_input_tokens_seen": 1228864,
      "step": 4390
    },
    {
      "epoch": 48.833333333333336,
      "grad_norm": 0.4966982305049896,
      "learning_rate": 4.8526007274092965e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 1230272,
      "step": 4395
    },
    {
      "epoch": 48.888888888888886,
      "grad_norm": 1.5019574165344238,
      "learning_rate": 4.852268426165822e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 1231696,
      "step": 4400
    },
    {
      "epoch": 48.888888888888886,
      "eval_loss": 0.20535536110401154,
      "eval_runtime": 1.2479,
      "eval_samples_per_second": 32.055,
      "eval_steps_per_second": 16.027,
      "num_input_tokens_seen": 1231696,
      "step": 4400
    },
    {
      "epoch": 48.94444444444444,
      "grad_norm": 1.2010035514831543,
      "learning_rate": 4.851935762172995e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 1233136,
      "step": 4405
    },
    {
      "epoch": 49.0,
      "grad_norm": 0.8377609848976135,
      "learning_rate": 4.8516027354821175e-05,
      "loss": 0.0753,
      "num_input_tokens_seen": 1234528,
      "step": 4410
    },
    {
      "epoch": 49.05555555555556,
      "grad_norm": 2.06184983253479,
      "learning_rate": 4.851269346144546e-05,
      "loss": 0.1438,
      "num_input_tokens_seen": 1235904,
      "step": 4415
    },
    {
      "epoch": 49.111111111111114,
      "grad_norm": 4.848985195159912,
      "learning_rate": 4.850935594211693e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 1237312,
      "step": 4420
    },
    {
      "epoch": 49.166666666666664,
      "grad_norm": 1.403273105621338,
      "learning_rate": 4.850601479735029e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 1238736,
      "step": 4425
    },
    {
      "epoch": 49.22222222222222,
      "grad_norm": 0.820608377456665,
      "learning_rate": 4.850267002766076e-05,
      "loss": 0.098,
      "num_input_tokens_seen": 1240128,
      "step": 4430
    },
    {
      "epoch": 49.27777777777778,
      "grad_norm": 1.3404521942138672,
      "learning_rate": 4.849932163356417e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 1241568,
      "step": 4435
    },
    {
      "epoch": 49.333333333333336,
      "grad_norm": 0.4912288188934326,
      "learning_rate": 4.8495969615576864e-05,
      "loss": 0.0856,
      "num_input_tokens_seen": 1242976,
      "step": 4440
    },
    {
      "epoch": 49.388888888888886,
      "grad_norm": 2.309706687927246,
      "learning_rate": 4.849261397421577e-05,
      "loss": 0.0538,
      "num_input_tokens_seen": 1244384,
      "step": 4445
    },
    {
      "epoch": 49.44444444444444,
      "grad_norm": 1.3176604509353638,
      "learning_rate": 4.848925470999839e-05,
      "loss": 0.0498,
      "num_input_tokens_seen": 1245776,
      "step": 4450
    },
    {
      "epoch": 49.5,
      "grad_norm": 2.064929246902466,
      "learning_rate": 4.848589182344273e-05,
      "loss": 0.0976,
      "num_input_tokens_seen": 1247200,
      "step": 4455
    },
    {
      "epoch": 49.55555555555556,
      "grad_norm": 1.7772520780563354,
      "learning_rate": 4.848252531506742e-05,
      "loss": 0.1166,
      "num_input_tokens_seen": 1248656,
      "step": 4460
    },
    {
      "epoch": 49.611111111111114,
      "grad_norm": 1.2789725065231323,
      "learning_rate": 4.847915518539161e-05,
      "loss": 0.0564,
      "num_input_tokens_seen": 1250048,
      "step": 4465
    },
    {
      "epoch": 49.666666666666664,
      "grad_norm": 0.5534557700157166,
      "learning_rate": 4.847578143493501e-05,
      "loss": 0.0901,
      "num_input_tokens_seen": 1251424,
      "step": 4470
    },
    {
      "epoch": 49.72222222222222,
      "grad_norm": 1.2693595886230469,
      "learning_rate": 4.847240406421789e-05,
      "loss": 0.0914,
      "num_input_tokens_seen": 1252816,
      "step": 4475
    },
    {
      "epoch": 49.77777777777778,
      "grad_norm": 2.0274205207824707,
      "learning_rate": 4.84690230737611e-05,
      "loss": 0.0588,
      "num_input_tokens_seen": 1254160,
      "step": 4480
    },
    {
      "epoch": 49.833333333333336,
      "grad_norm": 1.7658131122589111,
      "learning_rate": 4.846563846408602e-05,
      "loss": 0.068,
      "num_input_tokens_seen": 1255536,
      "step": 4485
    },
    {
      "epoch": 49.888888888888886,
      "grad_norm": 1.2530781030654907,
      "learning_rate": 4.84622502357146e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 1256960,
      "step": 4490
    },
    {
      "epoch": 49.94444444444444,
      "grad_norm": 2.1876773834228516,
      "learning_rate": 4.8458858389169345e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 1258352,
      "step": 4495
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.25376978516578674,
      "learning_rate": 4.8455462924973334e-05,
      "loss": 0.0985,
      "num_input_tokens_seen": 1259744,
      "step": 4500
    },
    {
      "epoch": 50.05555555555556,
      "grad_norm": 1.0328949689865112,
      "learning_rate": 4.845206384365018e-05,
      "loss": 0.0658,
      "num_input_tokens_seen": 1261152,
      "step": 4505
    },
    {
      "epoch": 50.111111111111114,
      "grad_norm": 1.3957518339157104,
      "learning_rate": 4.844866114572405e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 1262528,
      "step": 4510
    },
    {
      "epoch": 50.166666666666664,
      "grad_norm": 2.470111846923828,
      "learning_rate": 4.8445254831719706e-05,
      "loss": 0.0946,
      "num_input_tokens_seen": 1263936,
      "step": 4515
    },
    {
      "epoch": 50.22222222222222,
      "grad_norm": 0.9915376305580139,
      "learning_rate": 4.8441844902162434e-05,
      "loss": 0.0577,
      "num_input_tokens_seen": 1265312,
      "step": 4520
    },
    {
      "epoch": 50.27777777777778,
      "grad_norm": 2.166788339614868,
      "learning_rate": 4.843843135757809e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 1266720,
      "step": 4525
    },
    {
      "epoch": 50.333333333333336,
      "grad_norm": 0.30248185992240906,
      "learning_rate": 4.843501419849308e-05,
      "loss": 0.0852,
      "num_input_tokens_seen": 1268144,
      "step": 4530
    },
    {
      "epoch": 50.388888888888886,
      "grad_norm": 0.600077748298645,
      "learning_rate": 4.8431593425434386e-05,
      "loss": 0.0637,
      "num_input_tokens_seen": 1269568,
      "step": 4535
    },
    {
      "epoch": 50.44444444444444,
      "grad_norm": 0.38289138674736023,
      "learning_rate": 4.8428169038929526e-05,
      "loss": 0.1285,
      "num_input_tokens_seen": 1270896,
      "step": 4540
    },
    {
      "epoch": 50.5,
      "grad_norm": 0.2765878736972809,
      "learning_rate": 4.8424741039506575e-05,
      "loss": 0.0634,
      "num_input_tokens_seen": 1272304,
      "step": 4545
    },
    {
      "epoch": 50.55555555555556,
      "grad_norm": 0.5004063844680786,
      "learning_rate": 4.842130942769419e-05,
      "loss": 0.0631,
      "num_input_tokens_seen": 1273712,
      "step": 4550
    },
    {
      "epoch": 50.611111111111114,
      "grad_norm": 0.9484299421310425,
      "learning_rate": 4.841787420402156e-05,
      "loss": 0.1117,
      "num_input_tokens_seen": 1275120,
      "step": 4555
    },
    {
      "epoch": 50.666666666666664,
      "grad_norm": 0.06372679024934769,
      "learning_rate": 4.841443536901844e-05,
      "loss": 0.0695,
      "num_input_tokens_seen": 1276528,
      "step": 4560
    },
    {
      "epoch": 50.72222222222222,
      "grad_norm": 2.794954776763916,
      "learning_rate": 4.841099292321514e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 1277952,
      "step": 4565
    },
    {
      "epoch": 50.77777777777778,
      "grad_norm": 0.32044273614883423,
      "learning_rate": 4.8407546867142525e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 1279312,
      "step": 4570
    },
    {
      "epoch": 50.833333333333336,
      "grad_norm": 2.1522367000579834,
      "learning_rate": 4.840409720133203e-05,
      "loss": 0.051,
      "num_input_tokens_seen": 1280672,
      "step": 4575
    },
    {
      "epoch": 50.888888888888886,
      "grad_norm": 2.210143566131592,
      "learning_rate": 4.8400643926315634e-05,
      "loss": 0.1551,
      "num_input_tokens_seen": 1282096,
      "step": 4580
    },
    {
      "epoch": 50.94444444444444,
      "grad_norm": 1.405190110206604,
      "learning_rate": 4.839718704262587e-05,
      "loss": 0.0707,
      "num_input_tokens_seen": 1283520,
      "step": 4585
    },
    {
      "epoch": 51.0,
      "grad_norm": 3.3873836994171143,
      "learning_rate": 4.839372655079585e-05,
      "loss": 0.109,
      "num_input_tokens_seen": 1284912,
      "step": 4590
    },
    {
      "epoch": 51.05555555555556,
      "grad_norm": 1.0904093980789185,
      "learning_rate": 4.83902624513592e-05,
      "loss": 0.1274,
      "num_input_tokens_seen": 1286320,
      "step": 4595
    },
    {
      "epoch": 51.111111111111114,
      "grad_norm": 1.6669563055038452,
      "learning_rate": 4.838679474485014e-05,
      "loss": 0.0872,
      "num_input_tokens_seen": 1287744,
      "step": 4600
    },
    {
      "epoch": 51.111111111111114,
      "eval_loss": 0.21628537774085999,
      "eval_runtime": 1.2385,
      "eval_samples_per_second": 32.296,
      "eval_steps_per_second": 16.148,
      "num_input_tokens_seen": 1287744,
      "step": 4600
    },
    {
      "epoch": 51.166666666666664,
      "grad_norm": 2.7854740619659424,
      "learning_rate": 4.838332343180343e-05,
      "loss": 0.1399,
      "num_input_tokens_seen": 1289136,
      "step": 4605
    },
    {
      "epoch": 51.22222222222222,
      "grad_norm": 0.9639542698860168,
      "learning_rate": 4.83798485127544e-05,
      "loss": 0.1226,
      "num_input_tokens_seen": 1290544,
      "step": 4610
    },
    {
      "epoch": 51.27777777777778,
      "grad_norm": 1.8601151704788208,
      "learning_rate": 4.837636998823892e-05,
      "loss": 0.097,
      "num_input_tokens_seen": 1291920,
      "step": 4615
    },
    {
      "epoch": 51.333333333333336,
      "grad_norm": 4.659699440002441,
      "learning_rate": 4.8372887858793414e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 1293248,
      "step": 4620
    },
    {
      "epoch": 51.388888888888886,
      "grad_norm": 1.71848726272583,
      "learning_rate": 4.836940212495489e-05,
      "loss": 0.0833,
      "num_input_tokens_seen": 1294688,
      "step": 4625
    },
    {
      "epoch": 51.44444444444444,
      "grad_norm": 1.737235188484192,
      "learning_rate": 4.836591278726087e-05,
      "loss": 0.0704,
      "num_input_tokens_seen": 1296128,
      "step": 4630
    },
    {
      "epoch": 51.5,
      "grad_norm": 2.395540952682495,
      "learning_rate": 4.836241984624947e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 1297616,
      "step": 4635
    },
    {
      "epoch": 51.55555555555556,
      "grad_norm": 0.8722602128982544,
      "learning_rate": 4.8358923302459336e-05,
      "loss": 0.0752,
      "num_input_tokens_seen": 1298992,
      "step": 4640
    },
    {
      "epoch": 51.611111111111114,
      "grad_norm": 2.2494757175445557,
      "learning_rate": 4.835542315642968e-05,
      "loss": 0.0487,
      "num_input_tokens_seen": 1300336,
      "step": 4645
    },
    {
      "epoch": 51.666666666666664,
      "grad_norm": 1.2254189252853394,
      "learning_rate": 4.8351919408700274e-05,
      "loss": 0.0988,
      "num_input_tokens_seen": 1301728,
      "step": 4650
    },
    {
      "epoch": 51.72222222222222,
      "grad_norm": 0.45256033539772034,
      "learning_rate": 4.834841205981144e-05,
      "loss": 0.0444,
      "num_input_tokens_seen": 1303120,
      "step": 4655
    },
    {
      "epoch": 51.77777777777778,
      "grad_norm": 0.08028332144021988,
      "learning_rate": 4.8344901110304054e-05,
      "loss": 0.0824,
      "num_input_tokens_seen": 1304544,
      "step": 4660
    },
    {
      "epoch": 51.833333333333336,
      "grad_norm": 1.664504051208496,
      "learning_rate": 4.8341386560719534e-05,
      "loss": 0.0586,
      "num_input_tokens_seen": 1305920,
      "step": 4665
    },
    {
      "epoch": 51.888888888888886,
      "grad_norm": 2.447200298309326,
      "learning_rate": 4.833786841159989e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 1307312,
      "step": 4670
    },
    {
      "epoch": 51.94444444444444,
      "grad_norm": 1.8942556381225586,
      "learning_rate": 4.833434666348765e-05,
      "loss": 0.0956,
      "num_input_tokens_seen": 1308752,
      "step": 4675
    },
    {
      "epoch": 52.0,
      "grad_norm": 3.056776762008667,
      "learning_rate": 4.833082131692592e-05,
      "loss": 0.104,
      "num_input_tokens_seen": 1310128,
      "step": 4680
    },
    {
      "epoch": 52.05555555555556,
      "grad_norm": 0.9263203740119934,
      "learning_rate": 4.832729237245835e-05,
      "loss": 0.0879,
      "num_input_tokens_seen": 1311488,
      "step": 4685
    },
    {
      "epoch": 52.111111111111114,
      "grad_norm": 1.5393450260162354,
      "learning_rate": 4.8323759830629145e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 1312896,
      "step": 4690
    },
    {
      "epoch": 52.166666666666664,
      "grad_norm": 0.7428280711174011,
      "learning_rate": 4.8320223691983066e-05,
      "loss": 0.0602,
      "num_input_tokens_seen": 1314288,
      "step": 4695
    },
    {
      "epoch": 52.22222222222222,
      "grad_norm": 2.076357841491699,
      "learning_rate": 4.831668395706544e-05,
      "loss": 0.1791,
      "num_input_tokens_seen": 1315696,
      "step": 4700
    },
    {
      "epoch": 52.27777777777778,
      "grad_norm": 4.371750831604004,
      "learning_rate": 4.8313140626422125e-05,
      "loss": 0.115,
      "num_input_tokens_seen": 1317120,
      "step": 4705
    },
    {
      "epoch": 52.333333333333336,
      "grad_norm": 1.6633611917495728,
      "learning_rate": 4.830959370059956e-05,
      "loss": 0.1004,
      "num_input_tokens_seen": 1318528,
      "step": 4710
    },
    {
      "epoch": 52.388888888888886,
      "grad_norm": 1.1835992336273193,
      "learning_rate": 4.830604318014472e-05,
      "loss": 0.0627,
      "num_input_tokens_seen": 1319904,
      "step": 4715
    },
    {
      "epoch": 52.44444444444444,
      "grad_norm": 2.464583158493042,
      "learning_rate": 4.830248906560514e-05,
      "loss": 0.0808,
      "num_input_tokens_seen": 1321312,
      "step": 4720
    },
    {
      "epoch": 52.5,
      "grad_norm": 2.7314887046813965,
      "learning_rate": 4.829893135752891e-05,
      "loss": 0.1018,
      "num_input_tokens_seen": 1322784,
      "step": 4725
    },
    {
      "epoch": 52.55555555555556,
      "grad_norm": 0.8979439735412598,
      "learning_rate": 4.829537005646466e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 1324160,
      "step": 4730
    },
    {
      "epoch": 52.611111111111114,
      "grad_norm": 0.2769988477230072,
      "learning_rate": 4.8291805162961615e-05,
      "loss": 0.0646,
      "num_input_tokens_seen": 1325536,
      "step": 4735
    },
    {
      "epoch": 52.666666666666664,
      "grad_norm": 2.2416324615478516,
      "learning_rate": 4.82882366775695e-05,
      "loss": 0.1209,
      "num_input_tokens_seen": 1326880,
      "step": 4740
    },
    {
      "epoch": 52.72222222222222,
      "grad_norm": 3.6543211936950684,
      "learning_rate": 4.828466460083864e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 1328256,
      "step": 4745
    },
    {
      "epoch": 52.77777777777778,
      "grad_norm": 3.546560764312744,
      "learning_rate": 4.8281088933319877e-05,
      "loss": 0.0928,
      "num_input_tokens_seen": 1329648,
      "step": 4750
    },
    {
      "epoch": 52.833333333333336,
      "grad_norm": 2.1270530223846436,
      "learning_rate": 4.827750967556464e-05,
      "loss": 0.1029,
      "num_input_tokens_seen": 1331056,
      "step": 4755
    },
    {
      "epoch": 52.888888888888886,
      "grad_norm": 2.1031670570373535,
      "learning_rate": 4.827392682812488e-05,
      "loss": 0.0566,
      "num_input_tokens_seen": 1332464,
      "step": 4760
    },
    {
      "epoch": 52.94444444444444,
      "grad_norm": 0.15261203050613403,
      "learning_rate": 4.827034039155312e-05,
      "loss": 0.0553,
      "num_input_tokens_seen": 1333856,
      "step": 4765
    },
    {
      "epoch": 53.0,
      "grad_norm": 1.6633940935134888,
      "learning_rate": 4.8266750366402445e-05,
      "loss": 0.1752,
      "num_input_tokens_seen": 1335264,
      "step": 4770
    },
    {
      "epoch": 53.05555555555556,
      "grad_norm": 0.7385790348052979,
      "learning_rate": 4.8263156753226476e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 1336672,
      "step": 4775
    },
    {
      "epoch": 53.111111111111114,
      "grad_norm": 0.25429925322532654,
      "learning_rate": 4.8259559552579394e-05,
      "loss": 0.0667,
      "num_input_tokens_seen": 1338048,
      "step": 4780
    },
    {
      "epoch": 53.166666666666664,
      "grad_norm": 2.177635431289673,
      "learning_rate": 4.825595876501593e-05,
      "loss": 0.1059,
      "num_input_tokens_seen": 1339488,
      "step": 4785
    },
    {
      "epoch": 53.22222222222222,
      "grad_norm": 0.4570292830467224,
      "learning_rate": 4.825235439109137e-05,
      "loss": 0.0619,
      "num_input_tokens_seen": 1340912,
      "step": 4790
    },
    {
      "epoch": 53.27777777777778,
      "grad_norm": 2.9917218685150146,
      "learning_rate": 4.824874643136156e-05,
      "loss": 0.0728,
      "num_input_tokens_seen": 1342352,
      "step": 4795
    },
    {
      "epoch": 53.333333333333336,
      "grad_norm": 3.1133370399475098,
      "learning_rate": 4.824513488638288e-05,
      "loss": 0.0962,
      "num_input_tokens_seen": 1343760,
      "step": 4800
    },
    {
      "epoch": 53.333333333333336,
      "eval_loss": 0.21256950497627258,
      "eval_runtime": 1.2461,
      "eval_samples_per_second": 32.1,
      "eval_steps_per_second": 16.05,
      "num_input_tokens_seen": 1343760,
      "step": 4800
    },
    {
      "epoch": 53.388888888888886,
      "grad_norm": 2.122506618499756,
      "learning_rate": 4.8241519756712293e-05,
      "loss": 0.0626,
      "num_input_tokens_seen": 1345152,
      "step": 4805
    },
    {
      "epoch": 53.44444444444444,
      "grad_norm": 2.2062466144561768,
      "learning_rate": 4.8237901042907285e-05,
      "loss": 0.0649,
      "num_input_tokens_seen": 1346576,
      "step": 4810
    },
    {
      "epoch": 53.5,
      "grad_norm": 0.40410542488098145,
      "learning_rate": 4.823427874552591e-05,
      "loss": 0.0623,
      "num_input_tokens_seen": 1348016,
      "step": 4815
    },
    {
      "epoch": 53.55555555555556,
      "grad_norm": 0.912159264087677,
      "learning_rate": 4.823065286512677e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 1349408,
      "step": 4820
    },
    {
      "epoch": 53.611111111111114,
      "grad_norm": 1.0167083740234375,
      "learning_rate": 4.8227023402269025e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 1350816,
      "step": 4825
    },
    {
      "epoch": 53.666666666666664,
      "grad_norm": 1.0322893857955933,
      "learning_rate": 4.822339035751239e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 1352176,
      "step": 4830
    },
    {
      "epoch": 53.72222222222222,
      "grad_norm": 0.6603935956954956,
      "learning_rate": 4.8219753731417104e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 1353552,
      "step": 4835
    },
    {
      "epoch": 53.77777777777778,
      "grad_norm": 0.15842407941818237,
      "learning_rate": 4.821611352454401e-05,
      "loss": 0.0754,
      "num_input_tokens_seen": 1354928,
      "step": 4840
    },
    {
      "epoch": 53.833333333333336,
      "grad_norm": 1.325744867324829,
      "learning_rate": 4.8212469737454444e-05,
      "loss": 0.0885,
      "num_input_tokens_seen": 1356336,
      "step": 4845
    },
    {
      "epoch": 53.888888888888886,
      "grad_norm": 1.0145584344863892,
      "learning_rate": 4.820882237071035e-05,
      "loss": 0.109,
      "num_input_tokens_seen": 1357776,
      "step": 4850
    },
    {
      "epoch": 53.94444444444444,
      "grad_norm": 0.32572293281555176,
      "learning_rate": 4.820517142487417e-05,
      "loss": 0.1008,
      "num_input_tokens_seen": 1359184,
      "step": 4855
    },
    {
      "epoch": 54.0,
      "grad_norm": 1.468177318572998,
      "learning_rate": 4.8201516900508956e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 1360592,
      "step": 4860
    },
    {
      "epoch": 54.05555555555556,
      "grad_norm": 0.470329225063324,
      "learning_rate": 4.819785879817827e-05,
      "loss": 0.0796,
      "num_input_tokens_seen": 1361936,
      "step": 4865
    },
    {
      "epoch": 54.111111111111114,
      "grad_norm": 2.946504592895508,
      "learning_rate": 4.8194197118446226e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 1363360,
      "step": 4870
    },
    {
      "epoch": 54.166666666666664,
      "grad_norm": 2.6573643684387207,
      "learning_rate": 4.819053186187752e-05,
      "loss": 0.0623,
      "num_input_tokens_seen": 1364784,
      "step": 4875
    },
    {
      "epoch": 54.22222222222222,
      "grad_norm": 1.4528570175170898,
      "learning_rate": 4.818686302903736e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 1366176,
      "step": 4880
    },
    {
      "epoch": 54.27777777777778,
      "grad_norm": 1.1070032119750977,
      "learning_rate": 4.818319062049154e-05,
      "loss": 0.0471,
      "num_input_tokens_seen": 1367568,
      "step": 4885
    },
    {
      "epoch": 54.333333333333336,
      "grad_norm": 1.5042496919631958,
      "learning_rate": 4.817951463680639e-05,
      "loss": 0.0519,
      "num_input_tokens_seen": 1368944,
      "step": 4890
    },
    {
      "epoch": 54.388888888888886,
      "grad_norm": 1.5779855251312256,
      "learning_rate": 4.817583507854879e-05,
      "loss": 0.1686,
      "num_input_tokens_seen": 1370352,
      "step": 4895
    },
    {
      "epoch": 54.44444444444444,
      "grad_norm": 2.2517757415771484,
      "learning_rate": 4.817215194628617e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 1371760,
      "step": 4900
    },
    {
      "epoch": 54.5,
      "grad_norm": 3.4146130084991455,
      "learning_rate": 4.816846524058653e-05,
      "loss": 0.1166,
      "num_input_tokens_seen": 1373168,
      "step": 4905
    },
    {
      "epoch": 54.55555555555556,
      "grad_norm": 0.9532632827758789,
      "learning_rate": 4.816477496201839e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 1374544,
      "step": 4910
    },
    {
      "epoch": 54.611111111111114,
      "grad_norm": 1.6843609809875488,
      "learning_rate": 4.8161081111150845e-05,
      "loss": 0.069,
      "num_input_tokens_seen": 1375936,
      "step": 4915
    },
    {
      "epoch": 54.666666666666664,
      "grad_norm": 0.506683349609375,
      "learning_rate": 4.815738368855354e-05,
      "loss": 0.0911,
      "num_input_tokens_seen": 1377344,
      "step": 4920
    },
    {
      "epoch": 54.72222222222222,
      "grad_norm": 0.9220160841941833,
      "learning_rate": 4.815368269479664e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 1378752,
      "step": 4925
    },
    {
      "epoch": 54.77777777777778,
      "grad_norm": 1.7386606931686401,
      "learning_rate": 4.814997813045092e-05,
      "loss": 0.0683,
      "num_input_tokens_seen": 1380112,
      "step": 4930
    },
    {
      "epoch": 54.833333333333336,
      "grad_norm": 1.086064338684082,
      "learning_rate": 4.814626999608764e-05,
      "loss": 0.0459,
      "num_input_tokens_seen": 1381536,
      "step": 4935
    },
    {
      "epoch": 54.888888888888886,
      "grad_norm": 0.44000691175460815,
      "learning_rate": 4.814255829227865e-05,
      "loss": 0.0881,
      "num_input_tokens_seen": 1382944,
      "step": 4940
    },
    {
      "epoch": 54.94444444444444,
      "grad_norm": 3.311363697052002,
      "learning_rate": 4.813884301959635e-05,
      "loss": 0.077,
      "num_input_tokens_seen": 1384384,
      "step": 4945
    },
    {
      "epoch": 55.0,
      "grad_norm": 1.4069247245788574,
      "learning_rate": 4.813512417861368e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 1385760,
      "step": 4950
    },
    {
      "epoch": 55.05555555555556,
      "grad_norm": 0.48592787981033325,
      "learning_rate": 4.813140176990411e-05,
      "loss": 0.0779,
      "num_input_tokens_seen": 1387136,
      "step": 4955
    },
    {
      "epoch": 55.111111111111114,
      "grad_norm": 2.5375802516937256,
      "learning_rate": 4.8127675794041714e-05,
      "loss": 0.1157,
      "num_input_tokens_seen": 1388560,
      "step": 4960
    },
    {
      "epoch": 55.166666666666664,
      "grad_norm": 0.7719197869300842,
      "learning_rate": 4.812394625160107e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 1389952,
      "step": 4965
    },
    {
      "epoch": 55.22222222222222,
      "grad_norm": 0.6078088283538818,
      "learning_rate": 4.812021314315732e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 1391344,
      "step": 4970
    },
    {
      "epoch": 55.27777777777778,
      "grad_norm": 2.1539180278778076,
      "learning_rate": 4.811647646928616e-05,
      "loss": 0.0978,
      "num_input_tokens_seen": 1392752,
      "step": 4975
    },
    {
      "epoch": 55.333333333333336,
      "grad_norm": 1.623138666152954,
      "learning_rate": 4.8112736230563814e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 1394112,
      "step": 4980
    },
    {
      "epoch": 55.388888888888886,
      "grad_norm": 0.2597770094871521,
      "learning_rate": 4.81089924275671e-05,
      "loss": 0.0911,
      "num_input_tokens_seen": 1395520,
      "step": 4985
    },
    {
      "epoch": 55.44444444444444,
      "grad_norm": 3.3697521686553955,
      "learning_rate": 4.810524506087335e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 1396960,
      "step": 4990
    },
    {
      "epoch": 55.5,
      "grad_norm": 0.6103631854057312,
      "learning_rate": 4.810149413106044e-05,
      "loss": 0.0752,
      "num_input_tokens_seen": 1398416,
      "step": 4995
    },
    {
      "epoch": 55.55555555555556,
      "grad_norm": 0.717185378074646,
      "learning_rate": 4.809773963870684e-05,
      "loss": 0.0264,
      "num_input_tokens_seen": 1399856,
      "step": 5000
    },
    {
      "epoch": 55.55555555555556,
      "eval_loss": 0.23336979746818542,
      "eval_runtime": 1.2379,
      "eval_samples_per_second": 32.312,
      "eval_steps_per_second": 16.156,
      "num_input_tokens_seen": 1399856,
      "step": 5000
    },
    {
      "epoch": 55.611111111111114,
      "grad_norm": 3.658777952194214,
      "learning_rate": 4.809398158439151e-05,
      "loss": 0.1009,
      "num_input_tokens_seen": 1401264,
      "step": 5005
    },
    {
      "epoch": 55.666666666666664,
      "grad_norm": 0.41616150736808777,
      "learning_rate": 4.8090219968694005e-05,
      "loss": 0.0662,
      "num_input_tokens_seen": 1402656,
      "step": 5010
    },
    {
      "epoch": 55.72222222222222,
      "grad_norm": 0.7051559686660767,
      "learning_rate": 4.808645479219442e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 1404032,
      "step": 5015
    },
    {
      "epoch": 55.77777777777778,
      "grad_norm": 0.818858802318573,
      "learning_rate": 4.8082686055473375e-05,
      "loss": 0.0805,
      "num_input_tokens_seen": 1405424,
      "step": 5020
    },
    {
      "epoch": 55.833333333333336,
      "grad_norm": 1.8110592365264893,
      "learning_rate": 4.8078913759112066e-05,
      "loss": 0.06,
      "num_input_tokens_seen": 1406800,
      "step": 5025
    },
    {
      "epoch": 55.888888888888886,
      "grad_norm": 1.3464938402175903,
      "learning_rate": 4.807513790369223e-05,
      "loss": 0.0942,
      "num_input_tokens_seen": 1408192,
      "step": 5030
    },
    {
      "epoch": 55.94444444444444,
      "grad_norm": 0.7881632447242737,
      "learning_rate": 4.8071358489796145e-05,
      "loss": 0.055,
      "num_input_tokens_seen": 1409552,
      "step": 5035
    },
    {
      "epoch": 56.0,
      "grad_norm": 4.244215965270996,
      "learning_rate": 4.806757551800665e-05,
      "loss": 0.1148,
      "num_input_tokens_seen": 1410992,
      "step": 5040
    },
    {
      "epoch": 56.05555555555556,
      "grad_norm": 0.06587128341197968,
      "learning_rate": 4.806378898890713e-05,
      "loss": 0.0956,
      "num_input_tokens_seen": 1412400,
      "step": 5045
    },
    {
      "epoch": 56.111111111111114,
      "grad_norm": 0.9889267683029175,
      "learning_rate": 4.80599989030815e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 1413792,
      "step": 5050
    },
    {
      "epoch": 56.166666666666664,
      "grad_norm": 1.8771108388900757,
      "learning_rate": 4.805620526111426e-05,
      "loss": 0.0687,
      "num_input_tokens_seen": 1415200,
      "step": 5055
    },
    {
      "epoch": 56.22222222222222,
      "grad_norm": 1.724740743637085,
      "learning_rate": 4.805240806359042e-05,
      "loss": 0.1163,
      "num_input_tokens_seen": 1416608,
      "step": 5060
    },
    {
      "epoch": 56.27777777777778,
      "grad_norm": 0.5050414204597473,
      "learning_rate": 4.804860731109557e-05,
      "loss": 0.0861,
      "num_input_tokens_seen": 1418000,
      "step": 5065
    },
    {
      "epoch": 56.333333333333336,
      "grad_norm": 0.3606221377849579,
      "learning_rate": 4.804480300421581e-05,
      "loss": 0.0679,
      "num_input_tokens_seen": 1419376,
      "step": 5070
    },
    {
      "epoch": 56.388888888888886,
      "grad_norm": 0.08545849472284317,
      "learning_rate": 4.804099514353784e-05,
      "loss": 0.0848,
      "num_input_tokens_seen": 1420752,
      "step": 5075
    },
    {
      "epoch": 56.44444444444444,
      "grad_norm": 2.568150758743286,
      "learning_rate": 4.8037183729648867e-05,
      "loss": 0.0852,
      "num_input_tokens_seen": 1422112,
      "step": 5080
    },
    {
      "epoch": 56.5,
      "grad_norm": 1.7260783910751343,
      "learning_rate": 4.803336876313666e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 1423536,
      "step": 5085
    },
    {
      "epoch": 56.55555555555556,
      "grad_norm": 3.6505022048950195,
      "learning_rate": 4.802955024458953e-05,
      "loss": 0.0773,
      "num_input_tokens_seen": 1424944,
      "step": 5090
    },
    {
      "epoch": 56.611111111111114,
      "grad_norm": 3.4228479862213135,
      "learning_rate": 4.802572817459634e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 1426416,
      "step": 5095
    },
    {
      "epoch": 56.666666666666664,
      "grad_norm": 0.2931848466396332,
      "learning_rate": 4.802190255374651e-05,
      "loss": 0.0833,
      "num_input_tokens_seen": 1427856,
      "step": 5100
    },
    {
      "epoch": 56.72222222222222,
      "grad_norm": 1.5868149995803833,
      "learning_rate": 4.801807338263e-05,
      "loss": 0.0918,
      "num_input_tokens_seen": 1429168,
      "step": 5105
    },
    {
      "epoch": 56.77777777777778,
      "grad_norm": 0.5611855387687683,
      "learning_rate": 4.8014240661837306e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 1430576,
      "step": 5110
    },
    {
      "epoch": 56.833333333333336,
      "grad_norm": 0.7504281997680664,
      "learning_rate": 4.80104043919595e-05,
      "loss": 0.0841,
      "num_input_tokens_seen": 1431952,
      "step": 5115
    },
    {
      "epoch": 56.888888888888886,
      "grad_norm": 0.2520594000816345,
      "learning_rate": 4.800656457358815e-05,
      "loss": 0.095,
      "num_input_tokens_seen": 1433344,
      "step": 5120
    },
    {
      "epoch": 56.94444444444444,
      "grad_norm": 0.8620799779891968,
      "learning_rate": 4.800272120731544e-05,
      "loss": 0.0562,
      "num_input_tokens_seen": 1434752,
      "step": 5125
    },
    {
      "epoch": 57.0,
      "grad_norm": 2.683835983276367,
      "learning_rate": 4.799887429373404e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 1436160,
      "step": 5130
    },
    {
      "epoch": 57.05555555555556,
      "grad_norm": 1.799622654914856,
      "learning_rate": 4.79950238334372e-05,
      "loss": 0.1206,
      "num_input_tokens_seen": 1437600,
      "step": 5135
    },
    {
      "epoch": 57.111111111111114,
      "grad_norm": 0.996192216873169,
      "learning_rate": 4.799116982701872e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 1439024,
      "step": 5140
    },
    {
      "epoch": 57.166666666666664,
      "grad_norm": 2.5836024284362793,
      "learning_rate": 4.7987312275072926e-05,
      "loss": 0.1023,
      "num_input_tokens_seen": 1440432,
      "step": 5145
    },
    {
      "epoch": 57.22222222222222,
      "grad_norm": 0.4656190276145935,
      "learning_rate": 4.79834511781947e-05,
      "loss": 0.0154,
      "num_input_tokens_seen": 1441824,
      "step": 5150
    },
    {
      "epoch": 57.27777777777778,
      "grad_norm": 1.2707066535949707,
      "learning_rate": 4.797958653697947e-05,
      "loss": 0.0799,
      "num_input_tokens_seen": 1443200,
      "step": 5155
    },
    {
      "epoch": 57.333333333333336,
      "grad_norm": 1.6682170629501343,
      "learning_rate": 4.7975718352023225e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 1444592,
      "step": 5160
    },
    {
      "epoch": 57.388888888888886,
      "grad_norm": 3.74247145652771,
      "learning_rate": 4.7971846623922476e-05,
      "loss": 0.1129,
      "num_input_tokens_seen": 1446016,
      "step": 5165
    },
    {
      "epoch": 57.44444444444444,
      "grad_norm": 1.0405728816986084,
      "learning_rate": 4.7967971353274294e-05,
      "loss": 0.0533,
      "num_input_tokens_seen": 1447408,
      "step": 5170
    },
    {
      "epoch": 57.5,
      "grad_norm": 1.430204153060913,
      "learning_rate": 4.79640925406763e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 1448800,
      "step": 5175
    },
    {
      "epoch": 57.55555555555556,
      "grad_norm": 2.114457130432129,
      "learning_rate": 4.796021018672664e-05,
      "loss": 0.0713,
      "num_input_tokens_seen": 1450176,
      "step": 5180
    },
    {
      "epoch": 57.611111111111114,
      "grad_norm": 0.994190514087677,
      "learning_rate": 4.795632429202405e-05,
      "loss": 0.0518,
      "num_input_tokens_seen": 1451568,
      "step": 5185
    },
    {
      "epoch": 57.666666666666664,
      "grad_norm": 1.7087445259094238,
      "learning_rate": 4.795243485716775e-05,
      "loss": 0.1054,
      "num_input_tokens_seen": 1452992,
      "step": 5190
    },
    {
      "epoch": 57.72222222222222,
      "grad_norm": 2.401366710662842,
      "learning_rate": 4.794854188275757e-05,
      "loss": 0.0997,
      "num_input_tokens_seen": 1454416,
      "step": 5195
    },
    {
      "epoch": 57.77777777777778,
      "grad_norm": 0.4840102791786194,
      "learning_rate": 4.794464536939384e-05,
      "loss": 0.0622,
      "num_input_tokens_seen": 1455808,
      "step": 5200
    },
    {
      "epoch": 57.77777777777778,
      "eval_loss": 0.23133797943592072,
      "eval_runtime": 1.2388,
      "eval_samples_per_second": 32.288,
      "eval_steps_per_second": 16.144,
      "num_input_tokens_seen": 1455808,
      "step": 5200
    },
    {
      "epoch": 57.833333333333336,
      "grad_norm": 4.6461286544799805,
      "learning_rate": 4.794074531767745e-05,
      "loss": 0.1003,
      "num_input_tokens_seen": 1457200,
      "step": 5205
    },
    {
      "epoch": 57.888888888888886,
      "grad_norm": 1.9763147830963135,
      "learning_rate": 4.7936841728209834e-05,
      "loss": 0.0927,
      "num_input_tokens_seen": 1458560,
      "step": 5210
    },
    {
      "epoch": 57.94444444444444,
      "grad_norm": 1.883452296257019,
      "learning_rate": 4.7932934601593e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 1460000,
      "step": 5215
    },
    {
      "epoch": 58.0,
      "grad_norm": 4.005064010620117,
      "learning_rate": 4.792902393842943e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 1461440,
      "step": 5220
    },
    {
      "epoch": 58.05555555555556,
      "grad_norm": 2.373411178588867,
      "learning_rate": 4.792510973932225e-05,
      "loss": 0.0847,
      "num_input_tokens_seen": 1462800,
      "step": 5225
    },
    {
      "epoch": 58.111111111111114,
      "grad_norm": 0.74868243932724,
      "learning_rate": 4.7921192004875036e-05,
      "loss": 0.074,
      "num_input_tokens_seen": 1464192,
      "step": 5230
    },
    {
      "epoch": 58.166666666666664,
      "grad_norm": 0.9059299230575562,
      "learning_rate": 4.791727073569198e-05,
      "loss": 0.0266,
      "num_input_tokens_seen": 1465568,
      "step": 5235
    },
    {
      "epoch": 58.22222222222222,
      "grad_norm": 2.0080385208129883,
      "learning_rate": 4.7913345932377775e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 1466976,
      "step": 5240
    },
    {
      "epoch": 58.27777777777778,
      "grad_norm": 1.8157627582550049,
      "learning_rate": 4.790941759553769e-05,
      "loss": 0.0853,
      "num_input_tokens_seen": 1468368,
      "step": 5245
    },
    {
      "epoch": 58.333333333333336,
      "grad_norm": 1.5004974603652954,
      "learning_rate": 4.79054857257775e-05,
      "loss": 0.0942,
      "num_input_tokens_seen": 1469760,
      "step": 5250
    },
    {
      "epoch": 58.388888888888886,
      "grad_norm": 0.04137592390179634,
      "learning_rate": 4.790155032370357e-05,
      "loss": 0.0318,
      "num_input_tokens_seen": 1471152,
      "step": 5255
    },
    {
      "epoch": 58.44444444444444,
      "grad_norm": 1.5195258855819702,
      "learning_rate": 4.789761138992278e-05,
      "loss": 0.0642,
      "num_input_tokens_seen": 1472608,
      "step": 5260
    },
    {
      "epoch": 58.5,
      "grad_norm": 1.9193085432052612,
      "learning_rate": 4.7893668925042565e-05,
      "loss": 0.1014,
      "num_input_tokens_seen": 1474016,
      "step": 5265
    },
    {
      "epoch": 58.55555555555556,
      "grad_norm": 5.04272985458374,
      "learning_rate": 4.78897229296709e-05,
      "loss": 0.0926,
      "num_input_tokens_seen": 1475424,
      "step": 5270
    },
    {
      "epoch": 58.611111111111114,
      "grad_norm": 2.997624635696411,
      "learning_rate": 4.7885773404416315e-05,
      "loss": 0.0865,
      "num_input_tokens_seen": 1476832,
      "step": 5275
    },
    {
      "epoch": 58.666666666666664,
      "grad_norm": 0.9231474995613098,
      "learning_rate": 4.788182034988786e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 1478240,
      "step": 5280
    },
    {
      "epoch": 58.72222222222222,
      "grad_norm": 1.668399691581726,
      "learning_rate": 4.787786376669516e-05,
      "loss": 0.0935,
      "num_input_tokens_seen": 1479648,
      "step": 5285
    },
    {
      "epoch": 58.77777777777778,
      "grad_norm": 2.8627092838287354,
      "learning_rate": 4.787390365544837e-05,
      "loss": 0.102,
      "num_input_tokens_seen": 1481088,
      "step": 5290
    },
    {
      "epoch": 58.833333333333336,
      "grad_norm": 3.2807352542877197,
      "learning_rate": 4.786994001675818e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 1482512,
      "step": 5295
    },
    {
      "epoch": 58.888888888888886,
      "grad_norm": 1.880031943321228,
      "learning_rate": 4.786597285123584e-05,
      "loss": 0.0981,
      "num_input_tokens_seen": 1483888,
      "step": 5300
    },
    {
      "epoch": 58.94444444444444,
      "grad_norm": 1.3082422018051147,
      "learning_rate": 4.7862002159493135e-05,
      "loss": 0.1124,
      "num_input_tokens_seen": 1485280,
      "step": 5305
    },
    {
      "epoch": 59.0,
      "grad_norm": 1.9729368686676025,
      "learning_rate": 4.785802794214239e-05,
      "loss": 0.0587,
      "num_input_tokens_seen": 1486688,
      "step": 5310
    },
    {
      "epoch": 59.05555555555556,
      "grad_norm": 1.103089690208435,
      "learning_rate": 4.7854050199796495e-05,
      "loss": 0.0752,
      "num_input_tokens_seen": 1488080,
      "step": 5315
    },
    {
      "epoch": 59.111111111111114,
      "grad_norm": 2.6735739707946777,
      "learning_rate": 4.7850068933068845e-05,
      "loss": 0.0641,
      "num_input_tokens_seen": 1489472,
      "step": 5320
    },
    {
      "epoch": 59.166666666666664,
      "grad_norm": 1.312289834022522,
      "learning_rate": 4.7846084142573425e-05,
      "loss": 0.0733,
      "num_input_tokens_seen": 1490880,
      "step": 5325
    },
    {
      "epoch": 59.22222222222222,
      "grad_norm": 2.2388319969177246,
      "learning_rate": 4.7842095828924725e-05,
      "loss": 0.0403,
      "num_input_tokens_seen": 1492240,
      "step": 5330
    },
    {
      "epoch": 59.27777777777778,
      "grad_norm": 2.937765121459961,
      "learning_rate": 4.783810399273779e-05,
      "loss": 0.0584,
      "num_input_tokens_seen": 1493664,
      "step": 5335
    },
    {
      "epoch": 59.333333333333336,
      "grad_norm": 1.1288886070251465,
      "learning_rate": 4.7834108634628226e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 1495072,
      "step": 5340
    },
    {
      "epoch": 59.388888888888886,
      "grad_norm": 1.9531636238098145,
      "learning_rate": 4.783010975521216e-05,
      "loss": 0.1031,
      "num_input_tokens_seen": 1496512,
      "step": 5345
    },
    {
      "epoch": 59.44444444444444,
      "grad_norm": 0.33826687932014465,
      "learning_rate": 4.782610735510626e-05,
      "loss": 0.0872,
      "num_input_tokens_seen": 1497904,
      "step": 5350
    },
    {
      "epoch": 59.5,
      "grad_norm": 0.318584680557251,
      "learning_rate": 4.782210143492776e-05,
      "loss": 0.0753,
      "num_input_tokens_seen": 1499280,
      "step": 5355
    },
    {
      "epoch": 59.55555555555556,
      "grad_norm": 3.6429808139801025,
      "learning_rate": 4.781809199529442e-05,
      "loss": 0.1121,
      "num_input_tokens_seen": 1500688,
      "step": 5360
    },
    {
      "epoch": 59.611111111111114,
      "grad_norm": 1.0353327989578247,
      "learning_rate": 4.781407903682454e-05,
      "loss": 0.0554,
      "num_input_tokens_seen": 1502112,
      "step": 5365
    },
    {
      "epoch": 59.666666666666664,
      "grad_norm": 1.9566301107406616,
      "learning_rate": 4.781006256013698e-05,
      "loss": 0.1978,
      "num_input_tokens_seen": 1503520,
      "step": 5370
    },
    {
      "epoch": 59.72222222222222,
      "grad_norm": 2.101923704147339,
      "learning_rate": 4.7806042565851115e-05,
      "loss": 0.0611,
      "num_input_tokens_seen": 1504928,
      "step": 5375
    },
    {
      "epoch": 59.77777777777778,
      "grad_norm": 0.47708094120025635,
      "learning_rate": 4.7802019054586895e-05,
      "loss": 0.076,
      "num_input_tokens_seen": 1506288,
      "step": 5380
    },
    {
      "epoch": 59.833333333333336,
      "grad_norm": 0.2827202081680298,
      "learning_rate": 4.779799202696479e-05,
      "loss": 0.0662,
      "num_input_tokens_seen": 1507712,
      "step": 5385
    },
    {
      "epoch": 59.888888888888886,
      "grad_norm": 1.497269630432129,
      "learning_rate": 4.779396148360581e-05,
      "loss": 0.0674,
      "num_input_tokens_seen": 1509088,
      "step": 5390
    },
    {
      "epoch": 59.94444444444444,
      "grad_norm": 2.5426928997039795,
      "learning_rate": 4.7789927425131517e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 1510464,
      "step": 5395
    },
    {
      "epoch": 60.0,
      "grad_norm": 0.8169997930526733,
      "learning_rate": 4.778588985216403e-05,
      "loss": 0.0771,
      "num_input_tokens_seen": 1511856,
      "step": 5400
    },
    {
      "epoch": 60.0,
      "eval_loss": 0.24149923026561737,
      "eval_runtime": 1.242,
      "eval_samples_per_second": 32.206,
      "eval_steps_per_second": 16.103,
      "num_input_tokens_seen": 1511856,
      "step": 5400
    },
    {
      "epoch": 60.05555555555556,
      "grad_norm": 0.5576716661453247,
      "learning_rate": 4.778184876532598e-05,
      "loss": 0.0979,
      "num_input_tokens_seen": 1513280,
      "step": 5405
    },
    {
      "epoch": 60.111111111111114,
      "grad_norm": 0.9223631620407104,
      "learning_rate": 4.7777804165240556e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 1514656,
      "step": 5410
    },
    {
      "epoch": 60.166666666666664,
      "grad_norm": 1.408202886581421,
      "learning_rate": 4.7773756052531485e-05,
      "loss": 0.0548,
      "num_input_tokens_seen": 1516064,
      "step": 5415
    },
    {
      "epoch": 60.22222222222222,
      "grad_norm": 0.7158132195472717,
      "learning_rate": 4.7769704427823035e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 1517456,
      "step": 5420
    },
    {
      "epoch": 60.27777777777778,
      "grad_norm": 0.5603363513946533,
      "learning_rate": 4.776564929174003e-05,
      "loss": 0.0813,
      "num_input_tokens_seen": 1518832,
      "step": 5425
    },
    {
      "epoch": 60.333333333333336,
      "grad_norm": 0.5547673106193542,
      "learning_rate": 4.7761590644907806e-05,
      "loss": 0.0693,
      "num_input_tokens_seen": 1520256,
      "step": 5430
    },
    {
      "epoch": 60.388888888888886,
      "grad_norm": 2.6685616970062256,
      "learning_rate": 4.7757528487952263e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 1521632,
      "step": 5435
    },
    {
      "epoch": 60.44444444444444,
      "grad_norm": 0.8559563159942627,
      "learning_rate": 4.7753462821499836e-05,
      "loss": 0.106,
      "num_input_tokens_seen": 1523056,
      "step": 5440
    },
    {
      "epoch": 60.5,
      "grad_norm": 2.2488911151885986,
      "learning_rate": 4.774939364617751e-05,
      "loss": 0.0797,
      "num_input_tokens_seen": 1524480,
      "step": 5445
    },
    {
      "epoch": 60.55555555555556,
      "grad_norm": 2.7250349521636963,
      "learning_rate": 4.7745320962612795e-05,
      "loss": 0.1298,
      "num_input_tokens_seen": 1525920,
      "step": 5450
    },
    {
      "epoch": 60.611111111111114,
      "grad_norm": 0.3933481276035309,
      "learning_rate": 4.7741244771433756e-05,
      "loss": 0.0839,
      "num_input_tokens_seen": 1527344,
      "step": 5455
    },
    {
      "epoch": 60.666666666666664,
      "grad_norm": 2.359812021255493,
      "learning_rate": 4.7737165073268985e-05,
      "loss": 0.0886,
      "num_input_tokens_seen": 1528704,
      "step": 5460
    },
    {
      "epoch": 60.72222222222222,
      "grad_norm": 1.6341310739517212,
      "learning_rate": 4.7733081868747626e-05,
      "loss": 0.024,
      "num_input_tokens_seen": 1530080,
      "step": 5465
    },
    {
      "epoch": 60.77777777777778,
      "grad_norm": 0.3725566864013672,
      "learning_rate": 4.772899515849936e-05,
      "loss": 0.0866,
      "num_input_tokens_seen": 1531440,
      "step": 5470
    },
    {
      "epoch": 60.833333333333336,
      "grad_norm": 1.6151831150054932,
      "learning_rate": 4.7724904943154414e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 1532832,
      "step": 5475
    },
    {
      "epoch": 60.888888888888886,
      "grad_norm": 0.8606706261634827,
      "learning_rate": 4.772081122334354e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 1534224,
      "step": 5480
    },
    {
      "epoch": 60.94444444444444,
      "grad_norm": 2.5960581302642822,
      "learning_rate": 4.771671399969806e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 1535632,
      "step": 5485
    },
    {
      "epoch": 61.0,
      "grad_norm": 0.6793287396430969,
      "learning_rate": 4.7712613272849794e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 1537008,
      "step": 5490
    },
    {
      "epoch": 61.05555555555556,
      "grad_norm": 1.383703351020813,
      "learning_rate": 4.770850904343114e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 1538432,
      "step": 5495
    },
    {
      "epoch": 61.111111111111114,
      "grad_norm": 3.478306293487549,
      "learning_rate": 4.770440131207502e-05,
      "loss": 0.1335,
      "num_input_tokens_seen": 1539840,
      "step": 5500
    },
    {
      "epoch": 61.166666666666664,
      "grad_norm": 0.5319934487342834,
      "learning_rate": 4.7700290079414896e-05,
      "loss": 0.0881,
      "num_input_tokens_seen": 1541248,
      "step": 5505
    },
    {
      "epoch": 61.22222222222222,
      "grad_norm": 0.2913488745689392,
      "learning_rate": 4.769617534608477e-05,
      "loss": 0.0893,
      "num_input_tokens_seen": 1542688,
      "step": 5510
    },
    {
      "epoch": 61.27777777777778,
      "grad_norm": 3.138859987258911,
      "learning_rate": 4.7692057112719193e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 1544128,
      "step": 5515
    },
    {
      "epoch": 61.333333333333336,
      "grad_norm": 1.750166654586792,
      "learning_rate": 4.7687935379953234e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 1545504,
      "step": 5520
    },
    {
      "epoch": 61.388888888888886,
      "grad_norm": 1.0282386541366577,
      "learning_rate": 4.7683810148422534e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 1546864,
      "step": 5525
    },
    {
      "epoch": 61.44444444444444,
      "grad_norm": 0.09918778389692307,
      "learning_rate": 4.767968141876324e-05,
      "loss": 0.0885,
      "num_input_tokens_seen": 1548272,
      "step": 5530
    },
    {
      "epoch": 61.5,
      "grad_norm": 3.0488431453704834,
      "learning_rate": 4.767554919161207e-05,
      "loss": 0.0674,
      "num_input_tokens_seen": 1549648,
      "step": 5535
    },
    {
      "epoch": 61.55555555555556,
      "grad_norm": 1.163031816482544,
      "learning_rate": 4.767141346760624e-05,
      "loss": 0.1131,
      "num_input_tokens_seen": 1551072,
      "step": 5540
    },
    {
      "epoch": 61.611111111111114,
      "grad_norm": 3.3304030895233154,
      "learning_rate": 4.766727424738356e-05,
      "loss": 0.1133,
      "num_input_tokens_seen": 1552464,
      "step": 5545
    },
    {
      "epoch": 61.666666666666664,
      "grad_norm": 0.5153053402900696,
      "learning_rate": 4.7663131531582325e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 1553776,
      "step": 5550
    },
    {
      "epoch": 61.72222222222222,
      "grad_norm": 0.1625465750694275,
      "learning_rate": 4.765898532084142e-05,
      "loss": 0.0146,
      "num_input_tokens_seen": 1555216,
      "step": 5555
    },
    {
      "epoch": 61.77777777777778,
      "grad_norm": 0.16486051678657532,
      "learning_rate": 4.765483561580022e-05,
      "loss": 0.1123,
      "num_input_tokens_seen": 1556608,
      "step": 5560
    },
    {
      "epoch": 61.833333333333336,
      "grad_norm": 1.6355011463165283,
      "learning_rate": 4.7650682417098666e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 1558000,
      "step": 5565
    },
    {
      "epoch": 61.888888888888886,
      "grad_norm": 0.7324963212013245,
      "learning_rate": 4.7646525725377244e-05,
      "loss": 0.0511,
      "num_input_tokens_seen": 1559408,
      "step": 5570
    },
    {
      "epoch": 61.94444444444444,
      "grad_norm": 0.5986668467521667,
      "learning_rate": 4.764236554127696e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 1560832,
      "step": 5575
    },
    {
      "epoch": 62.0,
      "grad_norm": 0.1460205763578415,
      "learning_rate": 4.7638201865439356e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 1562240,
      "step": 5580
    },
    {
      "epoch": 62.05555555555556,
      "grad_norm": 2.317314863204956,
      "learning_rate": 4.7634034698506545e-05,
      "loss": 0.0586,
      "num_input_tokens_seen": 1563680,
      "step": 5585
    },
    {
      "epoch": 62.111111111111114,
      "grad_norm": 1.1826190948486328,
      "learning_rate": 4.762986404112115e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 1565056,
      "step": 5590
    },
    {
      "epoch": 62.166666666666664,
      "grad_norm": 0.027724245563149452,
      "learning_rate": 4.762568989392633e-05,
      "loss": 0.0715,
      "num_input_tokens_seen": 1566432,
      "step": 5595
    },
    {
      "epoch": 62.22222222222222,
      "grad_norm": 0.16200849413871765,
      "learning_rate": 4.76215122575658e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 1567808,
      "step": 5600
    },
    {
      "epoch": 62.22222222222222,
      "eval_loss": 0.25329703092575073,
      "eval_runtime": 1.2343,
      "eval_samples_per_second": 32.407,
      "eval_steps_per_second": 16.203,
      "num_input_tokens_seen": 1567808,
      "step": 5600
    },
    {
      "epoch": 62.27777777777778,
      "grad_norm": 1.826702356338501,
      "learning_rate": 4.7617331132683795e-05,
      "loss": 0.0622,
      "num_input_tokens_seen": 1569248,
      "step": 5605
    },
    {
      "epoch": 62.333333333333336,
      "grad_norm": 0.7499408721923828,
      "learning_rate": 4.7613146519925105e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 1570640,
      "step": 5610
    },
    {
      "epoch": 62.388888888888886,
      "grad_norm": 2.47220516204834,
      "learning_rate": 4.7608958419935045e-05,
      "loss": 0.0591,
      "num_input_tokens_seen": 1572032,
      "step": 5615
    },
    {
      "epoch": 62.44444444444444,
      "grad_norm": 3.786708116531372,
      "learning_rate": 4.760476683335948e-05,
      "loss": 0.0592,
      "num_input_tokens_seen": 1573456,
      "step": 5620
    },
    {
      "epoch": 62.5,
      "grad_norm": 0.8252387046813965,
      "learning_rate": 4.760057176084479e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 1574896,
      "step": 5625
    },
    {
      "epoch": 62.55555555555556,
      "grad_norm": 2.169797420501709,
      "learning_rate": 4.759637320303793e-05,
      "loss": 0.0637,
      "num_input_tokens_seen": 1576320,
      "step": 5630
    },
    {
      "epoch": 62.611111111111114,
      "grad_norm": 5.429933071136475,
      "learning_rate": 4.759217116058635e-05,
      "loss": 0.0927,
      "num_input_tokens_seen": 1577680,
      "step": 5635
    },
    {
      "epoch": 62.666666666666664,
      "grad_norm": 1.615835428237915,
      "learning_rate": 4.758796563413807e-05,
      "loss": 0.0722,
      "num_input_tokens_seen": 1579040,
      "step": 5640
    },
    {
      "epoch": 62.72222222222222,
      "grad_norm": 6.0945329666137695,
      "learning_rate": 4.758375662434163e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 1580432,
      "step": 5645
    },
    {
      "epoch": 62.77777777777778,
      "grad_norm": 1.8375399112701416,
      "learning_rate": 4.7579544131846114e-05,
      "loss": 0.0454,
      "num_input_tokens_seen": 1581872,
      "step": 5650
    },
    {
      "epoch": 62.833333333333336,
      "grad_norm": 1.8637855052947998,
      "learning_rate": 4.757532815730114e-05,
      "loss": 0.1085,
      "num_input_tokens_seen": 1583296,
      "step": 5655
    },
    {
      "epoch": 62.888888888888886,
      "grad_norm": 4.107780456542969,
      "learning_rate": 4.7571108701356865e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 1584688,
      "step": 5660
    },
    {
      "epoch": 62.94444444444444,
      "grad_norm": 0.17723235487937927,
      "learning_rate": 4.756688576466398e-05,
      "loss": 0.0904,
      "num_input_tokens_seen": 1586096,
      "step": 5665
    },
    {
      "epoch": 63.0,
      "grad_norm": 4.5588836669921875,
      "learning_rate": 4.756265934787372e-05,
      "loss": 0.0695,
      "num_input_tokens_seen": 1587488,
      "step": 5670
    },
    {
      "epoch": 63.05555555555556,
      "grad_norm": 2.8241701126098633,
      "learning_rate": 4.755842945163785e-05,
      "loss": 0.064,
      "num_input_tokens_seen": 1588880,
      "step": 5675
    },
    {
      "epoch": 63.111111111111114,
      "grad_norm": 3.004133939743042,
      "learning_rate": 4.755419607660867e-05,
      "loss": 0.13,
      "num_input_tokens_seen": 1590272,
      "step": 5680
    },
    {
      "epoch": 63.166666666666664,
      "grad_norm": 0.7034093141555786,
      "learning_rate": 4.7549959223439016e-05,
      "loss": 0.0387,
      "num_input_tokens_seen": 1591696,
      "step": 5685
    },
    {
      "epoch": 63.22222222222222,
      "grad_norm": 2.3472073078155518,
      "learning_rate": 4.754571889278228e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 1593024,
      "step": 5690
    },
    {
      "epoch": 63.27777777777778,
      "grad_norm": 1.395923137664795,
      "learning_rate": 4.754147508529235e-05,
      "loss": 0.069,
      "num_input_tokens_seen": 1594432,
      "step": 5695
    },
    {
      "epoch": 63.333333333333336,
      "grad_norm": 1.7951332330703735,
      "learning_rate": 4.75372278016237e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 1595808,
      "step": 5700
    },
    {
      "epoch": 63.388888888888886,
      "grad_norm": 0.6739165186882019,
      "learning_rate": 4.753297704243129e-05,
      "loss": 0.0861,
      "num_input_tokens_seen": 1597216,
      "step": 5705
    },
    {
      "epoch": 63.44444444444444,
      "grad_norm": 3.321620225906372,
      "learning_rate": 4.752872280837066e-05,
      "loss": 0.0543,
      "num_input_tokens_seen": 1598624,
      "step": 5710
    },
    {
      "epoch": 63.5,
      "grad_norm": 1.6638177633285522,
      "learning_rate": 4.752446510009786e-05,
      "loss": 0.0491,
      "num_input_tokens_seen": 1600000,
      "step": 5715
    },
    {
      "epoch": 63.55555555555556,
      "grad_norm": 3.3396098613739014,
      "learning_rate": 4.7520203918269476e-05,
      "loss": 0.0315,
      "num_input_tokens_seen": 1601392,
      "step": 5720
    },
    {
      "epoch": 63.611111111111114,
      "grad_norm": 0.9249308705329895,
      "learning_rate": 4.751593926354265e-05,
      "loss": 0.0793,
      "num_input_tokens_seen": 1602768,
      "step": 5725
    },
    {
      "epoch": 63.666666666666664,
      "grad_norm": 0.22054006159305573,
      "learning_rate": 4.751167113657503e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 1604160,
      "step": 5730
    },
    {
      "epoch": 63.72222222222222,
      "grad_norm": 1.4404717683792114,
      "learning_rate": 4.7507399538024834e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 1605584,
      "step": 5735
    },
    {
      "epoch": 63.77777777777778,
      "grad_norm": 3.0976412296295166,
      "learning_rate": 4.750312446855077e-05,
      "loss": 0.098,
      "num_input_tokens_seen": 1607008,
      "step": 5740
    },
    {
      "epoch": 63.833333333333336,
      "grad_norm": 3.0237481594085693,
      "learning_rate": 4.749884592881212e-05,
      "loss": 0.1245,
      "num_input_tokens_seen": 1608400,
      "step": 5745
    },
    {
      "epoch": 63.888888888888886,
      "grad_norm": 1.4394094944000244,
      "learning_rate": 4.74945639194687e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 1609808,
      "step": 5750
    },
    {
      "epoch": 63.94444444444444,
      "grad_norm": 3.0131452083587646,
      "learning_rate": 4.749027844118083e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 1611232,
      "step": 5755
    },
    {
      "epoch": 64.0,
      "grad_norm": 7.362742900848389,
      "learning_rate": 4.7485989494609395e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 1612624,
      "step": 5760
    },
    {
      "epoch": 64.05555555555556,
      "grad_norm": 2.3307688236236572,
      "learning_rate": 4.748169708041581e-05,
      "loss": 0.0951,
      "num_input_tokens_seen": 1614048,
      "step": 5765
    },
    {
      "epoch": 64.11111111111111,
      "grad_norm": 5.516408920288086,
      "learning_rate": 4.7477401199262004e-05,
      "loss": 0.1717,
      "num_input_tokens_seen": 1615408,
      "step": 5770
    },
    {
      "epoch": 64.16666666666667,
      "grad_norm": 0.04532832279801369,
      "learning_rate": 4.747310185181048e-05,
      "loss": 0.0695,
      "num_input_tokens_seen": 1616800,
      "step": 5775
    },
    {
      "epoch": 64.22222222222223,
      "grad_norm": 1.623472809791565,
      "learning_rate": 4.746879903872422e-05,
      "loss": 0.0356,
      "num_input_tokens_seen": 1618192,
      "step": 5780
    },
    {
      "epoch": 64.27777777777777,
      "grad_norm": 0.40643659234046936,
      "learning_rate": 4.746449276066679e-05,
      "loss": 0.0935,
      "num_input_tokens_seen": 1619568,
      "step": 5785
    },
    {
      "epoch": 64.33333333333333,
      "grad_norm": 1.7069239616394043,
      "learning_rate": 4.746018301830227e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 1620944,
      "step": 5790
    },
    {
      "epoch": 64.38888888888889,
      "grad_norm": 0.07316430658102036,
      "learning_rate": 4.7455869812295275e-05,
      "loss": 0.0542,
      "num_input_tokens_seen": 1622384,
      "step": 5795
    },
    {
      "epoch": 64.44444444444444,
      "grad_norm": 3.162832736968994,
      "learning_rate": 4.7451553143310964e-05,
      "loss": 0.0505,
      "num_input_tokens_seen": 1623744,
      "step": 5800
    },
    {
      "epoch": 64.44444444444444,
      "eval_loss": 0.26427003741264343,
      "eval_runtime": 1.235,
      "eval_samples_per_second": 32.389,
      "eval_steps_per_second": 16.194,
      "num_input_tokens_seen": 1623744,
      "step": 5800
    },
    {
      "epoch": 64.5,
      "grad_norm": 0.5166583061218262,
      "learning_rate": 4.744723301201501e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 1625152,
      "step": 5805
    },
    {
      "epoch": 64.55555555555556,
      "grad_norm": 1.0865082740783691,
      "learning_rate": 4.744290941907364e-05,
      "loss": 0.0907,
      "num_input_tokens_seen": 1626560,
      "step": 5810
    },
    {
      "epoch": 64.61111111111111,
      "grad_norm": 0.1362117975950241,
      "learning_rate": 4.7438582365153594e-05,
      "loss": 0.0996,
      "num_input_tokens_seen": 1627952,
      "step": 5815
    },
    {
      "epoch": 64.66666666666667,
      "grad_norm": 0.5322148203849792,
      "learning_rate": 4.743425185092217e-05,
      "loss": 0.0215,
      "num_input_tokens_seen": 1629392,
      "step": 5820
    },
    {
      "epoch": 64.72222222222223,
      "grad_norm": 2.305398464202881,
      "learning_rate": 4.742991787704719e-05,
      "loss": 0.0851,
      "num_input_tokens_seen": 1630784,
      "step": 5825
    },
    {
      "epoch": 64.77777777777777,
      "grad_norm": 0.20517103374004364,
      "learning_rate": 4.7425580444196994e-05,
      "loss": 0.0873,
      "num_input_tokens_seen": 1632160,
      "step": 5830
    },
    {
      "epoch": 64.83333333333333,
      "grad_norm": 2.3160641193389893,
      "learning_rate": 4.742123955304048e-05,
      "loss": 0.1255,
      "num_input_tokens_seen": 1633600,
      "step": 5835
    },
    {
      "epoch": 64.88888888888889,
      "grad_norm": 0.4513762295246124,
      "learning_rate": 4.741689520424706e-05,
      "loss": 0.0861,
      "num_input_tokens_seen": 1635008,
      "step": 5840
    },
    {
      "epoch": 64.94444444444444,
      "grad_norm": 0.1724293828010559,
      "learning_rate": 4.741254739848669e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 1636416,
      "step": 5845
    },
    {
      "epoch": 65.0,
      "grad_norm": 3.8614907264709473,
      "learning_rate": 4.740819613642987e-05,
      "loss": 0.0596,
      "num_input_tokens_seen": 1637824,
      "step": 5850
    },
    {
      "epoch": 65.05555555555556,
      "grad_norm": 4.0227813720703125,
      "learning_rate": 4.74038414187476e-05,
      "loss": 0.1163,
      "num_input_tokens_seen": 1639216,
      "step": 5855
    },
    {
      "epoch": 65.11111111111111,
      "grad_norm": 0.7826488018035889,
      "learning_rate": 4.739948324611144e-05,
      "loss": 0.0249,
      "num_input_tokens_seen": 1640624,
      "step": 5860
    },
    {
      "epoch": 65.16666666666667,
      "grad_norm": 0.26211413741111755,
      "learning_rate": 4.7395121619193465e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 1642000,
      "step": 5865
    },
    {
      "epoch": 65.22222222222223,
      "grad_norm": 1.7419105768203735,
      "learning_rate": 4.7390756538666313e-05,
      "loss": 0.0818,
      "num_input_tokens_seen": 1643376,
      "step": 5870
    },
    {
      "epoch": 65.27777777777777,
      "grad_norm": 0.22114944458007812,
      "learning_rate": 4.738638800520311e-05,
      "loss": 0.0708,
      "num_input_tokens_seen": 1644832,
      "step": 5875
    },
    {
      "epoch": 65.33333333333333,
      "grad_norm": 2.5325727462768555,
      "learning_rate": 4.738201601947757e-05,
      "loss": 0.0646,
      "num_input_tokens_seen": 1646208,
      "step": 5880
    },
    {
      "epoch": 65.38888888888889,
      "grad_norm": 0.3094221353530884,
      "learning_rate": 4.7377640582163876e-05,
      "loss": 0.0231,
      "num_input_tokens_seen": 1647600,
      "step": 5885
    },
    {
      "epoch": 65.44444444444444,
      "grad_norm": 3.153142213821411,
      "learning_rate": 4.7373261693936786e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 1648976,
      "step": 5890
    },
    {
      "epoch": 65.5,
      "grad_norm": 3.9627139568328857,
      "learning_rate": 4.7368879355471595e-05,
      "loss": 0.0508,
      "num_input_tokens_seen": 1650352,
      "step": 5895
    },
    {
      "epoch": 65.55555555555556,
      "grad_norm": 0.16992738842964172,
      "learning_rate": 4.736449356744409e-05,
      "loss": 0.0126,
      "num_input_tokens_seen": 1651744,
      "step": 5900
    },
    {
      "epoch": 65.61111111111111,
      "grad_norm": 1.4800448417663574,
      "learning_rate": 4.736010433053064e-05,
      "loss": 0.0531,
      "num_input_tokens_seen": 1653152,
      "step": 5905
    },
    {
      "epoch": 65.66666666666667,
      "grad_norm": 0.15600045025348663,
      "learning_rate": 4.73557116454081e-05,
      "loss": 0.0742,
      "num_input_tokens_seen": 1654576,
      "step": 5910
    },
    {
      "epoch": 65.72222222222223,
      "grad_norm": 0.07147924602031708,
      "learning_rate": 4.735131551275389e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 1656032,
      "step": 5915
    },
    {
      "epoch": 65.77777777777777,
      "grad_norm": 3.6327767372131348,
      "learning_rate": 4.734691593324594e-05,
      "loss": 0.0695,
      "num_input_tokens_seen": 1657408,
      "step": 5920
    },
    {
      "epoch": 65.83333333333333,
      "grad_norm": 2.463275909423828,
      "learning_rate": 4.734251290756272e-05,
      "loss": 0.0678,
      "num_input_tokens_seen": 1658800,
      "step": 5925
    },
    {
      "epoch": 65.88888888888889,
      "grad_norm": 3.8663904666900635,
      "learning_rate": 4.7338106436383246e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 1660192,
      "step": 5930
    },
    {
      "epoch": 65.94444444444444,
      "grad_norm": 5.693742752075195,
      "learning_rate": 4.733369652038703e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 1661616,
      "step": 5935
    },
    {
      "epoch": 66.0,
      "grad_norm": 3.219804048538208,
      "learning_rate": 4.7329283160254156e-05,
      "loss": 0.1042,
      "num_input_tokens_seen": 1663072,
      "step": 5940
    },
    {
      "epoch": 66.05555555555556,
      "grad_norm": 3.0331308841705322,
      "learning_rate": 4.732486635666521e-05,
      "loss": 0.0721,
      "num_input_tokens_seen": 1664432,
      "step": 5945
    },
    {
      "epoch": 66.11111111111111,
      "grad_norm": 0.6742559671401978,
      "learning_rate": 4.732044611030132e-05,
      "loss": 0.1147,
      "num_input_tokens_seen": 1665808,
      "step": 5950
    },
    {
      "epoch": 66.16666666666667,
      "grad_norm": 0.7283286452293396,
      "learning_rate": 4.731602242184414e-05,
      "loss": 0.0827,
      "num_input_tokens_seen": 1667200,
      "step": 5955
    },
    {
      "epoch": 66.22222222222223,
      "grad_norm": 1.2014354467391968,
      "learning_rate": 4.7311595291975864e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 1668608,
      "step": 5960
    },
    {
      "epoch": 66.27777777777777,
      "grad_norm": 3.367736339569092,
      "learning_rate": 4.7307164721379216e-05,
      "loss": 0.1017,
      "num_input_tokens_seen": 1669984,
      "step": 5965
    },
    {
      "epoch": 66.33333333333333,
      "grad_norm": 2.9866721630096436,
      "learning_rate": 4.730273071073743e-05,
      "loss": 0.1225,
      "num_input_tokens_seen": 1671440,
      "step": 5970
    },
    {
      "epoch": 66.38888888888889,
      "grad_norm": 4.357430934906006,
      "learning_rate": 4.729829326073429e-05,
      "loss": 0.1079,
      "num_input_tokens_seen": 1672816,
      "step": 5975
    },
    {
      "epoch": 66.44444444444444,
      "grad_norm": 0.6511040329933167,
      "learning_rate": 4.7293852372054126e-05,
      "loss": 0.093,
      "num_input_tokens_seen": 1674240,
      "step": 5980
    },
    {
      "epoch": 66.5,
      "grad_norm": 2.631479501724243,
      "learning_rate": 4.728940804538176e-05,
      "loss": 0.04,
      "num_input_tokens_seen": 1675648,
      "step": 5985
    },
    {
      "epoch": 66.55555555555556,
      "grad_norm": 0.7323653101921082,
      "learning_rate": 4.7284960281402556e-05,
      "loss": 0.0329,
      "num_input_tokens_seen": 1677024,
      "step": 5990
    },
    {
      "epoch": 66.61111111111111,
      "grad_norm": 1.4739844799041748,
      "learning_rate": 4.728050908080244e-05,
      "loss": 0.0217,
      "num_input_tokens_seen": 1678464,
      "step": 5995
    },
    {
      "epoch": 66.66666666666667,
      "grad_norm": 5.306184768676758,
      "learning_rate": 4.727605444426782e-05,
      "loss": 0.064,
      "num_input_tokens_seen": 1679888,
      "step": 6000
    },
    {
      "epoch": 66.66666666666667,
      "eval_loss": 0.27099061012268066,
      "eval_runtime": 1.234,
      "eval_samples_per_second": 32.416,
      "eval_steps_per_second": 16.208,
      "num_input_tokens_seen": 1679888,
      "step": 6000
    },
    {
      "epoch": 66.72222222222223,
      "grad_norm": 0.07205846905708313,
      "learning_rate": 4.727159637248567e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 1681264,
      "step": 6005
    },
    {
      "epoch": 66.77777777777777,
      "grad_norm": 2.753242254257202,
      "learning_rate": 4.7267134866143474e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 1682688,
      "step": 6010
    },
    {
      "epoch": 66.83333333333333,
      "grad_norm": 0.09009075164794922,
      "learning_rate": 4.726266992592926e-05,
      "loss": 0.058,
      "num_input_tokens_seen": 1684064,
      "step": 6015
    },
    {
      "epoch": 66.88888888888889,
      "grad_norm": 1.4693952798843384,
      "learning_rate": 4.725820155253157e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 1685440,
      "step": 6020
    },
    {
      "epoch": 66.94444444444444,
      "grad_norm": 2.1228597164154053,
      "learning_rate": 4.725372974663948e-05,
      "loss": 0.0879,
      "num_input_tokens_seen": 1686832,
      "step": 6025
    },
    {
      "epoch": 67.0,
      "grad_norm": 2.78877592086792,
      "learning_rate": 4.724925450894262e-05,
      "loss": 0.0302,
      "num_input_tokens_seen": 1688272,
      "step": 6030
    },
    {
      "epoch": 67.05555555555556,
      "grad_norm": 0.9054118990898132,
      "learning_rate": 4.72447758401311e-05,
      "loss": 0.0322,
      "num_input_tokens_seen": 1689680,
      "step": 6035
    },
    {
      "epoch": 67.11111111111111,
      "grad_norm": 0.3880032002925873,
      "learning_rate": 4.7240293740895616e-05,
      "loss": 0.0602,
      "num_input_tokens_seen": 1691104,
      "step": 6040
    },
    {
      "epoch": 67.16666666666667,
      "grad_norm": 5.288763046264648,
      "learning_rate": 4.723580821192733e-05,
      "loss": 0.0803,
      "num_input_tokens_seen": 1692512,
      "step": 6045
    },
    {
      "epoch": 67.22222222222223,
      "grad_norm": 0.4908899962902069,
      "learning_rate": 4.7231319253917996e-05,
      "loss": 0.0302,
      "num_input_tokens_seen": 1693936,
      "step": 6050
    },
    {
      "epoch": 67.27777777777777,
      "grad_norm": 0.5134313702583313,
      "learning_rate": 4.722682686755986e-05,
      "loss": 0.0496,
      "num_input_tokens_seen": 1695344,
      "step": 6055
    },
    {
      "epoch": 67.33333333333333,
      "grad_norm": 0.045499905943870544,
      "learning_rate": 4.722233105354569e-05,
      "loss": 0.0339,
      "num_input_tokens_seen": 1696736,
      "step": 6060
    },
    {
      "epoch": 67.38888888888889,
      "grad_norm": 2.1490976810455322,
      "learning_rate": 4.7217831812568815e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 1698128,
      "step": 6065
    },
    {
      "epoch": 67.44444444444444,
      "grad_norm": 4.50119686126709,
      "learning_rate": 4.721332914532307e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 1699504,
      "step": 6070
    },
    {
      "epoch": 67.5,
      "grad_norm": 0.2443414330482483,
      "learning_rate": 4.720882305250281e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 1700896,
      "step": 6075
    },
    {
      "epoch": 67.55555555555556,
      "grad_norm": 3.1928811073303223,
      "learning_rate": 4.720431353480295e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 1702272,
      "step": 6080
    },
    {
      "epoch": 67.61111111111111,
      "grad_norm": 0.5989601016044617,
      "learning_rate": 4.719980059291891e-05,
      "loss": 0.1168,
      "num_input_tokens_seen": 1703680,
      "step": 6085
    },
    {
      "epoch": 67.66666666666667,
      "grad_norm": 1.1456820964813232,
      "learning_rate": 4.7195284227546634e-05,
      "loss": 0.0407,
      "num_input_tokens_seen": 1705088,
      "step": 6090
    },
    {
      "epoch": 67.72222222222223,
      "grad_norm": 1.5870180130004883,
      "learning_rate": 4.7190764439382604e-05,
      "loss": 0.0656,
      "num_input_tokens_seen": 1706496,
      "step": 6095
    },
    {
      "epoch": 67.77777777777777,
      "grad_norm": 5.279561996459961,
      "learning_rate": 4.7186241229123826e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 1707904,
      "step": 6100
    },
    {
      "epoch": 67.83333333333333,
      "grad_norm": 0.38072577118873596,
      "learning_rate": 4.718171459746785e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 1709296,
      "step": 6105
    },
    {
      "epoch": 67.88888888888889,
      "grad_norm": 3.7258737087249756,
      "learning_rate": 4.717718454511273e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 1710672,
      "step": 6110
    },
    {
      "epoch": 67.94444444444444,
      "grad_norm": 1.3949328660964966,
      "learning_rate": 4.7172651072757056e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 1712048,
      "step": 6115
    },
    {
      "epoch": 68.0,
      "grad_norm": 0.02202235534787178,
      "learning_rate": 4.7168114181099945e-05,
      "loss": 0.0331,
      "num_input_tokens_seen": 1713488,
      "step": 6120
    },
    {
      "epoch": 68.05555555555556,
      "grad_norm": 0.08326900005340576,
      "learning_rate": 4.716357387084105e-05,
      "loss": 0.1021,
      "num_input_tokens_seen": 1714864,
      "step": 6125
    },
    {
      "epoch": 68.11111111111111,
      "grad_norm": 3.290828227996826,
      "learning_rate": 4.715903014268054e-05,
      "loss": 0.0429,
      "num_input_tokens_seen": 1716240,
      "step": 6130
    },
    {
      "epoch": 68.16666666666667,
      "grad_norm": 1.388460397720337,
      "learning_rate": 4.715448299731911e-05,
      "loss": 0.0504,
      "num_input_tokens_seen": 1717648,
      "step": 6135
    },
    {
      "epoch": 68.22222222222223,
      "grad_norm": 2.5787785053253174,
      "learning_rate": 4.7149932435457986e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 1719072,
      "step": 6140
    },
    {
      "epoch": 68.27777777777777,
      "grad_norm": 0.825234591960907,
      "learning_rate": 4.714537845779894e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 1720480,
      "step": 6145
    },
    {
      "epoch": 68.33333333333333,
      "grad_norm": 0.2116880565881729,
      "learning_rate": 4.714082106504423e-05,
      "loss": 0.0755,
      "num_input_tokens_seen": 1721888,
      "step": 6150
    },
    {
      "epoch": 68.38888888888889,
      "grad_norm": 0.18922197818756104,
      "learning_rate": 4.713626025789667e-05,
      "loss": 0.0795,
      "num_input_tokens_seen": 1723312,
      "step": 6155
    },
    {
      "epoch": 68.44444444444444,
      "grad_norm": 3.3675873279571533,
      "learning_rate": 4.7131696037059606e-05,
      "loss": 0.0305,
      "num_input_tokens_seen": 1724672,
      "step": 6160
    },
    {
      "epoch": 68.5,
      "grad_norm": 2.048048973083496,
      "learning_rate": 4.712712840323689e-05,
      "loss": 0.0318,
      "num_input_tokens_seen": 1726048,
      "step": 6165
    },
    {
      "epoch": 68.55555555555556,
      "grad_norm": 1.6905754804611206,
      "learning_rate": 4.71225573571329e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 1727504,
      "step": 6170
    },
    {
      "epoch": 68.61111111111111,
      "grad_norm": 1.076322078704834,
      "learning_rate": 4.711798289945256e-05,
      "loss": 0.0694,
      "num_input_tokens_seen": 1728928,
      "step": 6175
    },
    {
      "epoch": 68.66666666666667,
      "grad_norm": 1.3232814073562622,
      "learning_rate": 4.71134050309013e-05,
      "loss": 0.0722,
      "num_input_tokens_seen": 1730368,
      "step": 6180
    },
    {
      "epoch": 68.72222222222223,
      "grad_norm": 0.7610331773757935,
      "learning_rate": 4.710882375218509e-05,
      "loss": 0.0202,
      "num_input_tokens_seen": 1731792,
      "step": 6185
    },
    {
      "epoch": 68.77777777777777,
      "grad_norm": 4.580672264099121,
      "learning_rate": 4.7104239064010424e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 1733184,
      "step": 6190
    },
    {
      "epoch": 68.83333333333333,
      "grad_norm": 0.1644664853811264,
      "learning_rate": 4.709965096708432e-05,
      "loss": 0.0728,
      "num_input_tokens_seen": 1734576,
      "step": 6195
    },
    {
      "epoch": 68.88888888888889,
      "grad_norm": 0.014138957485556602,
      "learning_rate": 4.709505946211431e-05,
      "loss": 0.0508,
      "num_input_tokens_seen": 1735952,
      "step": 6200
    },
    {
      "epoch": 68.88888888888889,
      "eval_loss": 0.2814213037490845,
      "eval_runtime": 1.2389,
      "eval_samples_per_second": 32.288,
      "eval_steps_per_second": 16.144,
      "num_input_tokens_seen": 1735952,
      "step": 6200
    },
    {
      "epoch": 68.94444444444444,
      "grad_norm": 2.764373302459717,
      "learning_rate": 4.709046454980846e-05,
      "loss": 0.0529,
      "num_input_tokens_seen": 1737344,
      "step": 6205
    },
    {
      "epoch": 69.0,
      "grad_norm": 4.016452312469482,
      "learning_rate": 4.708586623087538e-05,
      "loss": 0.0941,
      "num_input_tokens_seen": 1738752,
      "step": 6210
    },
    {
      "epoch": 69.05555555555556,
      "grad_norm": 0.14238303899765015,
      "learning_rate": 4.708126450602418e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 1740176,
      "step": 6215
    },
    {
      "epoch": 69.11111111111111,
      "grad_norm": 3.188753604888916,
      "learning_rate": 4.7076659375964495e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 1741536,
      "step": 6220
    },
    {
      "epoch": 69.16666666666667,
      "grad_norm": 4.7551984786987305,
      "learning_rate": 4.707205084140651e-05,
      "loss": 0.0489,
      "num_input_tokens_seen": 1742944,
      "step": 6225
    },
    {
      "epoch": 69.22222222222223,
      "grad_norm": 1.7434718608856201,
      "learning_rate": 4.7067438903060904e-05,
      "loss": 0.0173,
      "num_input_tokens_seen": 1744384,
      "step": 6230
    },
    {
      "epoch": 69.27777777777777,
      "grad_norm": 0.21638107299804688,
      "learning_rate": 4.70628235616389e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 1745760,
      "step": 6235
    },
    {
      "epoch": 69.33333333333333,
      "grad_norm": 1.7513903379440308,
      "learning_rate": 4.7058204817852256e-05,
      "loss": 0.0635,
      "num_input_tokens_seen": 1747136,
      "step": 6240
    },
    {
      "epoch": 69.38888888888889,
      "grad_norm": 5.628890037536621,
      "learning_rate": 4.705358267241322e-05,
      "loss": 0.0402,
      "num_input_tokens_seen": 1748544,
      "step": 6245
    },
    {
      "epoch": 69.44444444444444,
      "grad_norm": 0.8524031043052673,
      "learning_rate": 4.704895712603459e-05,
      "loss": 0.0364,
      "num_input_tokens_seen": 1749984,
      "step": 6250
    },
    {
      "epoch": 69.5,
      "grad_norm": 1.4765682220458984,
      "learning_rate": 4.704432817942969e-05,
      "loss": 0.0765,
      "num_input_tokens_seen": 1751392,
      "step": 6255
    },
    {
      "epoch": 69.55555555555556,
      "grad_norm": 0.045554447919130325,
      "learning_rate": 4.703969583331236e-05,
      "loss": 0.1004,
      "num_input_tokens_seen": 1752784,
      "step": 6260
    },
    {
      "epoch": 69.61111111111111,
      "grad_norm": 0.32162123918533325,
      "learning_rate": 4.7035060088396965e-05,
      "loss": 0.0712,
      "num_input_tokens_seen": 1754160,
      "step": 6265
    },
    {
      "epoch": 69.66666666666667,
      "grad_norm": 0.190348818898201,
      "learning_rate": 4.703042094539839e-05,
      "loss": 0.0507,
      "num_input_tokens_seen": 1755584,
      "step": 6270
    },
    {
      "epoch": 69.72222222222223,
      "grad_norm": 1.3274627923965454,
      "learning_rate": 4.702577840503206e-05,
      "loss": 0.0833,
      "num_input_tokens_seen": 1756992,
      "step": 6275
    },
    {
      "epoch": 69.77777777777777,
      "grad_norm": 1.264351487159729,
      "learning_rate": 4.70211324680139e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 1758384,
      "step": 6280
    },
    {
      "epoch": 69.83333333333333,
      "grad_norm": 0.819122850894928,
      "learning_rate": 4.7016483135060386e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 1759792,
      "step": 6285
    },
    {
      "epoch": 69.88888888888889,
      "grad_norm": 4.333892822265625,
      "learning_rate": 4.701183040688849e-05,
      "loss": 0.1,
      "num_input_tokens_seen": 1761152,
      "step": 6290
    },
    {
      "epoch": 69.94444444444444,
      "grad_norm": 3.2289340496063232,
      "learning_rate": 4.700717428421573e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 1762608,
      "step": 6295
    },
    {
      "epoch": 70.0,
      "grad_norm": 3.5966708660125732,
      "learning_rate": 4.700251476776014e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 1764000,
      "step": 6300
    },
    {
      "epoch": 70.05555555555556,
      "grad_norm": 1.1106194257736206,
      "learning_rate": 4.699785185824026e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 1765376,
      "step": 6305
    },
    {
      "epoch": 70.11111111111111,
      "grad_norm": 5.173810005187988,
      "learning_rate": 4.699318555637519e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 1766752,
      "step": 6310
    },
    {
      "epoch": 70.16666666666667,
      "grad_norm": 1.0195591449737549,
      "learning_rate": 4.6988515862884525e-05,
      "loss": 0.1191,
      "num_input_tokens_seen": 1768160,
      "step": 6315
    },
    {
      "epoch": 70.22222222222223,
      "grad_norm": 1.455989956855774,
      "learning_rate": 4.698384277848838e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 1769632,
      "step": 6320
    },
    {
      "epoch": 70.27777777777777,
      "grad_norm": 0.07283042371273041,
      "learning_rate": 4.6979166303907425e-05,
      "loss": 0.0104,
      "num_input_tokens_seen": 1771008,
      "step": 6325
    },
    {
      "epoch": 70.33333333333333,
      "grad_norm": 0.33313262462615967,
      "learning_rate": 4.697448643986281e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 1772368,
      "step": 6330
    },
    {
      "epoch": 70.38888888888889,
      "grad_norm": 3.065274238586426,
      "learning_rate": 4.696980318707624e-05,
      "loss": 0.0745,
      "num_input_tokens_seen": 1773744,
      "step": 6335
    },
    {
      "epoch": 70.44444444444444,
      "grad_norm": 5.375479698181152,
      "learning_rate": 4.6965116546269924e-05,
      "loss": 0.0293,
      "num_input_tokens_seen": 1775168,
      "step": 6340
    },
    {
      "epoch": 70.5,
      "grad_norm": 4.8052849769592285,
      "learning_rate": 4.6960426518166615e-05,
      "loss": 0.1226,
      "num_input_tokens_seen": 1776544,
      "step": 6345
    },
    {
      "epoch": 70.55555555555556,
      "grad_norm": 0.367395281791687,
      "learning_rate": 4.6955733103489556e-05,
      "loss": 0.0388,
      "num_input_tokens_seen": 1777952,
      "step": 6350
    },
    {
      "epoch": 70.61111111111111,
      "grad_norm": 1.6441925764083862,
      "learning_rate": 4.695103630296255e-05,
      "loss": 0.0569,
      "num_input_tokens_seen": 1779328,
      "step": 6355
    },
    {
      "epoch": 70.66666666666667,
      "grad_norm": 6.519750595092773,
      "learning_rate": 4.694633611730988e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 1780736,
      "step": 6360
    },
    {
      "epoch": 70.72222222222223,
      "grad_norm": 3.444936752319336,
      "learning_rate": 4.694163254725639e-05,
      "loss": 0.0528,
      "num_input_tokens_seen": 1782112,
      "step": 6365
    },
    {
      "epoch": 70.77777777777777,
      "grad_norm": 0.9987745881080627,
      "learning_rate": 4.693692559352743e-05,
      "loss": 0.051,
      "num_input_tokens_seen": 1783520,
      "step": 6370
    },
    {
      "epoch": 70.83333333333333,
      "grad_norm": 0.6965280175209045,
      "learning_rate": 4.693221525684886e-05,
      "loss": 0.0774,
      "num_input_tokens_seen": 1784928,
      "step": 6375
    },
    {
      "epoch": 70.88888888888889,
      "grad_norm": 1.372977375984192,
      "learning_rate": 4.6927501537947084e-05,
      "loss": 0.1154,
      "num_input_tokens_seen": 1786304,
      "step": 6380
    },
    {
      "epoch": 70.94444444444444,
      "grad_norm": 0.05043976753950119,
      "learning_rate": 4.692278443754901e-05,
      "loss": 0.0133,
      "num_input_tokens_seen": 1787712,
      "step": 6385
    },
    {
      "epoch": 71.0,
      "grad_norm": 1.5185389518737793,
      "learning_rate": 4.691806395638208e-05,
      "loss": 0.0748,
      "num_input_tokens_seen": 1789088,
      "step": 6390
    },
    {
      "epoch": 71.05555555555556,
      "grad_norm": 0.4600331485271454,
      "learning_rate": 4.6913340095174255e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 1790496,
      "step": 6395
    },
    {
      "epoch": 71.11111111111111,
      "grad_norm": 0.10578540712594986,
      "learning_rate": 4.690861285465399e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 1791904,
      "step": 6400
    },
    {
      "epoch": 71.11111111111111,
      "eval_loss": 0.27475041151046753,
      "eval_runtime": 1.2365,
      "eval_samples_per_second": 32.349,
      "eval_steps_per_second": 16.174,
      "num_input_tokens_seen": 1791904,
      "step": 6400
    },
    {
      "epoch": 71.16666666666667,
      "grad_norm": 2.2166948318481445,
      "learning_rate": 4.690388223555031e-05,
      "loss": 0.0828,
      "num_input_tokens_seen": 1793296,
      "step": 6405
    },
    {
      "epoch": 71.22222222222223,
      "grad_norm": 0.0439300537109375,
      "learning_rate": 4.689914823859273e-05,
      "loss": 0.0103,
      "num_input_tokens_seen": 1794688,
      "step": 6410
    },
    {
      "epoch": 71.27777777777777,
      "grad_norm": 2.272075653076172,
      "learning_rate": 4.689441086451129e-05,
      "loss": 0.1164,
      "num_input_tokens_seen": 1796112,
      "step": 6415
    },
    {
      "epoch": 71.33333333333333,
      "grad_norm": 2.560701847076416,
      "learning_rate": 4.688967011403655e-05,
      "loss": 0.1269,
      "num_input_tokens_seen": 1797520,
      "step": 6420
    },
    {
      "epoch": 71.38888888888889,
      "grad_norm": 4.408782482147217,
      "learning_rate": 4.68849259878996e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 1798896,
      "step": 6425
    },
    {
      "epoch": 71.44444444444444,
      "grad_norm": 0.8516421318054199,
      "learning_rate": 4.6880178486832036e-05,
      "loss": 0.0765,
      "num_input_tokens_seen": 1800304,
      "step": 6430
    },
    {
      "epoch": 71.5,
      "grad_norm": 2.712446928024292,
      "learning_rate": 4.687542761156598e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 1801664,
      "step": 6435
    },
    {
      "epoch": 71.55555555555556,
      "grad_norm": 0.26549404859542847,
      "learning_rate": 4.6870673362834096e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 1803056,
      "step": 6440
    },
    {
      "epoch": 71.61111111111111,
      "grad_norm": 0.44609299302101135,
      "learning_rate": 4.6865915741369526e-05,
      "loss": 0.0421,
      "num_input_tokens_seen": 1804432,
      "step": 6445
    },
    {
      "epoch": 71.66666666666667,
      "grad_norm": 2.2592997550964355,
      "learning_rate": 4.686115474790597e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 1805856,
      "step": 6450
    },
    {
      "epoch": 71.72222222222223,
      "grad_norm": 1.4803017377853394,
      "learning_rate": 4.685639038317762e-05,
      "loss": 0.0497,
      "num_input_tokens_seen": 1807264,
      "step": 6455
    },
    {
      "epoch": 71.77777777777777,
      "grad_norm": 5.984060764312744,
      "learning_rate": 4.685162264791921e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 1808688,
      "step": 6460
    },
    {
      "epoch": 71.83333333333333,
      "grad_norm": 0.032258063554763794,
      "learning_rate": 4.684685154286599e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 1810096,
      "step": 6465
    },
    {
      "epoch": 71.88888888888889,
      "grad_norm": 0.2690339982509613,
      "learning_rate": 4.684207706875371e-05,
      "loss": 0.0208,
      "num_input_tokens_seen": 1811488,
      "step": 6470
    },
    {
      "epoch": 71.94444444444444,
      "grad_norm": 0.2685427963733673,
      "learning_rate": 4.683729922631866e-05,
      "loss": 0.0459,
      "num_input_tokens_seen": 1812896,
      "step": 6475
    },
    {
      "epoch": 72.0,
      "grad_norm": 0.8292906284332275,
      "learning_rate": 4.683251801629765e-05,
      "loss": 0.0742,
      "num_input_tokens_seen": 1814288,
      "step": 6480
    },
    {
      "epoch": 72.05555555555556,
      "grad_norm": 1.731931209564209,
      "learning_rate": 4.6827733439428e-05,
      "loss": 0.0355,
      "num_input_tokens_seen": 1815696,
      "step": 6485
    },
    {
      "epoch": 72.11111111111111,
      "grad_norm": 1.189234972000122,
      "learning_rate": 4.682294549644754e-05,
      "loss": 0.0286,
      "num_input_tokens_seen": 1817104,
      "step": 6490
    },
    {
      "epoch": 72.16666666666667,
      "grad_norm": 5.001075744628906,
      "learning_rate": 4.681815418809464e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 1818528,
      "step": 6495
    },
    {
      "epoch": 72.22222222222223,
      "grad_norm": 3.7302324771881104,
      "learning_rate": 4.681335951510819e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 1819920,
      "step": 6500
    },
    {
      "epoch": 72.27777777777777,
      "grad_norm": 0.3433591425418854,
      "learning_rate": 4.6808561478227576e-05,
      "loss": 0.0604,
      "num_input_tokens_seen": 1821312,
      "step": 6505
    },
    {
      "epoch": 72.33333333333333,
      "grad_norm": 0.18287527561187744,
      "learning_rate": 4.680376007819271e-05,
      "loss": 0.0807,
      "num_input_tokens_seen": 1822720,
      "step": 6510
    },
    {
      "epoch": 72.38888888888889,
      "grad_norm": 0.6866089701652527,
      "learning_rate": 4.679895531574405e-05,
      "loss": 0.0866,
      "num_input_tokens_seen": 1824096,
      "step": 6515
    },
    {
      "epoch": 72.44444444444444,
      "grad_norm": 3.2061574459075928,
      "learning_rate": 4.679414719162253e-05,
      "loss": 0.0597,
      "num_input_tokens_seen": 1825472,
      "step": 6520
    },
    {
      "epoch": 72.5,
      "grad_norm": 0.18260042369365692,
      "learning_rate": 4.6789335706569635e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 1826832,
      "step": 6525
    },
    {
      "epoch": 72.55555555555556,
      "grad_norm": 0.27874070405960083,
      "learning_rate": 4.678452086132734e-05,
      "loss": 0.0349,
      "num_input_tokens_seen": 1828272,
      "step": 6530
    },
    {
      "epoch": 72.61111111111111,
      "grad_norm": 2.4175198078155518,
      "learning_rate": 4.677970265663818e-05,
      "loss": 0.0128,
      "num_input_tokens_seen": 1829680,
      "step": 6535
    },
    {
      "epoch": 72.66666666666667,
      "grad_norm": 0.3990218937397003,
      "learning_rate": 4.677488109324517e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 1831072,
      "step": 6540
    },
    {
      "epoch": 72.72222222222223,
      "grad_norm": 0.055495865643024445,
      "learning_rate": 4.6770056171891846e-05,
      "loss": 0.0183,
      "num_input_tokens_seen": 1832480,
      "step": 6545
    },
    {
      "epoch": 72.77777777777777,
      "grad_norm": 2.8697409629821777,
      "learning_rate": 4.6765227893322286e-05,
      "loss": 0.091,
      "num_input_tokens_seen": 1833872,
      "step": 6550
    },
    {
      "epoch": 72.83333333333333,
      "grad_norm": 0.5764212608337402,
      "learning_rate": 4.676039625828107e-05,
      "loss": 0.0491,
      "num_input_tokens_seen": 1835296,
      "step": 6555
    },
    {
      "epoch": 72.88888888888889,
      "grad_norm": 0.725125789642334,
      "learning_rate": 4.675556126751328e-05,
      "loss": 0.0099,
      "num_input_tokens_seen": 1836704,
      "step": 6560
    },
    {
      "epoch": 72.94444444444444,
      "grad_norm": 4.677392959594727,
      "learning_rate": 4.6750722921764556e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 1838112,
      "step": 6565
    },
    {
      "epoch": 73.0,
      "grad_norm": 0.3532140254974365,
      "learning_rate": 4.674588122178102e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 1839536,
      "step": 6570
    },
    {
      "epoch": 73.05555555555556,
      "grad_norm": 0.6134850978851318,
      "learning_rate": 4.674103616830931e-05,
      "loss": 0.0268,
      "num_input_tokens_seen": 1840944,
      "step": 6575
    },
    {
      "epoch": 73.11111111111111,
      "grad_norm": 1.1177492141723633,
      "learning_rate": 4.673618776209663e-05,
      "loss": 0.0798,
      "num_input_tokens_seen": 1842352,
      "step": 6580
    },
    {
      "epoch": 73.16666666666667,
      "grad_norm": 0.05788592994213104,
      "learning_rate": 4.673133600389063e-05,
      "loss": 0.0119,
      "num_input_tokens_seen": 1843776,
      "step": 6585
    },
    {
      "epoch": 73.22222222222223,
      "grad_norm": 0.8819052577018738,
      "learning_rate": 4.672648089443953e-05,
      "loss": 0.0656,
      "num_input_tokens_seen": 1845152,
      "step": 6590
    },
    {
      "epoch": 73.27777777777777,
      "grad_norm": 1.0444514751434326,
      "learning_rate": 4.672162243449204e-05,
      "loss": 0.0639,
      "num_input_tokens_seen": 1846512,
      "step": 6595
    },
    {
      "epoch": 73.33333333333333,
      "grad_norm": 0.8036683201789856,
      "learning_rate": 4.67167606247974e-05,
      "loss": 0.1238,
      "num_input_tokens_seen": 1847888,
      "step": 6600
    },
    {
      "epoch": 73.33333333333333,
      "eval_loss": 0.30527928471565247,
      "eval_runtime": 1.2342,
      "eval_samples_per_second": 32.409,
      "eval_steps_per_second": 16.205,
      "num_input_tokens_seen": 1847888,
      "step": 6600
    },
    {
      "epoch": 73.38888888888889,
      "grad_norm": 0.14255602657794952,
      "learning_rate": 4.671189546610536e-05,
      "loss": 0.0336,
      "num_input_tokens_seen": 1849264,
      "step": 6605
    },
    {
      "epoch": 73.44444444444444,
      "grad_norm": 2.1096227169036865,
      "learning_rate": 4.67070269591662e-05,
      "loss": 0.0755,
      "num_input_tokens_seen": 1850656,
      "step": 6610
    },
    {
      "epoch": 73.5,
      "grad_norm": 1.6777597665786743,
      "learning_rate": 4.670215510473068e-05,
      "loss": 0.1015,
      "num_input_tokens_seen": 1852064,
      "step": 6615
    },
    {
      "epoch": 73.55555555555556,
      "grad_norm": 7.874538421630859,
      "learning_rate": 4.669727990355013e-05,
      "loss": 0.0604,
      "num_input_tokens_seen": 1853472,
      "step": 6620
    },
    {
      "epoch": 73.61111111111111,
      "grad_norm": 1.130089521408081,
      "learning_rate": 4.669240135637635e-05,
      "loss": 0.0673,
      "num_input_tokens_seen": 1854832,
      "step": 6625
    },
    {
      "epoch": 73.66666666666667,
      "grad_norm": 0.035686150193214417,
      "learning_rate": 4.6687519463961675e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 1856224,
      "step": 6630
    },
    {
      "epoch": 73.72222222222223,
      "grad_norm": 1.3847863674163818,
      "learning_rate": 4.668263422705896e-05,
      "loss": 0.0109,
      "num_input_tokens_seen": 1857648,
      "step": 6635
    },
    {
      "epoch": 73.77777777777777,
      "grad_norm": 0.2363118976354599,
      "learning_rate": 4.667774564642156e-05,
      "loss": 0.0265,
      "num_input_tokens_seen": 1859072,
      "step": 6640
    },
    {
      "epoch": 73.83333333333333,
      "grad_norm": 0.12600275874137878,
      "learning_rate": 4.6672853722803365e-05,
      "loss": 0.0851,
      "num_input_tokens_seen": 1860496,
      "step": 6645
    },
    {
      "epoch": 73.88888888888889,
      "grad_norm": 1.259805679321289,
      "learning_rate": 4.666795845695877e-05,
      "loss": 0.027,
      "num_input_tokens_seen": 1861920,
      "step": 6650
    },
    {
      "epoch": 73.94444444444444,
      "grad_norm": 3.003251791000366,
      "learning_rate": 4.666305984964269e-05,
      "loss": 0.0752,
      "num_input_tokens_seen": 1863344,
      "step": 6655
    },
    {
      "epoch": 74.0,
      "grad_norm": 0.33793869614601135,
      "learning_rate": 4.6658157901610535e-05,
      "loss": 0.075,
      "num_input_tokens_seen": 1864752,
      "step": 6660
    },
    {
      "epoch": 74.05555555555556,
      "grad_norm": 0.16882531344890594,
      "learning_rate": 4.665325261361826e-05,
      "loss": 0.0624,
      "num_input_tokens_seen": 1866176,
      "step": 6665
    },
    {
      "epoch": 74.11111111111111,
      "grad_norm": 5.433872222900391,
      "learning_rate": 4.664834398642232e-05,
      "loss": 0.0692,
      "num_input_tokens_seen": 1867568,
      "step": 6670
    },
    {
      "epoch": 74.16666666666667,
      "grad_norm": 0.7773382067680359,
      "learning_rate": 4.6643432020779686e-05,
      "loss": 0.0154,
      "num_input_tokens_seen": 1868928,
      "step": 6675
    },
    {
      "epoch": 74.22222222222223,
      "grad_norm": 0.6612764000892639,
      "learning_rate": 4.663851671744786e-05,
      "loss": 0.0406,
      "num_input_tokens_seen": 1870336,
      "step": 6680
    },
    {
      "epoch": 74.27777777777777,
      "grad_norm": 1.0385850667953491,
      "learning_rate": 4.6633598077184815e-05,
      "loss": 0.0403,
      "num_input_tokens_seen": 1871744,
      "step": 6685
    },
    {
      "epoch": 74.33333333333333,
      "grad_norm": 0.8005903363227844,
      "learning_rate": 4.662867610074908e-05,
      "loss": 0.0816,
      "num_input_tokens_seen": 1873136,
      "step": 6690
    },
    {
      "epoch": 74.38888888888889,
      "grad_norm": 4.332681655883789,
      "learning_rate": 4.6623750788899696e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 1874528,
      "step": 6695
    },
    {
      "epoch": 74.44444444444444,
      "grad_norm": 2.6915032863616943,
      "learning_rate": 4.6618822142396195e-05,
      "loss": 0.1062,
      "num_input_tokens_seen": 1875888,
      "step": 6700
    },
    {
      "epoch": 74.5,
      "grad_norm": 0.660467803478241,
      "learning_rate": 4.661389016199864e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 1877312,
      "step": 6705
    },
    {
      "epoch": 74.55555555555556,
      "grad_norm": 2.5431737899780273,
      "learning_rate": 4.660895484846761e-05,
      "loss": 0.0987,
      "num_input_tokens_seen": 1878736,
      "step": 6710
    },
    {
      "epoch": 74.61111111111111,
      "grad_norm": 2.2537240982055664,
      "learning_rate": 4.660401620256418e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 1880176,
      "step": 6715
    },
    {
      "epoch": 74.66666666666667,
      "grad_norm": 0.053373806178569794,
      "learning_rate": 4.659907422504997e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 1881568,
      "step": 6720
    },
    {
      "epoch": 74.72222222222223,
      "grad_norm": 1.5663131475448608,
      "learning_rate": 4.6594128916687074e-05,
      "loss": 0.1063,
      "num_input_tokens_seen": 1882992,
      "step": 6725
    },
    {
      "epoch": 74.77777777777777,
      "grad_norm": 1.372054100036621,
      "learning_rate": 4.658918027823813e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 1884352,
      "step": 6730
    },
    {
      "epoch": 74.83333333333333,
      "grad_norm": 3.459674596786499,
      "learning_rate": 4.658422831046628e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 1885696,
      "step": 6735
    },
    {
      "epoch": 74.88888888888889,
      "grad_norm": 0.4436798691749573,
      "learning_rate": 4.657927301413518e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 1887104,
      "step": 6740
    },
    {
      "epoch": 74.94444444444444,
      "grad_norm": 3.5059099197387695,
      "learning_rate": 4.657431439000901e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 1888512,
      "step": 6745
    },
    {
      "epoch": 75.0,
      "grad_norm": 0.05748822167515755,
      "learning_rate": 4.656935243885243e-05,
      "loss": 0.0487,
      "num_input_tokens_seen": 1889904,
      "step": 6750
    },
    {
      "epoch": 75.05555555555556,
      "grad_norm": 0.26843512058258057,
      "learning_rate": 4.656438716143066e-05,
      "loss": 0.0509,
      "num_input_tokens_seen": 1891312,
      "step": 6755
    },
    {
      "epoch": 75.11111111111111,
      "grad_norm": 1.5014623403549194,
      "learning_rate": 4.6559418558509384e-05,
      "loss": 0.0397,
      "num_input_tokens_seen": 1892704,
      "step": 6760
    },
    {
      "epoch": 75.16666666666667,
      "grad_norm": 0.1279534250497818,
      "learning_rate": 4.6554446630854833e-05,
      "loss": 0.0799,
      "num_input_tokens_seen": 1894112,
      "step": 6765
    },
    {
      "epoch": 75.22222222222223,
      "grad_norm": 0.4853286147117615,
      "learning_rate": 4.654947137923374e-05,
      "loss": 0.057,
      "num_input_tokens_seen": 1895552,
      "step": 6770
    },
    {
      "epoch": 75.27777777777777,
      "grad_norm": 1.1650595664978027,
      "learning_rate": 4.654449280441335e-05,
      "loss": 0.052,
      "num_input_tokens_seen": 1896992,
      "step": 6775
    },
    {
      "epoch": 75.33333333333333,
      "grad_norm": 0.24042434990406036,
      "learning_rate": 4.653951090716143e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 1898384,
      "step": 6780
    },
    {
      "epoch": 75.38888888888889,
      "grad_norm": 0.038166966289281845,
      "learning_rate": 4.653452568824625e-05,
      "loss": 0.0425,
      "num_input_tokens_seen": 1899776,
      "step": 6785
    },
    {
      "epoch": 75.44444444444444,
      "grad_norm": 0.3959033489227295,
      "learning_rate": 4.6529537148436585e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 1901168,
      "step": 6790
    },
    {
      "epoch": 75.5,
      "grad_norm": 0.631869912147522,
      "learning_rate": 4.6524545288501734e-05,
      "loss": 0.0353,
      "num_input_tokens_seen": 1902592,
      "step": 6795
    },
    {
      "epoch": 75.55555555555556,
      "grad_norm": 0.07242665439844131,
      "learning_rate": 4.6519550109211506e-05,
      "loss": 0.0565,
      "num_input_tokens_seen": 1904000,
      "step": 6800
    },
    {
      "epoch": 75.55555555555556,
      "eval_loss": 0.27639350295066833,
      "eval_runtime": 1.2533,
      "eval_samples_per_second": 31.916,
      "eval_steps_per_second": 15.958,
      "num_input_tokens_seen": 1904000,
      "step": 6800
    },
    {
      "epoch": 75.61111111111111,
      "grad_norm": 3.0177347660064697,
      "learning_rate": 4.651455161133622e-05,
      "loss": 0.0675,
      "num_input_tokens_seen": 1905392,
      "step": 6805
    },
    {
      "epoch": 75.66666666666667,
      "grad_norm": 1.5695538520812988,
      "learning_rate": 4.6509549795646704e-05,
      "loss": 0.0595,
      "num_input_tokens_seen": 1906736,
      "step": 6810
    },
    {
      "epoch": 75.72222222222223,
      "grad_norm": 0.6053824424743652,
      "learning_rate": 4.6504544662914306e-05,
      "loss": 0.1085,
      "num_input_tokens_seen": 1908128,
      "step": 6815
    },
    {
      "epoch": 75.77777777777777,
      "grad_norm": 0.04693637415766716,
      "learning_rate": 4.6499536213910876e-05,
      "loss": 0.0152,
      "num_input_tokens_seen": 1909552,
      "step": 6820
    },
    {
      "epoch": 75.83333333333333,
      "grad_norm": 2.814873218536377,
      "learning_rate": 4.6494524449408786e-05,
      "loss": 0.0813,
      "num_input_tokens_seen": 1910960,
      "step": 6825
    },
    {
      "epoch": 75.88888888888889,
      "grad_norm": 3.2145891189575195,
      "learning_rate": 4.6489509370180903e-05,
      "loss": 0.113,
      "num_input_tokens_seen": 1912320,
      "step": 6830
    },
    {
      "epoch": 75.94444444444444,
      "grad_norm": 3.652317523956299,
      "learning_rate": 4.648449097700063e-05,
      "loss": 0.0752,
      "num_input_tokens_seen": 1913744,
      "step": 6835
    },
    {
      "epoch": 76.0,
      "grad_norm": 0.11998078227043152,
      "learning_rate": 4.647946927064185e-05,
      "loss": 0.0147,
      "num_input_tokens_seen": 1915072,
      "step": 6840
    },
    {
      "epoch": 76.05555555555556,
      "grad_norm": 4.526359558105469,
      "learning_rate": 4.647444425187898e-05,
      "loss": 0.039,
      "num_input_tokens_seen": 1916480,
      "step": 6845
    },
    {
      "epoch": 76.11111111111111,
      "grad_norm": 0.9173427820205688,
      "learning_rate": 4.646941592148695e-05,
      "loss": 0.0732,
      "num_input_tokens_seen": 1917888,
      "step": 6850
    },
    {
      "epoch": 76.16666666666667,
      "grad_norm": 2.001350164413452,
      "learning_rate": 4.646438428024117e-05,
      "loss": 0.02,
      "num_input_tokens_seen": 1919216,
      "step": 6855
    },
    {
      "epoch": 76.22222222222223,
      "grad_norm": 0.04205183684825897,
      "learning_rate": 4.64593493289176e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 1920624,
      "step": 6860
    },
    {
      "epoch": 76.27777777777777,
      "grad_norm": 0.8001420497894287,
      "learning_rate": 4.64543110682927e-05,
      "loss": 0.0658,
      "num_input_tokens_seen": 1922016,
      "step": 6865
    },
    {
      "epoch": 76.33333333333333,
      "grad_norm": 0.0267377607524395,
      "learning_rate": 4.644926949914341e-05,
      "loss": 0.0527,
      "num_input_tokens_seen": 1923488,
      "step": 6870
    },
    {
      "epoch": 76.38888888888889,
      "grad_norm": 1.7844890356063843,
      "learning_rate": 4.644422462224722e-05,
      "loss": 0.0738,
      "num_input_tokens_seen": 1924880,
      "step": 6875
    },
    {
      "epoch": 76.44444444444444,
      "grad_norm": 0.05602513998746872,
      "learning_rate": 4.643917643838211e-05,
      "loss": 0.0261,
      "num_input_tokens_seen": 1926288,
      "step": 6880
    },
    {
      "epoch": 76.5,
      "grad_norm": 0.3193373680114746,
      "learning_rate": 4.6434124948326564e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 1927616,
      "step": 6885
    },
    {
      "epoch": 76.55555555555556,
      "grad_norm": 0.06585303694009781,
      "learning_rate": 4.6429070152859594e-05,
      "loss": 0.0673,
      "num_input_tokens_seen": 1929056,
      "step": 6890
    },
    {
      "epoch": 76.61111111111111,
      "grad_norm": 0.48348015546798706,
      "learning_rate": 4.6424012052760714e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 1930400,
      "step": 6895
    },
    {
      "epoch": 76.66666666666667,
      "grad_norm": 1.3148106336593628,
      "learning_rate": 4.6418950648809945e-05,
      "loss": 0.0936,
      "num_input_tokens_seen": 1931840,
      "step": 6900
    },
    {
      "epoch": 76.72222222222223,
      "grad_norm": 0.9484290480613708,
      "learning_rate": 4.641388594178782e-05,
      "loss": 0.067,
      "num_input_tokens_seen": 1933216,
      "step": 6905
    },
    {
      "epoch": 76.77777777777777,
      "grad_norm": 0.940179169178009,
      "learning_rate": 4.640881793247538e-05,
      "loss": 0.0176,
      "num_input_tokens_seen": 1934640,
      "step": 6910
    },
    {
      "epoch": 76.83333333333333,
      "grad_norm": 0.15671059489250183,
      "learning_rate": 4.6403746621654173e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 1936000,
      "step": 6915
    },
    {
      "epoch": 76.88888888888889,
      "grad_norm": 5.062198638916016,
      "learning_rate": 4.639867201010626e-05,
      "loss": 0.0726,
      "num_input_tokens_seen": 1937392,
      "step": 6920
    },
    {
      "epoch": 76.94444444444444,
      "grad_norm": 1.0373902320861816,
      "learning_rate": 4.6393594098614204e-05,
      "loss": 0.0508,
      "num_input_tokens_seen": 1938832,
      "step": 6925
    },
    {
      "epoch": 77.0,
      "grad_norm": 6.379692077636719,
      "learning_rate": 4.63885128879611e-05,
      "loss": 0.0552,
      "num_input_tokens_seen": 1940288,
      "step": 6930
    },
    {
      "epoch": 77.05555555555556,
      "grad_norm": 0.6223801374435425,
      "learning_rate": 4.638342837893052e-05,
      "loss": 0.0605,
      "num_input_tokens_seen": 1941728,
      "step": 6935
    },
    {
      "epoch": 77.11111111111111,
      "grad_norm": 3.778485059738159,
      "learning_rate": 4.6378340572306565e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 1943152,
      "step": 6940
    },
    {
      "epoch": 77.16666666666667,
      "grad_norm": 0.15512005984783173,
      "learning_rate": 4.6373249468873833e-05,
      "loss": 0.057,
      "num_input_tokens_seen": 1944560,
      "step": 6945
    },
    {
      "epoch": 77.22222222222223,
      "grad_norm": 0.13116900622844696,
      "learning_rate": 4.636815506941744e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 1945952,
      "step": 6950
    },
    {
      "epoch": 77.27777777777777,
      "grad_norm": 0.13141606748104095,
      "learning_rate": 4.6363057374723004e-05,
      "loss": 0.011,
      "num_input_tokens_seen": 1947312,
      "step": 6955
    },
    {
      "epoch": 77.33333333333333,
      "grad_norm": 4.121311187744141,
      "learning_rate": 4.635795638557666e-05,
      "loss": 0.0915,
      "num_input_tokens_seen": 1948704,
      "step": 6960
    },
    {
      "epoch": 77.38888888888889,
      "grad_norm": 0.576884388923645,
      "learning_rate": 4.635285210276504e-05,
      "loss": 0.0874,
      "num_input_tokens_seen": 1950128,
      "step": 6965
    },
    {
      "epoch": 77.44444444444444,
      "grad_norm": 2.0695252418518066,
      "learning_rate": 4.6347744527075295e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 1951568,
      "step": 6970
    },
    {
      "epoch": 77.5,
      "grad_norm": 0.98436439037323,
      "learning_rate": 4.634263365929506e-05,
      "loss": 0.0226,
      "num_input_tokens_seen": 1953008,
      "step": 6975
    },
    {
      "epoch": 77.55555555555556,
      "grad_norm": 0.03024224378168583,
      "learning_rate": 4.6337519500212515e-05,
      "loss": 0.0639,
      "num_input_tokens_seen": 1954368,
      "step": 6980
    },
    {
      "epoch": 77.61111111111111,
      "grad_norm": 0.2653888761997223,
      "learning_rate": 4.633240205061632e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 1955728,
      "step": 6985
    },
    {
      "epoch": 77.66666666666667,
      "grad_norm": 1.1394259929656982,
      "learning_rate": 4.632728131129565e-05,
      "loss": 0.0619,
      "num_input_tokens_seen": 1957104,
      "step": 6990
    },
    {
      "epoch": 77.72222222222223,
      "grad_norm": 0.04745563492178917,
      "learning_rate": 4.632215728304018e-05,
      "loss": 0.0356,
      "num_input_tokens_seen": 1958528,
      "step": 6995
    },
    {
      "epoch": 77.77777777777777,
      "grad_norm": 3.1665971279144287,
      "learning_rate": 4.63170299666401e-05,
      "loss": 0.0486,
      "num_input_tokens_seen": 1959920,
      "step": 7000
    },
    {
      "epoch": 77.77777777777777,
      "eval_loss": 0.261984646320343,
      "eval_runtime": 1.2343,
      "eval_samples_per_second": 32.408,
      "eval_steps_per_second": 16.204,
      "num_input_tokens_seen": 1959920,
      "step": 7000
    },
    {
      "epoch": 77.83333333333333,
      "grad_norm": 0.4699676036834717,
      "learning_rate": 4.631189936288612e-05,
      "loss": 0.0241,
      "num_input_tokens_seen": 1961328,
      "step": 7005
    },
    {
      "epoch": 77.88888888888889,
      "grad_norm": 0.1651366651058197,
      "learning_rate": 4.630676547256944e-05,
      "loss": 0.0194,
      "num_input_tokens_seen": 1962720,
      "step": 7010
    },
    {
      "epoch": 77.94444444444444,
      "grad_norm": 0.19046613574028015,
      "learning_rate": 4.630162829648176e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 1964064,
      "step": 7015
    },
    {
      "epoch": 78.0,
      "grad_norm": 4.011109828948975,
      "learning_rate": 4.629648783541531e-05,
      "loss": 0.0541,
      "num_input_tokens_seen": 1965472,
      "step": 7020
    },
    {
      "epoch": 78.05555555555556,
      "grad_norm": 0.8682985901832581,
      "learning_rate": 4.6291344090162804e-05,
      "loss": 0.0622,
      "num_input_tokens_seen": 1966896,
      "step": 7025
    },
    {
      "epoch": 78.11111111111111,
      "grad_norm": 0.5108500719070435,
      "learning_rate": 4.628619706151748e-05,
      "loss": 0.0078,
      "num_input_tokens_seen": 1968272,
      "step": 7030
    },
    {
      "epoch": 78.16666666666667,
      "grad_norm": 0.06377578526735306,
      "learning_rate": 4.628104675027306e-05,
      "loss": 0.0905,
      "num_input_tokens_seen": 1969696,
      "step": 7035
    },
    {
      "epoch": 78.22222222222223,
      "grad_norm": 0.13553982973098755,
      "learning_rate": 4.6275893157223805e-05,
      "loss": 0.0241,
      "num_input_tokens_seen": 1971136,
      "step": 7040
    },
    {
      "epoch": 78.27777777777777,
      "grad_norm": 0.18495775759220123,
      "learning_rate": 4.627073628316445e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 1972528,
      "step": 7045
    },
    {
      "epoch": 78.33333333333333,
      "grad_norm": 0.040075741708278656,
      "learning_rate": 4.626557612889026e-05,
      "loss": 0.044,
      "num_input_tokens_seen": 1973920,
      "step": 7050
    },
    {
      "epoch": 78.38888888888889,
      "grad_norm": 0.22108067572116852,
      "learning_rate": 4.626041269519699e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 1975296,
      "step": 7055
    },
    {
      "epoch": 78.44444444444444,
      "grad_norm": 0.16289913654327393,
      "learning_rate": 4.6255245982880905e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 1976656,
      "step": 7060
    },
    {
      "epoch": 78.5,
      "grad_norm": 1.45285964012146,
      "learning_rate": 4.625007599273879e-05,
      "loss": 0.0557,
      "num_input_tokens_seen": 1978048,
      "step": 7065
    },
    {
      "epoch": 78.55555555555556,
      "grad_norm": 0.34744128584861755,
      "learning_rate": 4.6244902725567895e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 1979408,
      "step": 7070
    },
    {
      "epoch": 78.61111111111111,
      "grad_norm": 5.713740348815918,
      "learning_rate": 4.6239726182166024e-05,
      "loss": 0.0517,
      "num_input_tokens_seen": 1980800,
      "step": 7075
    },
    {
      "epoch": 78.66666666666667,
      "grad_norm": 3.5684773921966553,
      "learning_rate": 4.623454636333147e-05,
      "loss": 0.0691,
      "num_input_tokens_seen": 1982176,
      "step": 7080
    },
    {
      "epoch": 78.72222222222223,
      "grad_norm": 8.55652904510498,
      "learning_rate": 4.622936326986301e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 1983584,
      "step": 7085
    },
    {
      "epoch": 78.77777777777777,
      "grad_norm": 3.2484514713287354,
      "learning_rate": 4.6224176902559946e-05,
      "loss": 0.0597,
      "num_input_tokens_seen": 1984976,
      "step": 7090
    },
    {
      "epoch": 78.83333333333333,
      "grad_norm": 0.39806634187698364,
      "learning_rate": 4.621898726222209e-05,
      "loss": 0.0155,
      "num_input_tokens_seen": 1986384,
      "step": 7095
    },
    {
      "epoch": 78.88888888888889,
      "grad_norm": 1.1453273296356201,
      "learning_rate": 4.6213794349649744e-05,
      "loss": 0.0664,
      "num_input_tokens_seen": 1987792,
      "step": 7100
    },
    {
      "epoch": 78.94444444444444,
      "grad_norm": 0.786248505115509,
      "learning_rate": 4.6208598165643715e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 1989216,
      "step": 7105
    },
    {
      "epoch": 79.0,
      "grad_norm": 1.4244314432144165,
      "learning_rate": 4.620339871100533e-05,
      "loss": 0.0155,
      "num_input_tokens_seen": 1990624,
      "step": 7110
    },
    {
      "epoch": 79.05555555555556,
      "grad_norm": 5.31256628036499,
      "learning_rate": 4.6198195986536394e-05,
      "loss": 0.0638,
      "num_input_tokens_seen": 1992048,
      "step": 7115
    },
    {
      "epoch": 79.11111111111111,
      "grad_norm": 0.5838577747344971,
      "learning_rate": 4.619298999303926e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 1993440,
      "step": 7120
    },
    {
      "epoch": 79.16666666666667,
      "grad_norm": 0.11892479658126831,
      "learning_rate": 4.618778073131673e-05,
      "loss": 0.0329,
      "num_input_tokens_seen": 1994816,
      "step": 7125
    },
    {
      "epoch": 79.22222222222223,
      "grad_norm": 3.580540895462036,
      "learning_rate": 4.618256820217215e-05,
      "loss": 0.0485,
      "num_input_tokens_seen": 1996224,
      "step": 7130
    },
    {
      "epoch": 79.27777777777777,
      "grad_norm": 0.6766536831855774,
      "learning_rate": 4.617735240640936e-05,
      "loss": 0.0699,
      "num_input_tokens_seen": 1997616,
      "step": 7135
    },
    {
      "epoch": 79.33333333333333,
      "grad_norm": 0.20995980501174927,
      "learning_rate": 4.6172133344832705e-05,
      "loss": 0.0394,
      "num_input_tokens_seen": 1999008,
      "step": 7140
    },
    {
      "epoch": 79.38888888888889,
      "grad_norm": 0.3716764748096466,
      "learning_rate": 4.6166911018247004e-05,
      "loss": 0.0102,
      "num_input_tokens_seen": 2000368,
      "step": 7145
    },
    {
      "epoch": 79.44444444444444,
      "grad_norm": 0.4051329493522644,
      "learning_rate": 4.616168542745764e-05,
      "loss": 0.0261,
      "num_input_tokens_seen": 2001824,
      "step": 7150
    },
    {
      "epoch": 79.5,
      "grad_norm": 1.4047346115112305,
      "learning_rate": 4.6156456573270446e-05,
      "loss": 0.0522,
      "num_input_tokens_seen": 2003200,
      "step": 7155
    },
    {
      "epoch": 79.55555555555556,
      "grad_norm": 0.16677796840667725,
      "learning_rate": 4.615122445649177e-05,
      "loss": 0.0179,
      "num_input_tokens_seen": 2004576,
      "step": 7160
    },
    {
      "epoch": 79.61111111111111,
      "grad_norm": 1.1556059122085571,
      "learning_rate": 4.6145989077928486e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 2005952,
      "step": 7165
    },
    {
      "epoch": 79.66666666666667,
      "grad_norm": 0.13816381990909576,
      "learning_rate": 4.6140750438387953e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 2007344,
      "step": 7170
    },
    {
      "epoch": 79.72222222222223,
      "grad_norm": 0.5330780148506165,
      "learning_rate": 4.613550853867803e-05,
      "loss": 0.0812,
      "num_input_tokens_seen": 2008784,
      "step": 7175
    },
    {
      "epoch": 79.77777777777777,
      "grad_norm": 0.16196869313716888,
      "learning_rate": 4.613026337960708e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 2010208,
      "step": 7180
    },
    {
      "epoch": 79.83333333333333,
      "grad_norm": 1.0957139730453491,
      "learning_rate": 4.612501496198398e-05,
      "loss": 0.0459,
      "num_input_tokens_seen": 2011568,
      "step": 7185
    },
    {
      "epoch": 79.88888888888889,
      "grad_norm": 3.7996370792388916,
      "learning_rate": 4.61197632866181e-05,
      "loss": 0.1226,
      "num_input_tokens_seen": 2012976,
      "step": 7190
    },
    {
      "epoch": 79.94444444444444,
      "grad_norm": 0.07557634264230728,
      "learning_rate": 4.611450835431931e-05,
      "loss": 0.0114,
      "num_input_tokens_seen": 2014384,
      "step": 7195
    },
    {
      "epoch": 80.0,
      "grad_norm": 0.30444177985191345,
      "learning_rate": 4.6109250165898e-05,
      "loss": 0.0104,
      "num_input_tokens_seen": 2015792,
      "step": 7200
    },
    {
      "epoch": 80.0,
      "eval_loss": 0.2913602888584137,
      "eval_runtime": 1.2363,
      "eval_samples_per_second": 32.354,
      "eval_steps_per_second": 16.177,
      "num_input_tokens_seen": 2015792,
      "step": 7200
    },
    {
      "epoch": 80.05555555555556,
      "grad_norm": 0.26115477085113525,
      "learning_rate": 4.610398872216503e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 2017184,
      "step": 7205
    },
    {
      "epoch": 80.11111111111111,
      "grad_norm": 0.11415518075227737,
      "learning_rate": 4.6098724023931796e-05,
      "loss": 0.0128,
      "num_input_tokens_seen": 2018592,
      "step": 7210
    },
    {
      "epoch": 80.16666666666667,
      "grad_norm": 2.8869028091430664,
      "learning_rate": 4.609345607201017e-05,
      "loss": 0.0611,
      "num_input_tokens_seen": 2020000,
      "step": 7215
    },
    {
      "epoch": 80.22222222222223,
      "grad_norm": 1.0391175746917725,
      "learning_rate": 4.608818486721254e-05,
      "loss": 0.0395,
      "num_input_tokens_seen": 2021440,
      "step": 7220
    },
    {
      "epoch": 80.27777777777777,
      "grad_norm": 1.9723145961761475,
      "learning_rate": 4.608291041035179e-05,
      "loss": 0.0596,
      "num_input_tokens_seen": 2022832,
      "step": 7225
    },
    {
      "epoch": 80.33333333333333,
      "grad_norm": 1.285016417503357,
      "learning_rate": 4.607763270224132e-05,
      "loss": 0.0246,
      "num_input_tokens_seen": 2024224,
      "step": 7230
    },
    {
      "epoch": 80.38888888888889,
      "grad_norm": 0.3750348389148712,
      "learning_rate": 4.6072351743695e-05,
      "loss": 0.0082,
      "num_input_tokens_seen": 2025632,
      "step": 7235
    },
    {
      "epoch": 80.44444444444444,
      "grad_norm": 0.7051460146903992,
      "learning_rate": 4.606706753552723e-05,
      "loss": 0.0154,
      "num_input_tokens_seen": 2027008,
      "step": 7240
    },
    {
      "epoch": 80.5,
      "grad_norm": 0.2548629343509674,
      "learning_rate": 4.6061780078552906e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 2028448,
      "step": 7245
    },
    {
      "epoch": 80.55555555555556,
      "grad_norm": 3.02944016456604,
      "learning_rate": 4.605648937358742e-05,
      "loss": 0.1105,
      "num_input_tokens_seen": 2029824,
      "step": 7250
    },
    {
      "epoch": 80.61111111111111,
      "grad_norm": 0.7543566823005676,
      "learning_rate": 4.605119542144665e-05,
      "loss": 0.0857,
      "num_input_tokens_seen": 2031200,
      "step": 7255
    },
    {
      "epoch": 80.66666666666667,
      "grad_norm": 0.9777029156684875,
      "learning_rate": 4.604589822294701e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 2032624,
      "step": 7260
    },
    {
      "epoch": 80.72222222222223,
      "grad_norm": 0.0741032287478447,
      "learning_rate": 4.604059777890537e-05,
      "loss": 0.018,
      "num_input_tokens_seen": 2034016,
      "step": 7265
    },
    {
      "epoch": 80.77777777777777,
      "grad_norm": 0.7893369793891907,
      "learning_rate": 4.6035294090139145e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 2035440,
      "step": 7270
    },
    {
      "epoch": 80.83333333333333,
      "grad_norm": 3.007460832595825,
      "learning_rate": 4.6029987157466226e-05,
      "loss": 0.0938,
      "num_input_tokens_seen": 2036832,
      "step": 7275
    },
    {
      "epoch": 80.88888888888889,
      "grad_norm": 0.7389900088310242,
      "learning_rate": 4.602467698170502e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 2038192,
      "step": 7280
    },
    {
      "epoch": 80.94444444444444,
      "grad_norm": 0.4149748980998993,
      "learning_rate": 4.601936356367439e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 2039584,
      "step": 7285
    },
    {
      "epoch": 81.0,
      "grad_norm": 0.9684134721755981,
      "learning_rate": 4.601404690419377e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 2040992,
      "step": 7290
    },
    {
      "epoch": 81.05555555555556,
      "grad_norm": 0.5965757966041565,
      "learning_rate": 4.600872700408303e-05,
      "loss": 0.0184,
      "num_input_tokens_seen": 2042352,
      "step": 7295
    },
    {
      "epoch": 81.11111111111111,
      "grad_norm": 1.0395629405975342,
      "learning_rate": 4.600340386416258e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 2043776,
      "step": 7300
    },
    {
      "epoch": 81.16666666666667,
      "grad_norm": 0.031806062906980515,
      "learning_rate": 4.5998077485253296e-05,
      "loss": 0.0381,
      "num_input_tokens_seen": 2045168,
      "step": 7305
    },
    {
      "epoch": 81.22222222222223,
      "grad_norm": 4.503859996795654,
      "learning_rate": 4.59927478681766e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 2046592,
      "step": 7310
    },
    {
      "epoch": 81.27777777777777,
      "grad_norm": 1.3050763607025146,
      "learning_rate": 4.5987415013754366e-05,
      "loss": 0.0095,
      "num_input_tokens_seen": 2047984,
      "step": 7315
    },
    {
      "epoch": 81.33333333333333,
      "grad_norm": 0.05079346522688866,
      "learning_rate": 4.598207892280899e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 2049408,
      "step": 7320
    },
    {
      "epoch": 81.38888888888889,
      "grad_norm": 0.8401999473571777,
      "learning_rate": 4.597673959616337e-05,
      "loss": 0.0514,
      "num_input_tokens_seen": 2050800,
      "step": 7325
    },
    {
      "epoch": 81.44444444444444,
      "grad_norm": 5.691809177398682,
      "learning_rate": 4.597139703464089e-05,
      "loss": 0.0923,
      "num_input_tokens_seen": 2052224,
      "step": 7330
    },
    {
      "epoch": 81.5,
      "grad_norm": 3.263793706893921,
      "learning_rate": 4.596605123906545e-05,
      "loss": 0.0704,
      "num_input_tokens_seen": 2053616,
      "step": 7335
    },
    {
      "epoch": 81.55555555555556,
      "grad_norm": 0.6804566383361816,
      "learning_rate": 4.596070221026143e-05,
      "loss": 0.083,
      "num_input_tokens_seen": 2055008,
      "step": 7340
    },
    {
      "epoch": 81.61111111111111,
      "grad_norm": 0.33344510197639465,
      "learning_rate": 4.595534994905372e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 2056416,
      "step": 7345
    },
    {
      "epoch": 81.66666666666667,
      "grad_norm": 1.3658316135406494,
      "learning_rate": 4.594999445626771e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 2057856,
      "step": 7350
    },
    {
      "epoch": 81.72222222222223,
      "grad_norm": 2.394937753677368,
      "learning_rate": 4.5944635732729276e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 2059248,
      "step": 7355
    },
    {
      "epoch": 81.77777777777777,
      "grad_norm": 0.35157114267349243,
      "learning_rate": 4.5939273779264804e-05,
      "loss": 0.1232,
      "num_input_tokens_seen": 2060624,
      "step": 7360
    },
    {
      "epoch": 81.83333333333333,
      "grad_norm": 0.047742437571287155,
      "learning_rate": 4.593390859670118e-05,
      "loss": 0.0579,
      "num_input_tokens_seen": 2062000,
      "step": 7365
    },
    {
      "epoch": 81.88888888888889,
      "grad_norm": 1.5884450674057007,
      "learning_rate": 4.5928540185865776e-05,
      "loss": 0.0633,
      "num_input_tokens_seen": 2063408,
      "step": 7370
    },
    {
      "epoch": 81.94444444444444,
      "grad_norm": 0.8605611324310303,
      "learning_rate": 4.592316854758648e-05,
      "loss": 0.0103,
      "num_input_tokens_seen": 2064816,
      "step": 7375
    },
    {
      "epoch": 82.0,
      "grad_norm": 1.5427957773208618,
      "learning_rate": 4.5917793682691646e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 2066240,
      "step": 7380
    },
    {
      "epoch": 82.05555555555556,
      "grad_norm": 0.13009324669837952,
      "learning_rate": 4.5912415592010164e-05,
      "loss": 0.0188,
      "num_input_tokens_seen": 2067632,
      "step": 7385
    },
    {
      "epoch": 82.11111111111111,
      "grad_norm": 3.585653781890869,
      "learning_rate": 4.5907034276371386e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 2069008,
      "step": 7390
    },
    {
      "epoch": 82.16666666666667,
      "grad_norm": 0.057500846683979034,
      "learning_rate": 4.5901649736605196e-05,
      "loss": 0.0767,
      "num_input_tokens_seen": 2070400,
      "step": 7395
    },
    {
      "epoch": 82.22222222222223,
      "grad_norm": 5.371120929718018,
      "learning_rate": 4.589626197354195e-05,
      "loss": 0.0327,
      "num_input_tokens_seen": 2071808,
      "step": 7400
    },
    {
      "epoch": 82.22222222222223,
      "eval_loss": 0.2978220582008362,
      "eval_runtime": 1.2347,
      "eval_samples_per_second": 32.395,
      "eval_steps_per_second": 16.198,
      "num_input_tokens_seen": 2071808,
      "step": 7400
    },
    {
      "epoch": 82.27777777777777,
      "grad_norm": 0.5219595432281494,
      "learning_rate": 4.5890870988012504e-05,
      "loss": 0.0145,
      "num_input_tokens_seen": 2073216,
      "step": 7405
    },
    {
      "epoch": 82.33333333333333,
      "grad_norm": 0.22371527552604675,
      "learning_rate": 4.5885476780848226e-05,
      "loss": 0.0483,
      "num_input_tokens_seen": 2074624,
      "step": 7410
    },
    {
      "epoch": 82.38888888888889,
      "grad_norm": 5.316195964813232,
      "learning_rate": 4.5880079352880964e-05,
      "loss": 0.1707,
      "num_input_tokens_seen": 2076000,
      "step": 7415
    },
    {
      "epoch": 82.44444444444444,
      "grad_norm": 5.506561279296875,
      "learning_rate": 4.5874678704943065e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 2077424,
      "step": 7420
    },
    {
      "epoch": 82.5,
      "grad_norm": 0.10838832706212997,
      "learning_rate": 4.5869274837867394e-05,
      "loss": 0.0174,
      "num_input_tokens_seen": 2078832,
      "step": 7425
    },
    {
      "epoch": 82.55555555555556,
      "grad_norm": 2.853600263595581,
      "learning_rate": 4.5863867752487275e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 2080240,
      "step": 7430
    },
    {
      "epoch": 82.61111111111111,
      "grad_norm": 0.08770353347063065,
      "learning_rate": 4.5858457449636554e-05,
      "loss": 0.0668,
      "num_input_tokens_seen": 2081632,
      "step": 7435
    },
    {
      "epoch": 82.66666666666667,
      "grad_norm": 0.026440666988492012,
      "learning_rate": 4.5853043930149574e-05,
      "loss": 0.0932,
      "num_input_tokens_seen": 2083056,
      "step": 7440
    },
    {
      "epoch": 82.72222222222223,
      "grad_norm": 0.8145642280578613,
      "learning_rate": 4.584762719486117e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 2084416,
      "step": 7445
    },
    {
      "epoch": 82.77777777777777,
      "grad_norm": 1.0377923250198364,
      "learning_rate": 4.584220724460665e-05,
      "loss": 0.0311,
      "num_input_tokens_seen": 2085808,
      "step": 7450
    },
    {
      "epoch": 82.83333333333333,
      "grad_norm": 0.09003932029008865,
      "learning_rate": 4.5836784080221865e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 2087216,
      "step": 7455
    },
    {
      "epoch": 82.88888888888889,
      "grad_norm": 0.37713316082954407,
      "learning_rate": 4.583135770254312e-05,
      "loss": 0.0591,
      "num_input_tokens_seen": 2088544,
      "step": 7460
    },
    {
      "epoch": 82.94444444444444,
      "grad_norm": 0.16251195967197418,
      "learning_rate": 4.5825928112407236e-05,
      "loss": 0.0226,
      "num_input_tokens_seen": 2089920,
      "step": 7465
    },
    {
      "epoch": 83.0,
      "grad_norm": 0.17921499907970428,
      "learning_rate": 4.582049531065152e-05,
      "loss": 0.0927,
      "num_input_tokens_seen": 2091376,
      "step": 7470
    },
    {
      "epoch": 83.05555555555556,
      "grad_norm": 0.7174181342124939,
      "learning_rate": 4.5815059298113783e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 2092784,
      "step": 7475
    },
    {
      "epoch": 83.11111111111111,
      "grad_norm": 1.9205470085144043,
      "learning_rate": 4.580962007563232e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 2094192,
      "step": 7480
    },
    {
      "epoch": 83.16666666666667,
      "grad_norm": 2.1464123725891113,
      "learning_rate": 4.5804177644045935e-05,
      "loss": 0.0104,
      "num_input_tokens_seen": 2095568,
      "step": 7485
    },
    {
      "epoch": 83.22222222222223,
      "grad_norm": 0.0709713026881218,
      "learning_rate": 4.579873200419391e-05,
      "loss": 0.0389,
      "num_input_tokens_seen": 2096960,
      "step": 7490
    },
    {
      "epoch": 83.27777777777777,
      "grad_norm": 0.46895086765289307,
      "learning_rate": 4.5793283156916046e-05,
      "loss": 0.0078,
      "num_input_tokens_seen": 2098320,
      "step": 7495
    },
    {
      "epoch": 83.33333333333333,
      "grad_norm": 0.33778664469718933,
      "learning_rate": 4.578783110305261e-05,
      "loss": 0.1168,
      "num_input_tokens_seen": 2099744,
      "step": 7500
    },
    {
      "epoch": 83.38888888888889,
      "grad_norm": 0.12044449895620346,
      "learning_rate": 4.578237584344438e-05,
      "loss": 0.0171,
      "num_input_tokens_seen": 2101168,
      "step": 7505
    },
    {
      "epoch": 83.44444444444444,
      "grad_norm": 0.49000948667526245,
      "learning_rate": 4.577691737893263e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 2102576,
      "step": 7510
    },
    {
      "epoch": 83.5,
      "grad_norm": 0.327290415763855,
      "learning_rate": 4.577145571035912e-05,
      "loss": 0.0083,
      "num_input_tokens_seen": 2103984,
      "step": 7515
    },
    {
      "epoch": 83.55555555555556,
      "grad_norm": 0.45618966221809387,
      "learning_rate": 4.576599083856611e-05,
      "loss": 0.04,
      "num_input_tokens_seen": 2105392,
      "step": 7520
    },
    {
      "epoch": 83.61111111111111,
      "grad_norm": 0.013966218568384647,
      "learning_rate": 4.576052276439635e-05,
      "loss": 0.011,
      "num_input_tokens_seen": 2106784,
      "step": 7525
    },
    {
      "epoch": 83.66666666666667,
      "grad_norm": 7.259743690490723,
      "learning_rate": 4.575505148869308e-05,
      "loss": 0.1108,
      "num_input_tokens_seen": 2108128,
      "step": 7530
    },
    {
      "epoch": 83.72222222222223,
      "grad_norm": 0.06010343134403229,
      "learning_rate": 4.574957701230006e-05,
      "loss": 0.072,
      "num_input_tokens_seen": 2109536,
      "step": 7535
    },
    {
      "epoch": 83.77777777777777,
      "grad_norm": 0.0559895783662796,
      "learning_rate": 4.57440993360615e-05,
      "loss": 0.046,
      "num_input_tokens_seen": 2110944,
      "step": 7540
    },
    {
      "epoch": 83.83333333333333,
      "grad_norm": 1.945332646369934,
      "learning_rate": 4.5738618460822134e-05,
      "loss": 0.0614,
      "num_input_tokens_seen": 2112368,
      "step": 7545
    },
    {
      "epoch": 83.88888888888889,
      "grad_norm": 0.38141438364982605,
      "learning_rate": 4.573313438742719e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 2113792,
      "step": 7550
    },
    {
      "epoch": 83.94444444444444,
      "grad_norm": 0.6491513848304749,
      "learning_rate": 4.5727647116722374e-05,
      "loss": 0.0483,
      "num_input_tokens_seen": 2115168,
      "step": 7555
    },
    {
      "epoch": 84.0,
      "grad_norm": 1.425256609916687,
      "learning_rate": 4.5722156649553884e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 2116592,
      "step": 7560
    },
    {
      "epoch": 84.05555555555556,
      "grad_norm": 1.342366099357605,
      "learning_rate": 4.571666298676843e-05,
      "loss": 0.0125,
      "num_input_tokens_seen": 2117984,
      "step": 7565
    },
    {
      "epoch": 84.11111111111111,
      "grad_norm": 0.044689077883958817,
      "learning_rate": 4.571116612921321e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 2119392,
      "step": 7570
    },
    {
      "epoch": 84.16666666666667,
      "grad_norm": 0.04030932858586311,
      "learning_rate": 4.57056660777359e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 2120800,
      "step": 7575
    },
    {
      "epoch": 84.22222222222223,
      "grad_norm": 0.08591393381357193,
      "learning_rate": 4.5700162833184666e-05,
      "loss": 0.1091,
      "num_input_tokens_seen": 2122192,
      "step": 7580
    },
    {
      "epoch": 84.27777777777777,
      "grad_norm": 5.219066143035889,
      "learning_rate": 4.5694656396408195e-05,
      "loss": 0.014,
      "num_input_tokens_seen": 2123600,
      "step": 7585
    },
    {
      "epoch": 84.33333333333333,
      "grad_norm": 4.945002555847168,
      "learning_rate": 4.5689146768255646e-05,
      "loss": 0.0655,
      "num_input_tokens_seen": 2125056,
      "step": 7590
    },
    {
      "epoch": 84.38888888888889,
      "grad_norm": 0.02984856627881527,
      "learning_rate": 4.568363394957667e-05,
      "loss": 0.0567,
      "num_input_tokens_seen": 2126448,
      "step": 7595
    },
    {
      "epoch": 84.44444444444444,
      "grad_norm": 0.44269341230392456,
      "learning_rate": 4.567811794122141e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 2127808,
      "step": 7600
    },
    {
      "epoch": 84.44444444444444,
      "eval_loss": 0.3489944040775299,
      "eval_runtime": 1.2359,
      "eval_samples_per_second": 32.366,
      "eval_steps_per_second": 16.183,
      "num_input_tokens_seen": 2127808,
      "step": 7600
    },
    {
      "epoch": 84.5,
      "grad_norm": 1.8452600240707397,
      "learning_rate": 4.56725987440405e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 2129248,
      "step": 7605
    },
    {
      "epoch": 84.55555555555556,
      "grad_norm": 1.2805949449539185,
      "learning_rate": 4.566707635888508e-05,
      "loss": 0.0073,
      "num_input_tokens_seen": 2130688,
      "step": 7610
    },
    {
      "epoch": 84.61111111111111,
      "grad_norm": 0.6996960043907166,
      "learning_rate": 4.566155078660677e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 2132064,
      "step": 7615
    },
    {
      "epoch": 84.66666666666667,
      "grad_norm": 0.19927649199962616,
      "learning_rate": 4.565602202805768e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 2133408,
      "step": 7620
    },
    {
      "epoch": 84.72222222222223,
      "grad_norm": 0.2778151035308838,
      "learning_rate": 4.56504900840904e-05,
      "loss": 0.1083,
      "num_input_tokens_seen": 2134784,
      "step": 7625
    },
    {
      "epoch": 84.77777777777777,
      "grad_norm": 1.197373628616333,
      "learning_rate": 4.564495495555805e-05,
      "loss": 0.0175,
      "num_input_tokens_seen": 2136224,
      "step": 7630
    },
    {
      "epoch": 84.83333333333333,
      "grad_norm": 0.05425238236784935,
      "learning_rate": 4.5639416643314204e-05,
      "loss": 0.0068,
      "num_input_tokens_seen": 2137616,
      "step": 7635
    },
    {
      "epoch": 84.88888888888889,
      "grad_norm": 2.361281394958496,
      "learning_rate": 4.5633875148212946e-05,
      "loss": 0.01,
      "num_input_tokens_seen": 2139040,
      "step": 7640
    },
    {
      "epoch": 84.94444444444444,
      "grad_norm": 0.6245580911636353,
      "learning_rate": 4.562833047110883e-05,
      "loss": 0.087,
      "num_input_tokens_seen": 2140416,
      "step": 7645
    },
    {
      "epoch": 85.0,
      "grad_norm": 0.08936631679534912,
      "learning_rate": 4.5622782612856923e-05,
      "loss": 0.0185,
      "num_input_tokens_seen": 2141824,
      "step": 7650
    },
    {
      "epoch": 85.05555555555556,
      "grad_norm": 2.2738096714019775,
      "learning_rate": 4.561723157431278e-05,
      "loss": 0.0537,
      "num_input_tokens_seen": 2143248,
      "step": 7655
    },
    {
      "epoch": 85.11111111111111,
      "grad_norm": 0.08944666385650635,
      "learning_rate": 4.5611677356332435e-05,
      "loss": 0.0189,
      "num_input_tokens_seen": 2144656,
      "step": 7660
    },
    {
      "epoch": 85.16666666666667,
      "grad_norm": 0.11182302981615067,
      "learning_rate": 4.560611995977242e-05,
      "loss": 0.0706,
      "num_input_tokens_seen": 2146080,
      "step": 7665
    },
    {
      "epoch": 85.22222222222223,
      "grad_norm": 0.2362109273672104,
      "learning_rate": 4.560055938548975e-05,
      "loss": 0.0079,
      "num_input_tokens_seen": 2147488,
      "step": 7670
    },
    {
      "epoch": 85.27777777777777,
      "grad_norm": 0.49482521414756775,
      "learning_rate": 4.5594995634341944e-05,
      "loss": 0.01,
      "num_input_tokens_seen": 2148864,
      "step": 7675
    },
    {
      "epoch": 85.33333333333333,
      "grad_norm": 0.3898817002773285,
      "learning_rate": 4.5589428707187e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 2150272,
      "step": 7680
    },
    {
      "epoch": 85.38888888888889,
      "grad_norm": 4.288393020629883,
      "learning_rate": 4.55838586048834e-05,
      "loss": 0.058,
      "num_input_tokens_seen": 2151664,
      "step": 7685
    },
    {
      "epoch": 85.44444444444444,
      "grad_norm": 5.182615756988525,
      "learning_rate": 4.557828532829013e-05,
      "loss": 0.0519,
      "num_input_tokens_seen": 2153088,
      "step": 7690
    },
    {
      "epoch": 85.5,
      "grad_norm": 0.017723120748996735,
      "learning_rate": 4.557270887826667e-05,
      "loss": 0.0067,
      "num_input_tokens_seen": 2154512,
      "step": 7695
    },
    {
      "epoch": 85.55555555555556,
      "grad_norm": 0.09716616570949554,
      "learning_rate": 4.556712925567296e-05,
      "loss": 0.0084,
      "num_input_tokens_seen": 2155904,
      "step": 7700
    },
    {
      "epoch": 85.61111111111111,
      "grad_norm": 9.278054237365723,
      "learning_rate": 4.5561546461369454e-05,
      "loss": 0.0788,
      "num_input_tokens_seen": 2157264,
      "step": 7705
    },
    {
      "epoch": 85.66666666666667,
      "grad_norm": 0.541721761226654,
      "learning_rate": 4.55559604962171e-05,
      "loss": 0.0094,
      "num_input_tokens_seen": 2158688,
      "step": 7710
    },
    {
      "epoch": 85.72222222222223,
      "grad_norm": 0.3779246211051941,
      "learning_rate": 4.55503713610773e-05,
      "loss": 0.0091,
      "num_input_tokens_seen": 2160096,
      "step": 7715
    },
    {
      "epoch": 85.77777777777777,
      "grad_norm": 0.6116267442703247,
      "learning_rate": 4.5544779056812e-05,
      "loss": 0.0741,
      "num_input_tokens_seen": 2161520,
      "step": 7720
    },
    {
      "epoch": 85.83333333333333,
      "grad_norm": 0.8870729207992554,
      "learning_rate": 4.553918358428358e-05,
      "loss": 0.0857,
      "num_input_tokens_seen": 2162928,
      "step": 7725
    },
    {
      "epoch": 85.88888888888889,
      "grad_norm": 2.902494430541992,
      "learning_rate": 4.553358494435494e-05,
      "loss": 0.1191,
      "num_input_tokens_seen": 2164336,
      "step": 7730
    },
    {
      "epoch": 85.94444444444444,
      "grad_norm": 0.10399344563484192,
      "learning_rate": 4.5527983137889464e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 2165760,
      "step": 7735
    },
    {
      "epoch": 86.0,
      "grad_norm": 0.04140431433916092,
      "learning_rate": 4.5522378165751015e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 2167152,
      "step": 7740
    },
    {
      "epoch": 86.05555555555556,
      "grad_norm": 0.17327731847763062,
      "learning_rate": 4.5516770028803954e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 2168544,
      "step": 7745
    },
    {
      "epoch": 86.11111111111111,
      "grad_norm": 0.17158718407154083,
      "learning_rate": 4.5511158727913116e-05,
      "loss": 0.0321,
      "num_input_tokens_seen": 2169904,
      "step": 7750
    },
    {
      "epoch": 86.16666666666667,
      "grad_norm": 0.3454972505569458,
      "learning_rate": 4.5505544263943856e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 2171296,
      "step": 7755
    },
    {
      "epoch": 86.22222222222223,
      "grad_norm": 0.13582585752010345,
      "learning_rate": 4.549992663776197e-05,
      "loss": 0.0498,
      "num_input_tokens_seen": 2172704,
      "step": 7760
    },
    {
      "epoch": 86.27777777777777,
      "grad_norm": 0.53038489818573,
      "learning_rate": 4.5494305850233786e-05,
      "loss": 0.0155,
      "num_input_tokens_seen": 2174112,
      "step": 7765
    },
    {
      "epoch": 86.33333333333333,
      "grad_norm": 5.468447208404541,
      "learning_rate": 4.5488681902226094e-05,
      "loss": 0.0524,
      "num_input_tokens_seen": 2175520,
      "step": 7770
    },
    {
      "epoch": 86.38888888888889,
      "grad_norm": 0.06875092536211014,
      "learning_rate": 4.5483054794606174e-05,
      "loss": 0.0453,
      "num_input_tokens_seen": 2176928,
      "step": 7775
    },
    {
      "epoch": 86.44444444444444,
      "grad_norm": 1.2705844640731812,
      "learning_rate": 4.547742452824179e-05,
      "loss": 0.1131,
      "num_input_tokens_seen": 2178320,
      "step": 7780
    },
    {
      "epoch": 86.5,
      "grad_norm": 0.07551468163728714,
      "learning_rate": 4.5471791104001215e-05,
      "loss": 0.0621,
      "num_input_tokens_seen": 2179728,
      "step": 7785
    },
    {
      "epoch": 86.55555555555556,
      "grad_norm": 0.16361579298973083,
      "learning_rate": 4.546615452275319e-05,
      "loss": 0.0108,
      "num_input_tokens_seen": 2181088,
      "step": 7790
    },
    {
      "epoch": 86.61111111111111,
      "grad_norm": 6.19449520111084,
      "learning_rate": 4.5460514785366944e-05,
      "loss": 0.1009,
      "num_input_tokens_seen": 2182480,
      "step": 7795
    },
    {
      "epoch": 86.66666666666667,
      "grad_norm": 0.04207031801342964,
      "learning_rate": 4.545487189271219e-05,
      "loss": 0.0092,
      "num_input_tokens_seen": 2183888,
      "step": 7800
    },
    {
      "epoch": 86.66666666666667,
      "eval_loss": 0.2913166880607605,
      "eval_runtime": 1.2489,
      "eval_samples_per_second": 32.027,
      "eval_steps_per_second": 16.014,
      "num_input_tokens_seen": 2183888,
      "step": 7800
    },
    {
      "epoch": 86.72222222222223,
      "grad_norm": 0.117640919983387,
      "learning_rate": 4.544922584565914e-05,
      "loss": 0.0169,
      "num_input_tokens_seen": 2185296,
      "step": 7805
    },
    {
      "epoch": 86.77777777777777,
      "grad_norm": 5.272991180419922,
      "learning_rate": 4.544357664507848e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 2186672,
      "step": 7810
    },
    {
      "epoch": 86.83333333333333,
      "grad_norm": 2.933056592941284,
      "learning_rate": 4.54379242918414e-05,
      "loss": 0.0536,
      "num_input_tokens_seen": 2188128,
      "step": 7815
    },
    {
      "epoch": 86.88888888888889,
      "grad_norm": 0.4605761170387268,
      "learning_rate": 4.543226878681955e-05,
      "loss": 0.0121,
      "num_input_tokens_seen": 2189536,
      "step": 7820
    },
    {
      "epoch": 86.94444444444444,
      "grad_norm": 0.4482252895832062,
      "learning_rate": 4.5426610130885087e-05,
      "loss": 0.0122,
      "num_input_tokens_seen": 2190880,
      "step": 7825
    },
    {
      "epoch": 87.0,
      "grad_norm": 3.662970542907715,
      "learning_rate": 4.542094832491064e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 2192304,
      "step": 7830
    },
    {
      "epoch": 87.05555555555556,
      "grad_norm": 0.6232876181602478,
      "learning_rate": 4.541528336976934e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 2193696,
      "step": 7835
    },
    {
      "epoch": 87.11111111111111,
      "grad_norm": 3.6465156078338623,
      "learning_rate": 4.540961526633479e-05,
      "loss": 0.0102,
      "num_input_tokens_seen": 2195104,
      "step": 7840
    },
    {
      "epoch": 87.16666666666667,
      "grad_norm": 0.026303164660930634,
      "learning_rate": 4.540394401548108e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 2196480,
      "step": 7845
    },
    {
      "epoch": 87.22222222222223,
      "grad_norm": 0.2828761637210846,
      "learning_rate": 4.539826961808279e-05,
      "loss": 0.0208,
      "num_input_tokens_seen": 2197856,
      "step": 7850
    },
    {
      "epoch": 87.27777777777777,
      "grad_norm": 0.21683776378631592,
      "learning_rate": 4.5392592075014994e-05,
      "loss": 0.0183,
      "num_input_tokens_seen": 2199248,
      "step": 7855
    },
    {
      "epoch": 87.33333333333333,
      "grad_norm": 0.09211771935224533,
      "learning_rate": 4.538691138715322e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 2200640,
      "step": 7860
    },
    {
      "epoch": 87.38888888888889,
      "grad_norm": 0.13484185934066772,
      "learning_rate": 4.5381227555373516e-05,
      "loss": 0.0787,
      "num_input_tokens_seen": 2202048,
      "step": 7865
    },
    {
      "epoch": 87.44444444444444,
      "grad_norm": 6.227532863616943,
      "learning_rate": 4.537554058055239e-05,
      "loss": 0.073,
      "num_input_tokens_seen": 2203472,
      "step": 7870
    },
    {
      "epoch": 87.5,
      "grad_norm": 0.5693550705909729,
      "learning_rate": 4.5369850463566865e-05,
      "loss": 0.0227,
      "num_input_tokens_seen": 2204864,
      "step": 7875
    },
    {
      "epoch": 87.55555555555556,
      "grad_norm": 0.5628534555435181,
      "learning_rate": 4.5364157205294404e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 2206288,
      "step": 7880
    },
    {
      "epoch": 87.61111111111111,
      "grad_norm": 0.8837116956710815,
      "learning_rate": 4.5358460806612996e-05,
      "loss": 0.0457,
      "num_input_tokens_seen": 2207728,
      "step": 7885
    },
    {
      "epoch": 87.66666666666667,
      "grad_norm": 0.1440500169992447,
      "learning_rate": 4.535276126840109e-05,
      "loss": 0.0049,
      "num_input_tokens_seen": 2209136,
      "step": 7890
    },
    {
      "epoch": 87.72222222222223,
      "grad_norm": 3.543720245361328,
      "learning_rate": 4.5347058591537626e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 2210528,
      "step": 7895
    },
    {
      "epoch": 87.77777777777777,
      "grad_norm": 0.04739097133278847,
      "learning_rate": 4.534135277690203e-05,
      "loss": 0.0657,
      "num_input_tokens_seen": 2211904,
      "step": 7900
    },
    {
      "epoch": 87.83333333333333,
      "grad_norm": 1.1597081422805786,
      "learning_rate": 4.533564382537421e-05,
      "loss": 0.0565,
      "num_input_tokens_seen": 2213264,
      "step": 7905
    },
    {
      "epoch": 87.88888888888889,
      "grad_norm": 0.12793423235416412,
      "learning_rate": 4.532993173783456e-05,
      "loss": 0.0145,
      "num_input_tokens_seen": 2214688,
      "step": 7910
    },
    {
      "epoch": 87.94444444444444,
      "grad_norm": 0.8803775906562805,
      "learning_rate": 4.5324216515163954e-05,
      "loss": 0.1101,
      "num_input_tokens_seen": 2216080,
      "step": 7915
    },
    {
      "epoch": 88.0,
      "grad_norm": 0.19455119967460632,
      "learning_rate": 4.531849815824375e-05,
      "loss": 0.0115,
      "num_input_tokens_seen": 2217488,
      "step": 7920
    },
    {
      "epoch": 88.05555555555556,
      "grad_norm": 0.19425107538700104,
      "learning_rate": 4.5312776667955795e-05,
      "loss": 0.1137,
      "num_input_tokens_seen": 2218864,
      "step": 7925
    },
    {
      "epoch": 88.11111111111111,
      "grad_norm": 0.4260970652103424,
      "learning_rate": 4.5307052045182405e-05,
      "loss": 0.0851,
      "num_input_tokens_seen": 2220240,
      "step": 7930
    },
    {
      "epoch": 88.16666666666667,
      "grad_norm": 5.2164201736450195,
      "learning_rate": 4.53013242908064e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 2221616,
      "step": 7935
    },
    {
      "epoch": 88.22222222222223,
      "grad_norm": 0.6201134920120239,
      "learning_rate": 4.529559340571107e-05,
      "loss": 0.0647,
      "num_input_tokens_seen": 2222992,
      "step": 7940
    },
    {
      "epoch": 88.27777777777777,
      "grad_norm": 0.3933674097061157,
      "learning_rate": 4.528985939078018e-05,
      "loss": 0.0569,
      "num_input_tokens_seen": 2224368,
      "step": 7945
    },
    {
      "epoch": 88.33333333333333,
      "grad_norm": 1.5848501920700073,
      "learning_rate": 4.5284122246898e-05,
      "loss": 0.0084,
      "num_input_tokens_seen": 2225776,
      "step": 7950
    },
    {
      "epoch": 88.38888888888889,
      "grad_norm": 5.099526405334473,
      "learning_rate": 4.527838197494926e-05,
      "loss": 0.0194,
      "num_input_tokens_seen": 2227184,
      "step": 7955
    },
    {
      "epoch": 88.44444444444444,
      "grad_norm": 0.03792404755949974,
      "learning_rate": 4.527263857581918e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 2228608,
      "step": 7960
    },
    {
      "epoch": 88.5,
      "grad_norm": 0.029749037697911263,
      "learning_rate": 4.526689205039347e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 2230032,
      "step": 7965
    },
    {
      "epoch": 88.55555555555556,
      "grad_norm": 3.5816650390625,
      "learning_rate": 4.5261142399558324e-05,
      "loss": 0.0098,
      "num_input_tokens_seen": 2231392,
      "step": 7970
    },
    {
      "epoch": 88.61111111111111,
      "grad_norm": 0.21663852035999298,
      "learning_rate": 4.525538962420041e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 2232816,
      "step": 7975
    },
    {
      "epoch": 88.66666666666667,
      "grad_norm": 1.8013049364089966,
      "learning_rate": 4.524963372520685e-05,
      "loss": 0.1058,
      "num_input_tokens_seen": 2234224,
      "step": 7980
    },
    {
      "epoch": 88.72222222222223,
      "grad_norm": 0.6317217946052551,
      "learning_rate": 4.524387470346531e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 2235664,
      "step": 7985
    },
    {
      "epoch": 88.77777777777777,
      "grad_norm": 0.025745216757059097,
      "learning_rate": 4.5238112559863885e-05,
      "loss": 0.0067,
      "num_input_tokens_seen": 2237040,
      "step": 7990
    },
    {
      "epoch": 88.83333333333333,
      "grad_norm": 0.18677446246147156,
      "learning_rate": 4.5232347295291175e-05,
      "loss": 0.1013,
      "num_input_tokens_seen": 2238432,
      "step": 7995
    },
    {
      "epoch": 88.88888888888889,
      "grad_norm": 0.16231967508792877,
      "learning_rate": 4.522657891063626e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 2239840,
      "step": 8000
    },
    {
      "epoch": 88.88888888888889,
      "eval_loss": 0.2781406342983246,
      "eval_runtime": 1.2422,
      "eval_samples_per_second": 32.201,
      "eval_steps_per_second": 16.1,
      "num_input_tokens_seen": 2239840,
      "step": 8000
    },
    {
      "epoch": 88.94444444444444,
      "grad_norm": 0.11868653446435928,
      "learning_rate": 4.52208074067887e-05,
      "loss": 0.0085,
      "num_input_tokens_seen": 2241200,
      "step": 8005
    },
    {
      "epoch": 89.0,
      "grad_norm": 0.042828574776649475,
      "learning_rate": 4.5215032784638516e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 2242608,
      "step": 8010
    },
    {
      "epoch": 89.05555555555556,
      "grad_norm": 0.33231619000434875,
      "learning_rate": 4.5209255045076245e-05,
      "loss": 0.064,
      "num_input_tokens_seen": 2244016,
      "step": 8015
    },
    {
      "epoch": 89.11111111111111,
      "grad_norm": 1.9947385787963867,
      "learning_rate": 4.5203474188992875e-05,
      "loss": 0.0092,
      "num_input_tokens_seen": 2245440,
      "step": 8020
    },
    {
      "epoch": 89.16666666666667,
      "grad_norm": 0.13118405640125275,
      "learning_rate": 4.51976902172799e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 2246832,
      "step": 8025
    },
    {
      "epoch": 89.22222222222223,
      "grad_norm": 9.775251388549805,
      "learning_rate": 4.519190313082927e-05,
      "loss": 0.124,
      "num_input_tokens_seen": 2248176,
      "step": 8030
    },
    {
      "epoch": 89.27777777777777,
      "grad_norm": 5.855961799621582,
      "learning_rate": 4.518611293053343e-05,
      "loss": 0.0815,
      "num_input_tokens_seen": 2249568,
      "step": 8035
    },
    {
      "epoch": 89.33333333333333,
      "grad_norm": 0.04057830944657326,
      "learning_rate": 4.51803196172853e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 2250976,
      "step": 8040
    },
    {
      "epoch": 89.38888888888889,
      "grad_norm": 5.95421028137207,
      "learning_rate": 4.517452319197828e-05,
      "loss": 0.0165,
      "num_input_tokens_seen": 2252384,
      "step": 8045
    },
    {
      "epoch": 89.44444444444444,
      "grad_norm": 0.0316832959651947,
      "learning_rate": 4.5168723655506265e-05,
      "loss": 0.0543,
      "num_input_tokens_seen": 2253792,
      "step": 8050
    },
    {
      "epoch": 89.5,
      "grad_norm": 2.1655983924865723,
      "learning_rate": 4.51629210087636e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 2255168,
      "step": 8055
    },
    {
      "epoch": 89.55555555555556,
      "grad_norm": 0.15970677137374878,
      "learning_rate": 4.515711525264513e-05,
      "loss": 0.0916,
      "num_input_tokens_seen": 2256624,
      "step": 8060
    },
    {
      "epoch": 89.61111111111111,
      "grad_norm": 0.027560163289308548,
      "learning_rate": 4.5151306388046175e-05,
      "loss": 0.0058,
      "num_input_tokens_seen": 2258016,
      "step": 8065
    },
    {
      "epoch": 89.66666666666667,
      "grad_norm": 2.3827106952667236,
      "learning_rate": 4.514549441586255e-05,
      "loss": 0.0094,
      "num_input_tokens_seen": 2259456,
      "step": 8070
    },
    {
      "epoch": 89.72222222222223,
      "grad_norm": 0.10553905367851257,
      "learning_rate": 4.513967933699051e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 2260880,
      "step": 8075
    },
    {
      "epoch": 89.77777777777777,
      "grad_norm": 0.26925837993621826,
      "learning_rate": 4.513386115232684e-05,
      "loss": 0.0082,
      "num_input_tokens_seen": 2262272,
      "step": 8080
    },
    {
      "epoch": 89.83333333333333,
      "grad_norm": 2.130699396133423,
      "learning_rate": 4.5128039862768745e-05,
      "loss": 0.0101,
      "num_input_tokens_seen": 2263632,
      "step": 8085
    },
    {
      "epoch": 89.88888888888889,
      "grad_norm": 0.2795388996601105,
      "learning_rate": 4.512221546921397e-05,
      "loss": 0.0175,
      "num_input_tokens_seen": 2265008,
      "step": 8090
    },
    {
      "epoch": 89.94444444444444,
      "grad_norm": 4.278561115264893,
      "learning_rate": 4.5116387972560694e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 2266400,
      "step": 8095
    },
    {
      "epoch": 90.0,
      "grad_norm": 0.080086350440979,
      "learning_rate": 4.511055737370759e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 2267824,
      "step": 8100
    },
    {
      "epoch": 90.05555555555556,
      "grad_norm": 0.1213769018650055,
      "learning_rate": 4.510472367355383e-05,
      "loss": 0.0556,
      "num_input_tokens_seen": 2269232,
      "step": 8105
    },
    {
      "epoch": 90.11111111111111,
      "grad_norm": 0.10914911329746246,
      "learning_rate": 4.509888687299901e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2270608,
      "step": 8110
    },
    {
      "epoch": 90.16666666666667,
      "grad_norm": 0.09412660449743271,
      "learning_rate": 4.5093046972943266e-05,
      "loss": 0.054,
      "num_input_tokens_seen": 2272048,
      "step": 8115
    },
    {
      "epoch": 90.22222222222223,
      "grad_norm": 0.06966418772935867,
      "learning_rate": 4.508720397428717e-05,
      "loss": 0.0078,
      "num_input_tokens_seen": 2273440,
      "step": 8120
    },
    {
      "epoch": 90.27777777777777,
      "grad_norm": 0.061090752482414246,
      "learning_rate": 4.508135787793178e-05,
      "loss": 0.0081,
      "num_input_tokens_seen": 2274832,
      "step": 8125
    },
    {
      "epoch": 90.33333333333333,
      "grad_norm": 2.301217555999756,
      "learning_rate": 4.5075508684778664e-05,
      "loss": 0.0776,
      "num_input_tokens_seen": 2276256,
      "step": 8130
    },
    {
      "epoch": 90.38888888888889,
      "grad_norm": 0.04372384026646614,
      "learning_rate": 4.506965639572982e-05,
      "loss": 0.0072,
      "num_input_tokens_seen": 2277648,
      "step": 8135
    },
    {
      "epoch": 90.44444444444444,
      "grad_norm": 1.3433705568313599,
      "learning_rate": 4.506380101168774e-05,
      "loss": 0.0683,
      "num_input_tokens_seen": 2279072,
      "step": 8140
    },
    {
      "epoch": 90.5,
      "grad_norm": 0.23888665437698364,
      "learning_rate": 4.505794253355542e-05,
      "loss": 0.0481,
      "num_input_tokens_seen": 2280464,
      "step": 8145
    },
    {
      "epoch": 90.55555555555556,
      "grad_norm": 7.858672142028809,
      "learning_rate": 4.5052080962236286e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 2281856,
      "step": 8150
    },
    {
      "epoch": 90.61111111111111,
      "grad_norm": 0.023388074710965157,
      "learning_rate": 4.504621629863428e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 2283280,
      "step": 8155
    },
    {
      "epoch": 90.66666666666667,
      "grad_norm": 0.21148748695850372,
      "learning_rate": 4.504034854365381e-05,
      "loss": 0.0098,
      "num_input_tokens_seen": 2284688,
      "step": 8160
    },
    {
      "epoch": 90.72222222222223,
      "grad_norm": 0.11227100342512131,
      "learning_rate": 4.503447769819974e-05,
      "loss": 0.0493,
      "num_input_tokens_seen": 2286080,
      "step": 8165
    },
    {
      "epoch": 90.77777777777777,
      "grad_norm": 0.5023769736289978,
      "learning_rate": 4.502860376317745e-05,
      "loss": 0.0098,
      "num_input_tokens_seen": 2287456,
      "step": 8170
    },
    {
      "epoch": 90.83333333333333,
      "grad_norm": 0.03941880539059639,
      "learning_rate": 4.502272673949276e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 2288832,
      "step": 8175
    },
    {
      "epoch": 90.88888888888889,
      "grad_norm": 0.1328078955411911,
      "learning_rate": 4.501684662805199e-05,
      "loss": 0.0531,
      "num_input_tokens_seen": 2290272,
      "step": 8180
    },
    {
      "epoch": 90.94444444444444,
      "grad_norm": 0.019440490752458572,
      "learning_rate": 4.5010963429761924e-05,
      "loss": 0.0687,
      "num_input_tokens_seen": 2291664,
      "step": 8185
    },
    {
      "epoch": 91.0,
      "grad_norm": 0.373207151889801,
      "learning_rate": 4.500507714552982e-05,
      "loss": 0.0047,
      "num_input_tokens_seen": 2293088,
      "step": 8190
    },
    {
      "epoch": 91.05555555555556,
      "grad_norm": 0.2351498007774353,
      "learning_rate": 4.499918777626342e-05,
      "loss": 0.0072,
      "num_input_tokens_seen": 2294544,
      "step": 8195
    },
    {
      "epoch": 91.11111111111111,
      "grad_norm": 7.993106842041016,
      "learning_rate": 4.499329532287093e-05,
      "loss": 0.0414,
      "num_input_tokens_seen": 2295888,
      "step": 8200
    },
    {
      "epoch": 91.11111111111111,
      "eval_loss": 0.3033084273338318,
      "eval_runtime": 1.2385,
      "eval_samples_per_second": 32.297,
      "eval_steps_per_second": 16.149,
      "num_input_tokens_seen": 2295888,
      "step": 8200
    },
    {
      "epoch": 91.16666666666667,
      "grad_norm": 0.6031044125556946,
      "learning_rate": 4.4987399786261064e-05,
      "loss": 0.0115,
      "num_input_tokens_seen": 2297232,
      "step": 8205
    },
    {
      "epoch": 91.22222222222223,
      "grad_norm": 1.2883752584457397,
      "learning_rate": 4.498150116734297e-05,
      "loss": 0.064,
      "num_input_tokens_seen": 2298608,
      "step": 8210
    },
    {
      "epoch": 91.27777777777777,
      "grad_norm": 6.47969913482666,
      "learning_rate": 4.4975599467026294e-05,
      "loss": 0.043,
      "num_input_tokens_seen": 2300016,
      "step": 8215
    },
    {
      "epoch": 91.33333333333333,
      "grad_norm": 0.05504029989242554,
      "learning_rate": 4.496969468622114e-05,
      "loss": 0.051,
      "num_input_tokens_seen": 2301408,
      "step": 8220
    },
    {
      "epoch": 91.38888888888889,
      "grad_norm": 0.6602217555046082,
      "learning_rate": 4.496378682583813e-05,
      "loss": 0.0504,
      "num_input_tokens_seen": 2302816,
      "step": 8225
    },
    {
      "epoch": 91.44444444444444,
      "grad_norm": 0.039086099714040756,
      "learning_rate": 4.495787588678829e-05,
      "loss": 0.0868,
      "num_input_tokens_seen": 2304240,
      "step": 8230
    },
    {
      "epoch": 91.5,
      "grad_norm": 0.04597949981689453,
      "learning_rate": 4.4951961869983196e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 2305664,
      "step": 8235
    },
    {
      "epoch": 91.55555555555556,
      "grad_norm": 2.297581911087036,
      "learning_rate": 4.494604477633485e-05,
      "loss": 0.0077,
      "num_input_tokens_seen": 2307024,
      "step": 8240
    },
    {
      "epoch": 91.61111111111111,
      "grad_norm": 4.591582298278809,
      "learning_rate": 4.4940124606755734e-05,
      "loss": 0.014,
      "num_input_tokens_seen": 2308416,
      "step": 8245
    },
    {
      "epoch": 91.66666666666667,
      "grad_norm": 0.09167524427175522,
      "learning_rate": 4.493420136215882e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 2309760,
      "step": 8250
    },
    {
      "epoch": 91.72222222222223,
      "grad_norm": 0.06471364200115204,
      "learning_rate": 4.492827504345756e-05,
      "loss": 0.08,
      "num_input_tokens_seen": 2311184,
      "step": 8255
    },
    {
      "epoch": 91.77777777777777,
      "grad_norm": 0.2073240429162979,
      "learning_rate": 4.492234565156584e-05,
      "loss": 0.028,
      "num_input_tokens_seen": 2312576,
      "step": 8260
    },
    {
      "epoch": 91.83333333333333,
      "grad_norm": 0.2898593246936798,
      "learning_rate": 4.491641318739807e-05,
      "loss": 0.0403,
      "num_input_tokens_seen": 2313952,
      "step": 8265
    },
    {
      "epoch": 91.88888888888889,
      "grad_norm": 0.05487434193491936,
      "learning_rate": 4.4910477651869096e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 2315360,
      "step": 8270
    },
    {
      "epoch": 91.94444444444444,
      "grad_norm": 0.03672988340258598,
      "learning_rate": 4.4904539045894254e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 2316832,
      "step": 8275
    },
    {
      "epoch": 92.0,
      "grad_norm": 0.22766724228858948,
      "learning_rate": 4.4898597370389364e-05,
      "loss": 0.0072,
      "num_input_tokens_seen": 2318256,
      "step": 8280
    },
    {
      "epoch": 92.05555555555556,
      "grad_norm": 0.8292459845542908,
      "learning_rate": 4.489265262627069e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 2319648,
      "step": 8285
    },
    {
      "epoch": 92.11111111111111,
      "grad_norm": 0.23142720758914948,
      "learning_rate": 4.488670481445499e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 2321056,
      "step": 8290
    },
    {
      "epoch": 92.16666666666667,
      "grad_norm": 0.3722792863845825,
      "learning_rate": 4.488075393585951e-05,
      "loss": 0.0594,
      "num_input_tokens_seen": 2322432,
      "step": 8295
    },
    {
      "epoch": 92.22222222222223,
      "grad_norm": 0.01903374120593071,
      "learning_rate": 4.487479999140193e-05,
      "loss": 0.0731,
      "num_input_tokens_seen": 2323840,
      "step": 8300
    },
    {
      "epoch": 92.27777777777777,
      "grad_norm": 0.7337058782577515,
      "learning_rate": 4.4868842982000425e-05,
      "loss": 0.0687,
      "num_input_tokens_seen": 2325216,
      "step": 8305
    },
    {
      "epoch": 92.33333333333333,
      "grad_norm": 0.124300017952919,
      "learning_rate": 4.486288290857365e-05,
      "loss": 0.0446,
      "num_input_tokens_seen": 2326608,
      "step": 8310
    },
    {
      "epoch": 92.38888888888889,
      "grad_norm": 0.5200842022895813,
      "learning_rate": 4.4856919772040715e-05,
      "loss": 0.006,
      "num_input_tokens_seen": 2328048,
      "step": 8315
    },
    {
      "epoch": 92.44444444444444,
      "grad_norm": 0.049474284052848816,
      "learning_rate": 4.485095357332122e-05,
      "loss": 0.0084,
      "num_input_tokens_seen": 2329440,
      "step": 8320
    },
    {
      "epoch": 92.5,
      "grad_norm": 0.05355818569660187,
      "learning_rate": 4.484498431333521e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 2330832,
      "step": 8325
    },
    {
      "epoch": 92.55555555555556,
      "grad_norm": 0.07047166675329208,
      "learning_rate": 4.4839011993003245e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 2332208,
      "step": 8330
    },
    {
      "epoch": 92.61111111111111,
      "grad_norm": 6.106523513793945,
      "learning_rate": 4.4833036613246305e-05,
      "loss": 0.016,
      "num_input_tokens_seen": 2333568,
      "step": 8335
    },
    {
      "epoch": 92.66666666666667,
      "grad_norm": 4.082818508148193,
      "learning_rate": 4.482705817498589e-05,
      "loss": 0.05,
      "num_input_tokens_seen": 2334960,
      "step": 8340
    },
    {
      "epoch": 92.72222222222223,
      "grad_norm": 1.0176461935043335,
      "learning_rate": 4.4821076679143934e-05,
      "loss": 0.0062,
      "num_input_tokens_seen": 2336368,
      "step": 8345
    },
    {
      "epoch": 92.77777777777777,
      "grad_norm": 0.7559652924537659,
      "learning_rate": 4.481509212664288e-05,
      "loss": 0.0528,
      "num_input_tokens_seen": 2337808,
      "step": 8350
    },
    {
      "epoch": 92.83333333333333,
      "grad_norm": 6.56707239151001,
      "learning_rate": 4.480910451840559e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 2339248,
      "step": 8355
    },
    {
      "epoch": 92.88888888888889,
      "grad_norm": 0.033242300152778625,
      "learning_rate": 4.480311385535546e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 2340640,
      "step": 8360
    },
    {
      "epoch": 92.94444444444444,
      "grad_norm": 6.083917140960693,
      "learning_rate": 4.47971201384163e-05,
      "loss": 0.0537,
      "num_input_tokens_seen": 2342032,
      "step": 8365
    },
    {
      "epoch": 93.0,
      "grad_norm": 0.08862270414829254,
      "learning_rate": 4.4791123368512446e-05,
      "loss": 0.0456,
      "num_input_tokens_seen": 2343424,
      "step": 8370
    },
    {
      "epoch": 93.05555555555556,
      "grad_norm": 0.05240608751773834,
      "learning_rate": 4.478512354656864e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 2344848,
      "step": 8375
    },
    {
      "epoch": 93.11111111111111,
      "grad_norm": 2.6214382648468018,
      "learning_rate": 4.477912067351016e-05,
      "loss": 0.0564,
      "num_input_tokens_seen": 2346256,
      "step": 8380
    },
    {
      "epoch": 93.16666666666667,
      "grad_norm": 7.370141506195068,
      "learning_rate": 4.477311475026271e-05,
      "loss": 0.0481,
      "num_input_tokens_seen": 2347696,
      "step": 8385
    },
    {
      "epoch": 93.22222222222223,
      "grad_norm": 0.07365325838327408,
      "learning_rate": 4.476710577775248e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2349072,
      "step": 8390
    },
    {
      "epoch": 93.27777777777777,
      "grad_norm": 0.05084284767508507,
      "learning_rate": 4.476109375690612e-05,
      "loss": 0.0519,
      "num_input_tokens_seen": 2350448,
      "step": 8395
    },
    {
      "epoch": 93.33333333333333,
      "grad_norm": 0.03862326964735985,
      "learning_rate": 4.4755078688650784e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 2351872,
      "step": 8400
    },
    {
      "epoch": 93.33333333333333,
      "eval_loss": 0.31914910674095154,
      "eval_runtime": 1.2353,
      "eval_samples_per_second": 32.382,
      "eval_steps_per_second": 16.191,
      "num_input_tokens_seen": 2351872,
      "step": 8400
    },
    {
      "epoch": 93.38888888888889,
      "grad_norm": 0.03319372981786728,
      "learning_rate": 4.474906057391406e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 2353280,
      "step": 8405
    },
    {
      "epoch": 93.44444444444444,
      "grad_norm": 0.7129984498023987,
      "learning_rate": 4.4743039413624e-05,
      "loss": 0.1041,
      "num_input_tokens_seen": 2354672,
      "step": 8410
    },
    {
      "epoch": 93.5,
      "grad_norm": 0.05739888176321983,
      "learning_rate": 4.473701520870916e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 2356080,
      "step": 8415
    },
    {
      "epoch": 93.55555555555556,
      "grad_norm": 0.39984649419784546,
      "learning_rate": 4.4730987960098544e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 2357456,
      "step": 8420
    },
    {
      "epoch": 93.61111111111111,
      "grad_norm": 0.049470338970422745,
      "learning_rate": 4.4724957668721635e-05,
      "loss": 0.0597,
      "num_input_tokens_seen": 2358816,
      "step": 8425
    },
    {
      "epoch": 93.66666666666667,
      "grad_norm": 0.13435491919517517,
      "learning_rate": 4.471892433550836e-05,
      "loss": 0.0841,
      "num_input_tokens_seen": 2360224,
      "step": 8430
    },
    {
      "epoch": 93.72222222222223,
      "grad_norm": 0.11594580113887787,
      "learning_rate": 4.471288796138916e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 2361616,
      "step": 8435
    },
    {
      "epoch": 93.77777777777777,
      "grad_norm": 0.047452181577682495,
      "learning_rate": 4.470684854729491e-05,
      "loss": 0.0133,
      "num_input_tokens_seen": 2363024,
      "step": 8440
    },
    {
      "epoch": 93.83333333333333,
      "grad_norm": 0.08485940843820572,
      "learning_rate": 4.4700806094156955e-05,
      "loss": 0.0514,
      "num_input_tokens_seen": 2364384,
      "step": 8445
    },
    {
      "epoch": 93.88888888888889,
      "grad_norm": 4.816036701202393,
      "learning_rate": 4.469476060290713e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 2365792,
      "step": 8450
    },
    {
      "epoch": 93.94444444444444,
      "grad_norm": 0.18202060461044312,
      "learning_rate": 4.468871207447772e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 2367152,
      "step": 8455
    },
    {
      "epoch": 94.0,
      "grad_norm": 0.06702545285224915,
      "learning_rate": 4.4682660509801486e-05,
      "loss": 0.01,
      "num_input_tokens_seen": 2368560,
      "step": 8460
    },
    {
      "epoch": 94.05555555555556,
      "grad_norm": 0.05473626032471657,
      "learning_rate": 4.467660590981165e-05,
      "loss": 0.0141,
      "num_input_tokens_seen": 2369952,
      "step": 8465
    },
    {
      "epoch": 94.11111111111111,
      "grad_norm": 0.60605388879776,
      "learning_rate": 4.467054827544191e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 2371344,
      "step": 8470
    },
    {
      "epoch": 94.16666666666667,
      "grad_norm": 3.453450918197632,
      "learning_rate": 4.4664487607626434e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 2372736,
      "step": 8475
    },
    {
      "epoch": 94.22222222222223,
      "grad_norm": 7.080897808074951,
      "learning_rate": 4.4658423907299845e-05,
      "loss": 0.0276,
      "num_input_tokens_seen": 2374096,
      "step": 8480
    },
    {
      "epoch": 94.27777777777777,
      "grad_norm": 2.1021430492401123,
      "learning_rate": 4.465235717539725e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 2375472,
      "step": 8485
    },
    {
      "epoch": 94.33333333333333,
      "grad_norm": 2.3582000732421875,
      "learning_rate": 4.464628741285421e-05,
      "loss": 0.052,
      "num_input_tokens_seen": 2376880,
      "step": 8490
    },
    {
      "epoch": 94.38888888888889,
      "grad_norm": 0.5261021256446838,
      "learning_rate": 4.4640214620606754e-05,
      "loss": 0.0531,
      "num_input_tokens_seen": 2378256,
      "step": 8495
    },
    {
      "epoch": 94.44444444444444,
      "grad_norm": 0.06367111951112747,
      "learning_rate": 4.46341387995914e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 2379648,
      "step": 8500
    },
    {
      "epoch": 94.5,
      "grad_norm": 0.14138640463352203,
      "learning_rate": 4.4628059950745106e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 2381088,
      "step": 8505
    },
    {
      "epoch": 94.55555555555556,
      "grad_norm": 0.7294134497642517,
      "learning_rate": 4.4621978075005297e-05,
      "loss": 0.0665,
      "num_input_tokens_seen": 2382464,
      "step": 8510
    },
    {
      "epoch": 94.61111111111111,
      "grad_norm": 0.14213338494300842,
      "learning_rate": 4.461589317330989e-05,
      "loss": 0.0639,
      "num_input_tokens_seen": 2383904,
      "step": 8515
    },
    {
      "epoch": 94.66666666666667,
      "grad_norm": 4.30581521987915,
      "learning_rate": 4.460980524659724e-05,
      "loss": 0.0673,
      "num_input_tokens_seen": 2385328,
      "step": 8520
    },
    {
      "epoch": 94.72222222222223,
      "grad_norm": 0.018560398370027542,
      "learning_rate": 4.46037142958062e-05,
      "loss": 0.0049,
      "num_input_tokens_seen": 2386720,
      "step": 8525
    },
    {
      "epoch": 94.77777777777777,
      "grad_norm": 0.3957129716873169,
      "learning_rate": 4.4597620321876046e-05,
      "loss": 0.0647,
      "num_input_tokens_seen": 2388144,
      "step": 8530
    },
    {
      "epoch": 94.83333333333333,
      "grad_norm": 0.06707819551229477,
      "learning_rate": 4.459152332574656e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 2389552,
      "step": 8535
    },
    {
      "epoch": 94.88888888888889,
      "grad_norm": 0.06268211454153061,
      "learning_rate": 4.4585423308357985e-05,
      "loss": 0.0114,
      "num_input_tokens_seen": 2390960,
      "step": 8540
    },
    {
      "epoch": 94.94444444444444,
      "grad_norm": 0.07094182074069977,
      "learning_rate": 4.457932027065102e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 2392368,
      "step": 8545
    },
    {
      "epoch": 95.0,
      "grad_norm": 1.92579984664917,
      "learning_rate": 4.45732142135668e-05,
      "loss": 0.0072,
      "num_input_tokens_seen": 2393808,
      "step": 8550
    },
    {
      "epoch": 95.05555555555556,
      "grad_norm": 0.05272843688726425,
      "learning_rate": 4.4567105138046986e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 2395232,
      "step": 8555
    },
    {
      "epoch": 95.11111111111111,
      "grad_norm": 0.07117616385221481,
      "learning_rate": 4.456099304503365e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 2396640,
      "step": 8560
    },
    {
      "epoch": 95.16666666666667,
      "grad_norm": 1.2559247016906738,
      "learning_rate": 4.455487793546939e-05,
      "loss": 0.0925,
      "num_input_tokens_seen": 2398080,
      "step": 8565
    },
    {
      "epoch": 95.22222222222223,
      "grad_norm": 0.03672585263848305,
      "learning_rate": 4.454875981029719e-05,
      "loss": 0.0088,
      "num_input_tokens_seen": 2399456,
      "step": 8570
    },
    {
      "epoch": 95.27777777777777,
      "grad_norm": 5.726057052612305,
      "learning_rate": 4.454263867046057e-05,
      "loss": 0.0194,
      "num_input_tokens_seen": 2400864,
      "step": 8575
    },
    {
      "epoch": 95.33333333333333,
      "grad_norm": 0.02724445052444935,
      "learning_rate": 4.4536514516903484e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 2402288,
      "step": 8580
    },
    {
      "epoch": 95.38888888888889,
      "grad_norm": 0.106313057243824,
      "learning_rate": 4.453038735057034e-05,
      "loss": 0.0431,
      "num_input_tokens_seen": 2403616,
      "step": 8585
    },
    {
      "epoch": 95.44444444444444,
      "grad_norm": 0.831264317035675,
      "learning_rate": 4.4524257172406034e-05,
      "loss": 0.01,
      "num_input_tokens_seen": 2404992,
      "step": 8590
    },
    {
      "epoch": 95.5,
      "grad_norm": 0.031380265951156616,
      "learning_rate": 4.451812398335592e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 2406400,
      "step": 8595
    },
    {
      "epoch": 95.55555555555556,
      "grad_norm": 1.5981940031051636,
      "learning_rate": 4.4511987784365805e-05,
      "loss": 0.0074,
      "num_input_tokens_seen": 2407824,
      "step": 8600
    },
    {
      "epoch": 95.55555555555556,
      "eval_loss": 0.3363414406776428,
      "eval_runtime": 1.2388,
      "eval_samples_per_second": 32.289,
      "eval_steps_per_second": 16.144,
      "num_input_tokens_seen": 2407824,
      "step": 8600
    },
    {
      "epoch": 95.61111111111111,
      "grad_norm": 0.02636515349149704,
      "learning_rate": 4.450584857638197e-05,
      "loss": 0.0634,
      "num_input_tokens_seen": 2409216,
      "step": 8605
    },
    {
      "epoch": 95.66666666666667,
      "grad_norm": 0.10790913552045822,
      "learning_rate": 4.449970636035116e-05,
      "loss": 0.046,
      "num_input_tokens_seen": 2410592,
      "step": 8610
    },
    {
      "epoch": 95.72222222222223,
      "grad_norm": 3.986255645751953,
      "learning_rate": 4.4493561137220574e-05,
      "loss": 0.1104,
      "num_input_tokens_seen": 2412000,
      "step": 8615
    },
    {
      "epoch": 95.77777777777777,
      "grad_norm": 6.336503982543945,
      "learning_rate": 4.44874129079379e-05,
      "loss": 0.0646,
      "num_input_tokens_seen": 2413408,
      "step": 8620
    },
    {
      "epoch": 95.83333333333333,
      "grad_norm": 0.057973865419626236,
      "learning_rate": 4.4481261673451255e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 2414800,
      "step": 8625
    },
    {
      "epoch": 95.88888888888889,
      "grad_norm": 3.9494056701660156,
      "learning_rate": 4.4475107434709245e-05,
      "loss": 0.0102,
      "num_input_tokens_seen": 2416208,
      "step": 8630
    },
    {
      "epoch": 95.94444444444444,
      "grad_norm": 0.03529202193021774,
      "learning_rate": 4.446895019266093e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 2417632,
      "step": 8635
    },
    {
      "epoch": 96.0,
      "grad_norm": 0.10738509893417358,
      "learning_rate": 4.446278994825583e-05,
      "loss": 0.0568,
      "num_input_tokens_seen": 2419008,
      "step": 8640
    },
    {
      "epoch": 96.05555555555556,
      "grad_norm": 1.0651994943618774,
      "learning_rate": 4.445662670244394e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 2420432,
      "step": 8645
    },
    {
      "epoch": 96.11111111111111,
      "grad_norm": 0.050989873707294464,
      "learning_rate": 4.44504604561757e-05,
      "loss": 0.0958,
      "num_input_tokens_seen": 2421808,
      "step": 8650
    },
    {
      "epoch": 96.16666666666667,
      "grad_norm": 0.24793432652950287,
      "learning_rate": 4.4444291210402035e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2423216,
      "step": 8655
    },
    {
      "epoch": 96.22222222222223,
      "grad_norm": 0.5171831250190735,
      "learning_rate": 4.443811896607431e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 2424592,
      "step": 8660
    },
    {
      "epoch": 96.27777777777777,
      "grad_norm": 0.6318588852882385,
      "learning_rate": 4.443194372414436e-05,
      "loss": 0.035,
      "num_input_tokens_seen": 2426048,
      "step": 8665
    },
    {
      "epoch": 96.33333333333333,
      "grad_norm": 0.13308019936084747,
      "learning_rate": 4.442576548556449e-05,
      "loss": 0.0606,
      "num_input_tokens_seen": 2427424,
      "step": 8670
    },
    {
      "epoch": 96.38888888888889,
      "grad_norm": 0.08617629110813141,
      "learning_rate": 4.441958425128747e-05,
      "loss": 0.009,
      "num_input_tokens_seen": 2428816,
      "step": 8675
    },
    {
      "epoch": 96.44444444444444,
      "grad_norm": 0.056330177932977676,
      "learning_rate": 4.4413400022266515e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 2430192,
      "step": 8680
    },
    {
      "epoch": 96.5,
      "grad_norm": 0.14672617614269257,
      "learning_rate": 4.4407212799455313e-05,
      "loss": 0.1105,
      "num_input_tokens_seen": 2431584,
      "step": 8685
    },
    {
      "epoch": 96.55555555555556,
      "grad_norm": 0.06510480493307114,
      "learning_rate": 4.4401022583808003e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 2432976,
      "step": 8690
    },
    {
      "epoch": 96.61111111111111,
      "grad_norm": 0.1040906235575676,
      "learning_rate": 4.439482937627921e-05,
      "loss": 0.0482,
      "num_input_tokens_seen": 2434368,
      "step": 8695
    },
    {
      "epoch": 96.66666666666667,
      "grad_norm": 3.9324469566345215,
      "learning_rate": 4.4388633177824004e-05,
      "loss": 0.0599,
      "num_input_tokens_seen": 2435792,
      "step": 8700
    },
    {
      "epoch": 96.72222222222223,
      "grad_norm": 0.049558307975530624,
      "learning_rate": 4.4382433989397895e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 2437216,
      "step": 8705
    },
    {
      "epoch": 96.77777777777777,
      "grad_norm": 0.04826352000236511,
      "learning_rate": 4.4376231811956895e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 2438608,
      "step": 8710
    },
    {
      "epoch": 96.83333333333333,
      "grad_norm": 0.03961774334311485,
      "learning_rate": 4.437002664645745e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2439968,
      "step": 8715
    },
    {
      "epoch": 96.88888888888889,
      "grad_norm": 3.190206527709961,
      "learning_rate": 4.436381849385649e-05,
      "loss": 0.0505,
      "num_input_tokens_seen": 2441392,
      "step": 8720
    },
    {
      "epoch": 96.94444444444444,
      "grad_norm": 0.04781443625688553,
      "learning_rate": 4.435760735511136e-05,
      "loss": 0.0079,
      "num_input_tokens_seen": 2442768,
      "step": 8725
    },
    {
      "epoch": 97.0,
      "grad_norm": 0.039050500839948654,
      "learning_rate": 4.435139323117992e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 2444160,
      "step": 8730
    },
    {
      "epoch": 97.05555555555556,
      "grad_norm": 0.7263879179954529,
      "learning_rate": 4.434517612302046e-05,
      "loss": 0.0509,
      "num_input_tokens_seen": 2445568,
      "step": 8735
    },
    {
      "epoch": 97.11111111111111,
      "grad_norm": 0.022861473262310028,
      "learning_rate": 4.433895603159174e-05,
      "loss": 0.0471,
      "num_input_tokens_seen": 2446976,
      "step": 8740
    },
    {
      "epoch": 97.16666666666667,
      "grad_norm": 0.7576487064361572,
      "learning_rate": 4.433273295785296e-05,
      "loss": 0.0161,
      "num_input_tokens_seen": 2448416,
      "step": 8745
    },
    {
      "epoch": 97.22222222222223,
      "grad_norm": 0.036938559263944626,
      "learning_rate": 4.432650690276382e-05,
      "loss": 0.0421,
      "num_input_tokens_seen": 2449808,
      "step": 8750
    },
    {
      "epoch": 97.27777777777777,
      "grad_norm": 3.6226212978363037,
      "learning_rate": 4.4320277867284435e-05,
      "loss": 0.0428,
      "num_input_tokens_seen": 2451216,
      "step": 8755
    },
    {
      "epoch": 97.33333333333333,
      "grad_norm": 0.027425255626440048,
      "learning_rate": 4.431404585237541e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 2452608,
      "step": 8760
    },
    {
      "epoch": 97.38888888888889,
      "grad_norm": 0.06150299310684204,
      "learning_rate": 4.43078108589978e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 2454000,
      "step": 8765
    },
    {
      "epoch": 97.44444444444444,
      "grad_norm": 0.16408877074718475,
      "learning_rate": 4.4301572888113116e-05,
      "loss": 0.0474,
      "num_input_tokens_seen": 2455408,
      "step": 8770
    },
    {
      "epoch": 97.5,
      "grad_norm": 1.155432939529419,
      "learning_rate": 4.4295331940683337e-05,
      "loss": 0.0091,
      "num_input_tokens_seen": 2456800,
      "step": 8775
    },
    {
      "epoch": 97.55555555555556,
      "grad_norm": 0.045117828994989395,
      "learning_rate": 4.428908801767089e-05,
      "loss": 0.046,
      "num_input_tokens_seen": 2458208,
      "step": 8780
    },
    {
      "epoch": 97.61111111111111,
      "grad_norm": 0.10589192062616348,
      "learning_rate": 4.428284112003868e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2459568,
      "step": 8785
    },
    {
      "epoch": 97.66666666666667,
      "grad_norm": 0.7588531374931335,
      "learning_rate": 4.4276591248750033e-05,
      "loss": 0.0723,
      "num_input_tokens_seen": 2460976,
      "step": 8790
    },
    {
      "epoch": 97.72222222222223,
      "grad_norm": 0.0403747633099556,
      "learning_rate": 4.4270338404768774e-05,
      "loss": 0.0046,
      "num_input_tokens_seen": 2462384,
      "step": 8795
    },
    {
      "epoch": 97.77777777777777,
      "grad_norm": 1.3845890760421753,
      "learning_rate": 4.426408258905917e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 2463744,
      "step": 8800
    },
    {
      "epoch": 97.77777777777777,
      "eval_loss": 0.2989887297153473,
      "eval_runtime": 1.2384,
      "eval_samples_per_second": 32.3,
      "eval_steps_per_second": 16.15,
      "num_input_tokens_seen": 2463744,
      "step": 8800
    },
    {
      "epoch": 97.83333333333333,
      "grad_norm": 0.16313910484313965,
      "learning_rate": 4.425782380258594e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 2465136,
      "step": 8805
    },
    {
      "epoch": 97.88888888888889,
      "grad_norm": 0.036389585584402084,
      "learning_rate": 4.425156204631427e-05,
      "loss": 0.0341,
      "num_input_tokens_seen": 2466512,
      "step": 8810
    },
    {
      "epoch": 97.94444444444444,
      "grad_norm": 0.8185663223266602,
      "learning_rate": 4.424529732120981e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 2467904,
      "step": 8815
    },
    {
      "epoch": 98.0,
      "grad_norm": 0.2773381769657135,
      "learning_rate": 4.423902962823864e-05,
      "loss": 0.0117,
      "num_input_tokens_seen": 2469296,
      "step": 8820
    },
    {
      "epoch": 98.05555555555556,
      "grad_norm": 0.24249771237373352,
      "learning_rate": 4.423275896836733e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 2470688,
      "step": 8825
    },
    {
      "epoch": 98.11111111111111,
      "grad_norm": 1.0113188028335571,
      "learning_rate": 4.42264853425629e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 2472080,
      "step": 8830
    },
    {
      "epoch": 98.16666666666667,
      "grad_norm": 0.05267757549881935,
      "learning_rate": 4.4220208751792816e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 2473488,
      "step": 8835
    },
    {
      "epoch": 98.22222222222223,
      "grad_norm": 10.236871719360352,
      "learning_rate": 4.421392919702499e-05,
      "loss": 0.0913,
      "num_input_tokens_seen": 2474896,
      "step": 8840
    },
    {
      "epoch": 98.27777777777777,
      "grad_norm": 0.039141733199357986,
      "learning_rate": 4.4207646679227846e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 2476272,
      "step": 8845
    },
    {
      "epoch": 98.33333333333333,
      "grad_norm": 0.321561336517334,
      "learning_rate": 4.42013611993702e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 2477696,
      "step": 8850
    },
    {
      "epoch": 98.38888888888889,
      "grad_norm": 0.025894692167639732,
      "learning_rate": 4.419507275842135e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 2479040,
      "step": 8855
    },
    {
      "epoch": 98.44444444444444,
      "grad_norm": 0.2157319337129593,
      "learning_rate": 4.418878135735106e-05,
      "loss": 0.0463,
      "num_input_tokens_seen": 2480496,
      "step": 8860
    },
    {
      "epoch": 98.5,
      "grad_norm": 0.03756893053650856,
      "learning_rate": 4.418248699712955e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 2481888,
      "step": 8865
    },
    {
      "epoch": 98.55555555555556,
      "grad_norm": 0.030577465891838074,
      "learning_rate": 4.417618967872748e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2483280,
      "step": 8870
    },
    {
      "epoch": 98.61111111111111,
      "grad_norm": 0.015237816609442234,
      "learning_rate": 4.4169889403115985e-05,
      "loss": 0.0062,
      "num_input_tokens_seen": 2484704,
      "step": 8875
    },
    {
      "epoch": 98.66666666666667,
      "grad_norm": 0.04272550716996193,
      "learning_rate": 4.4163586171266627e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 2486080,
      "step": 8880
    },
    {
      "epoch": 98.72222222222223,
      "grad_norm": 0.6528998017311096,
      "learning_rate": 4.415727998415147e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 2487520,
      "step": 8885
    },
    {
      "epoch": 98.77777777777777,
      "grad_norm": 0.026941100135445595,
      "learning_rate": 4.4150970842742985e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 2488944,
      "step": 8890
    },
    {
      "epoch": 98.83333333333333,
      "grad_norm": 0.05886133387684822,
      "learning_rate": 4.4144658748014134e-05,
      "loss": 0.0443,
      "num_input_tokens_seen": 2490320,
      "step": 8895
    },
    {
      "epoch": 98.88888888888889,
      "grad_norm": 0.7101027369499207,
      "learning_rate": 4.413834370093831e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 2491728,
      "step": 8900
    },
    {
      "epoch": 98.94444444444444,
      "grad_norm": 0.35725969076156616,
      "learning_rate": 4.413202570248939e-05,
      "loss": 0.0535,
      "num_input_tokens_seen": 2493088,
      "step": 8905
    },
    {
      "epoch": 99.0,
      "grad_norm": 1.3555128574371338,
      "learning_rate": 4.412570475364167e-05,
      "loss": 0.0453,
      "num_input_tokens_seen": 2494448,
      "step": 8910
    },
    {
      "epoch": 99.05555555555556,
      "grad_norm": 0.20640867948532104,
      "learning_rate": 4.411938085536994e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 2495888,
      "step": 8915
    },
    {
      "epoch": 99.11111111111111,
      "grad_norm": 0.05427011474967003,
      "learning_rate": 4.41130540086494e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 2497264,
      "step": 8920
    },
    {
      "epoch": 99.16666666666667,
      "grad_norm": 0.052744485437870026,
      "learning_rate": 4.4106724214455754e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2498624,
      "step": 8925
    },
    {
      "epoch": 99.22222222222223,
      "grad_norm": 0.2421903908252716,
      "learning_rate": 4.4100391473765115e-05,
      "loss": 0.0432,
      "num_input_tokens_seen": 2500000,
      "step": 8930
    },
    {
      "epoch": 99.27777777777777,
      "grad_norm": 0.3973543047904968,
      "learning_rate": 4.409405578755408e-05,
      "loss": 0.0983,
      "num_input_tokens_seen": 2501392,
      "step": 8935
    },
    {
      "epoch": 99.33333333333333,
      "grad_norm": 0.05009959638118744,
      "learning_rate": 4.4087717156799705e-05,
      "loss": 0.0147,
      "num_input_tokens_seen": 2502768,
      "step": 8940
    },
    {
      "epoch": 99.38888888888889,
      "grad_norm": 0.08322512358427048,
      "learning_rate": 4.408137558247946e-05,
      "loss": 0.0678,
      "num_input_tokens_seen": 2504160,
      "step": 8945
    },
    {
      "epoch": 99.44444444444444,
      "grad_norm": 0.7899781465530396,
      "learning_rate": 4.4075031065571306e-05,
      "loss": 0.0124,
      "num_input_tokens_seen": 2505552,
      "step": 8950
    },
    {
      "epoch": 99.5,
      "grad_norm": 0.016563668847084045,
      "learning_rate": 4.406868360705366e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 2506912,
      "step": 8955
    },
    {
      "epoch": 99.55555555555556,
      "grad_norm": 1.0949581861495972,
      "learning_rate": 4.406233320790536e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 2508384,
      "step": 8960
    },
    {
      "epoch": 99.61111111111111,
      "grad_norm": 0.9968736171722412,
      "learning_rate": 4.4055979869105734e-05,
      "loss": 0.078,
      "num_input_tokens_seen": 2509808,
      "step": 8965
    },
    {
      "epoch": 99.66666666666667,
      "grad_norm": 0.3959631025791168,
      "learning_rate": 4.404962359163454e-05,
      "loss": 0.0089,
      "num_input_tokens_seen": 2511216,
      "step": 8970
    },
    {
      "epoch": 99.72222222222223,
      "grad_norm": 0.8822899460792542,
      "learning_rate": 4.404326437647199e-05,
      "loss": 0.0152,
      "num_input_tokens_seen": 2512624,
      "step": 8975
    },
    {
      "epoch": 99.77777777777777,
      "grad_norm": 0.17726443707942963,
      "learning_rate": 4.403690222459877e-05,
      "loss": 0.0049,
      "num_input_tokens_seen": 2514032,
      "step": 8980
    },
    {
      "epoch": 99.83333333333333,
      "grad_norm": 0.11196363717317581,
      "learning_rate": 4.4030537136995984e-05,
      "loss": 0.0474,
      "num_input_tokens_seen": 2515456,
      "step": 8985
    },
    {
      "epoch": 99.88888888888889,
      "grad_norm": 0.040657855570316315,
      "learning_rate": 4.402416911464523e-05,
      "loss": 0.0898,
      "num_input_tokens_seen": 2516896,
      "step": 8990
    },
    {
      "epoch": 99.94444444444444,
      "grad_norm": 0.17010171711444855,
      "learning_rate": 4.4017798158528516e-05,
      "loss": 0.0071,
      "num_input_tokens_seen": 2518304,
      "step": 8995
    },
    {
      "epoch": 100.0,
      "grad_norm": 0.03811817616224289,
      "learning_rate": 4.401142426962834e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 2519680,
      "step": 9000
    },
    {
      "epoch": 100.0,
      "eval_loss": 0.3083083927631378,
      "eval_runtime": 1.2338,
      "eval_samples_per_second": 32.421,
      "eval_steps_per_second": 16.211,
      "num_input_tokens_seen": 2519680,
      "step": 9000
    },
    {
      "epoch": 100.05555555555556,
      "grad_norm": 0.8433417081832886,
      "learning_rate": 4.400504744892763e-05,
      "loss": 0.0691,
      "num_input_tokens_seen": 2521088,
      "step": 9005
    },
    {
      "epoch": 100.11111111111111,
      "grad_norm": 0.7715723514556885,
      "learning_rate": 4.399866769740975e-05,
      "loss": 0.0666,
      "num_input_tokens_seen": 2522512,
      "step": 9010
    },
    {
      "epoch": 100.16666666666667,
      "grad_norm": 0.08705484867095947,
      "learning_rate": 4.399228501605859e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 2523904,
      "step": 9015
    },
    {
      "epoch": 100.22222222222223,
      "grad_norm": 0.050680696964263916,
      "learning_rate": 4.398589940585839e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 2525264,
      "step": 9020
    },
    {
      "epoch": 100.27777777777777,
      "grad_norm": 0.32353949546813965,
      "learning_rate": 4.3979510867793917e-05,
      "loss": 0.0697,
      "num_input_tokens_seen": 2526704,
      "step": 9025
    },
    {
      "epoch": 100.33333333333333,
      "grad_norm": 0.15547195076942444,
      "learning_rate": 4.3973119402850346e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2528112,
      "step": 9030
    },
    {
      "epoch": 100.38888888888889,
      "grad_norm": 0.05971971154212952,
      "learning_rate": 4.396672501201334e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 2529504,
      "step": 9035
    },
    {
      "epoch": 100.44444444444444,
      "grad_norm": 0.283919095993042,
      "learning_rate": 4.396032769626899e-05,
      "loss": 0.0069,
      "num_input_tokens_seen": 2530880,
      "step": 9040
    },
    {
      "epoch": 100.5,
      "grad_norm": 0.09710641205310822,
      "learning_rate": 4.395392745660384e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 2532256,
      "step": 9045
    },
    {
      "epoch": 100.55555555555556,
      "grad_norm": 0.40946266055107117,
      "learning_rate": 4.394752429400488e-05,
      "loss": 0.0115,
      "num_input_tokens_seen": 2533616,
      "step": 9050
    },
    {
      "epoch": 100.61111111111111,
      "grad_norm": 4.544084548950195,
      "learning_rate": 4.394111820945957e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 2535024,
      "step": 9055
    },
    {
      "epoch": 100.66666666666667,
      "grad_norm": 0.27973753213882446,
      "learning_rate": 4.393470920395579e-05,
      "loss": 0.0467,
      "num_input_tokens_seen": 2536432,
      "step": 9060
    },
    {
      "epoch": 100.72222222222223,
      "grad_norm": 0.76377934217453,
      "learning_rate": 4.392829727848192e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 2537824,
      "step": 9065
    },
    {
      "epoch": 100.77777777777777,
      "grad_norm": 0.11635862290859222,
      "learning_rate": 4.392188243402673e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 2539232,
      "step": 9070
    },
    {
      "epoch": 100.83333333333333,
      "grad_norm": 0.22337110340595245,
      "learning_rate": 4.391546467157949e-05,
      "loss": 0.0465,
      "num_input_tokens_seen": 2540608,
      "step": 9075
    },
    {
      "epoch": 100.88888888888889,
      "grad_norm": 0.08251716941595078,
      "learning_rate": 4.390904399212988e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 2542000,
      "step": 9080
    },
    {
      "epoch": 100.94444444444444,
      "grad_norm": 0.040796615183353424,
      "learning_rate": 4.390262039666807e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 2543408,
      "step": 9085
    },
    {
      "epoch": 101.0,
      "grad_norm": 0.020083513110876083,
      "learning_rate": 4.389619388618464e-05,
      "loss": 0.0565,
      "num_input_tokens_seen": 2544800,
      "step": 9090
    },
    {
      "epoch": 101.05555555555556,
      "grad_norm": 0.06334606558084488,
      "learning_rate": 4.3889764461670655e-05,
      "loss": 0.0077,
      "num_input_tokens_seen": 2546176,
      "step": 9095
    },
    {
      "epoch": 101.11111111111111,
      "grad_norm": 0.36740642786026,
      "learning_rate": 4.38833321241176e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2547568,
      "step": 9100
    },
    {
      "epoch": 101.16666666666667,
      "grad_norm": 0.01632481813430786,
      "learning_rate": 4.3876896874517434e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 2548992,
      "step": 9105
    },
    {
      "epoch": 101.22222222222223,
      "grad_norm": 0.20958395302295685,
      "learning_rate": 4.3870458713862554e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 2550400,
      "step": 9110
    },
    {
      "epoch": 101.27777777777777,
      "grad_norm": 0.11295211315155029,
      "learning_rate": 4.386401764314579e-05,
      "loss": 0.0483,
      "num_input_tokens_seen": 2551776,
      "step": 9115
    },
    {
      "epoch": 101.33333333333333,
      "grad_norm": 6.188415050506592,
      "learning_rate": 4.385757366336045e-05,
      "loss": 0.0106,
      "num_input_tokens_seen": 2553184,
      "step": 9120
    },
    {
      "epoch": 101.38888888888889,
      "grad_norm": 0.03504262864589691,
      "learning_rate": 4.385112677550027e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 2554576,
      "step": 9125
    },
    {
      "epoch": 101.44444444444444,
      "grad_norm": 0.14000700414180756,
      "learning_rate": 4.384467698055945e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 2555968,
      "step": 9130
    },
    {
      "epoch": 101.5,
      "grad_norm": 0.06525950878858566,
      "learning_rate": 4.383822427953261e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 2557376,
      "step": 9135
    },
    {
      "epoch": 101.55555555555556,
      "grad_norm": 0.09077876061201096,
      "learning_rate": 4.3831768673414864e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 2558800,
      "step": 9140
    },
    {
      "epoch": 101.61111111111111,
      "grad_norm": 2.878632068634033,
      "learning_rate": 4.382531016320173e-05,
      "loss": 0.0096,
      "num_input_tokens_seen": 2560208,
      "step": 9145
    },
    {
      "epoch": 101.66666666666667,
      "grad_norm": 9.089218139648438,
      "learning_rate": 4.3818848749889184e-05,
      "loss": 0.0617,
      "num_input_tokens_seen": 2561648,
      "step": 9150
    },
    {
      "epoch": 101.72222222222223,
      "grad_norm": 0.08864381909370422,
      "learning_rate": 4.381238443447368e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 2563024,
      "step": 9155
    },
    {
      "epoch": 101.77777777777777,
      "grad_norm": 0.02051064558327198,
      "learning_rate": 4.380591721795208e-05,
      "loss": 0.0487,
      "num_input_tokens_seen": 2564416,
      "step": 9160
    },
    {
      "epoch": 101.83333333333333,
      "grad_norm": 0.04883555322885513,
      "learning_rate": 4.3799447101321723e-05,
      "loss": 0.0386,
      "num_input_tokens_seen": 2565856,
      "step": 9165
    },
    {
      "epoch": 101.88888888888889,
      "grad_norm": 0.03474656865000725,
      "learning_rate": 4.379297408558036e-05,
      "loss": 0.0685,
      "num_input_tokens_seen": 2567232,
      "step": 9170
    },
    {
      "epoch": 101.94444444444444,
      "grad_norm": 4.670870780944824,
      "learning_rate": 4.378649817172624e-05,
      "loss": 0.0657,
      "num_input_tokens_seen": 2568608,
      "step": 9175
    },
    {
      "epoch": 102.0,
      "grad_norm": 0.7874966263771057,
      "learning_rate": 4.378001936075801e-05,
      "loss": 0.0494,
      "num_input_tokens_seen": 2569968,
      "step": 9180
    },
    {
      "epoch": 102.05555555555556,
      "grad_norm": 0.03562683239579201,
      "learning_rate": 4.377353765367479e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2571392,
      "step": 9185
    },
    {
      "epoch": 102.11111111111111,
      "grad_norm": 0.03773824870586395,
      "learning_rate": 4.376705305147614e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 2572832,
      "step": 9190
    },
    {
      "epoch": 102.16666666666667,
      "grad_norm": 1.1300216913223267,
      "learning_rate": 4.376056555516206e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2574208,
      "step": 9195
    },
    {
      "epoch": 102.22222222222223,
      "grad_norm": 0.015963194891810417,
      "learning_rate": 4.375407516573302e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 2575584,
      "step": 9200
    },
    {
      "epoch": 102.22222222222223,
      "eval_loss": 0.3234058618545532,
      "eval_runtime": 1.2342,
      "eval_samples_per_second": 32.409,
      "eval_steps_per_second": 16.204,
      "num_input_tokens_seen": 2575584,
      "step": 9200
    },
    {
      "epoch": 102.27777777777777,
      "grad_norm": 0.0590968057513237,
      "learning_rate": 4.3747581884189913e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 2576976,
      "step": 9205
    },
    {
      "epoch": 102.33333333333333,
      "grad_norm": 2.250859498977661,
      "learning_rate": 4.374108571153408e-05,
      "loss": 0.007,
      "num_input_tokens_seen": 2578400,
      "step": 9210
    },
    {
      "epoch": 102.38888888888889,
      "grad_norm": 0.2421763688325882,
      "learning_rate": 4.3734586648767316e-05,
      "loss": 0.0387,
      "num_input_tokens_seen": 2579824,
      "step": 9215
    },
    {
      "epoch": 102.44444444444444,
      "grad_norm": 0.040467023849487305,
      "learning_rate": 4.372808469689186e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 2581248,
      "step": 9220
    },
    {
      "epoch": 102.5,
      "grad_norm": 0.1488858163356781,
      "learning_rate": 4.372157985691039e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 2582608,
      "step": 9225
    },
    {
      "epoch": 102.55555555555556,
      "grad_norm": 6.814762115478516,
      "learning_rate": 4.371507212982603e-05,
      "loss": 0.0987,
      "num_input_tokens_seen": 2584064,
      "step": 9230
    },
    {
      "epoch": 102.61111111111111,
      "grad_norm": 7.040499687194824,
      "learning_rate": 4.370856151664236e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 2585424,
      "step": 9235
    },
    {
      "epoch": 102.66666666666667,
      "grad_norm": 0.03536194935441017,
      "learning_rate": 4.3702048018363404e-05,
      "loss": 0.0416,
      "num_input_tokens_seen": 2586880,
      "step": 9240
    },
    {
      "epoch": 102.72222222222223,
      "grad_norm": 0.12946024537086487,
      "learning_rate": 4.369553163599362e-05,
      "loss": 0.0538,
      "num_input_tokens_seen": 2588240,
      "step": 9245
    },
    {
      "epoch": 102.77777777777777,
      "grad_norm": 0.15090347826480865,
      "learning_rate": 4.3689012370537904e-05,
      "loss": 0.0231,
      "num_input_tokens_seen": 2589616,
      "step": 9250
    },
    {
      "epoch": 102.83333333333333,
      "grad_norm": 0.049277178943157196,
      "learning_rate": 4.368249022300164e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2591008,
      "step": 9255
    },
    {
      "epoch": 102.88888888888889,
      "grad_norm": 0.46329256892204285,
      "learning_rate": 4.367596519439059e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 2592432,
      "step": 9260
    },
    {
      "epoch": 102.94444444444444,
      "grad_norm": 0.5220759510993958,
      "learning_rate": 4.366943728571101e-05,
      "loss": 0.0454,
      "num_input_tokens_seen": 2593824,
      "step": 9265
    },
    {
      "epoch": 103.0,
      "grad_norm": 0.020833374932408333,
      "learning_rate": 4.366290649796959e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 2595200,
      "step": 9270
    },
    {
      "epoch": 103.05555555555556,
      "grad_norm": 0.019200073555111885,
      "learning_rate": 4.3656372832173456e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 2596576,
      "step": 9275
    },
    {
      "epoch": 103.11111111111111,
      "grad_norm": 0.016274401918053627,
      "learning_rate": 4.364983628933017e-05,
      "loss": 0.0395,
      "num_input_tokens_seen": 2597952,
      "step": 9280
    },
    {
      "epoch": 103.16666666666667,
      "grad_norm": 0.021578816697001457,
      "learning_rate": 4.364329687044777e-05,
      "loss": 0.0049,
      "num_input_tokens_seen": 2599376,
      "step": 9285
    },
    {
      "epoch": 103.22222222222223,
      "grad_norm": 0.08462841063737869,
      "learning_rate": 4.36367545765347e-05,
      "loss": 0.0926,
      "num_input_tokens_seen": 2600832,
      "step": 9290
    },
    {
      "epoch": 103.27777777777777,
      "grad_norm": 0.21792787313461304,
      "learning_rate": 4.363020940859988e-05,
      "loss": 0.0373,
      "num_input_tokens_seen": 2602240,
      "step": 9295
    },
    {
      "epoch": 103.33333333333333,
      "grad_norm": 0.21704378724098206,
      "learning_rate": 4.362366136765263e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2603696,
      "step": 9300
    },
    {
      "epoch": 103.38888888888889,
      "grad_norm": 0.03761091083288193,
      "learning_rate": 4.361711045470278e-05,
      "loss": 0.0057,
      "num_input_tokens_seen": 2605088,
      "step": 9305
    },
    {
      "epoch": 103.44444444444444,
      "grad_norm": 0.056548673659563065,
      "learning_rate": 4.3610556670760524e-05,
      "loss": 0.1022,
      "num_input_tokens_seen": 2606496,
      "step": 9310
    },
    {
      "epoch": 103.5,
      "grad_norm": 0.04846350848674774,
      "learning_rate": 4.360400001683657e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 2607920,
      "step": 9315
    },
    {
      "epoch": 103.55555555555556,
      "grad_norm": 0.12759020924568176,
      "learning_rate": 4.3597440493942e-05,
      "loss": 0.007,
      "num_input_tokens_seen": 2609328,
      "step": 9320
    },
    {
      "epoch": 103.61111111111111,
      "grad_norm": 0.09641629457473755,
      "learning_rate": 4.3590878103088405e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 2610720,
      "step": 9325
    },
    {
      "epoch": 103.66666666666667,
      "grad_norm": 0.10855954885482788,
      "learning_rate": 4.358431284528779e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 2612144,
      "step": 9330
    },
    {
      "epoch": 103.72222222222223,
      "grad_norm": 0.9355184435844421,
      "learning_rate": 4.357774472155257e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 2613536,
      "step": 9335
    },
    {
      "epoch": 103.77777777777777,
      "grad_norm": 0.0382961668074131,
      "learning_rate": 4.3571173732895664e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 2614944,
      "step": 9340
    },
    {
      "epoch": 103.83333333333333,
      "grad_norm": 0.05375034734606743,
      "learning_rate": 4.356459988033039e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 2616304,
      "step": 9345
    },
    {
      "epoch": 103.88888888888889,
      "grad_norm": 0.0842958614230156,
      "learning_rate": 4.355802316487051e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 2617648,
      "step": 9350
    },
    {
      "epoch": 103.94444444444444,
      "grad_norm": 0.08676561713218689,
      "learning_rate": 4.355144358753025e-05,
      "loss": 0.0702,
      "num_input_tokens_seen": 2619056,
      "step": 9355
    },
    {
      "epoch": 104.0,
      "grad_norm": 0.19498860836029053,
      "learning_rate": 4.354486114932425e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2620432,
      "step": 9360
    },
    {
      "epoch": 104.05555555555556,
      "grad_norm": 7.048811435699463,
      "learning_rate": 4.353827585126762e-05,
      "loss": 0.0498,
      "num_input_tokens_seen": 2621808,
      "step": 9365
    },
    {
      "epoch": 104.11111111111111,
      "grad_norm": 0.0828288123011589,
      "learning_rate": 4.353168769437588e-05,
      "loss": 0.0089,
      "num_input_tokens_seen": 2623264,
      "step": 9370
    },
    {
      "epoch": 104.16666666666667,
      "grad_norm": 0.10071263462305069,
      "learning_rate": 4.3525096679665014e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 2624624,
      "step": 9375
    },
    {
      "epoch": 104.22222222222223,
      "grad_norm": 0.04809629172086716,
      "learning_rate": 4.351850280815144e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2626064,
      "step": 9380
    },
    {
      "epoch": 104.27777777777777,
      "grad_norm": 0.14000627398490906,
      "learning_rate": 4.3511906080852014e-05,
      "loss": 0.0542,
      "num_input_tokens_seen": 2627472,
      "step": 9385
    },
    {
      "epoch": 104.33333333333333,
      "grad_norm": 0.5761666297912598,
      "learning_rate": 4.350530649878404e-05,
      "loss": 0.0074,
      "num_input_tokens_seen": 2628864,
      "step": 9390
    },
    {
      "epoch": 104.38888888888889,
      "grad_norm": 0.4124310612678528,
      "learning_rate": 4.3498704062965246e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 2630272,
      "step": 9395
    },
    {
      "epoch": 104.44444444444444,
      "grad_norm": 0.03845745325088501,
      "learning_rate": 4.3492098774413815e-05,
      "loss": 0.1017,
      "num_input_tokens_seen": 2631680,
      "step": 9400
    },
    {
      "epoch": 104.44444444444444,
      "eval_loss": 0.3195965886116028,
      "eval_runtime": 1.2375,
      "eval_samples_per_second": 32.322,
      "eval_steps_per_second": 16.161,
      "num_input_tokens_seen": 2631680,
      "step": 9400
    },
    {
      "epoch": 104.5,
      "grad_norm": 0.2677428722381592,
      "learning_rate": 4.3485490634148375e-05,
      "loss": 0.0577,
      "num_input_tokens_seen": 2633088,
      "step": 9405
    },
    {
      "epoch": 104.55555555555556,
      "grad_norm": 0.06797739118337631,
      "learning_rate": 4.347887964318797e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 2634480,
      "step": 9410
    },
    {
      "epoch": 104.61111111111111,
      "grad_norm": 0.2169237583875656,
      "learning_rate": 4.34722658025521e-05,
      "loss": 0.0436,
      "num_input_tokens_seen": 2635824,
      "step": 9415
    },
    {
      "epoch": 104.66666666666667,
      "grad_norm": 0.018711617216467857,
      "learning_rate": 4.346564911326071e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2637264,
      "step": 9420
    },
    {
      "epoch": 104.72222222222223,
      "grad_norm": 0.04307044297456741,
      "learning_rate": 4.345902957633418e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 2638688,
      "step": 9425
    },
    {
      "epoch": 104.77777777777777,
      "grad_norm": 2.6683950424194336,
      "learning_rate": 4.345240719279331e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 2640080,
      "step": 9430
    },
    {
      "epoch": 104.83333333333333,
      "grad_norm": 0.10171566903591156,
      "learning_rate": 4.3445781963659374e-05,
      "loss": 0.0927,
      "num_input_tokens_seen": 2641456,
      "step": 9435
    },
    {
      "epoch": 104.88888888888889,
      "grad_norm": 0.05882981792092323,
      "learning_rate": 4.3439153889954045e-05,
      "loss": 0.0046,
      "num_input_tokens_seen": 2642832,
      "step": 9440
    },
    {
      "epoch": 104.94444444444444,
      "grad_norm": 0.014842819422483444,
      "learning_rate": 4.343252297269946e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 2644240,
      "step": 9445
    },
    {
      "epoch": 105.0,
      "grad_norm": 4.5225629806518555,
      "learning_rate": 4.342588921291821e-05,
      "loss": 0.0087,
      "num_input_tokens_seen": 2645648,
      "step": 9450
    },
    {
      "epoch": 105.05555555555556,
      "grad_norm": 0.04512425512075424,
      "learning_rate": 4.341925261163328e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 2647056,
      "step": 9455
    },
    {
      "epoch": 105.11111111111111,
      "grad_norm": 0.06547736376523972,
      "learning_rate": 4.341261316986813e-05,
      "loss": 0.0073,
      "num_input_tokens_seen": 2648432,
      "step": 9460
    },
    {
      "epoch": 105.16666666666667,
      "grad_norm": 1.9915786981582642,
      "learning_rate": 4.340597088864664e-05,
      "loss": 0.0431,
      "num_input_tokens_seen": 2649840,
      "step": 9465
    },
    {
      "epoch": 105.22222222222223,
      "grad_norm": 0.5241715312004089,
      "learning_rate": 4.339932576899313e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2651296,
      "step": 9470
    },
    {
      "epoch": 105.27777777777777,
      "grad_norm": 0.06483233720064163,
      "learning_rate": 4.3392677811932375e-05,
      "loss": 0.008,
      "num_input_tokens_seen": 2652672,
      "step": 9475
    },
    {
      "epoch": 105.33333333333333,
      "grad_norm": 4.33317756652832,
      "learning_rate": 4.338602701848956e-05,
      "loss": 0.0614,
      "num_input_tokens_seen": 2654080,
      "step": 9480
    },
    {
      "epoch": 105.38888888888889,
      "grad_norm": 0.07889207452535629,
      "learning_rate": 4.337937338969033e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 2655504,
      "step": 9485
    },
    {
      "epoch": 105.44444444444444,
      "grad_norm": 0.03505353629589081,
      "learning_rate": 4.337271692656075e-05,
      "loss": 0.0402,
      "num_input_tokens_seen": 2656928,
      "step": 9490
    },
    {
      "epoch": 105.5,
      "grad_norm": 0.08878660202026367,
      "learning_rate": 4.336605763012733e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 2658320,
      "step": 9495
    },
    {
      "epoch": 105.55555555555556,
      "grad_norm": 0.5757137537002563,
      "learning_rate": 4.3359395501417026e-05,
      "loss": 0.0094,
      "num_input_tokens_seen": 2659696,
      "step": 9500
    },
    {
      "epoch": 105.61111111111111,
      "grad_norm": 0.06696310639381409,
      "learning_rate": 4.335273054145722e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 2661056,
      "step": 9505
    },
    {
      "epoch": 105.66666666666667,
      "grad_norm": 0.06764277815818787,
      "learning_rate": 4.334606275127572e-05,
      "loss": 0.0491,
      "num_input_tokens_seen": 2662480,
      "step": 9510
    },
    {
      "epoch": 105.72222222222223,
      "grad_norm": 0.06248727813363075,
      "learning_rate": 4.33393921319008e-05,
      "loss": 0.0425,
      "num_input_tokens_seen": 2663872,
      "step": 9515
    },
    {
      "epoch": 105.77777777777777,
      "grad_norm": 1.727818250656128,
      "learning_rate": 4.3332718684361146e-05,
      "loss": 0.0482,
      "num_input_tokens_seen": 2665264,
      "step": 9520
    },
    {
      "epoch": 105.83333333333333,
      "grad_norm": 0.2573845088481903,
      "learning_rate": 4.332604240968588e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 2666672,
      "step": 9525
    },
    {
      "epoch": 105.88888888888889,
      "grad_norm": 0.05246538668870926,
      "learning_rate": 4.331936330890459e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2668080,
      "step": 9530
    },
    {
      "epoch": 105.94444444444444,
      "grad_norm": 0.4305064082145691,
      "learning_rate": 4.331268138304725e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 2669472,
      "step": 9535
    },
    {
      "epoch": 106.0,
      "grad_norm": 0.044286761432886124,
      "learning_rate": 4.330599663314431e-05,
      "loss": 0.1136,
      "num_input_tokens_seen": 2670864,
      "step": 9540
    },
    {
      "epoch": 106.05555555555556,
      "grad_norm": 0.046738967299461365,
      "learning_rate": 4.329930906022665e-05,
      "loss": 0.0322,
      "num_input_tokens_seen": 2672288,
      "step": 9545
    },
    {
      "epoch": 106.11111111111111,
      "grad_norm": 0.11084245890378952,
      "learning_rate": 4.3292618665325564e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 2673712,
      "step": 9550
    },
    {
      "epoch": 106.16666666666667,
      "grad_norm": 0.18425849080085754,
      "learning_rate": 4.3285925449472796e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 2675104,
      "step": 9555
    },
    {
      "epoch": 106.22222222222223,
      "grad_norm": 0.177049458026886,
      "learning_rate": 4.327922941370054e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 2676464,
      "step": 9560
    },
    {
      "epoch": 106.27777777777777,
      "grad_norm": 0.05609956383705139,
      "learning_rate": 4.3272530559041384e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2677904,
      "step": 9565
    },
    {
      "epoch": 106.33333333333333,
      "grad_norm": 0.4893060624599457,
      "learning_rate": 4.32658288865284e-05,
      "loss": 0.0046,
      "num_input_tokens_seen": 2679280,
      "step": 9570
    },
    {
      "epoch": 106.38888888888889,
      "grad_norm": 0.08910650759935379,
      "learning_rate": 4.325912439719505e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2680672,
      "step": 9575
    },
    {
      "epoch": 106.44444444444444,
      "grad_norm": 7.399765968322754,
      "learning_rate": 4.3252417092075266e-05,
      "loss": 0.0184,
      "num_input_tokens_seen": 2682064,
      "step": 9580
    },
    {
      "epoch": 106.5,
      "grad_norm": 0.05228490009903908,
      "learning_rate": 4.3245706972203385e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 2683488,
      "step": 9585
    },
    {
      "epoch": 106.55555555555556,
      "grad_norm": 0.5593752861022949,
      "learning_rate": 4.323899403861421e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 2684896,
      "step": 9590
    },
    {
      "epoch": 106.61111111111111,
      "grad_norm": 0.8429399132728577,
      "learning_rate": 4.3232278292342935e-05,
      "loss": 0.0621,
      "num_input_tokens_seen": 2686272,
      "step": 9595
    },
    {
      "epoch": 106.66666666666667,
      "grad_norm": 0.09929218143224716,
      "learning_rate": 4.322555973442524e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 2687728,
      "step": 9600
    },
    {
      "epoch": 106.66666666666667,
      "eval_loss": 0.3666020929813385,
      "eval_runtime": 1.2475,
      "eval_samples_per_second": 32.065,
      "eval_steps_per_second": 16.033,
      "num_input_tokens_seen": 2687728,
      "step": 9600
    },
    {
      "epoch": 106.72222222222223,
      "grad_norm": 0.3283708989620209,
      "learning_rate": 4.3218838365897184e-05,
      "loss": 0.0492,
      "num_input_tokens_seen": 2689136,
      "step": 9605
    },
    {
      "epoch": 106.77777777777777,
      "grad_norm": 0.015223176218569279,
      "learning_rate": 4.3212114187795306e-05,
      "loss": 0.0644,
      "num_input_tokens_seen": 2690544,
      "step": 9610
    },
    {
      "epoch": 106.83333333333333,
      "grad_norm": 0.024633722379803658,
      "learning_rate": 4.320538720115656e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 2691936,
      "step": 9615
    },
    {
      "epoch": 106.88888888888889,
      "grad_norm": 0.058470409363508224,
      "learning_rate": 4.319865740701831e-05,
      "loss": 0.0097,
      "num_input_tokens_seen": 2693344,
      "step": 9620
    },
    {
      "epoch": 106.94444444444444,
      "grad_norm": 3.6278555393218994,
      "learning_rate": 4.3191924806418396e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 2694720,
      "step": 9625
    },
    {
      "epoch": 107.0,
      "grad_norm": 0.8226507306098938,
      "learning_rate": 4.318518940039507e-05,
      "loss": 0.0476,
      "num_input_tokens_seen": 2696112,
      "step": 9630
    },
    {
      "epoch": 107.05555555555556,
      "grad_norm": 2.4026291370391846,
      "learning_rate": 4.3178451189987e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 2697536,
      "step": 9635
    },
    {
      "epoch": 107.11111111111111,
      "grad_norm": 0.3539431095123291,
      "learning_rate": 4.3171710176233315e-05,
      "loss": 0.037,
      "num_input_tokens_seen": 2698976,
      "step": 9640
    },
    {
      "epoch": 107.16666666666667,
      "grad_norm": 0.06794818490743637,
      "learning_rate": 4.316496636017355e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 2700384,
      "step": 9645
    },
    {
      "epoch": 107.22222222222223,
      "grad_norm": 1.120903491973877,
      "learning_rate": 4.315821974284771e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 2701792,
      "step": 9650
    },
    {
      "epoch": 107.27777777777777,
      "grad_norm": 0.14962612092494965,
      "learning_rate": 4.315147032529619e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 2703184,
      "step": 9655
    },
    {
      "epoch": 107.33333333333333,
      "grad_norm": 0.14310309290885925,
      "learning_rate": 4.3144718108559845e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 2704560,
      "step": 9660
    },
    {
      "epoch": 107.38888888888889,
      "grad_norm": 0.5003029108047485,
      "learning_rate": 4.3137963093679945e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 2705968,
      "step": 9665
    },
    {
      "epoch": 107.44444444444444,
      "grad_norm": 0.022811533883213997,
      "learning_rate": 4.31312052816982e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 2707376,
      "step": 9670
    },
    {
      "epoch": 107.5,
      "grad_norm": 0.35733872652053833,
      "learning_rate": 4.312444467365675e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 2708768,
      "step": 9675
    },
    {
      "epoch": 107.55555555555556,
      "grad_norm": 0.0862656757235527,
      "learning_rate": 4.311768127059816e-05,
      "loss": 0.0126,
      "num_input_tokens_seen": 2710160,
      "step": 9680
    },
    {
      "epoch": 107.61111111111111,
      "grad_norm": 0.01054824236780405,
      "learning_rate": 4.3110915073565444e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 2711584,
      "step": 9685
    },
    {
      "epoch": 107.66666666666667,
      "grad_norm": 3.6166446208953857,
      "learning_rate": 4.310414608360203e-05,
      "loss": 0.0319,
      "num_input_tokens_seen": 2712960,
      "step": 9690
    },
    {
      "epoch": 107.72222222222223,
      "grad_norm": 0.09258484840393066,
      "learning_rate": 4.309737430175177e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 2714368,
      "step": 9695
    },
    {
      "epoch": 107.77777777777777,
      "grad_norm": 0.15295188128948212,
      "learning_rate": 4.309059972905897e-05,
      "loss": 0.0587,
      "num_input_tokens_seen": 2715760,
      "step": 9700
    },
    {
      "epoch": 107.83333333333333,
      "grad_norm": 0.0354779027402401,
      "learning_rate": 4.308382236656836e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2717200,
      "step": 9705
    },
    {
      "epoch": 107.88888888888889,
      "grad_norm": 0.049514707177877426,
      "learning_rate": 4.307704221532507e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 2718624,
      "step": 9710
    },
    {
      "epoch": 107.94444444444444,
      "grad_norm": 4.531957149505615,
      "learning_rate": 4.307025927637471e-05,
      "loss": 0.0626,
      "num_input_tokens_seen": 2720032,
      "step": 9715
    },
    {
      "epoch": 108.0,
      "grad_norm": 0.028617417439818382,
      "learning_rate": 4.306347355076328e-05,
      "loss": 0.0388,
      "num_input_tokens_seen": 2721392,
      "step": 9720
    },
    {
      "epoch": 108.05555555555556,
      "grad_norm": 0.07500846683979034,
      "learning_rate": 4.305668503953724e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 2722784,
      "step": 9725
    },
    {
      "epoch": 108.11111111111111,
      "grad_norm": 1.4555811882019043,
      "learning_rate": 4.3049893743743436e-05,
      "loss": 0.0083,
      "num_input_tokens_seen": 2724192,
      "step": 9730
    },
    {
      "epoch": 108.16666666666667,
      "grad_norm": 5.864162921905518,
      "learning_rate": 4.304309966442919e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 2725600,
      "step": 9735
    },
    {
      "epoch": 108.22222222222223,
      "grad_norm": 0.04134637862443924,
      "learning_rate": 4.303630280264224e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2726992,
      "step": 9740
    },
    {
      "epoch": 108.27777777777777,
      "grad_norm": 0.3759552538394928,
      "learning_rate": 4.302950315943074e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 2728416,
      "step": 9745
    },
    {
      "epoch": 108.33333333333333,
      "grad_norm": 0.3840722441673279,
      "learning_rate": 4.3022700735843275e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 2729792,
      "step": 9750
    },
    {
      "epoch": 108.38888888888889,
      "grad_norm": 0.10974632948637009,
      "learning_rate": 4.301589553292887e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 2731200,
      "step": 9755
    },
    {
      "epoch": 108.44444444444444,
      "grad_norm": 0.011812030337750912,
      "learning_rate": 4.300908755173697e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 2732640,
      "step": 9760
    },
    {
      "epoch": 108.5,
      "grad_norm": 0.9177294969558716,
      "learning_rate": 4.300227679331745e-05,
      "loss": 0.0085,
      "num_input_tokens_seen": 2734016,
      "step": 9765
    },
    {
      "epoch": 108.55555555555556,
      "grad_norm": 0.056173477321863174,
      "learning_rate": 4.299546325872063e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 2735376,
      "step": 9770
    },
    {
      "epoch": 108.61111111111111,
      "grad_norm": 0.08160650730133057,
      "learning_rate": 4.2988646948997225e-05,
      "loss": 0.0058,
      "num_input_tokens_seen": 2736800,
      "step": 9775
    },
    {
      "epoch": 108.66666666666667,
      "grad_norm": 6.756847381591797,
      "learning_rate": 4.29818278651984e-05,
      "loss": 0.0112,
      "num_input_tokens_seen": 2738176,
      "step": 9780
    },
    {
      "epoch": 108.72222222222223,
      "grad_norm": 0.2062033861875534,
      "learning_rate": 4.297500600837574e-05,
      "loss": 0.0733,
      "num_input_tokens_seen": 2739552,
      "step": 9785
    },
    {
      "epoch": 108.77777777777777,
      "grad_norm": 0.06588231772184372,
      "learning_rate": 4.2968181379581276e-05,
      "loss": 0.0382,
      "num_input_tokens_seen": 2740960,
      "step": 9790
    },
    {
      "epoch": 108.83333333333333,
      "grad_norm": 0.030483702197670937,
      "learning_rate": 4.296135397986743e-05,
      "loss": 0.0758,
      "num_input_tokens_seen": 2742368,
      "step": 9795
    },
    {
      "epoch": 108.88888888888889,
      "grad_norm": 0.03212720528244972,
      "learning_rate": 4.295452381028709e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 2743792,
      "step": 9800
    },
    {
      "epoch": 108.88888888888889,
      "eval_loss": 0.2994301915168762,
      "eval_runtime": 1.247,
      "eval_samples_per_second": 32.077,
      "eval_steps_per_second": 16.039,
      "num_input_tokens_seen": 2743792,
      "step": 9800
    },
    {
      "epoch": 108.94444444444444,
      "grad_norm": 0.0608208104968071,
      "learning_rate": 4.294769087189354e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 2745184,
      "step": 9805
    },
    {
      "epoch": 109.0,
      "grad_norm": 0.03961048647761345,
      "learning_rate": 4.294085516574052e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2746592,
      "step": 9810
    },
    {
      "epoch": 109.05555555555556,
      "grad_norm": 0.1278054267168045,
      "learning_rate": 4.2934016692882176e-05,
      "loss": 0.0386,
      "num_input_tokens_seen": 2748016,
      "step": 9815
    },
    {
      "epoch": 109.11111111111111,
      "grad_norm": 4.340808868408203,
      "learning_rate": 4.292717545437308e-05,
      "loss": 0.0086,
      "num_input_tokens_seen": 2749408,
      "step": 9820
    },
    {
      "epoch": 109.16666666666667,
      "grad_norm": 1.0840847492218018,
      "learning_rate": 4.292033145126825e-05,
      "loss": 0.0335,
      "num_input_tokens_seen": 2750816,
      "step": 9825
    },
    {
      "epoch": 109.22222222222223,
      "grad_norm": 0.07237670570611954,
      "learning_rate": 4.29134846846231e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 2752224,
      "step": 9830
    },
    {
      "epoch": 109.27777777777777,
      "grad_norm": 0.023615075275301933,
      "learning_rate": 4.29066351554935e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2753616,
      "step": 9835
    },
    {
      "epoch": 109.33333333333333,
      "grad_norm": 0.05121598020195961,
      "learning_rate": 4.289978286493574e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 2754992,
      "step": 9840
    },
    {
      "epoch": 109.38888888888889,
      "grad_norm": 0.17053674161434174,
      "learning_rate": 4.28929278140065e-05,
      "loss": 0.016,
      "num_input_tokens_seen": 2756400,
      "step": 9845
    },
    {
      "epoch": 109.44444444444444,
      "grad_norm": 3.3443965911865234,
      "learning_rate": 4.288607000376295e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 2757760,
      "step": 9850
    },
    {
      "epoch": 109.5,
      "grad_norm": 0.013281348161399364,
      "learning_rate": 4.2879209435262624e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 2759216,
      "step": 9855
    },
    {
      "epoch": 109.55555555555556,
      "grad_norm": 0.04490074887871742,
      "learning_rate": 4.287234610956353e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 2760576,
      "step": 9860
    },
    {
      "epoch": 109.61111111111111,
      "grad_norm": 0.1046503409743309,
      "learning_rate": 4.2865480027724056e-05,
      "loss": 0.0642,
      "num_input_tokens_seen": 2761968,
      "step": 9865
    },
    {
      "epoch": 109.66666666666667,
      "grad_norm": 0.050052471458911896,
      "learning_rate": 4.285861119080306e-05,
      "loss": 0.0451,
      "num_input_tokens_seen": 2763376,
      "step": 9870
    },
    {
      "epoch": 109.72222222222223,
      "grad_norm": 0.023944802582263947,
      "learning_rate": 4.2851739599859784e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 2764768,
      "step": 9875
    },
    {
      "epoch": 109.77777777777777,
      "grad_norm": 0.35358065366744995,
      "learning_rate": 4.2844865255953934e-05,
      "loss": 0.0361,
      "num_input_tokens_seen": 2766208,
      "step": 9880
    },
    {
      "epoch": 109.83333333333333,
      "grad_norm": 0.2820456922054291,
      "learning_rate": 4.2837988160145605e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 2767616,
      "step": 9885
    },
    {
      "epoch": 109.88888888888889,
      "grad_norm": 0.16596360504627228,
      "learning_rate": 4.2831108313495336e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 2769008,
      "step": 9890
    },
    {
      "epoch": 109.94444444444444,
      "grad_norm": 0.0757044330239296,
      "learning_rate": 4.282422571706408e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2770416,
      "step": 9895
    },
    {
      "epoch": 110.0,
      "grad_norm": 0.06280745565891266,
      "learning_rate": 4.281734037191323e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 2771824,
      "step": 9900
    },
    {
      "epoch": 110.05555555555556,
      "grad_norm": 2.98471999168396,
      "learning_rate": 4.281045227910459e-05,
      "loss": 0.0569,
      "num_input_tokens_seen": 2773216,
      "step": 9905
    },
    {
      "epoch": 110.11111111111111,
      "grad_norm": 0.02303474210202694,
      "learning_rate": 4.280356143970038e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 2774608,
      "step": 9910
    },
    {
      "epoch": 110.16666666666667,
      "grad_norm": 0.1131405383348465,
      "learning_rate": 4.279666785476327e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 2776064,
      "step": 9915
    },
    {
      "epoch": 110.22222222222223,
      "grad_norm": 0.07358348369598389,
      "learning_rate": 4.2789771525356325e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 2777472,
      "step": 9920
    },
    {
      "epoch": 110.27777777777777,
      "grad_norm": 0.04604744166135788,
      "learning_rate": 4.2782872452543056e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 2778896,
      "step": 9925
    },
    {
      "epoch": 110.33333333333333,
      "grad_norm": 0.06458190083503723,
      "learning_rate": 4.2775970637387376e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 2780272,
      "step": 9930
    },
    {
      "epoch": 110.38888888888889,
      "grad_norm": 0.041224751621484756,
      "learning_rate": 4.276906608095363e-05,
      "loss": 0.0656,
      "num_input_tokens_seen": 2781680,
      "step": 9935
    },
    {
      "epoch": 110.44444444444444,
      "grad_norm": 1.0473823547363281,
      "learning_rate": 4.276215878430661e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 2783072,
      "step": 9940
    },
    {
      "epoch": 110.5,
      "grad_norm": 0.5123445391654968,
      "learning_rate": 4.275524874851149e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 2784480,
      "step": 9945
    },
    {
      "epoch": 110.55555555555556,
      "grad_norm": 8.053047180175781,
      "learning_rate": 4.274833597463388e-05,
      "loss": 0.0173,
      "num_input_tokens_seen": 2785904,
      "step": 9950
    },
    {
      "epoch": 110.61111111111111,
      "grad_norm": 0.10567926615476608,
      "learning_rate": 4.2741420463739824e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 2787328,
      "step": 9955
    },
    {
      "epoch": 110.66666666666667,
      "grad_norm": 0.2918130159378052,
      "learning_rate": 4.273450221689578e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 2788720,
      "step": 9960
    },
    {
      "epoch": 110.72222222222223,
      "grad_norm": 1.0730671882629395,
      "learning_rate": 4.272758123516863e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 2790160,
      "step": 9965
    },
    {
      "epoch": 110.77777777777777,
      "grad_norm": 0.8056572675704956,
      "learning_rate": 4.272065751962567e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 2791552,
      "step": 9970
    },
    {
      "epoch": 110.83333333333333,
      "grad_norm": 0.13064098358154297,
      "learning_rate": 4.271373107133464e-05,
      "loss": 0.0385,
      "num_input_tokens_seen": 2792928,
      "step": 9975
    },
    {
      "epoch": 110.88888888888889,
      "grad_norm": 0.14278151094913483,
      "learning_rate": 4.270680189136366e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 2794304,
      "step": 9980
    },
    {
      "epoch": 110.94444444444444,
      "grad_norm": 0.03691628947854042,
      "learning_rate": 4.269986998078132e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 2795648,
      "step": 9985
    },
    {
      "epoch": 111.0,
      "grad_norm": 0.45458897948265076,
      "learning_rate": 4.2692935340656595e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 2797056,
      "step": 9990
    },
    {
      "epoch": 111.05555555555556,
      "grad_norm": 0.012741127982735634,
      "learning_rate": 4.26859979720589e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 2798432,
      "step": 9995
    },
    {
      "epoch": 111.11111111111111,
      "grad_norm": 0.02861490286886692,
      "learning_rate": 4.267905787605806e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 2799840,
      "step": 10000
    },
    {
      "epoch": 111.11111111111111,
      "eval_loss": 0.32432129979133606,
      "eval_runtime": 1.2511,
      "eval_samples_per_second": 31.971,
      "eval_steps_per_second": 15.986,
      "num_input_tokens_seen": 2799840,
      "step": 10000
    },
    {
      "epoch": 111.16666666666667,
      "grad_norm": 0.08306031674146652,
      "learning_rate": 4.267211505372433e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 2801232,
      "step": 10005
    },
    {
      "epoch": 111.22222222222223,
      "grad_norm": 0.044105369597673416,
      "learning_rate": 4.266516950612837e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 2802608,
      "step": 10010
    },
    {
      "epoch": 111.27777777777777,
      "grad_norm": 0.10700410604476929,
      "learning_rate": 4.265822123434128e-05,
      "loss": 0.0349,
      "num_input_tokens_seen": 2804032,
      "step": 10015
    },
    {
      "epoch": 111.33333333333333,
      "grad_norm": 0.06296148896217346,
      "learning_rate": 4.265127023943457e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 2805440,
      "step": 10020
    },
    {
      "epoch": 111.38888888888889,
      "grad_norm": 0.9111348986625671,
      "learning_rate": 4.2644316522480176e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2806864,
      "step": 10025
    },
    {
      "epoch": 111.44444444444444,
      "grad_norm": 0.30371686816215515,
      "learning_rate": 4.263736008455044e-05,
      "loss": 0.0417,
      "num_input_tokens_seen": 2808272,
      "step": 10030
    },
    {
      "epoch": 111.5,
      "grad_norm": 0.05492471531033516,
      "learning_rate": 4.2630400926718125e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 2809664,
      "step": 10035
    },
    {
      "epoch": 111.55555555555556,
      "grad_norm": 0.048918820917606354,
      "learning_rate": 4.262343905005644e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 2811072,
      "step": 10040
    },
    {
      "epoch": 111.61111111111111,
      "grad_norm": 0.03517802059650421,
      "learning_rate": 4.261647445563897e-05,
      "loss": 0.0043,
      "num_input_tokens_seen": 2812464,
      "step": 10045
    },
    {
      "epoch": 111.66666666666667,
      "grad_norm": 0.05828617513179779,
      "learning_rate": 4.260950714453976e-05,
      "loss": 0.011,
      "num_input_tokens_seen": 2813856,
      "step": 10050
    },
    {
      "epoch": 111.72222222222223,
      "grad_norm": 0.031171919777989388,
      "learning_rate": 4.2602537117833266e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 2815232,
      "step": 10055
    },
    {
      "epoch": 111.77777777777777,
      "grad_norm": 0.7477172613143921,
      "learning_rate": 4.259556437659433e-05,
      "loss": 0.0622,
      "num_input_tokens_seen": 2816608,
      "step": 10060
    },
    {
      "epoch": 111.83333333333333,
      "grad_norm": 0.04342026263475418,
      "learning_rate": 4.258858892189825e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 2818016,
      "step": 10065
    },
    {
      "epoch": 111.88888888888889,
      "grad_norm": 0.2020392268896103,
      "learning_rate": 4.2581610754820725e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 2819376,
      "step": 10070
    },
    {
      "epoch": 111.94444444444444,
      "grad_norm": 0.05071524903178215,
      "learning_rate": 4.2574629876437876e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 2820752,
      "step": 10075
    },
    {
      "epoch": 112.0,
      "grad_norm": 0.019734393805265427,
      "learning_rate": 4.256764628782625e-05,
      "loss": 0.0396,
      "num_input_tokens_seen": 2822160,
      "step": 10080
    },
    {
      "epoch": 112.05555555555556,
      "grad_norm": 0.025327319279313087,
      "learning_rate": 4.256065999006279e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 2823520,
      "step": 10085
    },
    {
      "epoch": 112.11111111111111,
      "grad_norm": 0.09475494176149368,
      "learning_rate": 4.2553670984224885e-05,
      "loss": 0.1122,
      "num_input_tokens_seen": 2824960,
      "step": 10090
    },
    {
      "epoch": 112.16666666666667,
      "grad_norm": 0.35623660683631897,
      "learning_rate": 4.254667927139032e-05,
      "loss": 0.0099,
      "num_input_tokens_seen": 2826368,
      "step": 10095
    },
    {
      "epoch": 112.22222222222223,
      "grad_norm": 0.2352205216884613,
      "learning_rate": 4.2539684852637295e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 2827760,
      "step": 10100
    },
    {
      "epoch": 112.27777777777777,
      "grad_norm": 0.12840689718723297,
      "learning_rate": 4.253268772904446e-05,
      "loss": 0.0347,
      "num_input_tokens_seen": 2829152,
      "step": 10105
    },
    {
      "epoch": 112.33333333333333,
      "grad_norm": 0.05855895206332207,
      "learning_rate": 4.252568790169085e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 2830576,
      "step": 10110
    },
    {
      "epoch": 112.38888888888889,
      "grad_norm": 0.15385712683200836,
      "learning_rate": 4.251868537165592e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 2831968,
      "step": 10115
    },
    {
      "epoch": 112.44444444444444,
      "grad_norm": 0.9982633590698242,
      "learning_rate": 4.251168014001955e-05,
      "loss": 0.0047,
      "num_input_tokens_seen": 2833376,
      "step": 10120
    },
    {
      "epoch": 112.5,
      "grad_norm": 0.07224177569150925,
      "learning_rate": 4.250467220786204e-05,
      "loss": 0.0462,
      "num_input_tokens_seen": 2834752,
      "step": 10125
    },
    {
      "epoch": 112.55555555555556,
      "grad_norm": 1.015007495880127,
      "learning_rate": 4.249766157626409e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 2836160,
      "step": 10130
    },
    {
      "epoch": 112.61111111111111,
      "grad_norm": 0.042412545531988144,
      "learning_rate": 4.249064824630684e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 2837568,
      "step": 10135
    },
    {
      "epoch": 112.66666666666667,
      "grad_norm": 9.665692329406738,
      "learning_rate": 4.248363221907183e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 2838928,
      "step": 10140
    },
    {
      "epoch": 112.72222222222223,
      "grad_norm": 1.5334535837173462,
      "learning_rate": 4.2476613495641026e-05,
      "loss": 0.0046,
      "num_input_tokens_seen": 2840272,
      "step": 10145
    },
    {
      "epoch": 112.77777777777777,
      "grad_norm": 1.2020920515060425,
      "learning_rate": 4.246959207709679e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 2841712,
      "step": 10150
    },
    {
      "epoch": 112.83333333333333,
      "grad_norm": 0.017242200672626495,
      "learning_rate": 4.246256796452192e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 2843136,
      "step": 10155
    },
    {
      "epoch": 112.88888888888889,
      "grad_norm": 0.9510799646377563,
      "learning_rate": 4.245554115899962e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2844528,
      "step": 10160
    },
    {
      "epoch": 112.94444444444444,
      "grad_norm": 0.02132437564432621,
      "learning_rate": 4.2448511661613514e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2845904,
      "step": 10165
    },
    {
      "epoch": 113.0,
      "grad_norm": 0.08225011080503464,
      "learning_rate": 4.2441479473447635e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 2847344,
      "step": 10170
    },
    {
      "epoch": 113.05555555555556,
      "grad_norm": 0.3802366256713867,
      "learning_rate": 4.243444459558644e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 2848736,
      "step": 10175
    },
    {
      "epoch": 113.11111111111111,
      "grad_norm": 0.13653898239135742,
      "learning_rate": 4.24274070291148e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 2850144,
      "step": 10180
    },
    {
      "epoch": 113.16666666666667,
      "grad_norm": 0.06455975770950317,
      "learning_rate": 4.242036677511798e-05,
      "loss": 0.0594,
      "num_input_tokens_seen": 2851568,
      "step": 10185
    },
    {
      "epoch": 113.22222222222223,
      "grad_norm": 0.19155587255954742,
      "learning_rate": 4.241332383468169e-05,
      "loss": 0.0121,
      "num_input_tokens_seen": 2852976,
      "step": 10190
    },
    {
      "epoch": 113.27777777777777,
      "grad_norm": 0.02459676004946232,
      "learning_rate": 4.2406278208892034e-05,
      "loss": 0.0434,
      "num_input_tokens_seen": 2854384,
      "step": 10195
    },
    {
      "epoch": 113.33333333333333,
      "grad_norm": 0.0855620801448822,
      "learning_rate": 4.2399229898835536e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 2855808,
      "step": 10200
    },
    {
      "epoch": 113.33333333333333,
      "eval_loss": 0.34162887930870056,
      "eval_runtime": 1.2376,
      "eval_samples_per_second": 32.321,
      "eval_steps_per_second": 16.161,
      "num_input_tokens_seen": 2855808,
      "step": 10200
    },
    {
      "epoch": 113.38888888888889,
      "grad_norm": 0.02958301268517971,
      "learning_rate": 4.239217890559914e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 2857184,
      "step": 10205
    },
    {
      "epoch": 113.44444444444444,
      "grad_norm": 0.021586384624242783,
      "learning_rate": 4.238512523027019e-05,
      "loss": 0.0358,
      "num_input_tokens_seen": 2858544,
      "step": 10210
    },
    {
      "epoch": 113.5,
      "grad_norm": 0.031129613518714905,
      "learning_rate": 4.237806887393645e-05,
      "loss": 0.0367,
      "num_input_tokens_seen": 2859936,
      "step": 10215
    },
    {
      "epoch": 113.55555555555556,
      "grad_norm": 0.2874327600002289,
      "learning_rate": 4.237100983768611e-05,
      "loss": 0.0058,
      "num_input_tokens_seen": 2861360,
      "step": 10220
    },
    {
      "epoch": 113.61111111111111,
      "grad_norm": 3.6743741035461426,
      "learning_rate": 4.2363948122607756e-05,
      "loss": 0.0453,
      "num_input_tokens_seen": 2862752,
      "step": 10225
    },
    {
      "epoch": 113.66666666666667,
      "grad_norm": 0.1414078027009964,
      "learning_rate": 4.235688372979039e-05,
      "loss": 0.0345,
      "num_input_tokens_seen": 2864064,
      "step": 10230
    },
    {
      "epoch": 113.72222222222223,
      "grad_norm": 0.07550901174545288,
      "learning_rate": 4.234981666032343e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 2865440,
      "step": 10235
    },
    {
      "epoch": 113.77777777777777,
      "grad_norm": 0.03705446794629097,
      "learning_rate": 4.2342746915296704e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 2866800,
      "step": 10240
    },
    {
      "epoch": 113.83333333333333,
      "grad_norm": 0.07860402762889862,
      "learning_rate": 4.233567449580047e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2868208,
      "step": 10245
    },
    {
      "epoch": 113.88888888888889,
      "grad_norm": 0.880873441696167,
      "learning_rate": 4.232859940292537e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 2869600,
      "step": 10250
    },
    {
      "epoch": 113.94444444444444,
      "grad_norm": 1.6943532228469849,
      "learning_rate": 4.232152163776248e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2871040,
      "step": 10255
    },
    {
      "epoch": 114.0,
      "grad_norm": 0.05955064296722412,
      "learning_rate": 4.231444120140328e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 2872448,
      "step": 10260
    },
    {
      "epoch": 114.05555555555556,
      "grad_norm": 0.10945514589548111,
      "learning_rate": 4.230735809493967e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 2873840,
      "step": 10265
    },
    {
      "epoch": 114.11111111111111,
      "grad_norm": 0.054482635110616684,
      "learning_rate": 4.2300272319463926e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 2875216,
      "step": 10270
    },
    {
      "epoch": 114.16666666666667,
      "grad_norm": 0.06047813966870308,
      "learning_rate": 4.2293183876068786e-05,
      "loss": 0.0442,
      "num_input_tokens_seen": 2876656,
      "step": 10275
    },
    {
      "epoch": 114.22222222222223,
      "grad_norm": 1.363086223602295,
      "learning_rate": 4.228609276584737e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 2878016,
      "step": 10280
    },
    {
      "epoch": 114.27777777777777,
      "grad_norm": 0.039529334753751755,
      "learning_rate": 4.227899898989323e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 2879376,
      "step": 10285
    },
    {
      "epoch": 114.33333333333333,
      "grad_norm": 0.06847165524959564,
      "learning_rate": 4.2271902549300293e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 2880768,
      "step": 10290
    },
    {
      "epoch": 114.38888888888889,
      "grad_norm": 0.045846860855817795,
      "learning_rate": 4.226480344516294e-05,
      "loss": 0.0495,
      "num_input_tokens_seen": 2882176,
      "step": 10295
    },
    {
      "epoch": 114.44444444444444,
      "grad_norm": 1.1900455951690674,
      "learning_rate": 4.2257701678575925e-05,
      "loss": 0.0383,
      "num_input_tokens_seen": 2883600,
      "step": 10300
    },
    {
      "epoch": 114.5,
      "grad_norm": 1.2559516429901123,
      "learning_rate": 4.225059725063444e-05,
      "loss": 0.0383,
      "num_input_tokens_seen": 2884976,
      "step": 10305
    },
    {
      "epoch": 114.55555555555556,
      "grad_norm": 0.018504228442907333,
      "learning_rate": 4.2243490162434074e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2886384,
      "step": 10310
    },
    {
      "epoch": 114.61111111111111,
      "grad_norm": 0.016500001773238182,
      "learning_rate": 4.223638041507083e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2887776,
      "step": 10315
    },
    {
      "epoch": 114.66666666666667,
      "grad_norm": 0.017323549836874008,
      "learning_rate": 4.2229268009641124e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 2889200,
      "step": 10320
    },
    {
      "epoch": 114.72222222222223,
      "grad_norm": 0.12853379547595978,
      "learning_rate": 4.222215294724177e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 2890624,
      "step": 10325
    },
    {
      "epoch": 114.77777777777777,
      "grad_norm": 0.14272119104862213,
      "learning_rate": 4.2215035228970005e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 2892032,
      "step": 10330
    },
    {
      "epoch": 114.83333333333333,
      "grad_norm": 0.2447228878736496,
      "learning_rate": 4.2207914855923464e-05,
      "loss": 0.0611,
      "num_input_tokens_seen": 2893472,
      "step": 10335
    },
    {
      "epoch": 114.88888888888889,
      "grad_norm": 0.20745891332626343,
      "learning_rate": 4.220079182920021e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 2894880,
      "step": 10340
    },
    {
      "epoch": 114.94444444444444,
      "grad_norm": 0.14942947030067444,
      "learning_rate": 4.2193666149898705e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 2896288,
      "step": 10345
    },
    {
      "epoch": 115.0,
      "grad_norm": 0.10265742987394333,
      "learning_rate": 4.21865378191178e-05,
      "loss": 0.0101,
      "num_input_tokens_seen": 2897648,
      "step": 10350
    },
    {
      "epoch": 115.05555555555556,
      "grad_norm": 0.03561759740114212,
      "learning_rate": 4.217940683795678e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 2899088,
      "step": 10355
    },
    {
      "epoch": 115.11111111111111,
      "grad_norm": 0.023727888241410255,
      "learning_rate": 4.217227320751534e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2900448,
      "step": 10360
    },
    {
      "epoch": 115.16666666666667,
      "grad_norm": 0.11586957424879074,
      "learning_rate": 4.216513692889358e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 2901856,
      "step": 10365
    },
    {
      "epoch": 115.22222222222223,
      "grad_norm": 0.05830713361501694,
      "learning_rate": 4.215799800319199e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 2903216,
      "step": 10370
    },
    {
      "epoch": 115.27777777777777,
      "grad_norm": 0.04649665206670761,
      "learning_rate": 4.2150856431511485e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 2904608,
      "step": 10375
    },
    {
      "epoch": 115.33333333333333,
      "grad_norm": 0.09990992397069931,
      "learning_rate": 4.214371221495339e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 2905984,
      "step": 10380
    },
    {
      "epoch": 115.38888888888889,
      "grad_norm": 0.012868402525782585,
      "learning_rate": 4.213656535461942e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 2907408,
      "step": 10385
    },
    {
      "epoch": 115.44444444444444,
      "grad_norm": 0.016251472756266594,
      "learning_rate": 4.2129415851611734e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 2908848,
      "step": 10390
    },
    {
      "epoch": 115.5,
      "grad_norm": 0.03224567696452141,
      "learning_rate": 4.2122263707032855e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 2910256,
      "step": 10395
    },
    {
      "epoch": 115.55555555555556,
      "grad_norm": 0.051119159907102585,
      "learning_rate": 4.211510892198574e-05,
      "loss": 0.0477,
      "num_input_tokens_seen": 2911648,
      "step": 10400
    },
    {
      "epoch": 115.55555555555556,
      "eval_loss": 0.3560078740119934,
      "eval_runtime": 1.2473,
      "eval_samples_per_second": 32.07,
      "eval_steps_per_second": 16.035,
      "num_input_tokens_seen": 2911648,
      "step": 10400
    },
    {
      "epoch": 115.61111111111111,
      "grad_norm": 0.016871972009539604,
      "learning_rate": 4.210795149757375e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 2913056,
      "step": 10405
    },
    {
      "epoch": 115.66666666666667,
      "grad_norm": 0.013140903785824776,
      "learning_rate": 4.210079143490065e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 2914448,
      "step": 10410
    },
    {
      "epoch": 115.72222222222223,
      "grad_norm": 0.025337636470794678,
      "learning_rate": 4.2093628735070604e-05,
      "loss": 0.0345,
      "num_input_tokens_seen": 2915856,
      "step": 10415
    },
    {
      "epoch": 115.77777777777777,
      "grad_norm": 0.029964357614517212,
      "learning_rate": 4.208646339918819e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 2917216,
      "step": 10420
    },
    {
      "epoch": 115.83333333333333,
      "grad_norm": 0.06450626254081726,
      "learning_rate": 4.2079295428358414e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 2918624,
      "step": 10425
    },
    {
      "epoch": 115.88888888888889,
      "grad_norm": 2.07902455329895,
      "learning_rate": 4.207212482368664e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 2920016,
      "step": 10430
    },
    {
      "epoch": 115.94444444444444,
      "grad_norm": 0.24898852407932281,
      "learning_rate": 4.206495158627867e-05,
      "loss": 0.0952,
      "num_input_tokens_seen": 2921440,
      "step": 10435
    },
    {
      "epoch": 116.0,
      "grad_norm": 0.04195913299918175,
      "learning_rate": 4.205777571724073e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 2922848,
      "step": 10440
    },
    {
      "epoch": 116.05555555555556,
      "grad_norm": 0.16449840366840363,
      "learning_rate": 4.20505972176794e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 2924272,
      "step": 10445
    },
    {
      "epoch": 116.11111111111111,
      "grad_norm": 0.7653898596763611,
      "learning_rate": 4.204341608870171e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 2925696,
      "step": 10450
    },
    {
      "epoch": 116.16666666666667,
      "grad_norm": 0.029619576409459114,
      "learning_rate": 4.203623233141508e-05,
      "loss": 0.0593,
      "num_input_tokens_seen": 2927104,
      "step": 10455
    },
    {
      "epoch": 116.22222222222223,
      "grad_norm": 0.016713934019207954,
      "learning_rate": 4.2029045946927334e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 2928560,
      "step": 10460
    },
    {
      "epoch": 116.27777777777777,
      "grad_norm": 0.09045770764350891,
      "learning_rate": 4.20218569363467e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 2929952,
      "step": 10465
    },
    {
      "epoch": 116.33333333333333,
      "grad_norm": 0.05236661434173584,
      "learning_rate": 4.2014665300781834e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 2931296,
      "step": 10470
    },
    {
      "epoch": 116.38888888888889,
      "grad_norm": 0.080479197204113,
      "learning_rate": 4.200747104134174e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 2932672,
      "step": 10475
    },
    {
      "epoch": 116.44444444444444,
      "grad_norm": 0.06805435568094254,
      "learning_rate": 4.200027415913588e-05,
      "loss": 0.035,
      "num_input_tokens_seen": 2934064,
      "step": 10480
    },
    {
      "epoch": 116.5,
      "grad_norm": 0.015032085590064526,
      "learning_rate": 4.1993074655274126e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 2935504,
      "step": 10485
    },
    {
      "epoch": 116.55555555555556,
      "grad_norm": 1.35832679271698,
      "learning_rate": 4.198587253086669e-05,
      "loss": 0.0465,
      "num_input_tokens_seen": 2936864,
      "step": 10490
    },
    {
      "epoch": 116.61111111111111,
      "grad_norm": 0.09040869772434235,
      "learning_rate": 4.197866778702426e-05,
      "loss": 0.0481,
      "num_input_tokens_seen": 2938256,
      "step": 10495
    },
    {
      "epoch": 116.66666666666667,
      "grad_norm": 0.040953416377305984,
      "learning_rate": 4.197146042485789e-05,
      "loss": 0.0335,
      "num_input_tokens_seen": 2939632,
      "step": 10500
    },
    {
      "epoch": 116.72222222222223,
      "grad_norm": 2.427051544189453,
      "learning_rate": 4.1964250445479046e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 2941056,
      "step": 10505
    },
    {
      "epoch": 116.77777777777777,
      "grad_norm": 0.04377278685569763,
      "learning_rate": 4.19570378499996e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 2942464,
      "step": 10510
    },
    {
      "epoch": 116.83333333333333,
      "grad_norm": 0.024317840114235878,
      "learning_rate": 4.194982263953182e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 2943920,
      "step": 10515
    },
    {
      "epoch": 116.88888888888889,
      "grad_norm": 0.014865126460790634,
      "learning_rate": 4.194260481518838e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 2945296,
      "step": 10520
    },
    {
      "epoch": 116.94444444444444,
      "grad_norm": 0.05182821676135063,
      "learning_rate": 4.1935384378082366e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 2946688,
      "step": 10525
    },
    {
      "epoch": 117.0,
      "grad_norm": 0.5076623558998108,
      "learning_rate": 4.1928161329327267e-05,
      "loss": 0.0049,
      "num_input_tokens_seen": 2948144,
      "step": 10530
    },
    {
      "epoch": 117.05555555555556,
      "grad_norm": 0.0654686763882637,
      "learning_rate": 4.1920935670036945e-05,
      "loss": 0.0271,
      "num_input_tokens_seen": 2949568,
      "step": 10535
    },
    {
      "epoch": 117.11111111111111,
      "grad_norm": 4.711313724517822,
      "learning_rate": 4.1913707401325705e-05,
      "loss": 0.0074,
      "num_input_tokens_seen": 2950976,
      "step": 10540
    },
    {
      "epoch": 117.16666666666667,
      "grad_norm": 0.05390075594186783,
      "learning_rate": 4.1906476524308235e-05,
      "loss": 0.0074,
      "num_input_tokens_seen": 2952368,
      "step": 10545
    },
    {
      "epoch": 117.22222222222223,
      "grad_norm": 0.1471150517463684,
      "learning_rate": 4.189924304009962e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 2953712,
      "step": 10550
    },
    {
      "epoch": 117.27777777777777,
      "grad_norm": 0.011929120868444443,
      "learning_rate": 4.189200694981537e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 2955104,
      "step": 10555
    },
    {
      "epoch": 117.33333333333333,
      "grad_norm": 0.6249105334281921,
      "learning_rate": 4.188476825457136e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 2956480,
      "step": 10560
    },
    {
      "epoch": 117.38888888888889,
      "grad_norm": 0.21178625524044037,
      "learning_rate": 4.18775269554839e-05,
      "loss": 0.0349,
      "num_input_tokens_seen": 2957920,
      "step": 10565
    },
    {
      "epoch": 117.44444444444444,
      "grad_norm": 0.02793416753411293,
      "learning_rate": 4.187028305366969e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 2959328,
      "step": 10570
    },
    {
      "epoch": 117.5,
      "grad_norm": 0.06807888299226761,
      "learning_rate": 4.1863036550245824e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2960784,
      "step": 10575
    },
    {
      "epoch": 117.55555555555556,
      "grad_norm": 1.0272268056869507,
      "learning_rate": 4.1855787446329806e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 2962208,
      "step": 10580
    },
    {
      "epoch": 117.61111111111111,
      "grad_norm": 0.08607570827007294,
      "learning_rate": 4.184853574303955e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 2963632,
      "step": 10585
    },
    {
      "epoch": 117.66666666666667,
      "grad_norm": 0.08444539457559586,
      "learning_rate": 4.184128144149334e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 2965056,
      "step": 10590
    },
    {
      "epoch": 117.72222222222223,
      "grad_norm": 0.08254580199718475,
      "learning_rate": 4.1834024542809896e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 2966480,
      "step": 10595
    },
    {
      "epoch": 117.77777777777777,
      "grad_norm": 1.594938039779663,
      "learning_rate": 4.1826765048108315e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 2967856,
      "step": 10600
    },
    {
      "epoch": 117.77777777777777,
      "eval_loss": 0.35155409574508667,
      "eval_runtime": 1.2324,
      "eval_samples_per_second": 32.457,
      "eval_steps_per_second": 16.228,
      "num_input_tokens_seen": 2967856,
      "step": 10600
    },
    {
      "epoch": 117.83333333333333,
      "grad_norm": 0.14947621524333954,
      "learning_rate": 4.181950295850811e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 2969232,
      "step": 10605
    },
    {
      "epoch": 117.88888888888889,
      "grad_norm": 0.08535938709974289,
      "learning_rate": 4.181223827512918e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 2970608,
      "step": 10610
    },
    {
      "epoch": 117.94444444444444,
      "grad_norm": 0.4395146369934082,
      "learning_rate": 4.180497099909183e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 2972000,
      "step": 10615
    },
    {
      "epoch": 118.0,
      "grad_norm": 0.07691317796707153,
      "learning_rate": 4.179770113151677e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 2973360,
      "step": 10620
    },
    {
      "epoch": 118.05555555555556,
      "grad_norm": 0.028592148795723915,
      "learning_rate": 4.179042867352511e-05,
      "loss": 0.0072,
      "num_input_tokens_seen": 2974816,
      "step": 10625
    },
    {
      "epoch": 118.11111111111111,
      "grad_norm": 0.08564648777246475,
      "learning_rate": 4.1783153626238334e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 2976192,
      "step": 10630
    },
    {
      "epoch": 118.16666666666667,
      "grad_norm": 0.9409798979759216,
      "learning_rate": 4.177587599077836e-05,
      "loss": 0.0559,
      "num_input_tokens_seen": 2977584,
      "step": 10635
    },
    {
      "epoch": 118.22222222222223,
      "grad_norm": 0.044715143740177155,
      "learning_rate": 4.1768595768267494e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 2978992,
      "step": 10640
    },
    {
      "epoch": 118.27777777777777,
      "grad_norm": 0.05061964690685272,
      "learning_rate": 4.176131295982843e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 2980400,
      "step": 10645
    },
    {
      "epoch": 118.33333333333333,
      "grad_norm": 0.12271451950073242,
      "learning_rate": 4.1754027566584276e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 2981760,
      "step": 10650
    },
    {
      "epoch": 118.38888888888889,
      "grad_norm": 7.82412576675415,
      "learning_rate": 4.174673958965852e-05,
      "loss": 0.0139,
      "num_input_tokens_seen": 2983168,
      "step": 10655
    },
    {
      "epoch": 118.44444444444444,
      "grad_norm": 0.10806886106729507,
      "learning_rate": 4.173944903017507e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 2984608,
      "step": 10660
    },
    {
      "epoch": 118.5,
      "grad_norm": 0.03443077951669693,
      "learning_rate": 4.173215588925822e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 2986000,
      "step": 10665
    },
    {
      "epoch": 118.55555555555556,
      "grad_norm": 0.03693753108382225,
      "learning_rate": 4.172486016803266e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 2987408,
      "step": 10670
    },
    {
      "epoch": 118.61111111111111,
      "grad_norm": 0.01773088425397873,
      "learning_rate": 4.171756186762349e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 2988848,
      "step": 10675
    },
    {
      "epoch": 118.66666666666667,
      "grad_norm": 0.10741207003593445,
      "learning_rate": 4.171026098915619e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 2990272,
      "step": 10680
    },
    {
      "epoch": 118.72222222222223,
      "grad_norm": 0.34780874848365784,
      "learning_rate": 4.170295753375665e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 2991664,
      "step": 10685
    },
    {
      "epoch": 118.77777777777777,
      "grad_norm": 0.03654148057103157,
      "learning_rate": 4.169565150255117e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 2993072,
      "step": 10690
    },
    {
      "epoch": 118.83333333333333,
      "grad_norm": 0.01164152566343546,
      "learning_rate": 4.16883428966664e-05,
      "loss": 0.035,
      "num_input_tokens_seen": 2994416,
      "step": 10695
    },
    {
      "epoch": 118.88888888888889,
      "grad_norm": 0.1035209521651268,
      "learning_rate": 4.168103171722944e-05,
      "loss": 0.0403,
      "num_input_tokens_seen": 2995824,
      "step": 10700
    },
    {
      "epoch": 118.94444444444444,
      "grad_norm": 0.19525747001171112,
      "learning_rate": 4.167371796536777e-05,
      "loss": 0.0428,
      "num_input_tokens_seen": 2997232,
      "step": 10705
    },
    {
      "epoch": 119.0,
      "grad_norm": 0.06947451084852219,
      "learning_rate": 4.166640164220924e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 2998624,
      "step": 10710
    },
    {
      "epoch": 119.05555555555556,
      "grad_norm": 0.11513161659240723,
      "learning_rate": 4.1659082748882144e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3000032,
      "step": 10715
    },
    {
      "epoch": 119.11111111111111,
      "grad_norm": 0.08787686377763748,
      "learning_rate": 4.1651761286515135e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 3001424,
      "step": 10720
    },
    {
      "epoch": 119.16666666666667,
      "grad_norm": 0.05675964057445526,
      "learning_rate": 4.164443725623728e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 3002848,
      "step": 10725
    },
    {
      "epoch": 119.22222222222223,
      "grad_norm": 1.0271410942077637,
      "learning_rate": 4.163711065917802e-05,
      "loss": 0.0708,
      "num_input_tokens_seen": 3004256,
      "step": 10730
    },
    {
      "epoch": 119.27777777777777,
      "grad_norm": 0.006219383329153061,
      "learning_rate": 4.1629781496467234e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3005632,
      "step": 10735
    },
    {
      "epoch": 119.33333333333333,
      "grad_norm": 0.03818584978580475,
      "learning_rate": 4.1622449769235164e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3007040,
      "step": 10740
    },
    {
      "epoch": 119.38888888888889,
      "grad_norm": 0.10516492277383804,
      "learning_rate": 4.161511547861243e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3008432,
      "step": 10745
    },
    {
      "epoch": 119.44444444444444,
      "grad_norm": 0.3229607343673706,
      "learning_rate": 4.1607778625730104e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 3009776,
      "step": 10750
    },
    {
      "epoch": 119.5,
      "grad_norm": 0.08866322785615921,
      "learning_rate": 4.160043921171961e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 3011168,
      "step": 10755
    },
    {
      "epoch": 119.55555555555556,
      "grad_norm": 0.051162250339984894,
      "learning_rate": 4.159309723771276e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3012528,
      "step": 10760
    },
    {
      "epoch": 119.61111111111111,
      "grad_norm": 0.274418443441391,
      "learning_rate": 4.158575270484181e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 3013952,
      "step": 10765
    },
    {
      "epoch": 119.66666666666667,
      "grad_norm": 8.894527435302734,
      "learning_rate": 4.157840561423936e-05,
      "loss": 0.0455,
      "num_input_tokens_seen": 3015360,
      "step": 10770
    },
    {
      "epoch": 119.72222222222223,
      "grad_norm": 0.5639622807502747,
      "learning_rate": 4.1571055967038416e-05,
      "loss": 0.0049,
      "num_input_tokens_seen": 3016752,
      "step": 10775
    },
    {
      "epoch": 119.77777777777777,
      "grad_norm": 0.6595057249069214,
      "learning_rate": 4.156370376437241e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 3018144,
      "step": 10780
    },
    {
      "epoch": 119.83333333333333,
      "grad_norm": 0.011230802163481712,
      "learning_rate": 4.155634900737513e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3019552,
      "step": 10785
    },
    {
      "epoch": 119.88888888888889,
      "grad_norm": 0.058229200541973114,
      "learning_rate": 4.1548991697180764e-05,
      "loss": 0.0047,
      "num_input_tokens_seen": 3020928,
      "step": 10790
    },
    {
      "epoch": 119.94444444444444,
      "grad_norm": 0.044473979622125626,
      "learning_rate": 4.1541631834923914e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3022384,
      "step": 10795
    },
    {
      "epoch": 120.0,
      "grad_norm": 0.01229845266789198,
      "learning_rate": 4.153426942173956e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3023792,
      "step": 10800
    },
    {
      "epoch": 120.0,
      "eval_loss": 0.3121834993362427,
      "eval_runtime": 1.2348,
      "eval_samples_per_second": 32.394,
      "eval_steps_per_second": 16.197,
      "num_input_tokens_seen": 3023792,
      "step": 10800
    },
    {
      "epoch": 120.05555555555556,
      "grad_norm": 2.7843925952911377,
      "learning_rate": 4.152690445876308e-05,
      "loss": 0.0352,
      "num_input_tokens_seen": 3025184,
      "step": 10805
    },
    {
      "epoch": 120.11111111111111,
      "grad_norm": 0.11164937168359756,
      "learning_rate": 4.1519536947130245e-05,
      "loss": 0.0046,
      "num_input_tokens_seen": 3026560,
      "step": 10810
    },
    {
      "epoch": 120.16666666666667,
      "grad_norm": 0.047469619661569595,
      "learning_rate": 4.151216688797722e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3028000,
      "step": 10815
    },
    {
      "epoch": 120.22222222222223,
      "grad_norm": 6.650125980377197,
      "learning_rate": 4.150479428244054e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 3029424,
      "step": 10820
    },
    {
      "epoch": 120.27777777777777,
      "grad_norm": 0.024651149287819862,
      "learning_rate": 4.1497419131657176e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 3030848,
      "step": 10825
    },
    {
      "epoch": 120.33333333333333,
      "grad_norm": 0.02268024906516075,
      "learning_rate": 4.149004143676447e-05,
      "loss": 0.0528,
      "num_input_tokens_seen": 3032288,
      "step": 10830
    },
    {
      "epoch": 120.38888888888889,
      "grad_norm": 0.06573624908924103,
      "learning_rate": 4.148266119890015e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3033728,
      "step": 10835
    },
    {
      "epoch": 120.44444444444444,
      "grad_norm": 0.07728659361600876,
      "learning_rate": 4.1475278419202324e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3035168,
      "step": 10840
    },
    {
      "epoch": 120.5,
      "grad_norm": 0.0661531314253807,
      "learning_rate": 4.146789309880953e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3036576,
      "step": 10845
    },
    {
      "epoch": 120.55555555555556,
      "grad_norm": 0.052616994827985764,
      "learning_rate": 4.146050523886068e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3037984,
      "step": 10850
    },
    {
      "epoch": 120.61111111111111,
      "grad_norm": 0.01939227432012558,
      "learning_rate": 4.1453114840495055e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3039328,
      "step": 10855
    },
    {
      "epoch": 120.66666666666667,
      "grad_norm": 0.41956213116645813,
      "learning_rate": 4.1445721904852364e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3040784,
      "step": 10860
    },
    {
      "epoch": 120.72222222222223,
      "grad_norm": 1.402909517288208,
      "learning_rate": 4.143832643307269e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 3042176,
      "step": 10865
    },
    {
      "epoch": 120.77777777777777,
      "grad_norm": 0.05879025161266327,
      "learning_rate": 4.1430928426296503e-05,
      "loss": 0.0729,
      "num_input_tokens_seen": 3043552,
      "step": 10870
    },
    {
      "epoch": 120.83333333333333,
      "grad_norm": 0.2505069673061371,
      "learning_rate": 4.142352788566466e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 3044880,
      "step": 10875
    },
    {
      "epoch": 120.88888888888889,
      "grad_norm": 0.14710335433483124,
      "learning_rate": 4.1416124812318424e-05,
      "loss": 0.0043,
      "num_input_tokens_seen": 3046256,
      "step": 10880
    },
    {
      "epoch": 120.94444444444444,
      "grad_norm": 0.04522951692342758,
      "learning_rate": 4.1408719207399453e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3047648,
      "step": 10885
    },
    {
      "epoch": 121.0,
      "grad_norm": 0.05621521919965744,
      "learning_rate": 4.140131107204978e-05,
      "loss": 0.0392,
      "num_input_tokens_seen": 3049024,
      "step": 10890
    },
    {
      "epoch": 121.05555555555556,
      "grad_norm": 0.22658008337020874,
      "learning_rate": 4.139390040741182e-05,
      "loss": 0.0766,
      "num_input_tokens_seen": 3050416,
      "step": 10895
    },
    {
      "epoch": 121.11111111111111,
      "grad_norm": 0.022637145593762398,
      "learning_rate": 4.1386487214628396e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3051808,
      "step": 10900
    },
    {
      "epoch": 121.16666666666667,
      "grad_norm": 0.07676830142736435,
      "learning_rate": 4.137907149484272e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3053216,
      "step": 10905
    },
    {
      "epoch": 121.22222222222223,
      "grad_norm": 0.04834340885281563,
      "learning_rate": 4.137165324919839e-05,
      "loss": 0.0617,
      "num_input_tokens_seen": 3054608,
      "step": 10910
    },
    {
      "epoch": 121.27777777777777,
      "grad_norm": 0.22578614950180054,
      "learning_rate": 4.136423247883939e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 3055984,
      "step": 10915
    },
    {
      "epoch": 121.33333333333333,
      "grad_norm": 0.05175184831023216,
      "learning_rate": 4.135680918491009e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 3057424,
      "step": 10920
    },
    {
      "epoch": 121.38888888888889,
      "grad_norm": 1.7012550830841064,
      "learning_rate": 4.1349383368555265e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 3058832,
      "step": 10925
    },
    {
      "epoch": 121.44444444444444,
      "grad_norm": 0.022323355078697205,
      "learning_rate": 4.1341955030920065e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3060240,
      "step": 10930
    },
    {
      "epoch": 121.5,
      "grad_norm": 0.02105037122964859,
      "learning_rate": 4.1334524173150036e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3061616,
      "step": 10935
    },
    {
      "epoch": 121.55555555555556,
      "grad_norm": 8.176363945007324,
      "learning_rate": 4.13270907963911e-05,
      "loss": 0.016,
      "num_input_tokens_seen": 3063008,
      "step": 10940
    },
    {
      "epoch": 121.61111111111111,
      "grad_norm": 0.04055251181125641,
      "learning_rate": 4.131965490178959e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3064368,
      "step": 10945
    },
    {
      "epoch": 121.66666666666667,
      "grad_norm": 0.17564457654953003,
      "learning_rate": 4.131221649049222e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3065792,
      "step": 10950
    },
    {
      "epoch": 121.72222222222223,
      "grad_norm": 0.018532341346144676,
      "learning_rate": 4.130477556364606e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 3067216,
      "step": 10955
    },
    {
      "epoch": 121.77777777777777,
      "grad_norm": 0.016523873433470726,
      "learning_rate": 4.129733212239861e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3068608,
      "step": 10960
    },
    {
      "epoch": 121.83333333333333,
      "grad_norm": 0.07357289642095566,
      "learning_rate": 4.128988616789774e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3070016,
      "step": 10965
    },
    {
      "epoch": 121.88888888888889,
      "grad_norm": 0.5128329992294312,
      "learning_rate": 4.1282437701291724e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3071424,
      "step": 10970
    },
    {
      "epoch": 121.94444444444444,
      "grad_norm": 0.08337633311748505,
      "learning_rate": 4.1274986723729184e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 3072848,
      "step": 10975
    },
    {
      "epoch": 122.0,
      "grad_norm": 0.042160067707300186,
      "learning_rate": 4.126753323635917e-05,
      "loss": 0.0366,
      "num_input_tokens_seen": 3074272,
      "step": 10980
    },
    {
      "epoch": 122.05555555555556,
      "grad_norm": 0.16574214398860931,
      "learning_rate": 4.12600772403311e-05,
      "loss": 0.0506,
      "num_input_tokens_seen": 3075664,
      "step": 10985
    },
    {
      "epoch": 122.11111111111111,
      "grad_norm": 0.38212165236473083,
      "learning_rate": 4.125261873679479e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3077120,
      "step": 10990
    },
    {
      "epoch": 122.16666666666667,
      "grad_norm": 0.06486321985721588,
      "learning_rate": 4.124515772690042e-05,
      "loss": 0.0351,
      "num_input_tokens_seen": 3078544,
      "step": 10995
    },
    {
      "epoch": 122.22222222222223,
      "grad_norm": 0.032676760107278824,
      "learning_rate": 4.123769421179858e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3079920,
      "step": 11000
    },
    {
      "epoch": 122.22222222222223,
      "eval_loss": 0.34335482120513916,
      "eval_runtime": 1.2336,
      "eval_samples_per_second": 32.426,
      "eval_steps_per_second": 16.213,
      "num_input_tokens_seen": 3079920,
      "step": 11000
    },
    {
      "epoch": 122.27777777777777,
      "grad_norm": 2.2346198558807373,
      "learning_rate": 4.1230228192640236e-05,
      "loss": 0.0394,
      "num_input_tokens_seen": 3081312,
      "step": 11005
    },
    {
      "epoch": 122.33333333333333,
      "grad_norm": 0.03839247301220894,
      "learning_rate": 4.122275967057675e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3082720,
      "step": 11010
    },
    {
      "epoch": 122.38888888888889,
      "grad_norm": 0.06526731699705124,
      "learning_rate": 4.1215288646759846e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3084112,
      "step": 11015
    },
    {
      "epoch": 122.44444444444444,
      "grad_norm": 9.837249755859375,
      "learning_rate": 4.120781512234166e-05,
      "loss": 0.0257,
      "num_input_tokens_seen": 3085456,
      "step": 11020
    },
    {
      "epoch": 122.5,
      "grad_norm": 0.07826729863882065,
      "learning_rate": 4.120033909847471e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 3086880,
      "step": 11025
    },
    {
      "epoch": 122.55555555555556,
      "grad_norm": 0.14304783940315247,
      "learning_rate": 4.119286057631187e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3088304,
      "step": 11030
    },
    {
      "epoch": 122.61111111111111,
      "grad_norm": 0.09197106212377548,
      "learning_rate": 4.118537955700646e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3089680,
      "step": 11035
    },
    {
      "epoch": 122.66666666666667,
      "grad_norm": 0.0499754473567009,
      "learning_rate": 4.11778960417121e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3091072,
      "step": 11040
    },
    {
      "epoch": 122.72222222222223,
      "grad_norm": 0.07389654964208603,
      "learning_rate": 4.117041003158288e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3092480,
      "step": 11045
    },
    {
      "epoch": 122.77777777777777,
      "grad_norm": 0.6234204769134521,
      "learning_rate": 4.1162921527773215e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3093888,
      "step": 11050
    },
    {
      "epoch": 122.83333333333333,
      "grad_norm": 0.03078206442296505,
      "learning_rate": 4.115543053143794e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3095296,
      "step": 11055
    },
    {
      "epoch": 122.88888888888889,
      "grad_norm": 0.03199067711830139,
      "learning_rate": 4.114793704373226e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 3096688,
      "step": 11060
    },
    {
      "epoch": 122.94444444444444,
      "grad_norm": 1.3160241842269897,
      "learning_rate": 4.114044106581175e-05,
      "loss": 0.0288,
      "num_input_tokens_seen": 3098112,
      "step": 11065
    },
    {
      "epoch": 123.0,
      "grad_norm": 0.12089758366346359,
      "learning_rate": 4.11329425988324e-05,
      "loss": 0.0043,
      "num_input_tokens_seen": 3099536,
      "step": 11070
    },
    {
      "epoch": 123.05555555555556,
      "grad_norm": 0.07184556871652603,
      "learning_rate": 4.112544164395056e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3100960,
      "step": 11075
    },
    {
      "epoch": 123.11111111111111,
      "grad_norm": 0.01735205203294754,
      "learning_rate": 4.111793820232297e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3102336,
      "step": 11080
    },
    {
      "epoch": 123.16666666666667,
      "grad_norm": 0.0564606711268425,
      "learning_rate": 4.1110432275106767e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 3103728,
      "step": 11085
    },
    {
      "epoch": 123.22222222222223,
      "grad_norm": 2.7219767570495605,
      "learning_rate": 4.110292386345944e-05,
      "loss": 0.0435,
      "num_input_tokens_seen": 3105136,
      "step": 11090
    },
    {
      "epoch": 123.27777777777777,
      "grad_norm": 0.1404159665107727,
      "learning_rate": 4.109541296853891e-05,
      "loss": 0.0286,
      "num_input_tokens_seen": 3106544,
      "step": 11095
    },
    {
      "epoch": 123.33333333333333,
      "grad_norm": 0.12032794207334518,
      "learning_rate": 4.108789959150341e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3107936,
      "step": 11100
    },
    {
      "epoch": 123.38888888888889,
      "grad_norm": 0.07227794826030731,
      "learning_rate": 4.108038373351163e-05,
      "loss": 0.052,
      "num_input_tokens_seen": 3109376,
      "step": 11105
    },
    {
      "epoch": 123.44444444444444,
      "grad_norm": 1.3064824342727661,
      "learning_rate": 4.10728653957226e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 3110784,
      "step": 11110
    },
    {
      "epoch": 123.5,
      "grad_norm": 0.04597649723291397,
      "learning_rate": 4.106534457929575e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3112224,
      "step": 11115
    },
    {
      "epoch": 123.55555555555556,
      "grad_norm": 0.24859586358070374,
      "learning_rate": 4.105782128539086e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3113648,
      "step": 11120
    },
    {
      "epoch": 123.61111111111111,
      "grad_norm": 0.11397011578083038,
      "learning_rate": 4.1050295515168144e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3114976,
      "step": 11125
    },
    {
      "epoch": 123.66666666666667,
      "grad_norm": 0.03498399257659912,
      "learning_rate": 4.1042767269788155e-05,
      "loss": 0.0371,
      "num_input_tokens_seen": 3116368,
      "step": 11130
    },
    {
      "epoch": 123.72222222222223,
      "grad_norm": 0.028425827622413635,
      "learning_rate": 4.103523655041185e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3117760,
      "step": 11135
    },
    {
      "epoch": 123.77777777777777,
      "grad_norm": 0.14292429387569427,
      "learning_rate": 4.102770335820055e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 3119168,
      "step": 11140
    },
    {
      "epoch": 123.83333333333333,
      "grad_norm": 0.12830479443073273,
      "learning_rate": 4.1020167694315984e-05,
      "loss": 0.0046,
      "num_input_tokens_seen": 3120544,
      "step": 11145
    },
    {
      "epoch": 123.88888888888889,
      "grad_norm": 0.13374431431293488,
      "learning_rate": 4.101262955992023e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3121952,
      "step": 11150
    },
    {
      "epoch": 123.94444444444444,
      "grad_norm": 0.04734085127711296,
      "learning_rate": 4.100508895617578e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3123312,
      "step": 11155
    },
    {
      "epoch": 124.0,
      "grad_norm": 0.06752769649028778,
      "learning_rate": 4.099754588424547e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3124720,
      "step": 11160
    },
    {
      "epoch": 124.05555555555556,
      "grad_norm": 0.02130618505179882,
      "learning_rate": 4.0990000345292546e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 3126128,
      "step": 11165
    },
    {
      "epoch": 124.11111111111111,
      "grad_norm": 1.805056095123291,
      "learning_rate": 4.098245234048064e-05,
      "loss": 0.0514,
      "num_input_tokens_seen": 3127552,
      "step": 11170
    },
    {
      "epoch": 124.16666666666667,
      "grad_norm": 0.08145522326231003,
      "learning_rate": 4.0974901870973726e-05,
      "loss": 0.0344,
      "num_input_tokens_seen": 3128960,
      "step": 11175
    },
    {
      "epoch": 124.22222222222223,
      "grad_norm": 0.06992384046316147,
      "learning_rate": 4.096734893793619e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 3130352,
      "step": 11180
    },
    {
      "epoch": 124.27777777777777,
      "grad_norm": 0.10212922096252441,
      "learning_rate": 4.095979354253279e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 3131712,
      "step": 11185
    },
    {
      "epoch": 124.33333333333333,
      "grad_norm": 0.12550660967826843,
      "learning_rate": 4.0952235685928656e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 3133088,
      "step": 11190
    },
    {
      "epoch": 124.38888888888889,
      "grad_norm": 0.06808155030012131,
      "learning_rate": 4.094467536928932e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 3134512,
      "step": 11195
    },
    {
      "epoch": 124.44444444444444,
      "grad_norm": 0.061751849949359894,
      "learning_rate": 4.093711259378067e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3135904,
      "step": 11200
    },
    {
      "epoch": 124.44444444444444,
      "eval_loss": 0.3747292459011078,
      "eval_runtime": 1.2361,
      "eval_samples_per_second": 32.359,
      "eval_steps_per_second": 16.179,
      "num_input_tokens_seen": 3135904,
      "step": 11200
    },
    {
      "epoch": 124.5,
      "grad_norm": 0.03419704735279083,
      "learning_rate": 4.092954736056897e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 3137344,
      "step": 11205
    },
    {
      "epoch": 124.55555555555556,
      "grad_norm": 0.018427107483148575,
      "learning_rate": 4.09219796708209e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 3138752,
      "step": 11210
    },
    {
      "epoch": 124.61111111111111,
      "grad_norm": 0.08116088807582855,
      "learning_rate": 4.0914409525703464e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3140160,
      "step": 11215
    },
    {
      "epoch": 124.66666666666667,
      "grad_norm": 0.05779992789030075,
      "learning_rate": 4.090683692638408e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3141536,
      "step": 11220
    },
    {
      "epoch": 124.72222222222223,
      "grad_norm": 0.0273369699716568,
      "learning_rate": 4.089926187403056e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3142992,
      "step": 11225
    },
    {
      "epoch": 124.77777777777777,
      "grad_norm": 0.3069452941417694,
      "learning_rate": 4.0891684369811044e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3144384,
      "step": 11230
    },
    {
      "epoch": 124.83333333333333,
      "grad_norm": 0.04549495875835419,
      "learning_rate": 4.0884104414894107e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 3145760,
      "step": 11235
    },
    {
      "epoch": 124.88888888888889,
      "grad_norm": 0.09461475908756256,
      "learning_rate": 4.087652201044864e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3147152,
      "step": 11240
    },
    {
      "epoch": 124.94444444444444,
      "grad_norm": 0.04290250316262245,
      "learning_rate": 4.086893715764397e-05,
      "loss": 0.0495,
      "num_input_tokens_seen": 3148560,
      "step": 11245
    },
    {
      "epoch": 125.0,
      "grad_norm": 0.1523188352584839,
      "learning_rate": 4.086134985764977e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 3149952,
      "step": 11250
    },
    {
      "epoch": 125.05555555555556,
      "grad_norm": 0.05399790778756142,
      "learning_rate": 4.0853760111636085e-05,
      "loss": 0.0405,
      "num_input_tokens_seen": 3151344,
      "step": 11255
    },
    {
      "epoch": 125.11111111111111,
      "grad_norm": 0.05009378492832184,
      "learning_rate": 4.084616792077337e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 3152736,
      "step": 11260
    },
    {
      "epoch": 125.16666666666667,
      "grad_norm": 0.06503690034151077,
      "learning_rate": 4.083857328623243e-05,
      "loss": 0.0311,
      "num_input_tokens_seen": 3154144,
      "step": 11265
    },
    {
      "epoch": 125.22222222222223,
      "grad_norm": 0.019885096698999405,
      "learning_rate": 4.083097620918444e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3155536,
      "step": 11270
    },
    {
      "epoch": 125.27777777777777,
      "grad_norm": 0.10807617008686066,
      "learning_rate": 4.082337669080097e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 3156912,
      "step": 11275
    },
    {
      "epoch": 125.33333333333333,
      "grad_norm": 0.03045574389398098,
      "learning_rate": 4.081577473225398e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3158304,
      "step": 11280
    },
    {
      "epoch": 125.38888888888889,
      "grad_norm": 0.05323393642902374,
      "learning_rate": 4.080817033471577e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3159776,
      "step": 11285
    },
    {
      "epoch": 125.44444444444444,
      "grad_norm": 0.033246129751205444,
      "learning_rate": 4.080056349935903e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 3161136,
      "step": 11290
    },
    {
      "epoch": 125.5,
      "grad_norm": 0.06261087208986282,
      "learning_rate": 4.079295422735684e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3162512,
      "step": 11295
    },
    {
      "epoch": 125.55555555555556,
      "grad_norm": 0.11130686104297638,
      "learning_rate": 4.078534251988264e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 3163920,
      "step": 11300
    },
    {
      "epoch": 125.61111111111111,
      "grad_norm": 0.018370861187577248,
      "learning_rate": 4.077772837811025e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3165280,
      "step": 11305
    },
    {
      "epoch": 125.66666666666667,
      "grad_norm": 0.2842008173465729,
      "learning_rate": 4.0770111803213874e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3166704,
      "step": 11310
    },
    {
      "epoch": 125.72222222222223,
      "grad_norm": 0.03744323179125786,
      "learning_rate": 4.076249279636807e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3168064,
      "step": 11315
    },
    {
      "epoch": 125.77777777777777,
      "grad_norm": 0.015187899582087994,
      "learning_rate": 4.075487135874781e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3169440,
      "step": 11320
    },
    {
      "epoch": 125.83333333333333,
      "grad_norm": 0.07003583014011383,
      "learning_rate": 4.074724749152837e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 3170880,
      "step": 11325
    },
    {
      "epoch": 125.88888888888889,
      "grad_norm": 0.035952016711235046,
      "learning_rate": 4.07396211958855e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 3172256,
      "step": 11330
    },
    {
      "epoch": 125.94444444444444,
      "grad_norm": 0.055059049278497696,
      "learning_rate": 4.073199247299523e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3173712,
      "step": 11335
    },
    {
      "epoch": 126.0,
      "grad_norm": 0.031124712899327278,
      "learning_rate": 4.072436132403403e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3175056,
      "step": 11340
    },
    {
      "epoch": 126.05555555555556,
      "grad_norm": 0.11699189990758896,
      "learning_rate": 4.0716727750178704e-05,
      "loss": 0.0372,
      "num_input_tokens_seen": 3176432,
      "step": 11345
    },
    {
      "epoch": 126.11111111111111,
      "grad_norm": 0.0785767063498497,
      "learning_rate": 4.0709091752606455e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3177840,
      "step": 11350
    },
    {
      "epoch": 126.16666666666667,
      "grad_norm": 0.025656215846538544,
      "learning_rate": 4.070145333249484e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3179232,
      "step": 11355
    },
    {
      "epoch": 126.22222222222223,
      "grad_norm": 0.07197126746177673,
      "learning_rate": 4.069381249102181e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 3180672,
      "step": 11360
    },
    {
      "epoch": 126.27777777777777,
      "grad_norm": 0.053250424563884735,
      "learning_rate": 4.0686169229365665e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3182080,
      "step": 11365
    },
    {
      "epoch": 126.33333333333333,
      "grad_norm": 0.07347400486469269,
      "learning_rate": 4.067852354870511e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3183472,
      "step": 11370
    },
    {
      "epoch": 126.38888888888889,
      "grad_norm": 7.606632232666016,
      "learning_rate": 4.067087545021919e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 3184880,
      "step": 11375
    },
    {
      "epoch": 126.44444444444444,
      "grad_norm": 0.39918598532676697,
      "learning_rate": 4.066322493508734e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 3186272,
      "step": 11380
    },
    {
      "epoch": 126.5,
      "grad_norm": 0.09750892966985703,
      "learning_rate": 4.065557200448937e-05,
      "loss": 0.0237,
      "num_input_tokens_seen": 3187648,
      "step": 11385
    },
    {
      "epoch": 126.55555555555556,
      "grad_norm": 0.041615165770053864,
      "learning_rate": 4.064791665960546e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3189040,
      "step": 11390
    },
    {
      "epoch": 126.61111111111111,
      "grad_norm": 1.9299057722091675,
      "learning_rate": 4.064025890161615e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 3190432,
      "step": 11395
    },
    {
      "epoch": 126.66666666666667,
      "grad_norm": 0.02910917066037655,
      "learning_rate": 4.0632598731702373e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 3191808,
      "step": 11400
    },
    {
      "epoch": 126.66666666666667,
      "eval_loss": 0.40956950187683105,
      "eval_runtime": 1.2504,
      "eval_samples_per_second": 31.99,
      "eval_steps_per_second": 15.995,
      "num_input_tokens_seen": 3191808,
      "step": 11400
    },
    {
      "epoch": 126.72222222222223,
      "grad_norm": 0.18580372631549835,
      "learning_rate": 4.0624936151045426e-05,
      "loss": 0.0477,
      "num_input_tokens_seen": 3193168,
      "step": 11405
    },
    {
      "epoch": 126.77777777777777,
      "grad_norm": 0.22865280508995056,
      "learning_rate": 4.061727116082696e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 3194576,
      "step": 11410
    },
    {
      "epoch": 126.83333333333333,
      "grad_norm": 0.0796361193060875,
      "learning_rate": 4.060960376222903e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 3196000,
      "step": 11415
    },
    {
      "epoch": 126.88888888888889,
      "grad_norm": 0.015758145600557327,
      "learning_rate": 4.0601933956434034e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 3197376,
      "step": 11420
    },
    {
      "epoch": 126.94444444444444,
      "grad_norm": 0.04947539418935776,
      "learning_rate": 4.059426174462476e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 3198816,
      "step": 11425
    },
    {
      "epoch": 127.0,
      "grad_norm": 0.03008107654750347,
      "learning_rate": 4.058658712798435e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3200208,
      "step": 11430
    },
    {
      "epoch": 127.05555555555556,
      "grad_norm": 3.6770663261413574,
      "learning_rate": 4.0578910107696336e-05,
      "loss": 0.0353,
      "num_input_tokens_seen": 3201584,
      "step": 11435
    },
    {
      "epoch": 127.11111111111111,
      "grad_norm": 0.060010187327861786,
      "learning_rate": 4.05712306849446e-05,
      "loss": 0.0179,
      "num_input_tokens_seen": 3203008,
      "step": 11440
    },
    {
      "epoch": 127.16666666666667,
      "grad_norm": 0.0695524662733078,
      "learning_rate": 4.0563548860913415e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3204384,
      "step": 11445
    },
    {
      "epoch": 127.22222222222223,
      "grad_norm": 0.03891254961490631,
      "learning_rate": 4.0555864636787414e-05,
      "loss": 0.0103,
      "num_input_tokens_seen": 3205776,
      "step": 11450
    },
    {
      "epoch": 127.27777777777777,
      "grad_norm": 0.04436562582850456,
      "learning_rate": 4.054817801375159e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 3207200,
      "step": 11455
    },
    {
      "epoch": 127.33333333333333,
      "grad_norm": 0.08256371319293976,
      "learning_rate": 4.054048899299134e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3208576,
      "step": 11460
    },
    {
      "epoch": 127.38888888888889,
      "grad_norm": 0.3906802237033844,
      "learning_rate": 4.0532797575692385e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3209936,
      "step": 11465
    },
    {
      "epoch": 127.44444444444444,
      "grad_norm": 0.049942389130592346,
      "learning_rate": 4.052510376304085e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3211344,
      "step": 11470
    },
    {
      "epoch": 127.5,
      "grad_norm": 0.18325626850128174,
      "learning_rate": 4.051740755622321e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3212720,
      "step": 11475
    },
    {
      "epoch": 127.55555555555556,
      "grad_norm": 0.06565779447555542,
      "learning_rate": 4.050970895642632e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 3214128,
      "step": 11480
    },
    {
      "epoch": 127.61111111111111,
      "grad_norm": 0.14183101058006287,
      "learning_rate": 4.050200796483741e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3215488,
      "step": 11485
    },
    {
      "epoch": 127.66666666666667,
      "grad_norm": 0.03292442485690117,
      "learning_rate": 4.049430458264405e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3216896,
      "step": 11490
    },
    {
      "epoch": 127.72222222222223,
      "grad_norm": 0.5680873990058899,
      "learning_rate": 4.048659881103422e-05,
      "loss": 0.0391,
      "num_input_tokens_seen": 3218240,
      "step": 11495
    },
    {
      "epoch": 127.77777777777777,
      "grad_norm": 0.34282854199409485,
      "learning_rate": 4.0478890651196235e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 3219696,
      "step": 11500
    },
    {
      "epoch": 127.83333333333333,
      "grad_norm": 0.10772424191236496,
      "learning_rate": 4.047118010431879e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3221104,
      "step": 11505
    },
    {
      "epoch": 127.88888888888889,
      "grad_norm": 0.04982682317495346,
      "learning_rate": 4.046346717159094e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3222544,
      "step": 11510
    },
    {
      "epoch": 127.94444444444444,
      "grad_norm": 0.09993588179349899,
      "learning_rate": 4.045575185420214e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3223968,
      "step": 11515
    },
    {
      "epoch": 128.0,
      "grad_norm": 0.10343945026397705,
      "learning_rate": 4.0448034153342165e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3225376,
      "step": 11520
    },
    {
      "epoch": 128.05555555555554,
      "grad_norm": 0.07733960449695587,
      "learning_rate": 4.0440314070201194e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3226752,
      "step": 11525
    },
    {
      "epoch": 128.11111111111111,
      "grad_norm": 0.015941955149173737,
      "learning_rate": 4.043259160596976e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3228128,
      "step": 11530
    },
    {
      "epoch": 128.16666666666666,
      "grad_norm": 0.01223111990839243,
      "learning_rate": 4.0424866761838767e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3229520,
      "step": 11535
    },
    {
      "epoch": 128.22222222222223,
      "grad_norm": 0.034866202622652054,
      "learning_rate": 4.041713953899948e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3230976,
      "step": 11540
    },
    {
      "epoch": 128.27777777777777,
      "grad_norm": 0.07145681232213974,
      "learning_rate": 4.0409409938643515e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3232400,
      "step": 11545
    },
    {
      "epoch": 128.33333333333334,
      "grad_norm": 0.030857305973768234,
      "learning_rate": 4.0401677961962904e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3233808,
      "step": 11550
    },
    {
      "epoch": 128.38888888888889,
      "grad_norm": 0.054038457572460175,
      "learning_rate": 4.039394361015001e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 3235168,
      "step": 11555
    },
    {
      "epoch": 128.44444444444446,
      "grad_norm": 1.0745117664337158,
      "learning_rate": 4.038620688439755e-05,
      "loss": 0.0375,
      "num_input_tokens_seen": 3236528,
      "step": 11560
    },
    {
      "epoch": 128.5,
      "grad_norm": 0.07507606595754623,
      "learning_rate": 4.037846778589862e-05,
      "loss": 0.0337,
      "num_input_tokens_seen": 3237936,
      "step": 11565
    },
    {
      "epoch": 128.55555555555554,
      "grad_norm": 0.27882957458496094,
      "learning_rate": 4.0370726315846715e-05,
      "loss": 0.01,
      "num_input_tokens_seen": 3239392,
      "step": 11570
    },
    {
      "epoch": 128.61111111111111,
      "grad_norm": 0.02822282537817955,
      "learning_rate": 4.036298247543565e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3240800,
      "step": 11575
    },
    {
      "epoch": 128.66666666666666,
      "grad_norm": 1.3714689016342163,
      "learning_rate": 4.035523626585962e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 3242176,
      "step": 11580
    },
    {
      "epoch": 128.72222222222223,
      "grad_norm": 0.0604931078851223,
      "learning_rate": 4.0347487688313194e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 3243600,
      "step": 11585
    },
    {
      "epoch": 128.77777777777777,
      "grad_norm": 0.04563498869538307,
      "learning_rate": 4.0339736743991296e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3244992,
      "step": 11590
    },
    {
      "epoch": 128.83333333333334,
      "grad_norm": 0.05939854681491852,
      "learning_rate": 4.0331983434089227e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3246400,
      "step": 11595
    },
    {
      "epoch": 128.88888888888889,
      "grad_norm": 0.026925530284643173,
      "learning_rate": 4.032422775980264e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 3247840,
      "step": 11600
    },
    {
      "epoch": 128.88888888888889,
      "eval_loss": 0.39908546209335327,
      "eval_runtime": 1.2378,
      "eval_samples_per_second": 32.317,
      "eval_steps_per_second": 16.158,
      "num_input_tokens_seen": 3247840,
      "step": 11600
    },
    {
      "epoch": 128.94444444444446,
      "grad_norm": 0.16110292077064514,
      "learning_rate": 4.031646972232754e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3249232,
      "step": 11605
    },
    {
      "epoch": 129.0,
      "grad_norm": 0.24966268241405487,
      "learning_rate": 4.0308709322860344e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3250624,
      "step": 11610
    },
    {
      "epoch": 129.05555555555554,
      "grad_norm": 0.11937909573316574,
      "learning_rate": 4.0300946562597784e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3252032,
      "step": 11615
    },
    {
      "epoch": 129.11111111111111,
      "grad_norm": 0.042341817170381546,
      "learning_rate": 4.029318144273698e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3253408,
      "step": 11620
    },
    {
      "epoch": 129.16666666666666,
      "grad_norm": 0.05593515932559967,
      "learning_rate": 4.0285413964475415e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 3254800,
      "step": 11625
    },
    {
      "epoch": 129.22222222222223,
      "grad_norm": 8.677777290344238,
      "learning_rate": 4.0277644129010927e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 3256192,
      "step": 11630
    },
    {
      "epoch": 129.27777777777777,
      "grad_norm": 0.10578588396310806,
      "learning_rate": 4.0269871937541724e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3257552,
      "step": 11635
    },
    {
      "epoch": 129.33333333333334,
      "grad_norm": 0.06985794752836227,
      "learning_rate": 4.026209739126637e-05,
      "loss": 0.0137,
      "num_input_tokens_seen": 3258976,
      "step": 11640
    },
    {
      "epoch": 129.38888888888889,
      "grad_norm": 0.026301885023713112,
      "learning_rate": 4.025432049138381e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3260384,
      "step": 11645
    },
    {
      "epoch": 129.44444444444446,
      "grad_norm": 0.24106574058532715,
      "learning_rate": 4.0246541239093325e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 3261792,
      "step": 11650
    },
    {
      "epoch": 129.5,
      "grad_norm": 1.3296881914138794,
      "learning_rate": 4.023875963559459e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 3263184,
      "step": 11655
    },
    {
      "epoch": 129.55555555555554,
      "grad_norm": 0.011030678637325764,
      "learning_rate": 4.023097568208761e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3264560,
      "step": 11660
    },
    {
      "epoch": 129.61111111111111,
      "grad_norm": 0.017964564263820648,
      "learning_rate": 4.022318937977277e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3265968,
      "step": 11665
    },
    {
      "epoch": 129.66666666666666,
      "grad_norm": 0.14960378408432007,
      "learning_rate": 4.021540072985084e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3267360,
      "step": 11670
    },
    {
      "epoch": 129.72222222222223,
      "grad_norm": 0.06621088087558746,
      "learning_rate": 4.020760973352289e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3268768,
      "step": 11675
    },
    {
      "epoch": 129.77777777777777,
      "grad_norm": 0.11553477495908737,
      "learning_rate": 4.019981639199042e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3270160,
      "step": 11680
    },
    {
      "epoch": 129.83333333333334,
      "grad_norm": 0.022450048476457596,
      "learning_rate": 4.0192020706455245e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3271536,
      "step": 11685
    },
    {
      "epoch": 129.88888888888889,
      "grad_norm": 0.08638250082731247,
      "learning_rate": 4.018422267811956e-05,
      "loss": 0.0249,
      "num_input_tokens_seen": 3272944,
      "step": 11690
    },
    {
      "epoch": 129.94444444444446,
      "grad_norm": 0.14514736831188202,
      "learning_rate": 4.017642230818592e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3274320,
      "step": 11695
    },
    {
      "epoch": 130.0,
      "grad_norm": 0.035482462495565414,
      "learning_rate": 4.0168619597857246e-05,
      "loss": 0.0743,
      "num_input_tokens_seen": 3275712,
      "step": 11700
    },
    {
      "epoch": 130.05555555555554,
      "grad_norm": 0.06210021674633026,
      "learning_rate": 4.016081454833681e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3277152,
      "step": 11705
    },
    {
      "epoch": 130.11111111111111,
      "grad_norm": 0.040221650153398514,
      "learning_rate": 4.0153007160828245e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3278592,
      "step": 11710
    },
    {
      "epoch": 130.16666666666666,
      "grad_norm": 0.03937218338251114,
      "learning_rate": 4.0145197436535555e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 3280016,
      "step": 11715
    },
    {
      "epoch": 130.22222222222223,
      "grad_norm": 0.18208159506320953,
      "learning_rate": 4.0137385376663095e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 3281360,
      "step": 11720
    },
    {
      "epoch": 130.27777777777777,
      "grad_norm": 0.10445526242256165,
      "learning_rate": 4.012957098241558e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3282816,
      "step": 11725
    },
    {
      "epoch": 130.33333333333334,
      "grad_norm": 0.6433157324790955,
      "learning_rate": 4.0121754254998076e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3284208,
      "step": 11730
    },
    {
      "epoch": 130.38888888888889,
      "grad_norm": 1.1098058223724365,
      "learning_rate": 4.011393519561606e-05,
      "loss": 0.0434,
      "num_input_tokens_seen": 3285584,
      "step": 11735
    },
    {
      "epoch": 130.44444444444446,
      "grad_norm": 0.04983939230442047,
      "learning_rate": 4.010611380547529e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3286976,
      "step": 11740
    },
    {
      "epoch": 130.5,
      "grad_norm": 0.028228813782334328,
      "learning_rate": 4.009829008578192e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 3288352,
      "step": 11745
    },
    {
      "epoch": 130.55555555555554,
      "grad_norm": 1.9022678136825562,
      "learning_rate": 4.00904640377425e-05,
      "loss": 0.0395,
      "num_input_tokens_seen": 3289760,
      "step": 11750
    },
    {
      "epoch": 130.61111111111111,
      "grad_norm": 0.05429571866989136,
      "learning_rate": 4.0082635662563886e-05,
      "loss": 0.0105,
      "num_input_tokens_seen": 3291200,
      "step": 11755
    },
    {
      "epoch": 130.66666666666666,
      "grad_norm": 0.6917058229446411,
      "learning_rate": 4.007480496145331e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 3292608,
      "step": 11760
    },
    {
      "epoch": 130.72222222222223,
      "grad_norm": 0.2968350052833557,
      "learning_rate": 4.006697193561837e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 3294032,
      "step": 11765
    },
    {
      "epoch": 130.77777777777777,
      "grad_norm": 0.4350326359272003,
      "learning_rate": 4.005913658626701e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3295360,
      "step": 11770
    },
    {
      "epoch": 130.83333333333334,
      "grad_norm": 0.18578492105007172,
      "learning_rate": 4.005129891460754e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3296768,
      "step": 11775
    },
    {
      "epoch": 130.88888888888889,
      "grad_norm": 0.2259959727525711,
      "learning_rate": 4.004345892184864e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3298160,
      "step": 11780
    },
    {
      "epoch": 130.94444444444446,
      "grad_norm": 0.0341925248503685,
      "learning_rate": 4.003561660919932e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3299536,
      "step": 11785
    },
    {
      "epoch": 131.0,
      "grad_norm": 0.5714805126190186,
      "learning_rate": 4.002777197786897e-05,
      "loss": 0.0086,
      "num_input_tokens_seen": 3300896,
      "step": 11790
    },
    {
      "epoch": 131.05555555555554,
      "grad_norm": 0.023223884403705597,
      "learning_rate": 4.0019925029067326e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3302272,
      "step": 11795
    },
    {
      "epoch": 131.11111111111111,
      "grad_norm": 0.015760883688926697,
      "learning_rate": 4.0012075764004495e-05,
      "loss": 0.028,
      "num_input_tokens_seen": 3303712,
      "step": 11800
    },
    {
      "epoch": 131.11111111111111,
      "eval_loss": 0.3551647663116455,
      "eval_runtime": 1.2529,
      "eval_samples_per_second": 31.927,
      "eval_steps_per_second": 15.963,
      "num_input_tokens_seen": 3303712,
      "step": 11800
    },
    {
      "epoch": 131.16666666666666,
      "grad_norm": 0.03619011119008064,
      "learning_rate": 4.000422418389094e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 3305152,
      "step": 11805
    },
    {
      "epoch": 131.22222222222223,
      "grad_norm": 0.04694085568189621,
      "learning_rate": 3.999637028993744e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3306576,
      "step": 11810
    },
    {
      "epoch": 131.27777777777777,
      "grad_norm": 0.04827222228050232,
      "learning_rate": 3.99885140833552e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3307984,
      "step": 11815
    },
    {
      "epoch": 131.33333333333334,
      "grad_norm": 0.004336053039878607,
      "learning_rate": 3.998065556535572e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3309392,
      "step": 11820
    },
    {
      "epoch": 131.38888888888889,
      "grad_norm": 0.013496151193976402,
      "learning_rate": 3.9972794737150895e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 3310736,
      "step": 11825
    },
    {
      "epoch": 131.44444444444446,
      "grad_norm": 0.18720942735671997,
      "learning_rate": 3.996493159995297e-05,
      "loss": 0.012,
      "num_input_tokens_seen": 3312096,
      "step": 11830
    },
    {
      "epoch": 131.5,
      "grad_norm": 0.1945410817861557,
      "learning_rate": 3.995706615497453e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3313536,
      "step": 11835
    },
    {
      "epoch": 131.55555555555554,
      "grad_norm": 0.01671007089316845,
      "learning_rate": 3.994919840342852e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3314944,
      "step": 11840
    },
    {
      "epoch": 131.61111111111111,
      "grad_norm": 1.3276809453964233,
      "learning_rate": 3.994132834652825e-05,
      "loss": 0.0446,
      "num_input_tokens_seen": 3316304,
      "step": 11845
    },
    {
      "epoch": 131.66666666666666,
      "grad_norm": 0.03642260655760765,
      "learning_rate": 3.99334559854874e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3317680,
      "step": 11850
    },
    {
      "epoch": 131.72222222222223,
      "grad_norm": 0.06578918546438217,
      "learning_rate": 3.9925581321519955e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3319056,
      "step": 11855
    },
    {
      "epoch": 131.77777777777777,
      "grad_norm": 0.07295611500740051,
      "learning_rate": 3.991770435584031e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3320480,
      "step": 11860
    },
    {
      "epoch": 131.83333333333334,
      "grad_norm": 0.04846571013331413,
      "learning_rate": 3.990982508966319e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3321840,
      "step": 11865
    },
    {
      "epoch": 131.88888888888889,
      "grad_norm": 0.07680318504571915,
      "learning_rate": 3.990194352420367e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3323248,
      "step": 11870
    },
    {
      "epoch": 131.94444444444446,
      "grad_norm": 0.031286660581827164,
      "learning_rate": 3.9894059660677184e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 3324640,
      "step": 11875
    },
    {
      "epoch": 132.0,
      "grad_norm": 0.1422128826379776,
      "learning_rate": 3.9886173500299526e-05,
      "loss": 0.0375,
      "num_input_tokens_seen": 3326064,
      "step": 11880
    },
    {
      "epoch": 132.05555555555554,
      "grad_norm": 0.06111743301153183,
      "learning_rate": 3.987828504428685e-05,
      "loss": 0.0399,
      "num_input_tokens_seen": 3327456,
      "step": 11885
    },
    {
      "epoch": 132.11111111111111,
      "grad_norm": 0.028062893077731133,
      "learning_rate": 3.987039429385565e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3328928,
      "step": 11890
    },
    {
      "epoch": 132.16666666666666,
      "grad_norm": 0.06525691598653793,
      "learning_rate": 3.986250125022277e-05,
      "loss": 0.0179,
      "num_input_tokens_seen": 3330352,
      "step": 11895
    },
    {
      "epoch": 132.22222222222223,
      "grad_norm": 0.5480440258979797,
      "learning_rate": 3.985460591460544e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3331760,
      "step": 11900
    },
    {
      "epoch": 132.27777777777777,
      "grad_norm": 0.059182245284318924,
      "learning_rate": 3.984670828822118e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3333152,
      "step": 11905
    },
    {
      "epoch": 132.33333333333334,
      "grad_norm": 0.024349462240934372,
      "learning_rate": 3.983880837228794e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3334560,
      "step": 11910
    },
    {
      "epoch": 132.38888888888889,
      "grad_norm": 0.075358547270298,
      "learning_rate": 3.983090616802396e-05,
      "loss": 0.01,
      "num_input_tokens_seen": 3335952,
      "step": 11915
    },
    {
      "epoch": 132.44444444444446,
      "grad_norm": 0.020014403387904167,
      "learning_rate": 3.982300167664788e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3337328,
      "step": 11920
    },
    {
      "epoch": 132.5,
      "grad_norm": 0.03963126242160797,
      "learning_rate": 3.981509489937868e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 3338704,
      "step": 11925
    },
    {
      "epoch": 132.55555555555554,
      "grad_norm": 0.05296626314520836,
      "learning_rate": 3.9807185837435643e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3340128,
      "step": 11930
    },
    {
      "epoch": 132.61111111111111,
      "grad_norm": 0.032944049686193466,
      "learning_rate": 3.9799274492038484e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3341552,
      "step": 11935
    },
    {
      "epoch": 132.66666666666666,
      "grad_norm": 0.04903354123234749,
      "learning_rate": 3.979136086440722e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 3342896,
      "step": 11940
    },
    {
      "epoch": 132.72222222222223,
      "grad_norm": 0.38885462284088135,
      "learning_rate": 3.9783444955762226e-05,
      "loss": 0.0103,
      "num_input_tokens_seen": 3344256,
      "step": 11945
    },
    {
      "epoch": 132.77777777777777,
      "grad_norm": 0.5127944350242615,
      "learning_rate": 3.977552676732424e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 3345632,
      "step": 11950
    },
    {
      "epoch": 132.83333333333334,
      "grad_norm": 0.029612645506858826,
      "learning_rate": 3.976760630031435e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3347056,
      "step": 11955
    },
    {
      "epoch": 132.88888888888889,
      "grad_norm": 0.08596263080835342,
      "learning_rate": 3.975968355595398e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3348480,
      "step": 11960
    },
    {
      "epoch": 132.94444444444446,
      "grad_norm": 0.32190677523612976,
      "learning_rate": 3.9751758535464935e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3349888,
      "step": 11965
    },
    {
      "epoch": 133.0,
      "grad_norm": 2.8843235969543457,
      "learning_rate": 3.9743831240069326e-05,
      "loss": 0.0387,
      "num_input_tokens_seen": 3351296,
      "step": 11970
    },
    {
      "epoch": 133.05555555555554,
      "grad_norm": 2.0641303062438965,
      "learning_rate": 3.9735901670989675e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 3352624,
      "step": 11975
    },
    {
      "epoch": 133.11111111111111,
      "grad_norm": 0.017290586605668068,
      "learning_rate": 3.97279698294488e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3354032,
      "step": 11980
    },
    {
      "epoch": 133.16666666666666,
      "grad_norm": 0.13035713136196136,
      "learning_rate": 3.9720035716669876e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3355440,
      "step": 11985
    },
    {
      "epoch": 133.22222222222223,
      "grad_norm": 0.1776961237192154,
      "learning_rate": 3.9712099333876474e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3356848,
      "step": 11990
    },
    {
      "epoch": 133.27777777777777,
      "grad_norm": 0.23860889673233032,
      "learning_rate": 3.9704160682292475e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3358272,
      "step": 11995
    },
    {
      "epoch": 133.33333333333334,
      "grad_norm": 0.014967321418225765,
      "learning_rate": 3.9696219763142106e-05,
      "loss": 0.0101,
      "num_input_tokens_seen": 3359680,
      "step": 12000
    },
    {
      "epoch": 133.33333333333334,
      "eval_loss": 0.41930827498435974,
      "eval_runtime": 1.2316,
      "eval_samples_per_second": 32.479,
      "eval_steps_per_second": 16.24,
      "num_input_tokens_seen": 3359680,
      "step": 12000
    },
    {
      "epoch": 133.38888888888889,
      "grad_norm": 0.07991684973239899,
      "learning_rate": 3.968827657764997e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3361088,
      "step": 12005
    },
    {
      "epoch": 133.44444444444446,
      "grad_norm": 4.516703128814697,
      "learning_rate": 3.9680331127041e-05,
      "loss": 0.0081,
      "num_input_tokens_seen": 3362528,
      "step": 12010
    },
    {
      "epoch": 133.5,
      "grad_norm": 0.03234417364001274,
      "learning_rate": 3.9672383412540495e-05,
      "loss": 0.0085,
      "num_input_tokens_seen": 3363920,
      "step": 12015
    },
    {
      "epoch": 133.55555555555554,
      "grad_norm": 0.2943073511123657,
      "learning_rate": 3.966443343537407e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 3365312,
      "step": 12020
    },
    {
      "epoch": 133.61111111111111,
      "grad_norm": 0.33792436122894287,
      "learning_rate": 3.965648119676772e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3366736,
      "step": 12025
    },
    {
      "epoch": 133.66666666666666,
      "grad_norm": 0.3400917649269104,
      "learning_rate": 3.96485266979478e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3368160,
      "step": 12030
    },
    {
      "epoch": 133.72222222222223,
      "grad_norm": 0.1969456523656845,
      "learning_rate": 3.9640569940140974e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 3369584,
      "step": 12035
    },
    {
      "epoch": 133.77777777777777,
      "grad_norm": 0.05374060943722725,
      "learning_rate": 3.963261092457428e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 3371008,
      "step": 12040
    },
    {
      "epoch": 133.83333333333334,
      "grad_norm": 1.2378852367401123,
      "learning_rate": 3.962464965247509e-05,
      "loss": 0.0427,
      "num_input_tokens_seen": 3372416,
      "step": 12045
    },
    {
      "epoch": 133.88888888888889,
      "grad_norm": 0.08312591165304184,
      "learning_rate": 3.9616686125071135e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 3373840,
      "step": 12050
    },
    {
      "epoch": 133.94444444444446,
      "grad_norm": 0.019849861040711403,
      "learning_rate": 3.9608720343590506e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3375216,
      "step": 12055
    },
    {
      "epoch": 134.0,
      "grad_norm": 0.0615706592798233,
      "learning_rate": 3.960075230926161e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3376560,
      "step": 12060
    },
    {
      "epoch": 134.05555555555554,
      "grad_norm": 0.42372927069664,
      "learning_rate": 3.959278202331322e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 3377968,
      "step": 12065
    },
    {
      "epoch": 134.11111111111111,
      "grad_norm": 0.4439363181591034,
      "learning_rate": 3.958480948697446e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3379312,
      "step": 12070
    },
    {
      "epoch": 134.16666666666666,
      "grad_norm": 0.07894574105739594,
      "learning_rate": 3.95768347014748e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3380736,
      "step": 12075
    },
    {
      "epoch": 134.22222222222223,
      "grad_norm": 0.021137293428182602,
      "learning_rate": 3.956885766804404e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3382144,
      "step": 12080
    },
    {
      "epoch": 134.27777777777777,
      "grad_norm": 0.019083041697740555,
      "learning_rate": 3.956087838791235e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3383504,
      "step": 12085
    },
    {
      "epoch": 134.33333333333334,
      "grad_norm": 0.011184210889041424,
      "learning_rate": 3.955289686231022e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3384928,
      "step": 12090
    },
    {
      "epoch": 134.38888888888889,
      "grad_norm": 1.1725445985794067,
      "learning_rate": 3.9544913092468504e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 3386336,
      "step": 12095
    },
    {
      "epoch": 134.44444444444446,
      "grad_norm": 0.05306538566946983,
      "learning_rate": 3.9536927079618425e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3387728,
      "step": 12100
    },
    {
      "epoch": 134.5,
      "grad_norm": 0.013956797309219837,
      "learning_rate": 3.9528938824991494e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3389120,
      "step": 12105
    },
    {
      "epoch": 134.55555555555554,
      "grad_norm": 0.15479113161563873,
      "learning_rate": 3.952094832981962e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3390544,
      "step": 12110
    },
    {
      "epoch": 134.61111111111111,
      "grad_norm": 0.15270423889160156,
      "learning_rate": 3.951295559533503e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3391920,
      "step": 12115
    },
    {
      "epoch": 134.66666666666666,
      "grad_norm": 0.029268812388181686,
      "learning_rate": 3.95049606227703e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3393296,
      "step": 12120
    },
    {
      "epoch": 134.72222222222223,
      "grad_norm": 0.06170087680220604,
      "learning_rate": 3.949696341335838e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 3394720,
      "step": 12125
    },
    {
      "epoch": 134.77777777777777,
      "grad_norm": 0.015794184058904648,
      "learning_rate": 3.9488963968332503e-05,
      "loss": 0.0391,
      "num_input_tokens_seen": 3396128,
      "step": 12130
    },
    {
      "epoch": 134.83333333333334,
      "grad_norm": 0.025179732590913773,
      "learning_rate": 3.948096228892631e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 3397536,
      "step": 12135
    },
    {
      "epoch": 134.88888888888889,
      "grad_norm": 0.07961688935756683,
      "learning_rate": 3.947295837637375e-05,
      "loss": 0.0156,
      "num_input_tokens_seen": 3398928,
      "step": 12140
    },
    {
      "epoch": 134.94444444444446,
      "grad_norm": 0.04068479314446449,
      "learning_rate": 3.9464952231909135e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3400336,
      "step": 12145
    },
    {
      "epoch": 135.0,
      "grad_norm": 0.19329747557640076,
      "learning_rate": 3.945694385676711e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3401776,
      "step": 12150
    },
    {
      "epoch": 135.05555555555554,
      "grad_norm": 0.12038015574216843,
      "learning_rate": 3.944893325218265e-05,
      "loss": 0.0047,
      "num_input_tokens_seen": 3403184,
      "step": 12155
    },
    {
      "epoch": 135.11111111111111,
      "grad_norm": 0.049572739750146866,
      "learning_rate": 3.944092041939112e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3404560,
      "step": 12160
    },
    {
      "epoch": 135.16666666666666,
      "grad_norm": 0.01440472062677145,
      "learning_rate": 3.943290535962818e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3406000,
      "step": 12165
    },
    {
      "epoch": 135.22222222222223,
      "grad_norm": 0.009750002063810825,
      "learning_rate": 3.942488807412985e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3407392,
      "step": 12170
    },
    {
      "epoch": 135.27777777777777,
      "grad_norm": 0.01793650910258293,
      "learning_rate": 3.941686856413251e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3408800,
      "step": 12175
    },
    {
      "epoch": 135.33333333333334,
      "grad_norm": 0.06125554442405701,
      "learning_rate": 3.9408846830872874e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3410240,
      "step": 12180
    },
    {
      "epoch": 135.38888888888889,
      "grad_norm": 0.012745431624352932,
      "learning_rate": 3.940082287558798e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 3411616,
      "step": 12185
    },
    {
      "epoch": 135.44444444444446,
      "grad_norm": 0.09626718610525131,
      "learning_rate": 3.939279669951522e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3413040,
      "step": 12190
    },
    {
      "epoch": 135.5,
      "grad_norm": 0.02887718565762043,
      "learning_rate": 3.938476830389234e-05,
      "loss": 0.0067,
      "num_input_tokens_seen": 3414432,
      "step": 12195
    },
    {
      "epoch": 135.55555555555554,
      "grad_norm": 0.176244854927063,
      "learning_rate": 3.937673768995742e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 3415824,
      "step": 12200
    },
    {
      "epoch": 135.55555555555554,
      "eval_loss": 0.4204668402671814,
      "eval_runtime": 1.2424,
      "eval_samples_per_second": 32.195,
      "eval_steps_per_second": 16.097,
      "num_input_tokens_seen": 3415824,
      "step": 12200
    },
    {
      "epoch": 135.61111111111111,
      "grad_norm": 0.15237653255462646,
      "learning_rate": 3.936870485894888e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3417184,
      "step": 12205
    },
    {
      "epoch": 135.66666666666666,
      "grad_norm": 0.9908055067062378,
      "learning_rate": 3.9360669812105475e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 3418576,
      "step": 12210
    },
    {
      "epoch": 135.72222222222223,
      "grad_norm": 0.06550995260477066,
      "learning_rate": 3.9352632550666325e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3419984,
      "step": 12215
    },
    {
      "epoch": 135.77777777777777,
      "grad_norm": 0.34644854068756104,
      "learning_rate": 3.9344593075870866e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 3421360,
      "step": 12220
    },
    {
      "epoch": 135.83333333333334,
      "grad_norm": 0.17042852938175201,
      "learning_rate": 3.933655138895889e-05,
      "loss": 0.0383,
      "num_input_tokens_seen": 3422720,
      "step": 12225
    },
    {
      "epoch": 135.88888888888889,
      "grad_norm": 0.011130996979773045,
      "learning_rate": 3.932850749117053e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 3424112,
      "step": 12230
    },
    {
      "epoch": 135.94444444444446,
      "grad_norm": 0.027703600004315376,
      "learning_rate": 3.932046138374624e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3425488,
      "step": 12235
    },
    {
      "epoch": 136.0,
      "grad_norm": 0.1605251282453537,
      "learning_rate": 3.9312413067926854e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3426912,
      "step": 12240
    },
    {
      "epoch": 136.05555555555554,
      "grad_norm": 0.053453586995601654,
      "learning_rate": 3.9304362544953506e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3428304,
      "step": 12245
    },
    {
      "epoch": 136.11111111111111,
      "grad_norm": 0.06509539484977722,
      "learning_rate": 3.929630981606769e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3429712,
      "step": 12250
    },
    {
      "epoch": 136.16666666666666,
      "grad_norm": 0.1711602807044983,
      "learning_rate": 3.928825488251124e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3431088,
      "step": 12255
    },
    {
      "epoch": 136.22222222222223,
      "grad_norm": 0.0544453039765358,
      "learning_rate": 3.9280197745526344e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3432448,
      "step": 12260
    },
    {
      "epoch": 136.27777777777777,
      "grad_norm": 0.026512006297707558,
      "learning_rate": 3.9272138406355495e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3433872,
      "step": 12265
    },
    {
      "epoch": 136.33333333333334,
      "grad_norm": 0.030493734404444695,
      "learning_rate": 3.926407686624154e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 3435248,
      "step": 12270
    },
    {
      "epoch": 136.38888888888889,
      "grad_norm": 4.25136661529541,
      "learning_rate": 3.9256013126427684e-05,
      "loss": 0.0145,
      "num_input_tokens_seen": 3436592,
      "step": 12275
    },
    {
      "epoch": 136.44444444444446,
      "grad_norm": 0.05122780054807663,
      "learning_rate": 3.9247947188157455e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3438016,
      "step": 12280
    },
    {
      "epoch": 136.5,
      "grad_norm": 2.984325647354126,
      "learning_rate": 3.9239879052674715e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 3439392,
      "step": 12285
    },
    {
      "epoch": 136.55555555555554,
      "grad_norm": 0.05481431260704994,
      "learning_rate": 3.9231808721223673e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3440816,
      "step": 12290
    },
    {
      "epoch": 136.61111111111111,
      "grad_norm": 0.024960795417428017,
      "learning_rate": 3.9223736195048886e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3442208,
      "step": 12295
    },
    {
      "epoch": 136.66666666666666,
      "grad_norm": 0.08229300379753113,
      "learning_rate": 3.921566147539523e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3443632,
      "step": 12300
    },
    {
      "epoch": 136.72222222222223,
      "grad_norm": 0.05548291653394699,
      "learning_rate": 3.920758456350792e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3445040,
      "step": 12305
    },
    {
      "epoch": 136.77777777777777,
      "grad_norm": 0.004340479150414467,
      "learning_rate": 3.919950546063253e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3446352,
      "step": 12310
    },
    {
      "epoch": 136.83333333333334,
      "grad_norm": 0.8936319351196289,
      "learning_rate": 3.919142416801496e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 3447792,
      "step": 12315
    },
    {
      "epoch": 136.88888888888889,
      "grad_norm": 2.0436394214630127,
      "learning_rate": 3.918334068690144e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 3449152,
      "step": 12320
    },
    {
      "epoch": 136.94444444444446,
      "grad_norm": 0.011840008199214935,
      "learning_rate": 3.917525501853855e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 3450576,
      "step": 12325
    },
    {
      "epoch": 137.0,
      "grad_norm": 0.36878907680511475,
      "learning_rate": 3.916716716417319e-05,
      "loss": 0.0169,
      "num_input_tokens_seen": 3451984,
      "step": 12330
    },
    {
      "epoch": 137.05555555555554,
      "grad_norm": 0.2494126409292221,
      "learning_rate": 3.915907712505263e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3453360,
      "step": 12335
    },
    {
      "epoch": 137.11111111111111,
      "grad_norm": 0.04592215642333031,
      "learning_rate": 3.915098490242444e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3454736,
      "step": 12340
    },
    {
      "epoch": 137.16666666666666,
      "grad_norm": 0.026002954691648483,
      "learning_rate": 3.914289049753654e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3456160,
      "step": 12345
    },
    {
      "epoch": 137.22222222222223,
      "grad_norm": 0.09073606133460999,
      "learning_rate": 3.913479391163719e-05,
      "loss": 0.012,
      "num_input_tokens_seen": 3457536,
      "step": 12350
    },
    {
      "epoch": 137.27777777777777,
      "grad_norm": 0.16268211603164673,
      "learning_rate": 3.9126695145975e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3458976,
      "step": 12355
    },
    {
      "epoch": 137.33333333333334,
      "grad_norm": 0.011572883464396,
      "learning_rate": 3.911859420179889e-05,
      "loss": 0.0288,
      "num_input_tokens_seen": 3460336,
      "step": 12360
    },
    {
      "epoch": 137.38888888888889,
      "grad_norm": 0.028479693457484245,
      "learning_rate": 3.911049108035813e-05,
      "loss": 0.0275,
      "num_input_tokens_seen": 3461760,
      "step": 12365
    },
    {
      "epoch": 137.44444444444446,
      "grad_norm": 0.060944996774196625,
      "learning_rate": 3.910238578290232e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3463200,
      "step": 12370
    },
    {
      "epoch": 137.5,
      "grad_norm": 0.051628343760967255,
      "learning_rate": 3.90942783106814e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3464576,
      "step": 12375
    },
    {
      "epoch": 137.55555555555554,
      "grad_norm": 0.07545940577983856,
      "learning_rate": 3.908616866494564e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3465936,
      "step": 12380
    },
    {
      "epoch": 137.61111111111111,
      "grad_norm": 0.025925597175955772,
      "learning_rate": 3.907805684694566e-05,
      "loss": 0.0135,
      "num_input_tokens_seen": 3467344,
      "step": 12385
    },
    {
      "epoch": 137.66666666666666,
      "grad_norm": 0.03368198499083519,
      "learning_rate": 3.90699428579324e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3468720,
      "step": 12390
    },
    {
      "epoch": 137.72222222222223,
      "grad_norm": 0.07948169857263565,
      "learning_rate": 3.906182669915713e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 3470128,
      "step": 12395
    },
    {
      "epoch": 137.77777777777777,
      "grad_norm": 0.42584410309791565,
      "learning_rate": 3.9053708371871476e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 3471520,
      "step": 12400
    },
    {
      "epoch": 137.77777777777777,
      "eval_loss": 0.4430425763130188,
      "eval_runtime": 1.2358,
      "eval_samples_per_second": 32.368,
      "eval_steps_per_second": 16.184,
      "num_input_tokens_seen": 3471520,
      "step": 12400
    },
    {
      "epoch": 137.83333333333334,
      "grad_norm": 0.06594458222389221,
      "learning_rate": 3.904558787732738e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3472944,
      "step": 12405
    },
    {
      "epoch": 137.88888888888889,
      "grad_norm": 0.07237505912780762,
      "learning_rate": 3.9037465216777135e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 3474352,
      "step": 12410
    },
    {
      "epoch": 137.94444444444446,
      "grad_norm": 0.07612795382738113,
      "learning_rate": 3.902934039147334e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 3475776,
      "step": 12415
    },
    {
      "epoch": 138.0,
      "grad_norm": 0.19909843802452087,
      "learning_rate": 3.902121340266894e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3477184,
      "step": 12420
    },
    {
      "epoch": 138.05555555555554,
      "grad_norm": 0.040738124400377274,
      "learning_rate": 3.9013084251617246e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 3478608,
      "step": 12425
    },
    {
      "epoch": 138.11111111111111,
      "grad_norm": 2.031125783920288,
      "learning_rate": 3.9004952939571865e-05,
      "loss": 0.0231,
      "num_input_tokens_seen": 3479968,
      "step": 12430
    },
    {
      "epoch": 138.16666666666666,
      "grad_norm": 0.009414182044565678,
      "learning_rate": 3.899681946778673e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 3481392,
      "step": 12435
    },
    {
      "epoch": 138.22222222222223,
      "grad_norm": 0.010179141536355019,
      "learning_rate": 3.898868383751615e-05,
      "loss": 0.0086,
      "num_input_tokens_seen": 3482832,
      "step": 12440
    },
    {
      "epoch": 138.27777777777777,
      "grad_norm": 0.1910523772239685,
      "learning_rate": 3.8980546050014724e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 3484192,
      "step": 12445
    },
    {
      "epoch": 138.33333333333334,
      "grad_norm": 0.1646217405796051,
      "learning_rate": 3.897240610653741e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3485584,
      "step": 12450
    },
    {
      "epoch": 138.38888888888889,
      "grad_norm": 0.07302524149417877,
      "learning_rate": 3.896426400833948e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3486992,
      "step": 12455
    },
    {
      "epoch": 138.44444444444446,
      "grad_norm": 0.04567598178982735,
      "learning_rate": 3.895611975667656e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 3488368,
      "step": 12460
    },
    {
      "epoch": 138.5,
      "grad_norm": 0.8927302360534668,
      "learning_rate": 3.8947973352804584e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 3489792,
      "step": 12465
    },
    {
      "epoch": 138.55555555555554,
      "grad_norm": 0.08859757333993912,
      "learning_rate": 3.893982479797984e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 3491200,
      "step": 12470
    },
    {
      "epoch": 138.61111111111111,
      "grad_norm": 0.06615463644266129,
      "learning_rate": 3.8931674093458926e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3492608,
      "step": 12475
    },
    {
      "epoch": 138.66666666666666,
      "grad_norm": 0.03679698333144188,
      "learning_rate": 3.89235212404988e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3494016,
      "step": 12480
    },
    {
      "epoch": 138.72222222222223,
      "grad_norm": 0.33051639795303345,
      "learning_rate": 3.891536624035672e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 3495424,
      "step": 12485
    },
    {
      "epoch": 138.77777777777777,
      "grad_norm": 0.13997000455856323,
      "learning_rate": 3.8907209094290295e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3496864,
      "step": 12490
    },
    {
      "epoch": 138.83333333333334,
      "grad_norm": 0.047157976776361465,
      "learning_rate": 3.8899049803557466e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3498240,
      "step": 12495
    },
    {
      "epoch": 138.88888888888889,
      "grad_norm": 0.014191296882927418,
      "learning_rate": 3.889088836941648e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3499680,
      "step": 12500
    },
    {
      "epoch": 138.94444444444446,
      "grad_norm": 0.054439477622509,
      "learning_rate": 3.8882724793125946e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3501056,
      "step": 12505
    },
    {
      "epoch": 139.0,
      "grad_norm": 0.01303825993090868,
      "learning_rate": 3.8874559075944794e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3502448,
      "step": 12510
    },
    {
      "epoch": 139.05555555555554,
      "grad_norm": 0.010344025678932667,
      "learning_rate": 3.886639121913227e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3503872,
      "step": 12515
    },
    {
      "epoch": 139.11111111111111,
      "grad_norm": 0.10030224919319153,
      "learning_rate": 3.885822122394797e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3505280,
      "step": 12520
    },
    {
      "epoch": 139.16666666666666,
      "grad_norm": 0.0035984779242426157,
      "learning_rate": 3.8850049091651794e-05,
      "loss": 0.0298,
      "num_input_tokens_seen": 3506672,
      "step": 12525
    },
    {
      "epoch": 139.22222222222223,
      "grad_norm": 0.003991794306784868,
      "learning_rate": 3.8841874823504e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3508048,
      "step": 12530
    },
    {
      "epoch": 139.27777777777777,
      "grad_norm": 0.018903391435742378,
      "learning_rate": 3.8833698420765157e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3509408,
      "step": 12535
    },
    {
      "epoch": 139.33333333333334,
      "grad_norm": 0.12362518906593323,
      "learning_rate": 3.882551988469618e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3510800,
      "step": 12540
    },
    {
      "epoch": 139.38888888888889,
      "grad_norm": 0.010305922478437424,
      "learning_rate": 3.881733921655829e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3512224,
      "step": 12545
    },
    {
      "epoch": 139.44444444444446,
      "grad_norm": 0.07117374986410141,
      "learning_rate": 3.8809156417613054e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3513664,
      "step": 12550
    },
    {
      "epoch": 139.5,
      "grad_norm": 0.026475539430975914,
      "learning_rate": 3.8800971489122364e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3515072,
      "step": 12555
    },
    {
      "epoch": 139.55555555555554,
      "grad_norm": 0.298136442899704,
      "learning_rate": 3.8792784432348434e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3516464,
      "step": 12560
    },
    {
      "epoch": 139.61111111111111,
      "grad_norm": 0.010251945815980434,
      "learning_rate": 3.878459524855381e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3517824,
      "step": 12565
    },
    {
      "epoch": 139.66666666666666,
      "grad_norm": 0.015613695606589317,
      "learning_rate": 3.8776403939001384e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3519216,
      "step": 12570
    },
    {
      "epoch": 139.72222222222223,
      "grad_norm": 0.0320529080927372,
      "learning_rate": 3.876821050495433e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 3520624,
      "step": 12575
    },
    {
      "epoch": 139.77777777777777,
      "grad_norm": 0.034432318061590195,
      "learning_rate": 3.87600149476762e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3522048,
      "step": 12580
    },
    {
      "epoch": 139.83333333333334,
      "grad_norm": 0.33762815594673157,
      "learning_rate": 3.8751817268430843e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 3523440,
      "step": 12585
    },
    {
      "epoch": 139.88888888888889,
      "grad_norm": 0.1997912973165512,
      "learning_rate": 3.8743617468482464e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3524832,
      "step": 12590
    },
    {
      "epoch": 139.94444444444446,
      "grad_norm": 1.2290313243865967,
      "learning_rate": 3.8735415549095535e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 3526224,
      "step": 12595
    },
    {
      "epoch": 140.0,
      "grad_norm": 0.04245232418179512,
      "learning_rate": 3.8727211511534934e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 3527664,
      "step": 12600
    },
    {
      "epoch": 140.0,
      "eval_loss": 0.3998602032661438,
      "eval_runtime": 1.2321,
      "eval_samples_per_second": 32.464,
      "eval_steps_per_second": 16.232,
      "num_input_tokens_seen": 3527664,
      "step": 12600
    },
    {
      "epoch": 140.05555555555554,
      "grad_norm": 0.021507736295461655,
      "learning_rate": 3.8719005357065804e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3529040,
      "step": 12605
    },
    {
      "epoch": 140.11111111111111,
      "grad_norm": 2.253793954849243,
      "learning_rate": 3.8710797086953645e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 3530448,
      "step": 12610
    },
    {
      "epoch": 140.16666666666666,
      "grad_norm": 0.08849595487117767,
      "learning_rate": 3.870258670246427e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3531824,
      "step": 12615
    },
    {
      "epoch": 140.22222222222223,
      "grad_norm": 0.5255588293075562,
      "learning_rate": 3.869437420486384e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3533280,
      "step": 12620
    },
    {
      "epoch": 140.27777777777777,
      "grad_norm": 0.01166149415075779,
      "learning_rate": 3.8686159595418805e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3534688,
      "step": 12625
    },
    {
      "epoch": 140.33333333333334,
      "grad_norm": 0.05848929286003113,
      "learning_rate": 3.867794287539597e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3536080,
      "step": 12630
    },
    {
      "epoch": 140.38888888888889,
      "grad_norm": 0.016653049737215042,
      "learning_rate": 3.866972404606245e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3537424,
      "step": 12635
    },
    {
      "epoch": 140.44444444444446,
      "grad_norm": 0.021209266036748886,
      "learning_rate": 3.866150310868571e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3538832,
      "step": 12640
    },
    {
      "epoch": 140.5,
      "grad_norm": 0.019932324066758156,
      "learning_rate": 3.8653280064533506e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3540256,
      "step": 12645
    },
    {
      "epoch": 140.55555555555554,
      "grad_norm": 0.022285420447587967,
      "learning_rate": 3.864505491487394e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3541648,
      "step": 12650
    },
    {
      "epoch": 140.61111111111111,
      "grad_norm": 0.07299376279115677,
      "learning_rate": 3.8636827660975414e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 3543120,
      "step": 12655
    },
    {
      "epoch": 140.66666666666666,
      "grad_norm": 0.025254493579268456,
      "learning_rate": 3.862859830410671e-05,
      "loss": 0.0066,
      "num_input_tokens_seen": 3544528,
      "step": 12660
    },
    {
      "epoch": 140.72222222222223,
      "grad_norm": 0.024454446509480476,
      "learning_rate": 3.862036684553688e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3545904,
      "step": 12665
    },
    {
      "epoch": 140.77777777777777,
      "grad_norm": 0.03202951326966286,
      "learning_rate": 3.8612133286535314e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 3547328,
      "step": 12670
    },
    {
      "epoch": 140.83333333333334,
      "grad_norm": 0.44497817754745483,
      "learning_rate": 3.860389762837173e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 3548720,
      "step": 12675
    },
    {
      "epoch": 140.88888888888889,
      "grad_norm": 0.011060819029808044,
      "learning_rate": 3.859565987231618e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3550080,
      "step": 12680
    },
    {
      "epoch": 140.94444444444446,
      "grad_norm": 0.037802886217832565,
      "learning_rate": 3.858742001963902e-05,
      "loss": 0.035,
      "num_input_tokens_seen": 3551488,
      "step": 12685
    },
    {
      "epoch": 141.0,
      "grad_norm": 0.00888059101998806,
      "learning_rate": 3.857917807161094e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 3552864,
      "step": 12690
    },
    {
      "epoch": 141.05555555555554,
      "grad_norm": 0.0625678151845932,
      "learning_rate": 3.857093402950296e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3554272,
      "step": 12695
    },
    {
      "epoch": 141.11111111111111,
      "grad_norm": 1.9705435037612915,
      "learning_rate": 3.8562687894586414e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 3555696,
      "step": 12700
    },
    {
      "epoch": 141.16666666666666,
      "grad_norm": 0.11866556107997894,
      "learning_rate": 3.8554439668132946e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3557104,
      "step": 12705
    },
    {
      "epoch": 141.22222222222223,
      "grad_norm": 0.12121996283531189,
      "learning_rate": 3.854618935141455e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3558480,
      "step": 12710
    },
    {
      "epoch": 141.27777777777777,
      "grad_norm": 0.2989172041416168,
      "learning_rate": 3.8537936945703525e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 3559872,
      "step": 12715
    },
    {
      "epoch": 141.33333333333334,
      "grad_norm": 0.01752912811934948,
      "learning_rate": 3.852968245227249e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3561280,
      "step": 12720
    },
    {
      "epoch": 141.38888888888889,
      "grad_norm": 0.016795359551906586,
      "learning_rate": 3.85214258723944e-05,
      "loss": 0.0221,
      "num_input_tokens_seen": 3562688,
      "step": 12725
    },
    {
      "epoch": 141.44444444444446,
      "grad_norm": 0.04334472119808197,
      "learning_rate": 3.8513167207342524e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3564080,
      "step": 12730
    },
    {
      "epoch": 141.5,
      "grad_norm": 0.07229319214820862,
      "learning_rate": 3.850490645839044e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3565440,
      "step": 12735
    },
    {
      "epoch": 141.55555555555554,
      "grad_norm": 0.03972737491130829,
      "learning_rate": 3.849664362681207e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3566880,
      "step": 12740
    },
    {
      "epoch": 141.61111111111111,
      "grad_norm": 0.021660510450601578,
      "learning_rate": 3.848837871388165e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3568288,
      "step": 12745
    },
    {
      "epoch": 141.66666666666666,
      "grad_norm": 0.009898400865495205,
      "learning_rate": 3.848011172087371e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 3569648,
      "step": 12750
    },
    {
      "epoch": 141.72222222222223,
      "grad_norm": 0.15598109364509583,
      "learning_rate": 3.847184264906315e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 3571056,
      "step": 12755
    },
    {
      "epoch": 141.77777777777777,
      "grad_norm": 0.08639650791883469,
      "learning_rate": 3.846357149972516e-05,
      "loss": 0.0309,
      "num_input_tokens_seen": 3572464,
      "step": 12760
    },
    {
      "epoch": 141.83333333333334,
      "grad_norm": 6.392677307128906,
      "learning_rate": 3.8455298274135246e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 3573824,
      "step": 12765
    },
    {
      "epoch": 141.88888888888889,
      "grad_norm": 0.04781216010451317,
      "learning_rate": 3.8447022973569254e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3575248,
      "step": 12770
    },
    {
      "epoch": 141.94444444444446,
      "grad_norm": 0.08037317544221878,
      "learning_rate": 3.843874559930332e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 3576672,
      "step": 12775
    },
    {
      "epoch": 142.0,
      "grad_norm": 0.05537543073296547,
      "learning_rate": 3.843046615261394e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3578144,
      "step": 12780
    },
    {
      "epoch": 142.05555555555554,
      "grad_norm": 0.01860489696264267,
      "learning_rate": 3.842218463477791e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3579552,
      "step": 12785
    },
    {
      "epoch": 142.11111111111111,
      "grad_norm": 0.08323255181312561,
      "learning_rate": 3.841390104707233e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 3580944,
      "step": 12790
    },
    {
      "epoch": 142.16666666666666,
      "grad_norm": 0.6733118295669556,
      "learning_rate": 3.8405615390774643e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3582336,
      "step": 12795
    },
    {
      "epoch": 142.22222222222223,
      "grad_norm": 0.1761132776737213,
      "learning_rate": 3.839732766716259e-05,
      "loss": 0.0068,
      "num_input_tokens_seen": 3583696,
      "step": 12800
    },
    {
      "epoch": 142.22222222222223,
      "eval_loss": 0.426602840423584,
      "eval_runtime": 1.2363,
      "eval_samples_per_second": 32.353,
      "eval_steps_per_second": 16.177,
      "num_input_tokens_seen": 3583696,
      "step": 12800
    },
    {
      "epoch": 142.27777777777777,
      "grad_norm": 0.3083146810531616,
      "learning_rate": 3.838903787751425e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 3585120,
      "step": 12805
    },
    {
      "epoch": 142.33333333333334,
      "grad_norm": 0.08341438323259354,
      "learning_rate": 3.838074602310802e-05,
      "loss": 0.0273,
      "num_input_tokens_seen": 3586496,
      "step": 12810
    },
    {
      "epoch": 142.38888888888889,
      "grad_norm": 0.27907976508140564,
      "learning_rate": 3.837245210522258e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3587888,
      "step": 12815
    },
    {
      "epoch": 142.44444444444446,
      "grad_norm": 0.0057378290221095085,
      "learning_rate": 3.8364156125136996e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3589280,
      "step": 12820
    },
    {
      "epoch": 142.5,
      "grad_norm": 1.389970302581787,
      "learning_rate": 3.835585808413059e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 3590704,
      "step": 12825
    },
    {
      "epoch": 142.55555555555554,
      "grad_norm": 0.5267683267593384,
      "learning_rate": 3.8347557983483024e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3592096,
      "step": 12830
    },
    {
      "epoch": 142.61111111111111,
      "grad_norm": 0.04194064810872078,
      "learning_rate": 3.833925582447428e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 3593472,
      "step": 12835
    },
    {
      "epoch": 142.66666666666666,
      "grad_norm": 0.011633186601102352,
      "learning_rate": 3.8330951608384656e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 3594880,
      "step": 12840
    },
    {
      "epoch": 142.72222222222223,
      "grad_norm": 0.3698907196521759,
      "learning_rate": 3.832264533649477e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3596304,
      "step": 12845
    },
    {
      "epoch": 142.77777777777777,
      "grad_norm": 0.08721329271793365,
      "learning_rate": 3.8314337010085555e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3597696,
      "step": 12850
    },
    {
      "epoch": 142.83333333333334,
      "grad_norm": 0.010365593247115612,
      "learning_rate": 3.830602663043824e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3599120,
      "step": 12855
    },
    {
      "epoch": 142.88888888888889,
      "grad_norm": 0.028627069666981697,
      "learning_rate": 3.8297714198834414e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3600528,
      "step": 12860
    },
    {
      "epoch": 142.94444444444446,
      "grad_norm": 0.1515538990497589,
      "learning_rate": 3.828939971655595e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 3601888,
      "step": 12865
    },
    {
      "epoch": 143.0,
      "grad_norm": 0.005986401345580816,
      "learning_rate": 3.828108318488505e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3603312,
      "step": 12870
    },
    {
      "epoch": 143.05555555555554,
      "grad_norm": 0.0628935843706131,
      "learning_rate": 3.8272764605104216e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3604704,
      "step": 12875
    },
    {
      "epoch": 143.11111111111111,
      "grad_norm": 0.011623067781329155,
      "learning_rate": 3.826444397849628e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3606096,
      "step": 12880
    },
    {
      "epoch": 143.16666666666666,
      "grad_norm": 0.019302526488900185,
      "learning_rate": 3.825612130634439e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 3607440,
      "step": 12885
    },
    {
      "epoch": 143.22222222222223,
      "grad_norm": 0.08214659243822098,
      "learning_rate": 3.824779658993202e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3608848,
      "step": 12890
    },
    {
      "epoch": 143.27777777777777,
      "grad_norm": 0.12360211461782455,
      "learning_rate": 3.823946983054292e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3610288,
      "step": 12895
    },
    {
      "epoch": 143.33333333333334,
      "grad_norm": 0.011767538264393806,
      "learning_rate": 3.82311410294612e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3611664,
      "step": 12900
    },
    {
      "epoch": 143.38888888888889,
      "grad_norm": 0.5565090179443359,
      "learning_rate": 3.822281018797127e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 3613056,
      "step": 12905
    },
    {
      "epoch": 143.44444444444446,
      "grad_norm": 0.05394411087036133,
      "learning_rate": 3.821447730735783e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3614480,
      "step": 12910
    },
    {
      "epoch": 143.5,
      "grad_norm": 0.03404231742024422,
      "learning_rate": 3.820614238890592e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3615888,
      "step": 12915
    },
    {
      "epoch": 143.55555555555554,
      "grad_norm": 0.012736672535538673,
      "learning_rate": 3.819780543390091e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3617312,
      "step": 12920
    },
    {
      "epoch": 143.61111111111111,
      "grad_norm": 0.04788237810134888,
      "learning_rate": 3.818946644362844e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3618752,
      "step": 12925
    },
    {
      "epoch": 143.66666666666666,
      "grad_norm": 0.026630975306034088,
      "learning_rate": 3.81811254193745e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3620176,
      "step": 12930
    },
    {
      "epoch": 143.72222222222223,
      "grad_norm": 0.01861611381173134,
      "learning_rate": 3.8172782362425366e-05,
      "loss": 0.0133,
      "num_input_tokens_seen": 3621568,
      "step": 12935
    },
    {
      "epoch": 143.77777777777777,
      "grad_norm": 0.0212355088442564,
      "learning_rate": 3.816443727406765e-05,
      "loss": 0.0266,
      "num_input_tokens_seen": 3623008,
      "step": 12940
    },
    {
      "epoch": 143.83333333333334,
      "grad_norm": 0.005189375951886177,
      "learning_rate": 3.815609015558829e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 3624400,
      "step": 12945
    },
    {
      "epoch": 143.88888888888889,
      "grad_norm": 1.3318662643432617,
      "learning_rate": 3.814774100827448e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3625776,
      "step": 12950
    },
    {
      "epoch": 143.94444444444446,
      "grad_norm": 0.025557242333889008,
      "learning_rate": 3.813938983341379e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3627168,
      "step": 12955
    },
    {
      "epoch": 144.0,
      "grad_norm": 0.06739328801631927,
      "learning_rate": 3.813103663229407e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3628544,
      "step": 12960
    },
    {
      "epoch": 144.05555555555554,
      "grad_norm": 0.04621768742799759,
      "learning_rate": 3.812268140620349e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 3629968,
      "step": 12965
    },
    {
      "epoch": 144.11111111111111,
      "grad_norm": 0.032243721187114716,
      "learning_rate": 3.811432415643051e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 3631344,
      "step": 12970
    },
    {
      "epoch": 144.16666666666666,
      "grad_norm": 0.7297658920288086,
      "learning_rate": 3.8105964884263954e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 3632736,
      "step": 12975
    },
    {
      "epoch": 144.22222222222223,
      "grad_norm": 0.0037953131832182407,
      "learning_rate": 3.809760359099291e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3634144,
      "step": 12980
    },
    {
      "epoch": 144.27777777777777,
      "grad_norm": 0.08640396595001221,
      "learning_rate": 3.8089240277906804e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3635520,
      "step": 12985
    },
    {
      "epoch": 144.33333333333334,
      "grad_norm": 0.009387868456542492,
      "learning_rate": 3.808087494629535e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3636864,
      "step": 12990
    },
    {
      "epoch": 144.38888888888889,
      "grad_norm": 2.6446900367736816,
      "learning_rate": 3.8072507597448595e-05,
      "loss": 0.0308,
      "num_input_tokens_seen": 3638240,
      "step": 12995
    },
    {
      "epoch": 144.44444444444446,
      "grad_norm": 0.023002121597528458,
      "learning_rate": 3.806413823265689e-05,
      "loss": 0.0043,
      "num_input_tokens_seen": 3639680,
      "step": 13000
    },
    {
      "epoch": 144.44444444444446,
      "eval_loss": 0.4579645097255707,
      "eval_runtime": 1.2385,
      "eval_samples_per_second": 32.296,
      "eval_steps_per_second": 16.148,
      "num_input_tokens_seen": 3639680,
      "step": 13000
    },
    {
      "epoch": 144.5,
      "grad_norm": 0.02385997772216797,
      "learning_rate": 3.805576685321089e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3641088,
      "step": 13005
    },
    {
      "epoch": 144.55555555555554,
      "grad_norm": 0.024489574134349823,
      "learning_rate": 3.804739346040158e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3642464,
      "step": 13010
    },
    {
      "epoch": 144.61111111111111,
      "grad_norm": 0.06517381221055984,
      "learning_rate": 3.8039018055520234e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3643824,
      "step": 13015
    },
    {
      "epoch": 144.66666666666666,
      "grad_norm": 0.056034155189991,
      "learning_rate": 3.803064063985844e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3645232,
      "step": 13020
    },
    {
      "epoch": 144.72222222222223,
      "grad_norm": 0.8669463396072388,
      "learning_rate": 3.802226121470811e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3646640,
      "step": 13025
    },
    {
      "epoch": 144.77777777777777,
      "grad_norm": 2.370462417602539,
      "learning_rate": 3.801387978136145e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 3648032,
      "step": 13030
    },
    {
      "epoch": 144.83333333333334,
      "grad_norm": 2.495654582977295,
      "learning_rate": 3.800549634111099e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 3649456,
      "step": 13035
    },
    {
      "epoch": 144.88888888888889,
      "grad_norm": 0.13245819509029388,
      "learning_rate": 3.799711089524955e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3650864,
      "step": 13040
    },
    {
      "epoch": 144.94444444444446,
      "grad_norm": 0.009469658136367798,
      "learning_rate": 3.7988723445070285e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3652272,
      "step": 13045
    },
    {
      "epoch": 145.0,
      "grad_norm": 0.126252219080925,
      "learning_rate": 3.798033399186663e-05,
      "loss": 0.0253,
      "num_input_tokens_seen": 3653696,
      "step": 13050
    },
    {
      "epoch": 145.05555555555554,
      "grad_norm": 0.013958986848592758,
      "learning_rate": 3.797194253693237e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 3655120,
      "step": 13055
    },
    {
      "epoch": 145.11111111111111,
      "grad_norm": 0.01678682118654251,
      "learning_rate": 3.796354908156153e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3656512,
      "step": 13060
    },
    {
      "epoch": 145.16666666666666,
      "grad_norm": 0.016464663669466972,
      "learning_rate": 3.795515362704853e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3657872,
      "step": 13065
    },
    {
      "epoch": 145.22222222222223,
      "grad_norm": 0.031349409371614456,
      "learning_rate": 3.794675617468803e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3659280,
      "step": 13070
    },
    {
      "epoch": 145.27777777777777,
      "grad_norm": 0.07278244942426682,
      "learning_rate": 3.793835672577503e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3660720,
      "step": 13075
    },
    {
      "epoch": 145.33333333333334,
      "grad_norm": 0.03195759281516075,
      "learning_rate": 3.7929955281604826e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3662144,
      "step": 13080
    },
    {
      "epoch": 145.38888888888889,
      "grad_norm": 0.1755899339914322,
      "learning_rate": 3.7921551843473036e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3663552,
      "step": 13085
    },
    {
      "epoch": 145.44444444444446,
      "grad_norm": 0.06375882029533386,
      "learning_rate": 3.791314641267557e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 3664944,
      "step": 13090
    },
    {
      "epoch": 145.5,
      "grad_norm": 0.12454172968864441,
      "learning_rate": 3.790473899050864e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3666368,
      "step": 13095
    },
    {
      "epoch": 145.55555555555554,
      "grad_norm": 0.18290726840496063,
      "learning_rate": 3.7896329578268794e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3667776,
      "step": 13100
    },
    {
      "epoch": 145.61111111111111,
      "grad_norm": 0.08017271012067795,
      "learning_rate": 3.7887918177252855e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3669168,
      "step": 13105
    },
    {
      "epoch": 145.66666666666666,
      "grad_norm": 0.09389721602201462,
      "learning_rate": 3.787950478875798e-05,
      "loss": 0.0199,
      "num_input_tokens_seen": 3670576,
      "step": 13110
    },
    {
      "epoch": 145.72222222222223,
      "grad_norm": 0.05228535830974579,
      "learning_rate": 3.787108941408162e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3671968,
      "step": 13115
    },
    {
      "epoch": 145.77777777777777,
      "grad_norm": 0.020902976393699646,
      "learning_rate": 3.786267205452151e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3673344,
      "step": 13120
    },
    {
      "epoch": 145.83333333333334,
      "grad_norm": 2.1039860248565674,
      "learning_rate": 3.785425271137573e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 3674752,
      "step": 13125
    },
    {
      "epoch": 145.88888888888889,
      "grad_norm": 0.0066171386279165745,
      "learning_rate": 3.7845831385942655e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3676160,
      "step": 13130
    },
    {
      "epoch": 145.94444444444446,
      "grad_norm": 0.14140097796916962,
      "learning_rate": 3.7837408079520944e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3677568,
      "step": 13135
    },
    {
      "epoch": 146.0,
      "grad_norm": 0.06716878712177277,
      "learning_rate": 3.782898279340957e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3678992,
      "step": 13140
    },
    {
      "epoch": 146.05555555555554,
      "grad_norm": 0.042472466826438904,
      "learning_rate": 3.782055552890784e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3680368,
      "step": 13145
    },
    {
      "epoch": 146.11111111111111,
      "grad_norm": 0.012548373080790043,
      "learning_rate": 3.781212628731534e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3681776,
      "step": 13150
    },
    {
      "epoch": 146.16666666666666,
      "grad_norm": 0.024183019995689392,
      "learning_rate": 3.7803695069931946e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3683184,
      "step": 13155
    },
    {
      "epoch": 146.22222222222223,
      "grad_norm": 0.015416059643030167,
      "learning_rate": 3.779526187805789e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3684640,
      "step": 13160
    },
    {
      "epoch": 146.27777777777777,
      "grad_norm": 0.7095667719841003,
      "learning_rate": 3.778682671299364e-05,
      "loss": 0.0288,
      "num_input_tokens_seen": 3686000,
      "step": 13165
    },
    {
      "epoch": 146.33333333333334,
      "grad_norm": 0.006984783802181482,
      "learning_rate": 3.777838957604003e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3687344,
      "step": 13170
    },
    {
      "epoch": 146.38888888888889,
      "grad_norm": 0.02276059426367283,
      "learning_rate": 3.776995046849816e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3688720,
      "step": 13175
    },
    {
      "epoch": 146.44444444444446,
      "grad_norm": 0.004272219724953175,
      "learning_rate": 3.776150939166945e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3690112,
      "step": 13180
    },
    {
      "epoch": 146.5,
      "grad_norm": 0.21709105372428894,
      "learning_rate": 3.775306634685562e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3691472,
      "step": 13185
    },
    {
      "epoch": 146.55555555555554,
      "grad_norm": 0.011026831343770027,
      "learning_rate": 3.7744621335358696e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3692896,
      "step": 13190
    },
    {
      "epoch": 146.61111111111111,
      "grad_norm": 0.0030232174322009087,
      "learning_rate": 3.7736174358481e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 3694320,
      "step": 13195
    },
    {
      "epoch": 146.66666666666666,
      "grad_norm": 0.023872708901762962,
      "learning_rate": 3.7727725417525175e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 3695712,
      "step": 13200
    },
    {
      "epoch": 146.66666666666666,
      "eval_loss": 0.514919638633728,
      "eval_runtime": 1.2378,
      "eval_samples_per_second": 32.315,
      "eval_steps_per_second": 16.158,
      "num_input_tokens_seen": 3695712,
      "step": 13200
    },
    {
      "epoch": 146.72222222222223,
      "grad_norm": 0.028826244175434113,
      "learning_rate": 3.771927451379414e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 3697088,
      "step": 13205
    },
    {
      "epoch": 146.77777777777777,
      "grad_norm": 0.3557609021663666,
      "learning_rate": 3.7710821648591135e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3698544,
      "step": 13210
    },
    {
      "epoch": 146.83333333333334,
      "grad_norm": 0.013385373167693615,
      "learning_rate": 3.7702366823219694e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3699904,
      "step": 13215
    },
    {
      "epoch": 146.88888888888889,
      "grad_norm": 0.0715845450758934,
      "learning_rate": 3.769391003898366e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 3701296,
      "step": 13220
    },
    {
      "epoch": 146.94444444444446,
      "grad_norm": 0.014233303256332874,
      "learning_rate": 3.768545129718718e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3702656,
      "step": 13225
    },
    {
      "epoch": 147.0,
      "grad_norm": 0.07945820689201355,
      "learning_rate": 3.7676990599134686e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3704064,
      "step": 13230
    },
    {
      "epoch": 147.05555555555554,
      "grad_norm": 0.04315363988280296,
      "learning_rate": 3.766852794613095e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3705440,
      "step": 13235
    },
    {
      "epoch": 147.11111111111111,
      "grad_norm": 0.011165213771164417,
      "learning_rate": 3.766006333948099e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3706816,
      "step": 13240
    },
    {
      "epoch": 147.16666666666666,
      "grad_norm": 0.011860559694468975,
      "learning_rate": 3.765159678049017e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3708192,
      "step": 13245
    },
    {
      "epoch": 147.22222222222223,
      "grad_norm": 0.09621091187000275,
      "learning_rate": 3.7643128270464134e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3709616,
      "step": 13250
    },
    {
      "epoch": 147.27777777777777,
      "grad_norm": 0.06915857642889023,
      "learning_rate": 3.763465781070884e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3710992,
      "step": 13255
    },
    {
      "epoch": 147.33333333333334,
      "grad_norm": 0.053173646330833435,
      "learning_rate": 3.762618540253052e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3712400,
      "step": 13260
    },
    {
      "epoch": 147.38888888888889,
      "grad_norm": 0.03508806228637695,
      "learning_rate": 3.761771104723576e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3713824,
      "step": 13265
    },
    {
      "epoch": 147.44444444444446,
      "grad_norm": 0.050781670957803726,
      "learning_rate": 3.7609234746131386e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3715264,
      "step": 13270
    },
    {
      "epoch": 147.5,
      "grad_norm": 0.043136704713106155,
      "learning_rate": 3.7600756500524556e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 3716656,
      "step": 13275
    },
    {
      "epoch": 147.55555555555554,
      "grad_norm": 0.2668047547340393,
      "learning_rate": 3.759227631172271e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 3718064,
      "step": 13280
    },
    {
      "epoch": 147.61111111111111,
      "grad_norm": 0.028775300830602646,
      "learning_rate": 3.758379418103363e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3719456,
      "step": 13285
    },
    {
      "epoch": 147.66666666666666,
      "grad_norm": 0.1000927984714508,
      "learning_rate": 3.757531010976534e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3720848,
      "step": 13290
    },
    {
      "epoch": 147.72222222222223,
      "grad_norm": 0.3715973496437073,
      "learning_rate": 3.75668240992262e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3722224,
      "step": 13295
    },
    {
      "epoch": 147.77777777777777,
      "grad_norm": 0.2269083708524704,
      "learning_rate": 3.7558336150724865e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3723648,
      "step": 13300
    },
    {
      "epoch": 147.83333333333334,
      "grad_norm": 0.021085944026708603,
      "learning_rate": 3.754984626557028e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3725024,
      "step": 13305
    },
    {
      "epoch": 147.88888888888889,
      "grad_norm": 0.01694333367049694,
      "learning_rate": 3.754135444507168e-05,
      "loss": 0.0329,
      "num_input_tokens_seen": 3726416,
      "step": 13310
    },
    {
      "epoch": 147.94444444444446,
      "grad_norm": 0.005560481455177069,
      "learning_rate": 3.753286069053863e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3727840,
      "step": 13315
    },
    {
      "epoch": 148.0,
      "grad_norm": 2.0241010189056396,
      "learning_rate": 3.7524365003280945e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 3729264,
      "step": 13320
    },
    {
      "epoch": 148.05555555555554,
      "grad_norm": 0.013879070058465004,
      "learning_rate": 3.75158673846088e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3730656,
      "step": 13325
    },
    {
      "epoch": 148.11111111111111,
      "grad_norm": 0.11771461367607117,
      "learning_rate": 3.750736783583262e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3732048,
      "step": 13330
    },
    {
      "epoch": 148.16666666666666,
      "grad_norm": 0.08704084903001785,
      "learning_rate": 3.7498866358263144e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3733472,
      "step": 13335
    },
    {
      "epoch": 148.22222222222223,
      "grad_norm": 0.5028125047683716,
      "learning_rate": 3.74903629532114e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 3734848,
      "step": 13340
    },
    {
      "epoch": 148.27777777777777,
      "grad_norm": 2.9901773929595947,
      "learning_rate": 3.748185762198873e-05,
      "loss": 0.0152,
      "num_input_tokens_seen": 3736288,
      "step": 13345
    },
    {
      "epoch": 148.33333333333334,
      "grad_norm": 0.024059899151325226,
      "learning_rate": 3.747335036590676e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3737712,
      "step": 13350
    },
    {
      "epoch": 148.38888888888889,
      "grad_norm": 0.0348508320748806,
      "learning_rate": 3.7464841186277405e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3739120,
      "step": 13355
    },
    {
      "epoch": 148.44444444444446,
      "grad_norm": 0.30866190791130066,
      "learning_rate": 3.7456330084412896e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 3740560,
      "step": 13360
    },
    {
      "epoch": 148.5,
      "grad_norm": 0.024098780006170273,
      "learning_rate": 3.744781706162576e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3741968,
      "step": 13365
    },
    {
      "epoch": 148.55555555555554,
      "grad_norm": 0.5892476439476013,
      "learning_rate": 3.743930211922879e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3743312,
      "step": 13370
    },
    {
      "epoch": 148.61111111111111,
      "grad_norm": 0.00800534337759018,
      "learning_rate": 3.743078525853513e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 3744688,
      "step": 13375
    },
    {
      "epoch": 148.66666666666666,
      "grad_norm": 0.9248530268669128,
      "learning_rate": 3.7422266480858154e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3746080,
      "step": 13380
    },
    {
      "epoch": 148.72222222222223,
      "grad_norm": 0.07265818119049072,
      "learning_rate": 3.741374578751158e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3747504,
      "step": 13385
    },
    {
      "epoch": 148.77777777777777,
      "grad_norm": 0.01094350777566433,
      "learning_rate": 3.740522317980941e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3748928,
      "step": 13390
    },
    {
      "epoch": 148.83333333333334,
      "grad_norm": 0.024197399616241455,
      "learning_rate": 3.739669865906593e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 3750320,
      "step": 13395
    },
    {
      "epoch": 148.88888888888889,
      "grad_norm": 0.008320147171616554,
      "learning_rate": 3.738817222659573e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3751728,
      "step": 13400
    },
    {
      "epoch": 148.88888888888889,
      "eval_loss": 0.4668912887573242,
      "eval_runtime": 1.2331,
      "eval_samples_per_second": 32.44,
      "eval_steps_per_second": 16.22,
      "num_input_tokens_seen": 3751728,
      "step": 13400
    },
    {
      "epoch": 148.94444444444446,
      "grad_norm": 0.00837884284555912,
      "learning_rate": 3.73796438837137e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3753136,
      "step": 13405
    },
    {
      "epoch": 149.0,
      "grad_norm": 0.009917774237692356,
      "learning_rate": 3.7371113631735e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3754544,
      "step": 13410
    },
    {
      "epoch": 149.05555555555554,
      "grad_norm": 0.2588339149951935,
      "learning_rate": 3.736258147197512e-05,
      "loss": 0.0151,
      "num_input_tokens_seen": 3755920,
      "step": 13415
    },
    {
      "epoch": 149.11111111111111,
      "grad_norm": 1.6054434776306152,
      "learning_rate": 3.735404740574981e-05,
      "loss": 0.0344,
      "num_input_tokens_seen": 3757312,
      "step": 13420
    },
    {
      "epoch": 149.16666666666666,
      "grad_norm": 0.020532606169581413,
      "learning_rate": 3.7345511434375145e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3758720,
      "step": 13425
    },
    {
      "epoch": 149.22222222222223,
      "grad_norm": 0.033600013703107834,
      "learning_rate": 3.733697355916748e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3760144,
      "step": 13430
    },
    {
      "epoch": 149.27777777777777,
      "grad_norm": 0.058087609708309174,
      "learning_rate": 3.732843378144345e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 3761504,
      "step": 13435
    },
    {
      "epoch": 149.33333333333334,
      "grad_norm": 0.033969514071941376,
      "learning_rate": 3.7319892102519995e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3762912,
      "step": 13440
    },
    {
      "epoch": 149.38888888888889,
      "grad_norm": 0.020993979647755623,
      "learning_rate": 3.731134852371436e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3764336,
      "step": 13445
    },
    {
      "epoch": 149.44444444444446,
      "grad_norm": 0.006202940363436937,
      "learning_rate": 3.730280304634408e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3765744,
      "step": 13450
    },
    {
      "epoch": 149.5,
      "grad_norm": 0.05042724683880806,
      "learning_rate": 3.729425567172696e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3767104,
      "step": 13455
    },
    {
      "epoch": 149.55555555555554,
      "grad_norm": 0.05421210825443268,
      "learning_rate": 3.728570640118111e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3768512,
      "step": 13460
    },
    {
      "epoch": 149.61111111111111,
      "grad_norm": 0.013549835421144962,
      "learning_rate": 3.727715523602494e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3769920,
      "step": 13465
    },
    {
      "epoch": 149.66666666666666,
      "grad_norm": 0.0081245806068182,
      "learning_rate": 3.726860217757715e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3771328,
      "step": 13470
    },
    {
      "epoch": 149.72222222222223,
      "grad_norm": 0.04567393660545349,
      "learning_rate": 3.726004722715673e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3772720,
      "step": 13475
    },
    {
      "epoch": 149.77777777777777,
      "grad_norm": 0.015359323471784592,
      "learning_rate": 3.725149038608296e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3774144,
      "step": 13480
    },
    {
      "epoch": 149.83333333333334,
      "grad_norm": 0.032566703855991364,
      "learning_rate": 3.7242931655675404e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3775536,
      "step": 13485
    },
    {
      "epoch": 149.88888888888889,
      "grad_norm": 0.00871114619076252,
      "learning_rate": 3.7234371037253937e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3776912,
      "step": 13490
    },
    {
      "epoch": 149.94444444444446,
      "grad_norm": 0.029611574485898018,
      "learning_rate": 3.7225808532138705e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3778304,
      "step": 13495
    },
    {
      "epoch": 150.0,
      "grad_norm": 0.013342730700969696,
      "learning_rate": 3.721724414165016e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 3779696,
      "step": 13500
    },
    {
      "epoch": 150.05555555555554,
      "grad_norm": 0.011000074446201324,
      "learning_rate": 3.720867786710904e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3781056,
      "step": 13505
    },
    {
      "epoch": 150.11111111111111,
      "grad_norm": 0.1213960349559784,
      "learning_rate": 3.7200109709836366e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 3782448,
      "step": 13510
    },
    {
      "epoch": 150.16666666666666,
      "grad_norm": 0.006408501882106066,
      "learning_rate": 3.7191539671153465e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3783824,
      "step": 13515
    },
    {
      "epoch": 150.22222222222223,
      "grad_norm": 0.0071240696124732494,
      "learning_rate": 3.718296775238193e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 3785216,
      "step": 13520
    },
    {
      "epoch": 150.27777777777777,
      "grad_norm": 0.014129114337265491,
      "learning_rate": 3.7174393954843675e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3786656,
      "step": 13525
    },
    {
      "epoch": 150.33333333333334,
      "grad_norm": 0.0068581136874854565,
      "learning_rate": 3.716581827986087e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3788064,
      "step": 13530
    },
    {
      "epoch": 150.38888888888889,
      "grad_norm": 0.011368085630238056,
      "learning_rate": 3.7157240728756004e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 3789488,
      "step": 13535
    },
    {
      "epoch": 150.44444444444446,
      "grad_norm": 0.019848758354783058,
      "learning_rate": 3.714866130285184e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3790848,
      "step": 13540
    },
    {
      "epoch": 150.5,
      "grad_norm": 0.025280149653553963,
      "learning_rate": 3.714008000347143e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3792224,
      "step": 13545
    },
    {
      "epoch": 150.55555555555554,
      "grad_norm": 0.05540541186928749,
      "learning_rate": 3.7131496831938126e-05,
      "loss": 0.0149,
      "num_input_tokens_seen": 3793664,
      "step": 13550
    },
    {
      "epoch": 150.61111111111111,
      "grad_norm": 0.06868914514780045,
      "learning_rate": 3.7122911789575565e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3795072,
      "step": 13555
    },
    {
      "epoch": 150.66666666666666,
      "grad_norm": 0.014258590526878834,
      "learning_rate": 3.711432487770765e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3796464,
      "step": 13560
    },
    {
      "epoch": 150.72222222222223,
      "grad_norm": 0.01161752175539732,
      "learning_rate": 3.710573609765861e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 3797904,
      "step": 13565
    },
    {
      "epoch": 150.77777777777777,
      "grad_norm": 0.07032182067632675,
      "learning_rate": 3.709714545075292e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3799312,
      "step": 13570
    },
    {
      "epoch": 150.83333333333334,
      "grad_norm": 0.01632436364889145,
      "learning_rate": 3.708855293831538e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3800736,
      "step": 13575
    },
    {
      "epoch": 150.88888888888889,
      "grad_norm": 0.06245862692594528,
      "learning_rate": 3.707995856167107e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 3802144,
      "step": 13580
    },
    {
      "epoch": 150.94444444444446,
      "grad_norm": 1.7424378395080566,
      "learning_rate": 3.707136232214534e-05,
      "loss": 0.0108,
      "num_input_tokens_seen": 3803520,
      "step": 13585
    },
    {
      "epoch": 151.0,
      "grad_norm": 0.029404016211628914,
      "learning_rate": 3.7062764221063844e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 3804896,
      "step": 13590
    },
    {
      "epoch": 151.05555555555554,
      "grad_norm": 0.077729232609272,
      "learning_rate": 3.705416425975252e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3806304,
      "step": 13595
    },
    {
      "epoch": 151.11111111111111,
      "grad_norm": 0.005876823794096708,
      "learning_rate": 3.704556243953758e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3807744,
      "step": 13600
    },
    {
      "epoch": 151.11111111111111,
      "eval_loss": 0.47362518310546875,
      "eval_runtime": 1.2481,
      "eval_samples_per_second": 32.048,
      "eval_steps_per_second": 16.024,
      "num_input_tokens_seen": 3807744,
      "step": 13600
    },
    {
      "epoch": 151.16666666666666,
      "grad_norm": 0.03162286430597305,
      "learning_rate": 3.7036958761745535e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 3809104,
      "step": 13605
    },
    {
      "epoch": 151.22222222222223,
      "grad_norm": 0.030604815110564232,
      "learning_rate": 3.702835322770318e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3810480,
      "step": 13610
    },
    {
      "epoch": 151.27777777777777,
      "grad_norm": 0.09330737590789795,
      "learning_rate": 3.701974583873761e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3811920,
      "step": 13615
    },
    {
      "epoch": 151.33333333333334,
      "grad_norm": 0.04862061142921448,
      "learning_rate": 3.701113659617618e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 3813328,
      "step": 13620
    },
    {
      "epoch": 151.38888888888889,
      "grad_norm": 0.08583187311887741,
      "learning_rate": 3.7002525501346535e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3814752,
      "step": 13625
    },
    {
      "epoch": 151.44444444444446,
      "grad_norm": 0.0017061978578567505,
      "learning_rate": 3.699391255557664e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3816096,
      "step": 13630
    },
    {
      "epoch": 151.5,
      "grad_norm": 0.016951406374573708,
      "learning_rate": 3.69852977601947e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3817488,
      "step": 13635
    },
    {
      "epoch": 151.55555555555554,
      "grad_norm": 0.10625593364238739,
      "learning_rate": 3.697668111652922e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3818880,
      "step": 13640
    },
    {
      "epoch": 151.61111111111111,
      "grad_norm": 0.3606511354446411,
      "learning_rate": 3.6968062625909005e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3820272,
      "step": 13645
    },
    {
      "epoch": 151.66666666666666,
      "grad_norm": 0.0111155416816473,
      "learning_rate": 3.6959442289663135e-05,
      "loss": 0.0175,
      "num_input_tokens_seen": 3821664,
      "step": 13650
    },
    {
      "epoch": 151.72222222222223,
      "grad_norm": 0.015013080090284348,
      "learning_rate": 3.695082010912098e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3823072,
      "step": 13655
    },
    {
      "epoch": 151.77777777777777,
      "grad_norm": 0.0022852043621242046,
      "learning_rate": 3.694219608561217e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3824464,
      "step": 13660
    },
    {
      "epoch": 151.83333333333334,
      "grad_norm": 0.019199438393115997,
      "learning_rate": 3.693357022046665e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3825888,
      "step": 13665
    },
    {
      "epoch": 151.88888888888889,
      "grad_norm": 0.011504649184644222,
      "learning_rate": 3.6924942515014644e-05,
      "loss": 0.0138,
      "num_input_tokens_seen": 3827280,
      "step": 13670
    },
    {
      "epoch": 151.94444444444446,
      "grad_norm": 0.014987872913479805,
      "learning_rate": 3.691631297058664e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3828720,
      "step": 13675
    },
    {
      "epoch": 152.0,
      "grad_norm": 0.03241609036922455,
      "learning_rate": 3.6907681588513424e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3830096,
      "step": 13680
    },
    {
      "epoch": 152.05555555555554,
      "grad_norm": 0.09393493086099625,
      "learning_rate": 3.689904837012606e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3831520,
      "step": 13685
    },
    {
      "epoch": 152.11111111111111,
      "grad_norm": 0.00808445643633604,
      "learning_rate": 3.689041331675591e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3832864,
      "step": 13690
    },
    {
      "epoch": 152.16666666666666,
      "grad_norm": 0.013497444801032543,
      "learning_rate": 3.688177642973461e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3834224,
      "step": 13695
    },
    {
      "epoch": 152.22222222222223,
      "grad_norm": 0.006993724964559078,
      "learning_rate": 3.687313771039406e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3835680,
      "step": 13700
    },
    {
      "epoch": 152.27777777777777,
      "grad_norm": 0.03309626132249832,
      "learning_rate": 3.686449716006647e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3837072,
      "step": 13705
    },
    {
      "epoch": 152.33333333333334,
      "grad_norm": 0.02198244258761406,
      "learning_rate": 3.685585478008432e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3838480,
      "step": 13710
    },
    {
      "epoch": 152.38888888888889,
      "grad_norm": 0.10746368020772934,
      "learning_rate": 3.6847210571780364e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 3839840,
      "step": 13715
    },
    {
      "epoch": 152.44444444444446,
      "grad_norm": 0.005832733120769262,
      "learning_rate": 3.683856453648767e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3841200,
      "step": 13720
    },
    {
      "epoch": 152.5,
      "grad_norm": 0.012480447068810463,
      "learning_rate": 3.682991667553954e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 3842576,
      "step": 13725
    },
    {
      "epoch": 152.55555555555554,
      "grad_norm": 0.06548471748828888,
      "learning_rate": 3.6821266990269606e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3843968,
      "step": 13730
    },
    {
      "epoch": 152.61111111111111,
      "grad_norm": 1.8378793001174927,
      "learning_rate": 3.681261548201174e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 3845376,
      "step": 13735
    },
    {
      "epoch": 152.66666666666666,
      "grad_norm": 0.0034822230227291584,
      "learning_rate": 3.6803962152100125e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3846816,
      "step": 13740
    },
    {
      "epoch": 152.72222222222223,
      "grad_norm": 0.05154639109969139,
      "learning_rate": 3.67953070018692e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3848208,
      "step": 13745
    },
    {
      "epoch": 152.77777777777777,
      "grad_norm": 0.0038211923092603683,
      "learning_rate": 3.678665003265371e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3849600,
      "step": 13750
    },
    {
      "epoch": 152.83333333333334,
      "grad_norm": 0.0032776957377791405,
      "learning_rate": 3.677799124578867e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3851040,
      "step": 13755
    },
    {
      "epoch": 152.88888888888889,
      "grad_norm": 0.002329333918169141,
      "learning_rate": 3.676933064260937e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3852432,
      "step": 13760
    },
    {
      "epoch": 152.94444444444446,
      "grad_norm": 0.09900345653295517,
      "learning_rate": 3.6760668224451365e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3853872,
      "step": 13765
    },
    {
      "epoch": 153.0,
      "grad_norm": 0.0021476156543940306,
      "learning_rate": 3.675200399265054e-05,
      "loss": 0.0104,
      "num_input_tokens_seen": 3855264,
      "step": 13770
    },
    {
      "epoch": 153.05555555555554,
      "grad_norm": 0.009853934869170189,
      "learning_rate": 3.6743337948543014e-05,
      "loss": 0.0158,
      "num_input_tokens_seen": 3856656,
      "step": 13775
    },
    {
      "epoch": 153.11111111111111,
      "grad_norm": 0.05827099829912186,
      "learning_rate": 3.6734670093465204e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3858048,
      "step": 13780
    },
    {
      "epoch": 153.16666666666666,
      "grad_norm": 0.01043347641825676,
      "learning_rate": 3.672600042875379e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3859456,
      "step": 13785
    },
    {
      "epoch": 153.22222222222223,
      "grad_norm": 0.02227214351296425,
      "learning_rate": 3.671732895574575e-05,
      "loss": 0.0057,
      "num_input_tokens_seen": 3860848,
      "step": 13790
    },
    {
      "epoch": 153.27777777777777,
      "grad_norm": 0.010003415867686272,
      "learning_rate": 3.670865567577834e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3862304,
      "step": 13795
    },
    {
      "epoch": 153.33333333333334,
      "grad_norm": 0.07038789987564087,
      "learning_rate": 3.669998059018909e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3863664,
      "step": 13800
    },
    {
      "epoch": 153.33333333333334,
      "eval_loss": 0.48506394028663635,
      "eval_runtime": 1.2311,
      "eval_samples_per_second": 32.491,
      "eval_steps_per_second": 16.245,
      "num_input_tokens_seen": 3863664,
      "step": 13800
    },
    {
      "epoch": 153.38888888888889,
      "grad_norm": 0.03988673537969589,
      "learning_rate": 3.6691303700315796e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3865024,
      "step": 13805
    },
    {
      "epoch": 153.44444444444446,
      "grad_norm": 0.03924393281340599,
      "learning_rate": 3.668262500749655e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3866448,
      "step": 13810
    },
    {
      "epoch": 153.5,
      "grad_norm": 0.02342815324664116,
      "learning_rate": 3.667394451306971e-05,
      "loss": 0.0148,
      "num_input_tokens_seen": 3867856,
      "step": 13815
    },
    {
      "epoch": 153.55555555555554,
      "grad_norm": 0.22015252709388733,
      "learning_rate": 3.666526221837393e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3869232,
      "step": 13820
    },
    {
      "epoch": 153.61111111111111,
      "grad_norm": 0.10990491509437561,
      "learning_rate": 3.665657812474812e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3870640,
      "step": 13825
    },
    {
      "epoch": 153.66666666666666,
      "grad_norm": 0.05183480307459831,
      "learning_rate": 3.664789223353147e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 3872032,
      "step": 13830
    },
    {
      "epoch": 153.72222222222223,
      "grad_norm": 0.2020704746246338,
      "learning_rate": 3.663920454606347e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3873424,
      "step": 13835
    },
    {
      "epoch": 153.77777777777777,
      "grad_norm": 0.2967427372932434,
      "learning_rate": 3.6630515063683856e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3874800,
      "step": 13840
    },
    {
      "epoch": 153.83333333333334,
      "grad_norm": 0.011029688641428947,
      "learning_rate": 3.662182378773267e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3876176,
      "step": 13845
    },
    {
      "epoch": 153.88888888888889,
      "grad_norm": 0.023720914497971535,
      "learning_rate": 3.66131307195502e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3877584,
      "step": 13850
    },
    {
      "epoch": 153.94444444444446,
      "grad_norm": 0.0409511923789978,
      "learning_rate": 3.6604435860477034e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3878944,
      "step": 13855
    },
    {
      "epoch": 154.0,
      "grad_norm": 0.018461719155311584,
      "learning_rate": 3.6595739211854025e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3880400,
      "step": 13860
    },
    {
      "epoch": 154.05555555555554,
      "grad_norm": 0.005385678727179766,
      "learning_rate": 3.658704077502231e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 3881808,
      "step": 13865
    },
    {
      "epoch": 154.11111111111111,
      "grad_norm": 0.006448786240071058,
      "learning_rate": 3.65783405513233e-05,
      "loss": 0.0046,
      "num_input_tokens_seen": 3883264,
      "step": 13870
    },
    {
      "epoch": 154.16666666666666,
      "grad_norm": 0.3455326557159424,
      "learning_rate": 3.656963854209867e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 3884608,
      "step": 13875
    },
    {
      "epoch": 154.22222222222223,
      "grad_norm": 0.10698522627353668,
      "learning_rate": 3.656093474869038e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3885952,
      "step": 13880
    },
    {
      "epoch": 154.27777777777777,
      "grad_norm": 0.01955278217792511,
      "learning_rate": 3.655222917244068e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3887360,
      "step": 13885
    },
    {
      "epoch": 154.33333333333334,
      "grad_norm": 0.05388588085770607,
      "learning_rate": 3.6543521814692054e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3888736,
      "step": 13890
    },
    {
      "epoch": 154.38888888888889,
      "grad_norm": 0.013769922778010368,
      "learning_rate": 3.653481267678731e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3890160,
      "step": 13895
    },
    {
      "epoch": 154.44444444444446,
      "grad_norm": 0.04555009305477142,
      "learning_rate": 3.652610176006949e-05,
      "loss": 0.0141,
      "num_input_tokens_seen": 3891520,
      "step": 13900
    },
    {
      "epoch": 154.5,
      "grad_norm": 0.006007580552250147,
      "learning_rate": 3.6517389065881925e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3892912,
      "step": 13905
    },
    {
      "epoch": 154.55555555555554,
      "grad_norm": 0.003078807843849063,
      "learning_rate": 3.650867459556824e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3894320,
      "step": 13910
    },
    {
      "epoch": 154.61111111111111,
      "grad_norm": 0.02016676776111126,
      "learning_rate": 3.64999583504723e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 3895728,
      "step": 13915
    },
    {
      "epoch": 154.66666666666666,
      "grad_norm": 0.06275039166212082,
      "learning_rate": 3.649124033193827e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3897136,
      "step": 13920
    },
    {
      "epoch": 154.72222222222223,
      "grad_norm": 0.0030572619289159775,
      "learning_rate": 3.648252054131057e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 3898560,
      "step": 13925
    },
    {
      "epoch": 154.77777777777777,
      "grad_norm": 0.0038478130009025335,
      "learning_rate": 3.647379897993391e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3899968,
      "step": 13930
    },
    {
      "epoch": 154.83333333333334,
      "grad_norm": 0.004852957092225552,
      "learning_rate": 3.646507564915325e-05,
      "loss": 0.0095,
      "num_input_tokens_seen": 3901360,
      "step": 13935
    },
    {
      "epoch": 154.88888888888889,
      "grad_norm": 0.013995690271258354,
      "learning_rate": 3.645635055031385e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3902752,
      "step": 13940
    },
    {
      "epoch": 154.94444444444446,
      "grad_norm": 0.08828818053007126,
      "learning_rate": 3.6447623684761224e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3904128,
      "step": 13945
    },
    {
      "epoch": 155.0,
      "grad_norm": 0.04591338336467743,
      "learning_rate": 3.643889505384117e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 3905552,
      "step": 13950
    },
    {
      "epoch": 155.05555555555554,
      "grad_norm": 0.0725151002407074,
      "learning_rate": 3.6430164658899744e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3906928,
      "step": 13955
    },
    {
      "epoch": 155.11111111111111,
      "grad_norm": 0.02902669832110405,
      "learning_rate": 3.642143250128329e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 3908256,
      "step": 13960
    },
    {
      "epoch": 155.16666666666666,
      "grad_norm": 0.003573743160814047,
      "learning_rate": 3.641269858233841e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3909648,
      "step": 13965
    },
    {
      "epoch": 155.22222222222223,
      "grad_norm": 0.03258763626217842,
      "learning_rate": 3.640396290341199e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3911088,
      "step": 13970
    },
    {
      "epoch": 155.27777777777777,
      "grad_norm": 0.012597373686730862,
      "learning_rate": 3.639522546585118e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3912528,
      "step": 13975
    },
    {
      "epoch": 155.33333333333334,
      "grad_norm": 2.7898354530334473,
      "learning_rate": 3.6386486271003404e-05,
      "loss": 0.0091,
      "num_input_tokens_seen": 3913952,
      "step": 13980
    },
    {
      "epoch": 155.38888888888889,
      "grad_norm": 0.3598289489746094,
      "learning_rate": 3.6377745320216346e-05,
      "loss": 0.0095,
      "num_input_tokens_seen": 3915408,
      "step": 13985
    },
    {
      "epoch": 155.44444444444446,
      "grad_norm": 0.26154765486717224,
      "learning_rate": 3.636900261483798e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3916768,
      "step": 13990
    },
    {
      "epoch": 155.5,
      "grad_norm": 0.014426599256694317,
      "learning_rate": 3.636025815621654e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3918160,
      "step": 13995
    },
    {
      "epoch": 155.55555555555554,
      "grad_norm": 0.04352566972374916,
      "learning_rate": 3.635151194570054e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3919584,
      "step": 14000
    },
    {
      "epoch": 155.55555555555554,
      "eval_loss": 0.4694768786430359,
      "eval_runtime": 1.2621,
      "eval_samples_per_second": 31.694,
      "eval_steps_per_second": 15.847,
      "num_input_tokens_seen": 3919584,
      "step": 14000
    },
    {
      "epoch": 155.61111111111111,
      "grad_norm": 0.023990098387002945,
      "learning_rate": 3.634276398463873e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 3920992,
      "step": 14005
    },
    {
      "epoch": 155.66666666666666,
      "grad_norm": 0.15114493668079376,
      "learning_rate": 3.633401427438018e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3922384,
      "step": 14010
    },
    {
      "epoch": 155.72222222222223,
      "grad_norm": 0.060055941343307495,
      "learning_rate": 3.63252628162742e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3923776,
      "step": 14015
    },
    {
      "epoch": 155.77777777777777,
      "grad_norm": 0.027745908126235008,
      "learning_rate": 3.6316509611670364e-05,
      "loss": 0.0066,
      "num_input_tokens_seen": 3925184,
      "step": 14020
    },
    {
      "epoch": 155.83333333333334,
      "grad_norm": 0.02426561340689659,
      "learning_rate": 3.630775466191854e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3926560,
      "step": 14025
    },
    {
      "epoch": 155.88888888888889,
      "grad_norm": 0.2323189377784729,
      "learning_rate": 3.629899796836884e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 3927936,
      "step": 14030
    },
    {
      "epoch": 155.94444444444446,
      "grad_norm": 0.0063311439007520676,
      "learning_rate": 3.6290239532371666e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3929344,
      "step": 14035
    },
    {
      "epoch": 156.0,
      "grad_norm": 0.005756983999162912,
      "learning_rate": 3.628147935527767e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 3930784,
      "step": 14040
    },
    {
      "epoch": 156.05555555555554,
      "grad_norm": 0.00262288935482502,
      "learning_rate": 3.627271743843779e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3932176,
      "step": 14045
    },
    {
      "epoch": 156.11111111111111,
      "grad_norm": 0.018075723201036453,
      "learning_rate": 3.626395378320321e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3933552,
      "step": 14050
    },
    {
      "epoch": 156.16666666666666,
      "grad_norm": 0.015178445726633072,
      "learning_rate": 3.625518839092541e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3934912,
      "step": 14055
    },
    {
      "epoch": 156.22222222222223,
      "grad_norm": 0.03470059484243393,
      "learning_rate": 3.624642126295612e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3936272,
      "step": 14060
    },
    {
      "epoch": 156.27777777777777,
      "grad_norm": 0.004635073244571686,
      "learning_rate": 3.6237652400647345e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 3937712,
      "step": 14065
    },
    {
      "epoch": 156.33333333333334,
      "grad_norm": 0.006357308477163315,
      "learning_rate": 3.622888180535134e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3939152,
      "step": 14070
    },
    {
      "epoch": 156.38888888888889,
      "grad_norm": 0.02285430207848549,
      "learning_rate": 3.6220109478420655e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3940544,
      "step": 14075
    },
    {
      "epoch": 156.44444444444446,
      "grad_norm": 0.003663771552965045,
      "learning_rate": 3.6211335421208084e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3941952,
      "step": 14080
    },
    {
      "epoch": 156.5,
      "grad_norm": 0.05446252226829529,
      "learning_rate": 3.62025596350667e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3943376,
      "step": 14085
    },
    {
      "epoch": 156.55555555555554,
      "grad_norm": 0.029580308124423027,
      "learning_rate": 3.619378212134984e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3944768,
      "step": 14090
    },
    {
      "epoch": 156.61111111111111,
      "grad_norm": 0.004782118834555149,
      "learning_rate": 3.618500288141111e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 3946160,
      "step": 14095
    },
    {
      "epoch": 156.66666666666666,
      "grad_norm": 0.0352398119866848,
      "learning_rate": 3.617622191660438e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3947552,
      "step": 14100
    },
    {
      "epoch": 156.72222222222223,
      "grad_norm": 0.00965357106178999,
      "learning_rate": 3.616743922828377e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3948992,
      "step": 14105
    },
    {
      "epoch": 156.77777777777777,
      "grad_norm": 0.020452309399843216,
      "learning_rate": 3.615865481780371e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 3950368,
      "step": 14110
    },
    {
      "epoch": 156.83333333333334,
      "grad_norm": 1.3256193399429321,
      "learning_rate": 3.614986868651883e-05,
      "loss": 0.0081,
      "num_input_tokens_seen": 3951824,
      "step": 14115
    },
    {
      "epoch": 156.88888888888889,
      "grad_norm": 0.03973802179098129,
      "learning_rate": 3.614108083578409e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 3953200,
      "step": 14120
    },
    {
      "epoch": 156.94444444444446,
      "grad_norm": 0.0025490177795290947,
      "learning_rate": 3.613229126695467e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 3954592,
      "step": 14125
    },
    {
      "epoch": 157.0,
      "grad_norm": 0.5749077796936035,
      "learning_rate": 3.612349998138605e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3956000,
      "step": 14130
    },
    {
      "epoch": 157.05555555555554,
      "grad_norm": 0.02295614406466484,
      "learning_rate": 3.6114706980433946e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 3957360,
      "step": 14135
    },
    {
      "epoch": 157.11111111111111,
      "grad_norm": 0.05804285407066345,
      "learning_rate": 3.610591226545435e-05,
      "loss": 0.0072,
      "num_input_tokens_seen": 3958736,
      "step": 14140
    },
    {
      "epoch": 157.16666666666666,
      "grad_norm": 0.25955066084861755,
      "learning_rate": 3.6097115837803505e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 3960112,
      "step": 14145
    },
    {
      "epoch": 157.22222222222223,
      "grad_norm": 0.022782927379012108,
      "learning_rate": 3.608831769883795e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3961536,
      "step": 14150
    },
    {
      "epoch": 157.27777777777777,
      "grad_norm": 0.02008640766143799,
      "learning_rate": 3.607951784991446e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3962944,
      "step": 14155
    },
    {
      "epoch": 157.33333333333334,
      "grad_norm": 0.03169367089867592,
      "learning_rate": 3.6070716292390085e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 3964368,
      "step": 14160
    },
    {
      "epoch": 157.38888888888889,
      "grad_norm": 0.01893976703286171,
      "learning_rate": 3.606191302762213e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 3965728,
      "step": 14165
    },
    {
      "epoch": 157.44444444444446,
      "grad_norm": 0.010337891057133675,
      "learning_rate": 3.605310805696818e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3967168,
      "step": 14170
    },
    {
      "epoch": 157.5,
      "grad_norm": 1.1412477493286133,
      "learning_rate": 3.6044301381786067e-05,
      "loss": 0.009,
      "num_input_tokens_seen": 3968528,
      "step": 14175
    },
    {
      "epoch": 157.55555555555554,
      "grad_norm": 0.12827514111995697,
      "learning_rate": 3.6035493003433883e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 3969952,
      "step": 14180
    },
    {
      "epoch": 157.61111111111111,
      "grad_norm": 0.06492944806814194,
      "learning_rate": 3.6026682923269994e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 3971328,
      "step": 14185
    },
    {
      "epoch": 157.66666666666666,
      "grad_norm": 0.013309750705957413,
      "learning_rate": 3.6017871142653034e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 3972752,
      "step": 14190
    },
    {
      "epoch": 157.72222222222223,
      "grad_norm": 0.4305080771446228,
      "learning_rate": 3.600905766294189e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 3974144,
      "step": 14195
    },
    {
      "epoch": 157.77777777777777,
      "grad_norm": 0.09606809169054031,
      "learning_rate": 3.60002424854957e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3975568,
      "step": 14200
    },
    {
      "epoch": 157.77777777777777,
      "eval_loss": 0.536629855632782,
      "eval_runtime": 1.243,
      "eval_samples_per_second": 32.181,
      "eval_steps_per_second": 16.09,
      "num_input_tokens_seen": 3975568,
      "step": 14200
    },
    {
      "epoch": 157.83333333333334,
      "grad_norm": 0.03427869454026222,
      "learning_rate": 3.5991425611673876e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3977008,
      "step": 14205
    },
    {
      "epoch": 157.88888888888889,
      "grad_norm": 0.07253678143024445,
      "learning_rate": 3.5982607042836105e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 3978432,
      "step": 14210
    },
    {
      "epoch": 157.94444444444446,
      "grad_norm": 0.05773350968956947,
      "learning_rate": 3.597378678034231e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3979840,
      "step": 14215
    },
    {
      "epoch": 158.0,
      "grad_norm": 0.008493156172335148,
      "learning_rate": 3.596496482555269e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3981216,
      "step": 14220
    },
    {
      "epoch": 158.05555555555554,
      "grad_norm": 0.005526612047106028,
      "learning_rate": 3.595614117982769e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3982608,
      "step": 14225
    },
    {
      "epoch": 158.11111111111111,
      "grad_norm": 0.02972235530614853,
      "learning_rate": 3.594731584452805e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3983968,
      "step": 14230
    },
    {
      "epoch": 158.16666666666666,
      "grad_norm": 0.007802159991115332,
      "learning_rate": 3.593848882101472e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 3985328,
      "step": 14235
    },
    {
      "epoch": 158.22222222222223,
      "grad_norm": 0.04942718893289566,
      "learning_rate": 3.592966011064896e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 3986720,
      "step": 14240
    },
    {
      "epoch": 158.27777777777777,
      "grad_norm": 0.0383930578827858,
      "learning_rate": 3.592082971479226e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 3988128,
      "step": 14245
    },
    {
      "epoch": 158.33333333333334,
      "grad_norm": 0.03170041739940643,
      "learning_rate": 3.5911997634806385e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 3989536,
      "step": 14250
    },
    {
      "epoch": 158.38888888888889,
      "grad_norm": 0.0006215259782038629,
      "learning_rate": 3.5903163872053336e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 3990944,
      "step": 14255
    },
    {
      "epoch": 158.44444444444446,
      "grad_norm": 0.02909398265182972,
      "learning_rate": 3.58943284278954e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 3992368,
      "step": 14260
    },
    {
      "epoch": 158.5,
      "grad_norm": 0.003403332317247987,
      "learning_rate": 3.588549130369512e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 3993760,
      "step": 14265
    },
    {
      "epoch": 158.55555555555554,
      "grad_norm": 0.024465564638376236,
      "learning_rate": 3.5876652500815274e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 3995152,
      "step": 14270
    },
    {
      "epoch": 158.61111111111111,
      "grad_norm": 0.028607988730072975,
      "learning_rate": 3.586781202061894e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 3996560,
      "step": 14275
    },
    {
      "epoch": 158.66666666666666,
      "grad_norm": 0.03256596252322197,
      "learning_rate": 3.585896986446942e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 3998016,
      "step": 14280
    },
    {
      "epoch": 158.72222222222223,
      "grad_norm": 0.011588110588490963,
      "learning_rate": 3.585012603373028e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 3999408,
      "step": 14285
    },
    {
      "epoch": 158.77777777777777,
      "grad_norm": 0.007070439867675304,
      "learning_rate": 3.584128052976535e-05,
      "loss": 0.0118,
      "num_input_tokens_seen": 4000832,
      "step": 14290
    },
    {
      "epoch": 158.83333333333334,
      "grad_norm": 0.1473616659641266,
      "learning_rate": 3.5832433353938724e-05,
      "loss": 0.0067,
      "num_input_tokens_seen": 4002224,
      "step": 14295
    },
    {
      "epoch": 158.88888888888889,
      "grad_norm": 0.009430550038814545,
      "learning_rate": 3.5823584507614746e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 4003632,
      "step": 14300
    },
    {
      "epoch": 158.94444444444446,
      "grad_norm": 0.20436224341392517,
      "learning_rate": 3.581473399215802e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4005056,
      "step": 14305
    },
    {
      "epoch": 159.0,
      "grad_norm": 0.08906110376119614,
      "learning_rate": 3.580588180893341e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 4006448,
      "step": 14310
    },
    {
      "epoch": 159.05555555555554,
      "grad_norm": 0.013160296715795994,
      "learning_rate": 3.579702795930602e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4007840,
      "step": 14315
    },
    {
      "epoch": 159.11111111111111,
      "grad_norm": 0.16472038626670837,
      "learning_rate": 3.578817244464125e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4009248,
      "step": 14320
    },
    {
      "epoch": 159.16666666666666,
      "grad_norm": 0.0049614207819104195,
      "learning_rate": 3.577931526630471e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4010640,
      "step": 14325
    },
    {
      "epoch": 159.22222222222223,
      "grad_norm": 8.206719398498535,
      "learning_rate": 3.577045642566229e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 4012032,
      "step": 14330
    },
    {
      "epoch": 159.27777777777777,
      "grad_norm": 0.004366980399936438,
      "learning_rate": 3.576159592408014e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4013456,
      "step": 14335
    },
    {
      "epoch": 159.33333333333334,
      "grad_norm": 0.01306853722780943,
      "learning_rate": 3.575273376292466e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 4014816,
      "step": 14340
    },
    {
      "epoch": 159.38888888888889,
      "grad_norm": 0.02931751310825348,
      "learning_rate": 3.574386994356251e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4016192,
      "step": 14345
    },
    {
      "epoch": 159.44444444444446,
      "grad_norm": 0.7582403421401978,
      "learning_rate": 3.573500446736059e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 4017600,
      "step": 14350
    },
    {
      "epoch": 159.5,
      "grad_norm": 0.04985349252820015,
      "learning_rate": 3.5726137335686094e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4019008,
      "step": 14355
    },
    {
      "epoch": 159.55555555555554,
      "grad_norm": 0.02240869402885437,
      "learning_rate": 3.571726854990642e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 4020416,
      "step": 14360
    },
    {
      "epoch": 159.61111111111111,
      "grad_norm": 0.018625544384121895,
      "learning_rate": 3.570839811138925e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4021808,
      "step": 14365
    },
    {
      "epoch": 159.66666666666666,
      "grad_norm": 1.1555827856063843,
      "learning_rate": 3.569952602150252e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4023216,
      "step": 14370
    },
    {
      "epoch": 159.72222222222223,
      "grad_norm": 0.15526333451271057,
      "learning_rate": 3.569065228161442e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4024608,
      "step": 14375
    },
    {
      "epoch": 159.77777777777777,
      "grad_norm": 0.0038185478188097477,
      "learning_rate": 3.5681776893093395e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4026032,
      "step": 14380
    },
    {
      "epoch": 159.83333333333334,
      "grad_norm": 0.058334752917289734,
      "learning_rate": 3.5672899857308134e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4027408,
      "step": 14385
    },
    {
      "epoch": 159.88888888888889,
      "grad_norm": 0.018868209794163704,
      "learning_rate": 3.566402117562759e-05,
      "loss": 0.0139,
      "num_input_tokens_seen": 4028800,
      "step": 14390
    },
    {
      "epoch": 159.94444444444446,
      "grad_norm": 0.01393749751150608,
      "learning_rate": 3.565514084942097e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 4030208,
      "step": 14395
    },
    {
      "epoch": 160.0,
      "grad_norm": 0.007478413172066212,
      "learning_rate": 3.564625888005773e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4031632,
      "step": 14400
    },
    {
      "epoch": 160.0,
      "eval_loss": 0.5121029615402222,
      "eval_runtime": 1.2372,
      "eval_samples_per_second": 32.33,
      "eval_steps_per_second": 16.165,
      "num_input_tokens_seen": 4031632,
      "step": 14400
    },
    {
      "epoch": 160.05555555555554,
      "grad_norm": 0.07247979193925858,
      "learning_rate": 3.563737526890759e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 4032992,
      "step": 14405
    },
    {
      "epoch": 160.11111111111111,
      "grad_norm": 0.00602383678779006,
      "learning_rate": 3.562849001734049e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 4034416,
      "step": 14410
    },
    {
      "epoch": 160.16666666666666,
      "grad_norm": 0.022235246375203133,
      "learning_rate": 3.561960312672667e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 4035808,
      "step": 14415
    },
    {
      "epoch": 160.22222222222223,
      "grad_norm": 0.05983844771981239,
      "learning_rate": 3.5610714598436596e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4037184,
      "step": 14420
    },
    {
      "epoch": 160.27777777777777,
      "grad_norm": 0.03927641734480858,
      "learning_rate": 3.5601824433840986e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4038608,
      "step": 14425
    },
    {
      "epoch": 160.33333333333334,
      "grad_norm": 0.044623877853155136,
      "learning_rate": 3.559293263431082e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4040016,
      "step": 14430
    },
    {
      "epoch": 160.38888888888889,
      "grad_norm": 0.009172419086098671,
      "learning_rate": 3.558403920121732e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4041408,
      "step": 14435
    },
    {
      "epoch": 160.44444444444446,
      "grad_norm": 0.004191087558865547,
      "learning_rate": 3.557514413593197e-05,
      "loss": 0.0075,
      "num_input_tokens_seen": 4042784,
      "step": 14440
    },
    {
      "epoch": 160.5,
      "grad_norm": 0.00969824381172657,
      "learning_rate": 3.55662474398265e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4044160,
      "step": 14445
    },
    {
      "epoch": 160.55555555555554,
      "grad_norm": 0.0016515403985977173,
      "learning_rate": 3.555734911427288e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4045520,
      "step": 14450
    },
    {
      "epoch": 160.61111111111111,
      "grad_norm": 0.006467317696660757,
      "learning_rate": 3.5548449160643363e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 4046928,
      "step": 14455
    },
    {
      "epoch": 160.66666666666666,
      "grad_norm": 0.02350197546184063,
      "learning_rate": 3.553954758031043e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4048304,
      "step": 14460
    },
    {
      "epoch": 160.72222222222223,
      "grad_norm": 0.1797129064798355,
      "learning_rate": 3.5530644374646815e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 4049728,
      "step": 14465
    },
    {
      "epoch": 160.77777777777777,
      "grad_norm": 0.05705767124891281,
      "learning_rate": 3.552173954502549e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4051136,
      "step": 14470
    },
    {
      "epoch": 160.83333333333334,
      "grad_norm": 0.04227885976433754,
      "learning_rate": 3.55128330928197e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4052544,
      "step": 14475
    },
    {
      "epoch": 160.88888888888889,
      "grad_norm": 1.814642310142517,
      "learning_rate": 3.550392501940294e-05,
      "loss": 0.0088,
      "num_input_tokens_seen": 4054000,
      "step": 14480
    },
    {
      "epoch": 160.94444444444446,
      "grad_norm": 0.05102952942252159,
      "learning_rate": 3.5495015326148945e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4055408,
      "step": 14485
    },
    {
      "epoch": 161.0,
      "grad_norm": 0.0455411858856678,
      "learning_rate": 3.548610401443169e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4056816,
      "step": 14490
    },
    {
      "epoch": 161.05555555555554,
      "grad_norm": 0.020429134368896484,
      "learning_rate": 3.547719108562543e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4058224,
      "step": 14495
    },
    {
      "epoch": 161.11111111111111,
      "grad_norm": 0.028893737122416496,
      "learning_rate": 3.546827654110464e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4059584,
      "step": 14500
    },
    {
      "epoch": 161.16666666666666,
      "grad_norm": 0.02053890749812126,
      "learning_rate": 3.545936038224405e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4061008,
      "step": 14505
    },
    {
      "epoch": 161.22222222222223,
      "grad_norm": 0.04258447512984276,
      "learning_rate": 3.545044261041864e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 4062448,
      "step": 14510
    },
    {
      "epoch": 161.27777777777777,
      "grad_norm": 0.09996345639228821,
      "learning_rate": 3.5441523227003657e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 4063824,
      "step": 14515
    },
    {
      "epoch": 161.33333333333334,
      "grad_norm": 0.14357030391693115,
      "learning_rate": 3.543260223337459e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 4065248,
      "step": 14520
    },
    {
      "epoch": 161.38888888888889,
      "grad_norm": 0.008412701077759266,
      "learning_rate": 3.542367963090714e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4066624,
      "step": 14525
    },
    {
      "epoch": 161.44444444444446,
      "grad_norm": 0.015014891512691975,
      "learning_rate": 3.5414755420977295e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 4068048,
      "step": 14530
    },
    {
      "epoch": 161.5,
      "grad_norm": 0.034227464348077774,
      "learning_rate": 3.54058296049613e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4069472,
      "step": 14535
    },
    {
      "epoch": 161.55555555555554,
      "grad_norm": 0.0543285608291626,
      "learning_rate": 3.53969021842356e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4070848,
      "step": 14540
    },
    {
      "epoch": 161.61111111111111,
      "grad_norm": 0.036200571805238724,
      "learning_rate": 3.5387973160176926e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4072256,
      "step": 14545
    },
    {
      "epoch": 161.66666666666666,
      "grad_norm": 0.004911370109766722,
      "learning_rate": 3.537904253416224e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 4073632,
      "step": 14550
    },
    {
      "epoch": 161.72222222222223,
      "grad_norm": 0.03068716451525688,
      "learning_rate": 3.537011030756878e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4075024,
      "step": 14555
    },
    {
      "epoch": 161.77777777777777,
      "grad_norm": 0.008704561740159988,
      "learning_rate": 3.536117648177399e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4076400,
      "step": 14560
    },
    {
      "epoch": 161.83333333333334,
      "grad_norm": 0.041607435792684555,
      "learning_rate": 3.535224105815558e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4077840,
      "step": 14565
    },
    {
      "epoch": 161.88888888888889,
      "grad_norm": 0.023338908329606056,
      "learning_rate": 3.5343304038091494e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 4079248,
      "step": 14570
    },
    {
      "epoch": 161.94444444444446,
      "grad_norm": 0.15485835075378418,
      "learning_rate": 3.5334365422959955e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4080656,
      "step": 14575
    },
    {
      "epoch": 162.0,
      "grad_norm": 0.01723659783601761,
      "learning_rate": 3.5325425214139396e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4082032,
      "step": 14580
    },
    {
      "epoch": 162.05555555555554,
      "grad_norm": 0.00750171672552824,
      "learning_rate": 3.531648341300851e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4083424,
      "step": 14585
    },
    {
      "epoch": 162.11111111111111,
      "grad_norm": 1.1276458501815796,
      "learning_rate": 3.530754002094623e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 4084768,
      "step": 14590
    },
    {
      "epoch": 162.16666666666666,
      "grad_norm": 0.008813825435936451,
      "learning_rate": 3.529859503933175e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4086160,
      "step": 14595
    },
    {
      "epoch": 162.22222222222223,
      "grad_norm": 0.024972913786768913,
      "learning_rate": 3.52896484695445e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4087632,
      "step": 14600
    },
    {
      "epoch": 162.22222222222223,
      "eval_loss": 0.5168201327323914,
      "eval_runtime": 1.235,
      "eval_samples_per_second": 32.389,
      "eval_steps_per_second": 16.195,
      "num_input_tokens_seen": 4087632,
      "step": 14600
    },
    {
      "epoch": 162.27777777777777,
      "grad_norm": 0.16264384984970093,
      "learning_rate": 3.528070031296414e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4089008,
      "step": 14605
    },
    {
      "epoch": 162.33333333333334,
      "grad_norm": 0.0021802657283842564,
      "learning_rate": 3.5271750570970605e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4090416,
      "step": 14610
    },
    {
      "epoch": 162.38888888888889,
      "grad_norm": 0.039909154176712036,
      "learning_rate": 3.526279924494405e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4091824,
      "step": 14615
    },
    {
      "epoch": 162.44444444444446,
      "grad_norm": 0.00858673732727766,
      "learning_rate": 3.5253846336264874e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4093248,
      "step": 14620
    },
    {
      "epoch": 162.5,
      "grad_norm": 0.11438267678022385,
      "learning_rate": 3.5244891846313736e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4094672,
      "step": 14625
    },
    {
      "epoch": 162.55555555555554,
      "grad_norm": 0.02060726098716259,
      "learning_rate": 3.5235935776471527e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 4096096,
      "step": 14630
    },
    {
      "epoch": 162.61111111111111,
      "grad_norm": 0.004686386324465275,
      "learning_rate": 3.522697812811939e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4097456,
      "step": 14635
    },
    {
      "epoch": 162.66666666666666,
      "grad_norm": 0.04077586904168129,
      "learning_rate": 3.521801890263871e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 4098848,
      "step": 14640
    },
    {
      "epoch": 162.72222222222223,
      "grad_norm": 0.12179247289896011,
      "learning_rate": 3.5209058101411114e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4100272,
      "step": 14645
    },
    {
      "epoch": 162.77777777777777,
      "grad_norm": 0.3454165756702423,
      "learning_rate": 3.520009572581845e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 4101696,
      "step": 14650
    },
    {
      "epoch": 162.83333333333334,
      "grad_norm": 0.17008021473884583,
      "learning_rate": 3.519113177724285e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4103120,
      "step": 14655
    },
    {
      "epoch": 162.88888888888889,
      "grad_norm": 0.3615027070045471,
      "learning_rate": 3.5182166257066656e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4104512,
      "step": 14660
    },
    {
      "epoch": 162.94444444444446,
      "grad_norm": 0.039361268281936646,
      "learning_rate": 3.517319916667247e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4105920,
      "step": 14665
    },
    {
      "epoch": 163.0,
      "grad_norm": 0.004255814477801323,
      "learning_rate": 3.516423050744313e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4107312,
      "step": 14670
    },
    {
      "epoch": 163.05555555555554,
      "grad_norm": 0.03052257001399994,
      "learning_rate": 3.5155260280761704e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4108768,
      "step": 14675
    },
    {
      "epoch": 163.11111111111111,
      "grad_norm": 0.0028242894914001226,
      "learning_rate": 3.514628848801154e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4110144,
      "step": 14680
    },
    {
      "epoch": 163.16666666666666,
      "grad_norm": 0.01739090494811535,
      "learning_rate": 3.5137315130576174e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4111520,
      "step": 14685
    },
    {
      "epoch": 163.22222222222223,
      "grad_norm": 0.10964835435152054,
      "learning_rate": 3.512834020983942e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4112944,
      "step": 14690
    },
    {
      "epoch": 163.27777777777777,
      "grad_norm": 0.17333123087882996,
      "learning_rate": 3.5119363727185334e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4114384,
      "step": 14695
    },
    {
      "epoch": 163.33333333333334,
      "grad_norm": 0.011323862709105015,
      "learning_rate": 3.511038568399819e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4115776,
      "step": 14700
    },
    {
      "epoch": 163.38888888888889,
      "grad_norm": 0.25830331444740295,
      "learning_rate": 3.510140608166251e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 4117152,
      "step": 14705
    },
    {
      "epoch": 163.44444444444446,
      "grad_norm": 0.0072546349838376045,
      "learning_rate": 3.509242492156308e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4118576,
      "step": 14710
    },
    {
      "epoch": 163.5,
      "grad_norm": 0.0021190610714256763,
      "learning_rate": 3.5083442205084896e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4119968,
      "step": 14715
    },
    {
      "epoch": 163.55555555555554,
      "grad_norm": 0.002696806564927101,
      "learning_rate": 3.507445793361321e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4121328,
      "step": 14720
    },
    {
      "epoch": 163.61111111111111,
      "grad_norm": 1.5685526132583618,
      "learning_rate": 3.5065472108533505e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4122736,
      "step": 14725
    },
    {
      "epoch": 163.66666666666666,
      "grad_norm": 0.04872581735253334,
      "learning_rate": 3.5056484731231504e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4124080,
      "step": 14730
    },
    {
      "epoch": 163.72222222222223,
      "grad_norm": 0.009730176068842411,
      "learning_rate": 3.504749580309319e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4125488,
      "step": 14735
    },
    {
      "epoch": 163.77777777777777,
      "grad_norm": 0.010382154025137424,
      "learning_rate": 3.5038505325504753e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4126896,
      "step": 14740
    },
    {
      "epoch": 163.83333333333334,
      "grad_norm": 0.02400226704776287,
      "learning_rate": 3.502951329985264e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4128304,
      "step": 14745
    },
    {
      "epoch": 163.88888888888889,
      "grad_norm": 0.04928573593497276,
      "learning_rate": 3.502051972752354e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4129680,
      "step": 14750
    },
    {
      "epoch": 163.94444444444446,
      "grad_norm": 0.11056260019540787,
      "learning_rate": 3.5011524609904374e-05,
      "loss": 0.0067,
      "num_input_tokens_seen": 4131088,
      "step": 14755
    },
    {
      "epoch": 164.0,
      "grad_norm": 0.002029515104368329,
      "learning_rate": 3.50025279483823e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 4132496,
      "step": 14760
    },
    {
      "epoch": 164.05555555555554,
      "grad_norm": 0.0030273261945694685,
      "learning_rate": 3.499352974434472e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4133872,
      "step": 14765
    },
    {
      "epoch": 164.11111111111111,
      "grad_norm": 0.017475226894021034,
      "learning_rate": 3.498452999917926e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 4135248,
      "step": 14770
    },
    {
      "epoch": 164.16666666666666,
      "grad_norm": 0.01482357457280159,
      "learning_rate": 3.4975528714273795e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4136608,
      "step": 14775
    },
    {
      "epoch": 164.22222222222223,
      "grad_norm": 0.026381798088550568,
      "learning_rate": 3.4966525891016454e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4138048,
      "step": 14780
    },
    {
      "epoch": 164.27777777777777,
      "grad_norm": 0.0029858718626201153,
      "learning_rate": 3.495752153079557e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4139456,
      "step": 14785
    },
    {
      "epoch": 164.33333333333334,
      "grad_norm": 0.016472429037094116,
      "learning_rate": 3.494851563499974e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4140864,
      "step": 14790
    },
    {
      "epoch": 164.38888888888889,
      "grad_norm": 0.06434419751167297,
      "learning_rate": 3.493950820501777e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4142272,
      "step": 14795
    },
    {
      "epoch": 164.44444444444446,
      "grad_norm": 0.01031546015292406,
      "learning_rate": 3.493049924223872e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 4143664,
      "step": 14800
    },
    {
      "epoch": 164.44444444444446,
      "eval_loss": 0.5454543828964233,
      "eval_runtime": 1.2378,
      "eval_samples_per_second": 32.316,
      "eval_steps_per_second": 16.158,
      "num_input_tokens_seen": 4143664,
      "step": 14800
    },
    {
      "epoch": 164.5,
      "grad_norm": 0.10884030908346176,
      "learning_rate": 3.49214887480519e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4145024,
      "step": 14805
    },
    {
      "epoch": 164.55555555555554,
      "grad_norm": 1.4120454788208008,
      "learning_rate": 3.4912476723846834e-05,
      "loss": 0.0076,
      "num_input_tokens_seen": 4146384,
      "step": 14810
    },
    {
      "epoch": 164.61111111111111,
      "grad_norm": 0.00915363896638155,
      "learning_rate": 3.490346317101328e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4147824,
      "step": 14815
    },
    {
      "epoch": 164.66666666666666,
      "grad_norm": 0.14268949627876282,
      "learning_rate": 3.4894448090941266e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4149184,
      "step": 14820
    },
    {
      "epoch": 164.72222222222223,
      "grad_norm": 0.002937538782134652,
      "learning_rate": 3.488543148502101e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4150592,
      "step": 14825
    },
    {
      "epoch": 164.77777777777777,
      "grad_norm": 0.018535098060965538,
      "learning_rate": 3.487641335464299e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4152000,
      "step": 14830
    },
    {
      "epoch": 164.83333333333334,
      "grad_norm": 0.01282710861414671,
      "learning_rate": 3.4867393701197914e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4153408,
      "step": 14835
    },
    {
      "epoch": 164.88888888888889,
      "grad_norm": 0.15475884079933167,
      "learning_rate": 3.485837252607673e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4154816,
      "step": 14840
    },
    {
      "epoch": 164.94444444444446,
      "grad_norm": 0.002750121522694826,
      "learning_rate": 3.4849349830670615e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4156208,
      "step": 14845
    },
    {
      "epoch": 165.0,
      "grad_norm": 0.09827806055545807,
      "learning_rate": 3.4840325616370976e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 4157600,
      "step": 14850
    },
    {
      "epoch": 165.05555555555554,
      "grad_norm": 0.05410649627447128,
      "learning_rate": 3.483129988456947e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4158992,
      "step": 14855
    },
    {
      "epoch": 165.11111111111111,
      "grad_norm": 0.08708824962377548,
      "learning_rate": 3.482227263665797e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4160432,
      "step": 14860
    },
    {
      "epoch": 165.16666666666666,
      "grad_norm": 0.02479914389550686,
      "learning_rate": 3.48132438740286e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4161856,
      "step": 14865
    },
    {
      "epoch": 165.22222222222223,
      "grad_norm": 0.04420514032244682,
      "learning_rate": 3.48042135980737e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4163232,
      "step": 14870
    },
    {
      "epoch": 165.27777777777777,
      "grad_norm": 0.01398772094398737,
      "learning_rate": 3.479518181018586e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4164608,
      "step": 14875
    },
    {
      "epoch": 165.33333333333334,
      "grad_norm": 0.0018219645135104656,
      "learning_rate": 3.4786148511757886e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4165984,
      "step": 14880
    },
    {
      "epoch": 165.38888888888889,
      "grad_norm": 0.0066678584553301334,
      "learning_rate": 3.477711370418284e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 4167360,
      "step": 14885
    },
    {
      "epoch": 165.44444444444446,
      "grad_norm": 0.060200035572052,
      "learning_rate": 3.476807738885399e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4168784,
      "step": 14890
    },
    {
      "epoch": 165.5,
      "grad_norm": 0.06473827362060547,
      "learning_rate": 3.475903956716485e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4170192,
      "step": 14895
    },
    {
      "epoch": 165.55555555555554,
      "grad_norm": 0.22188401222229004,
      "learning_rate": 3.475000024050917e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4171600,
      "step": 14900
    },
    {
      "epoch": 165.61111111111111,
      "grad_norm": 0.002108511747792363,
      "learning_rate": 3.4740959410280926e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4172992,
      "step": 14905
    },
    {
      "epoch": 165.66666666666666,
      "grad_norm": 0.010902821086347103,
      "learning_rate": 3.4731917077874324e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4174384,
      "step": 14910
    },
    {
      "epoch": 165.72222222222223,
      "grad_norm": 0.0028510906267911196,
      "learning_rate": 3.4722873244683816e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4175776,
      "step": 14915
    },
    {
      "epoch": 165.77777777777777,
      "grad_norm": 0.0020792572759091854,
      "learning_rate": 3.4713827912104065e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4177168,
      "step": 14920
    },
    {
      "epoch": 165.83333333333334,
      "grad_norm": 0.02285078354179859,
      "learning_rate": 3.470478108152998e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4178544,
      "step": 14925
    },
    {
      "epoch": 165.88888888888889,
      "grad_norm": 0.006456342060118914,
      "learning_rate": 3.4695732754356695e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4179984,
      "step": 14930
    },
    {
      "epoch": 165.94444444444446,
      "grad_norm": 0.0029749185778200626,
      "learning_rate": 3.4686682931979576e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 4181376,
      "step": 14935
    },
    {
      "epoch": 166.0,
      "grad_norm": 0.0669902116060257,
      "learning_rate": 3.467763161579422e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4182816,
      "step": 14940
    },
    {
      "epoch": 166.05555555555554,
      "grad_norm": 0.0180421881377697,
      "learning_rate": 3.466857880719645e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4184208,
      "step": 14945
    },
    {
      "epoch": 166.11111111111111,
      "grad_norm": 0.13029243052005768,
      "learning_rate": 3.465952450758233e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4185568,
      "step": 14950
    },
    {
      "epoch": 166.16666666666666,
      "grad_norm": 0.06894128024578094,
      "learning_rate": 3.4650468718348126e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4186944,
      "step": 14955
    },
    {
      "epoch": 166.22222222222223,
      "grad_norm": 0.0036209679674357176,
      "learning_rate": 3.464141144089038e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 4188336,
      "step": 14960
    },
    {
      "epoch": 166.27777777777777,
      "grad_norm": 0.0073242587968707085,
      "learning_rate": 3.463235267660583e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4189760,
      "step": 14965
    },
    {
      "epoch": 166.33333333333334,
      "grad_norm": 0.011653085239231586,
      "learning_rate": 3.462329242689145e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4191104,
      "step": 14970
    },
    {
      "epoch": 166.38888888888889,
      "grad_norm": 0.039942625910043716,
      "learning_rate": 3.461423069314444e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4192544,
      "step": 14975
    },
    {
      "epoch": 166.44444444444446,
      "grad_norm": 0.8097030520439148,
      "learning_rate": 3.460516747676224e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 4193952,
      "step": 14980
    },
    {
      "epoch": 166.5,
      "grad_norm": 0.004086535889655352,
      "learning_rate": 3.459610277914251e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4195360,
      "step": 14985
    },
    {
      "epoch": 166.55555555555554,
      "grad_norm": 0.0021369282621890306,
      "learning_rate": 3.458703660168314e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4196768,
      "step": 14990
    },
    {
      "epoch": 166.61111111111111,
      "grad_norm": 0.03218016400933266,
      "learning_rate": 3.457796894578224e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4198176,
      "step": 14995
    },
    {
      "epoch": 166.66666666666666,
      "grad_norm": 0.027310242876410484,
      "learning_rate": 3.456889981283817e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4199552,
      "step": 15000
    },
    {
      "epoch": 166.66666666666666,
      "eval_loss": 0.4630361497402191,
      "eval_runtime": 1.2427,
      "eval_samples_per_second": 32.187,
      "eval_steps_per_second": 16.094,
      "num_input_tokens_seen": 4199552,
      "step": 15000
    },
    {
      "epoch": 166.72222222222223,
      "grad_norm": 0.044003672897815704,
      "learning_rate": 3.45598292042495e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4200960,
      "step": 15005
    },
    {
      "epoch": 166.77777777777777,
      "grad_norm": 0.12453898042440414,
      "learning_rate": 3.4550757121415035e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4202400,
      "step": 15010
    },
    {
      "epoch": 166.83333333333334,
      "grad_norm": 0.8620405197143555,
      "learning_rate": 3.454168356573378e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 4203760,
      "step": 15015
    },
    {
      "epoch": 166.88888888888889,
      "grad_norm": 0.021931074559688568,
      "learning_rate": 3.453260853860503e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4205120,
      "step": 15020
    },
    {
      "epoch": 166.94444444444446,
      "grad_norm": 0.006938697770237923,
      "learning_rate": 3.452353204142824e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 4206496,
      "step": 15025
    },
    {
      "epoch": 167.0,
      "grad_norm": 0.0033863005228340626,
      "learning_rate": 3.4514454075603136e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4207936,
      "step": 15030
    },
    {
      "epoch": 167.05555555555554,
      "grad_norm": 0.29608508944511414,
      "learning_rate": 3.450537464252964e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4209312,
      "step": 15035
    },
    {
      "epoch": 167.11111111111111,
      "grad_norm": 0.013698254711925983,
      "learning_rate": 3.4496293743607925e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4210736,
      "step": 15040
    },
    {
      "epoch": 167.16666666666666,
      "grad_norm": 0.08747779577970505,
      "learning_rate": 3.448721138023838e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4212112,
      "step": 15045
    },
    {
      "epoch": 167.22222222222223,
      "grad_norm": 0.14547431468963623,
      "learning_rate": 3.447812755382162e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 4213488,
      "step": 15050
    },
    {
      "epoch": 167.27777777777777,
      "grad_norm": 0.014747737906873226,
      "learning_rate": 3.446904226575847e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4214880,
      "step": 15055
    },
    {
      "epoch": 167.33333333333334,
      "grad_norm": 0.0012970053358003497,
      "learning_rate": 3.445995551745002e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4216288,
      "step": 15060
    },
    {
      "epoch": 167.38888888888889,
      "grad_norm": 0.3043172061443329,
      "learning_rate": 3.445086731029753e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 4217696,
      "step": 15065
    },
    {
      "epoch": 167.44444444444446,
      "grad_norm": 0.04233643785119057,
      "learning_rate": 3.444177764570255e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4219056,
      "step": 15070
    },
    {
      "epoch": 167.5,
      "grad_norm": 0.0018331683240830898,
      "learning_rate": 3.44326865250668e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4220496,
      "step": 15075
    },
    {
      "epoch": 167.55555555555554,
      "grad_norm": 0.018627982586622238,
      "learning_rate": 3.442359394979225e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4221888,
      "step": 15080
    },
    {
      "epoch": 167.61111111111111,
      "grad_norm": 0.00192809640429914,
      "learning_rate": 3.441449992128108e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4223296,
      "step": 15085
    },
    {
      "epoch": 167.66666666666666,
      "grad_norm": 0.004122207406908274,
      "learning_rate": 3.440540444093573e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4224736,
      "step": 15090
    },
    {
      "epoch": 167.72222222222223,
      "grad_norm": 0.006841904018074274,
      "learning_rate": 3.43963075101588e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4226112,
      "step": 15095
    },
    {
      "epoch": 167.77777777777777,
      "grad_norm": 9.721311569213867,
      "learning_rate": 3.438720913035318e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 4227520,
      "step": 15100
    },
    {
      "epoch": 167.83333333333334,
      "grad_norm": 0.015612097457051277,
      "learning_rate": 3.437810930292195e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4228880,
      "step": 15105
    },
    {
      "epoch": 167.88888888888889,
      "grad_norm": 0.008329277858138084,
      "learning_rate": 3.43690080292684e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4230320,
      "step": 15110
    },
    {
      "epoch": 167.94444444444446,
      "grad_norm": 0.02076120115816593,
      "learning_rate": 3.435990531079608e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 4231728,
      "step": 15115
    },
    {
      "epoch": 168.0,
      "grad_norm": 0.01042510848492384,
      "learning_rate": 3.435080114890874e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4233152,
      "step": 15120
    },
    {
      "epoch": 168.05555555555554,
      "grad_norm": 0.03506092727184296,
      "learning_rate": 3.434169554501035e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4234544,
      "step": 15125
    },
    {
      "epoch": 168.11111111111111,
      "grad_norm": 0.47727417945861816,
      "learning_rate": 3.433258850050511e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 4235904,
      "step": 15130
    },
    {
      "epoch": 168.16666666666666,
      "grad_norm": 0.0028540072962641716,
      "learning_rate": 3.4323480016797446e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4237280,
      "step": 15135
    },
    {
      "epoch": 168.22222222222223,
      "grad_norm": 0.00136491097509861,
      "learning_rate": 3.4314370095291995e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4238688,
      "step": 15140
    },
    {
      "epoch": 168.27777777777777,
      "grad_norm": 0.3344740569591522,
      "learning_rate": 3.430525873739363e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4240112,
      "step": 15145
    },
    {
      "epoch": 168.33333333333334,
      "grad_norm": 0.1701337844133377,
      "learning_rate": 3.429614594450743e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4241520,
      "step": 15150
    },
    {
      "epoch": 168.38888888888889,
      "grad_norm": 0.006634527817368507,
      "learning_rate": 3.428703171803869e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4242912,
      "step": 15155
    },
    {
      "epoch": 168.44444444444446,
      "grad_norm": 0.003324233926832676,
      "learning_rate": 3.4277916059392964e-05,
      "loss": 0.0042,
      "num_input_tokens_seen": 4244352,
      "step": 15160
    },
    {
      "epoch": 168.5,
      "grad_norm": 0.06705982983112335,
      "learning_rate": 3.426879896997598e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4245744,
      "step": 15165
    },
    {
      "epoch": 168.55555555555554,
      "grad_norm": 0.0011952804634347558,
      "learning_rate": 3.425968045119372e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4247168,
      "step": 15170
    },
    {
      "epoch": 168.61111111111111,
      "grad_norm": 0.01772019825875759,
      "learning_rate": 3.425056050445237e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4248560,
      "step": 15175
    },
    {
      "epoch": 168.66666666666666,
      "grad_norm": 0.03590679541230202,
      "learning_rate": 3.4241439131158336e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4249968,
      "step": 15180
    },
    {
      "epoch": 168.72222222222223,
      "grad_norm": 0.0016176103381440043,
      "learning_rate": 3.423231633271825e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4251392,
      "step": 15185
    },
    {
      "epoch": 168.77777777777777,
      "grad_norm": 0.008847860619425774,
      "learning_rate": 3.4223192110538985e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4252736,
      "step": 15190
    },
    {
      "epoch": 168.83333333333334,
      "grad_norm": 0.25529250502586365,
      "learning_rate": 3.4214066466027575e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4254144,
      "step": 15195
    },
    {
      "epoch": 168.88888888888889,
      "grad_norm": 0.23003950715065002,
      "learning_rate": 3.4204939400591325e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4255584,
      "step": 15200
    },
    {
      "epoch": 168.88888888888889,
      "eval_loss": 0.5377890467643738,
      "eval_runtime": 1.2374,
      "eval_samples_per_second": 32.325,
      "eval_steps_per_second": 16.163,
      "num_input_tokens_seen": 4255584,
      "step": 15200
    },
    {
      "epoch": 168.94444444444446,
      "grad_norm": 0.213808074593544,
      "learning_rate": 3.419581091563775e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4256992,
      "step": 15205
    },
    {
      "epoch": 169.0,
      "grad_norm": 0.10129161924123764,
      "learning_rate": 3.418668101257456e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 4258384,
      "step": 15210
    },
    {
      "epoch": 169.05555555555554,
      "grad_norm": 0.003162229200825095,
      "learning_rate": 3.417754969280971e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4259808,
      "step": 15215
    },
    {
      "epoch": 169.11111111111111,
      "grad_norm": 0.008069261908531189,
      "learning_rate": 3.416841695775137e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4261216,
      "step": 15220
    },
    {
      "epoch": 169.16666666666666,
      "grad_norm": 0.02466074749827385,
      "learning_rate": 3.415928280880792e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4262608,
      "step": 15225
    },
    {
      "epoch": 169.22222222222223,
      "grad_norm": 0.028475245460867882,
      "learning_rate": 3.4150147247387965e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4264032,
      "step": 15230
    },
    {
      "epoch": 169.27777777777777,
      "grad_norm": 0.09464011341333389,
      "learning_rate": 3.4141010274900306e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4265424,
      "step": 15235
    },
    {
      "epoch": 169.33333333333334,
      "grad_norm": 0.007109987549483776,
      "learning_rate": 3.413187189275399e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4266880,
      "step": 15240
    },
    {
      "epoch": 169.38888888888889,
      "grad_norm": 0.01951112039387226,
      "learning_rate": 3.4122732102358265e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4268224,
      "step": 15245
    },
    {
      "epoch": 169.44444444444446,
      "grad_norm": 0.0038740800227969885,
      "learning_rate": 3.411359090512261e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4269600,
      "step": 15250
    },
    {
      "epoch": 169.5,
      "grad_norm": 0.011702184565365314,
      "learning_rate": 3.410444830245672e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4270992,
      "step": 15255
    },
    {
      "epoch": 169.55555555555554,
      "grad_norm": 0.08561573922634125,
      "learning_rate": 3.409530429577048e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4272368,
      "step": 15260
    },
    {
      "epoch": 169.61111111111111,
      "grad_norm": 0.16094525158405304,
      "learning_rate": 3.408615888647402e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 4273792,
      "step": 15265
    },
    {
      "epoch": 169.66666666666666,
      "grad_norm": 0.03022248111665249,
      "learning_rate": 3.4077012075977675e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4275200,
      "step": 15270
    },
    {
      "epoch": 169.72222222222223,
      "grad_norm": 0.027959715574979782,
      "learning_rate": 3.4067863865692e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4276592,
      "step": 15275
    },
    {
      "epoch": 169.77777777777777,
      "grad_norm": 0.05242697522044182,
      "learning_rate": 3.4058714257027755e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4277952,
      "step": 15280
    },
    {
      "epoch": 169.83333333333334,
      "grad_norm": 0.01434215810149908,
      "learning_rate": 3.404956325139594e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 4279360,
      "step": 15285
    },
    {
      "epoch": 169.88888888888889,
      "grad_norm": 0.1210647001862526,
      "learning_rate": 3.404041085020775e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4280752,
      "step": 15290
    },
    {
      "epoch": 169.94444444444446,
      "grad_norm": 0.1386735886335373,
      "learning_rate": 3.403125705487459e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4282160,
      "step": 15295
    },
    {
      "epoch": 170.0,
      "grad_norm": 0.002693566959351301,
      "learning_rate": 3.402210186680811e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4283584,
      "step": 15300
    },
    {
      "epoch": 170.05555555555554,
      "grad_norm": 0.002123862272128463,
      "learning_rate": 3.4012945287420137e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4284912,
      "step": 15305
    },
    {
      "epoch": 170.11111111111111,
      "grad_norm": 0.38868448138237,
      "learning_rate": 3.400378731812274e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 4286304,
      "step": 15310
    },
    {
      "epoch": 170.16666666666666,
      "grad_norm": 0.005643113981932402,
      "learning_rate": 3.399462796032817e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4287696,
      "step": 15315
    },
    {
      "epoch": 170.22222222222223,
      "grad_norm": 0.007987572811543941,
      "learning_rate": 3.3985467215448954e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4289104,
      "step": 15320
    },
    {
      "epoch": 170.27777777777777,
      "grad_norm": 0.8315901160240173,
      "learning_rate": 3.3976305084897776e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 4290496,
      "step": 15325
    },
    {
      "epoch": 170.33333333333334,
      "grad_norm": 0.03132127970457077,
      "learning_rate": 3.3967141570087544e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4291872,
      "step": 15330
    },
    {
      "epoch": 170.38888888888889,
      "grad_norm": 0.005460117477923632,
      "learning_rate": 3.39579766724314e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 4293248,
      "step": 15335
    },
    {
      "epoch": 170.44444444444446,
      "grad_norm": 0.0035228112246841192,
      "learning_rate": 3.3948810393342677e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4294640,
      "step": 15340
    },
    {
      "epoch": 170.5,
      "grad_norm": 0.04443833604454994,
      "learning_rate": 3.3939642734234936e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4296064,
      "step": 15345
    },
    {
      "epoch": 170.55555555555554,
      "grad_norm": 0.08055151253938675,
      "learning_rate": 3.393047369652194e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 4297472,
      "step": 15350
    },
    {
      "epoch": 170.61111111111111,
      "grad_norm": 0.05391110107302666,
      "learning_rate": 3.3921303281617664e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4298912,
      "step": 15355
    },
    {
      "epoch": 170.66666666666666,
      "grad_norm": 0.0029651953373104334,
      "learning_rate": 3.391213149093632e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4300320,
      "step": 15360
    },
    {
      "epoch": 170.72222222222223,
      "grad_norm": 0.15137478709220886,
      "learning_rate": 3.3902958325892303e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4301712,
      "step": 15365
    },
    {
      "epoch": 170.77777777777777,
      "grad_norm": 0.00789796095341444,
      "learning_rate": 3.389378378790023e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4303136,
      "step": 15370
    },
    {
      "epoch": 170.83333333333334,
      "grad_norm": 0.0015411929925903678,
      "learning_rate": 3.388460787837493e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4304512,
      "step": 15375
    },
    {
      "epoch": 170.88888888888889,
      "grad_norm": 0.01442545186728239,
      "learning_rate": 3.387543059873145e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4305920,
      "step": 15380
    },
    {
      "epoch": 170.94444444444446,
      "grad_norm": 0.01057231891900301,
      "learning_rate": 3.386625195038503e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4307312,
      "step": 15385
    },
    {
      "epoch": 171.0,
      "grad_norm": 0.0012991969706490636,
      "learning_rate": 3.3857071934751136e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4308752,
      "step": 15390
    },
    {
      "epoch": 171.05555555555554,
      "grad_norm": 0.025696100667119026,
      "learning_rate": 3.384789055324544e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 4310144,
      "step": 15395
    },
    {
      "epoch": 171.11111111111111,
      "grad_norm": 0.002690017456188798,
      "learning_rate": 3.3838707807283843e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4311504,
      "step": 15400
    },
    {
      "epoch": 171.11111111111111,
      "eval_loss": 0.5204831957817078,
      "eval_runtime": 1.2355,
      "eval_samples_per_second": 32.374,
      "eval_steps_per_second": 16.187,
      "num_input_tokens_seen": 4311504,
      "step": 15400
    },
    {
      "epoch": 171.16666666666666,
      "grad_norm": 0.0015035687247291207,
      "learning_rate": 3.382952369828243e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4312880,
      "step": 15405
    },
    {
      "epoch": 171.22222222222223,
      "grad_norm": 0.02048053964972496,
      "learning_rate": 3.38203382276575e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4314272,
      "step": 15410
    },
    {
      "epoch": 171.27777777777777,
      "grad_norm": 0.06112082302570343,
      "learning_rate": 3.381115139682557e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4315632,
      "step": 15415
    },
    {
      "epoch": 171.33333333333334,
      "grad_norm": 0.003632856532931328,
      "learning_rate": 3.3801963207203366e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4317008,
      "step": 15420
    },
    {
      "epoch": 171.38888888888889,
      "grad_norm": 1.2286453247070312,
      "learning_rate": 3.379277366020782e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4318448,
      "step": 15425
    },
    {
      "epoch": 171.44444444444446,
      "grad_norm": 0.0023725368082523346,
      "learning_rate": 3.3783582757256085e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4319872,
      "step": 15430
    },
    {
      "epoch": 171.5,
      "grad_norm": 0.013287794776260853,
      "learning_rate": 3.3774390499765504e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4321264,
      "step": 15435
    },
    {
      "epoch": 171.55555555555554,
      "grad_norm": 0.002244430361315608,
      "learning_rate": 3.376519688915364e-05,
      "loss": 0.0119,
      "num_input_tokens_seen": 4322656,
      "step": 15440
    },
    {
      "epoch": 171.61111111111111,
      "grad_norm": 0.1389235407114029,
      "learning_rate": 3.3756001926838273e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4324048,
      "step": 15445
    },
    {
      "epoch": 171.66666666666666,
      "grad_norm": 0.01373171154409647,
      "learning_rate": 3.374680561423737e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 4325424,
      "step": 15450
    },
    {
      "epoch": 171.72222222222223,
      "grad_norm": 1.0931564569473267,
      "learning_rate": 3.373760795276912e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4326848,
      "step": 15455
    },
    {
      "epoch": 171.77777777777777,
      "grad_norm": 0.11745523661375046,
      "learning_rate": 3.372840894385192e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 4328240,
      "step": 15460
    },
    {
      "epoch": 171.83333333333334,
      "grad_norm": 0.004665685351938009,
      "learning_rate": 3.3719208588904375e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4329616,
      "step": 15465
    },
    {
      "epoch": 171.88888888888889,
      "grad_norm": 0.0130905257537961,
      "learning_rate": 3.371000688934529e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4330992,
      "step": 15470
    },
    {
      "epoch": 171.94444444444446,
      "grad_norm": 0.014261077158153057,
      "learning_rate": 3.370080384659369e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4332400,
      "step": 15475
    },
    {
      "epoch": 172.0,
      "grad_norm": 0.0035059605725109577,
      "learning_rate": 3.36915994620688e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4333808,
      "step": 15480
    },
    {
      "epoch": 172.05555555555554,
      "grad_norm": 0.05549617111682892,
      "learning_rate": 3.3682393737190035e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4335216,
      "step": 15485
    },
    {
      "epoch": 172.11111111111111,
      "grad_norm": 0.02055756188929081,
      "learning_rate": 3.3673186673377054e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4336624,
      "step": 15490
    },
    {
      "epoch": 172.16666666666666,
      "grad_norm": 0.009399290196597576,
      "learning_rate": 3.366397827204969e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4338000,
      "step": 15495
    },
    {
      "epoch": 172.22222222222223,
      "grad_norm": 0.00970522128045559,
      "learning_rate": 3.3654768534628e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4339408,
      "step": 15500
    },
    {
      "epoch": 172.27777777777777,
      "grad_norm": 0.0011042143451049924,
      "learning_rate": 3.3645557462532245e-05,
      "loss": 0.0074,
      "num_input_tokens_seen": 4340864,
      "step": 15505
    },
    {
      "epoch": 172.33333333333334,
      "grad_norm": 0.013197608292102814,
      "learning_rate": 3.363634505718288e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4342272,
      "step": 15510
    },
    {
      "epoch": 172.38888888888889,
      "grad_norm": 0.014432283118367195,
      "learning_rate": 3.362713132000057e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4343680,
      "step": 15515
    },
    {
      "epoch": 172.44444444444446,
      "grad_norm": 0.007230627350509167,
      "learning_rate": 3.36179162524062e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4345104,
      "step": 15520
    },
    {
      "epoch": 172.5,
      "grad_norm": 0.024829115718603134,
      "learning_rate": 3.3608699855820846e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4346496,
      "step": 15525
    },
    {
      "epoch": 172.55555555555554,
      "grad_norm": 0.02822212129831314,
      "learning_rate": 3.359948213166578e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 4347872,
      "step": 15530
    },
    {
      "epoch": 172.61111111111111,
      "grad_norm": 0.022705992683768272,
      "learning_rate": 3.359026308136252e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4349248,
      "step": 15535
    },
    {
      "epoch": 172.66666666666666,
      "grad_norm": 0.018537160009145737,
      "learning_rate": 3.358104270633272e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4350656,
      "step": 15540
    },
    {
      "epoch": 172.72222222222223,
      "grad_norm": 0.004157602321356535,
      "learning_rate": 3.357182100799831e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4352032,
      "step": 15545
    },
    {
      "epoch": 172.77777777777777,
      "grad_norm": 0.0037482324987649918,
      "learning_rate": 3.3562597987781384e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4353456,
      "step": 15550
    },
    {
      "epoch": 172.83333333333334,
      "grad_norm": 0.008850251324474812,
      "learning_rate": 3.355337364710424e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 4354800,
      "step": 15555
    },
    {
      "epoch": 172.88888888888889,
      "grad_norm": 0.016971345990896225,
      "learning_rate": 3.354414798738939e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4356160,
      "step": 15560
    },
    {
      "epoch": 172.94444444444446,
      "grad_norm": 0.0020599698182195425,
      "learning_rate": 3.353492101005955e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4357616,
      "step": 15565
    },
    {
      "epoch": 173.0,
      "grad_norm": 0.012135219760239124,
      "learning_rate": 3.352569271653763e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4358992,
      "step": 15570
    },
    {
      "epoch": 173.05555555555554,
      "grad_norm": 0.01937481388449669,
      "learning_rate": 3.351646310824675e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4360464,
      "step": 15575
    },
    {
      "epoch": 173.11111111111111,
      "grad_norm": 0.039321910589933395,
      "learning_rate": 3.350723218661023e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 4361856,
      "step": 15580
    },
    {
      "epoch": 173.16666666666666,
      "grad_norm": 0.03312881290912628,
      "learning_rate": 3.349799995305162e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4363248,
      "step": 15585
    },
    {
      "epoch": 173.22222222222223,
      "grad_norm": 0.015230773948132992,
      "learning_rate": 3.348876640899461e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4364624,
      "step": 15590
    },
    {
      "epoch": 173.27777777777777,
      "grad_norm": 0.00984795019030571,
      "learning_rate": 3.3479531555863144e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4366048,
      "step": 15595
    },
    {
      "epoch": 173.33333333333334,
      "grad_norm": 0.03465268760919571,
      "learning_rate": 3.3470295395081344e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4367408,
      "step": 15600
    },
    {
      "epoch": 173.33333333333334,
      "eval_loss": 0.5626125335693359,
      "eval_runtime": 1.2369,
      "eval_samples_per_second": 32.339,
      "eval_steps_per_second": 16.169,
      "num_input_tokens_seen": 4367408,
      "step": 15600
    },
    {
      "epoch": 173.38888888888889,
      "grad_norm": 0.019332734867930412,
      "learning_rate": 3.3461057928073556e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4368832,
      "step": 15605
    },
    {
      "epoch": 173.44444444444446,
      "grad_norm": 0.06605934351682663,
      "learning_rate": 3.345181915626431e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4370224,
      "step": 15610
    },
    {
      "epoch": 173.5,
      "grad_norm": 0.11181621998548508,
      "learning_rate": 3.344257908107834e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4371648,
      "step": 15615
    },
    {
      "epoch": 173.55555555555554,
      "grad_norm": 0.0074147447012364864,
      "learning_rate": 3.343333770394058e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4373008,
      "step": 15620
    },
    {
      "epoch": 173.61111111111111,
      "grad_norm": 0.2319626808166504,
      "learning_rate": 3.342409502627616e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4374400,
      "step": 15625
    },
    {
      "epoch": 173.66666666666666,
      "grad_norm": 0.025903290137648582,
      "learning_rate": 3.341485104951043e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4375808,
      "step": 15630
    },
    {
      "epoch": 173.72222222222223,
      "grad_norm": 0.02390426956117153,
      "learning_rate": 3.340560577506892e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 4377200,
      "step": 15635
    },
    {
      "epoch": 173.77777777777777,
      "grad_norm": 0.00611894391477108,
      "learning_rate": 3.339635920437735e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4378560,
      "step": 15640
    },
    {
      "epoch": 173.83333333333334,
      "grad_norm": 0.0843362808227539,
      "learning_rate": 3.338711133886169e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4379936,
      "step": 15645
    },
    {
      "epoch": 173.88888888888889,
      "grad_norm": 0.01031921710819006,
      "learning_rate": 3.3377862179948064e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4381360,
      "step": 15650
    },
    {
      "epoch": 173.94444444444446,
      "grad_norm": 0.05165741965174675,
      "learning_rate": 3.336861172906281e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4382784,
      "step": 15655
    },
    {
      "epoch": 174.0,
      "grad_norm": 0.0012710756855085492,
      "learning_rate": 3.335935998763245e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4384176,
      "step": 15660
    },
    {
      "epoch": 174.05555555555554,
      "grad_norm": 0.47599396109580994,
      "learning_rate": 3.3350106957083744e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 4385536,
      "step": 15665
    },
    {
      "epoch": 174.11111111111111,
      "grad_norm": 0.07451626658439636,
      "learning_rate": 3.33408526388436e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4386928,
      "step": 15670
    },
    {
      "epoch": 174.16666666666666,
      "grad_norm": 0.03522241860628128,
      "learning_rate": 3.3331597034339166e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4388352,
      "step": 15675
    },
    {
      "epoch": 174.22222222222223,
      "grad_norm": 0.23807889223098755,
      "learning_rate": 3.3322340144997764e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4389760,
      "step": 15680
    },
    {
      "epoch": 174.27777777777777,
      "grad_norm": 0.013914369978010654,
      "learning_rate": 3.331308197224693e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4391136,
      "step": 15685
    },
    {
      "epoch": 174.33333333333334,
      "grad_norm": 0.12148417532444,
      "learning_rate": 3.330382251751438e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 4392528,
      "step": 15690
    },
    {
      "epoch": 174.38888888888889,
      "grad_norm": 0.03268413245677948,
      "learning_rate": 3.3294561782228054e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4393856,
      "step": 15695
    },
    {
      "epoch": 174.44444444444446,
      "grad_norm": 0.0027353924233466387,
      "learning_rate": 3.328529976781607e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4395232,
      "step": 15700
    },
    {
      "epoch": 174.5,
      "grad_norm": 0.002952778013423085,
      "learning_rate": 3.327603647570673e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4396688,
      "step": 15705
    },
    {
      "epoch": 174.55555555555554,
      "grad_norm": 0.08308279514312744,
      "learning_rate": 3.326677190732857e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4398096,
      "step": 15710
    },
    {
      "epoch": 174.61111111111111,
      "grad_norm": 0.014896481297910213,
      "learning_rate": 3.325750606411029e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4399520,
      "step": 15715
    },
    {
      "epoch": 174.66666666666666,
      "grad_norm": 0.019955888390541077,
      "learning_rate": 3.3248238947480804e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4400912,
      "step": 15720
    },
    {
      "epoch": 174.72222222222223,
      "grad_norm": 0.0032533928751945496,
      "learning_rate": 3.323897055886922e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4402336,
      "step": 15725
    },
    {
      "epoch": 174.77777777777777,
      "grad_norm": 0.08415235579013824,
      "learning_rate": 3.322970089970484e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4403712,
      "step": 15730
    },
    {
      "epoch": 174.83333333333334,
      "grad_norm": 0.006826568860560656,
      "learning_rate": 3.3220429971417165e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4405152,
      "step": 15735
    },
    {
      "epoch": 174.88888888888889,
      "grad_norm": 0.008493649773299694,
      "learning_rate": 3.321115777543588e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4406544,
      "step": 15740
    },
    {
      "epoch": 174.94444444444446,
      "grad_norm": 0.001521554309874773,
      "learning_rate": 3.320188431319088e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4408016,
      "step": 15745
    },
    {
      "epoch": 175.0,
      "grad_norm": 0.15707331895828247,
      "learning_rate": 3.319260958611224e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4409408,
      "step": 15750
    },
    {
      "epoch": 175.05555555555554,
      "grad_norm": 0.025018243119120598,
      "learning_rate": 3.3183333595630256e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4410848,
      "step": 15755
    },
    {
      "epoch": 175.11111111111111,
      "grad_norm": 0.015240645036101341,
      "learning_rate": 3.317405634317538e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4412272,
      "step": 15760
    },
    {
      "epoch": 175.16666666666666,
      "grad_norm": 0.4219651222229004,
      "learning_rate": 3.3164777830178315e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4413632,
      "step": 15765
    },
    {
      "epoch": 175.22222222222223,
      "grad_norm": 0.011653799563646317,
      "learning_rate": 3.315549805806989e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4415008,
      "step": 15770
    },
    {
      "epoch": 175.27777777777777,
      "grad_norm": 0.013424966484308243,
      "learning_rate": 3.314621702828118e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4416400,
      "step": 15775
    },
    {
      "epoch": 175.33333333333334,
      "grad_norm": 0.13728222250938416,
      "learning_rate": 3.313693474224342e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4417808,
      "step": 15780
    },
    {
      "epoch": 175.38888888888889,
      "grad_norm": 0.02594475820660591,
      "learning_rate": 3.312765120138809e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4419184,
      "step": 15785
    },
    {
      "epoch": 175.44444444444446,
      "grad_norm": 0.009449992328882217,
      "learning_rate": 3.311836640714679e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4420608,
      "step": 15790
    },
    {
      "epoch": 175.5,
      "grad_norm": 0.008150284178555012,
      "learning_rate": 3.310908036095137e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4422000,
      "step": 15795
    },
    {
      "epoch": 175.55555555555554,
      "grad_norm": 0.013020382262766361,
      "learning_rate": 3.309979306423386e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 4423376,
      "step": 15800
    },
    {
      "epoch": 175.55555555555554,
      "eval_loss": 0.6110655665397644,
      "eval_runtime": 1.2476,
      "eval_samples_per_second": 32.063,
      "eval_steps_per_second": 16.031,
      "num_input_tokens_seen": 4423376,
      "step": 15800
    },
    {
      "epoch": 175.61111111111111,
      "grad_norm": 0.0007762478198856115,
      "learning_rate": 3.309050451842647e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4424800,
      "step": 15805
    },
    {
      "epoch": 175.66666666666666,
      "grad_norm": 0.0013814406702294946,
      "learning_rate": 3.3081214724961604e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 4426208,
      "step": 15810
    },
    {
      "epoch": 175.72222222222223,
      "grad_norm": 0.0018558356678113341,
      "learning_rate": 3.307192368527188e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4427600,
      "step": 15815
    },
    {
      "epoch": 175.77777777777777,
      "grad_norm": 0.09168000519275665,
      "learning_rate": 3.306263140079008e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 4428976,
      "step": 15820
    },
    {
      "epoch": 175.83333333333334,
      "grad_norm": 0.0015328198205679655,
      "learning_rate": 3.30533378729492e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4430368,
      "step": 15825
    },
    {
      "epoch": 175.88888888888889,
      "grad_norm": 0.020845789462327957,
      "learning_rate": 3.304404310318242e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4431792,
      "step": 15830
    },
    {
      "epoch": 175.94444444444446,
      "grad_norm": 0.013882716186344624,
      "learning_rate": 3.3034747092923105e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4433168,
      "step": 15835
    },
    {
      "epoch": 176.0,
      "grad_norm": 0.07026425004005432,
      "learning_rate": 3.3025449843604806e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4434592,
      "step": 15840
    },
    {
      "epoch": 176.05555555555554,
      "grad_norm": 0.04296353831887245,
      "learning_rate": 3.30161513566613e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4435984,
      "step": 15845
    },
    {
      "epoch": 176.11111111111111,
      "grad_norm": 0.005266046151518822,
      "learning_rate": 3.3006851633526506e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4437376,
      "step": 15850
    },
    {
      "epoch": 176.16666666666666,
      "grad_norm": 0.015360617078840733,
      "learning_rate": 3.2997550675634584e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4438784,
      "step": 15855
    },
    {
      "epoch": 176.22222222222223,
      "grad_norm": 0.009975150227546692,
      "learning_rate": 3.2988248484419825e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4440224,
      "step": 15860
    },
    {
      "epoch": 176.27777777777777,
      "grad_norm": 0.0018019535345956683,
      "learning_rate": 3.2978945061316776e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 4441632,
      "step": 15865
    },
    {
      "epoch": 176.33333333333334,
      "grad_norm": 0.03253712132573128,
      "learning_rate": 3.296964040776013e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4443088,
      "step": 15870
    },
    {
      "epoch": 176.38888888888889,
      "grad_norm": 0.014267821796238422,
      "learning_rate": 3.296033452518478e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4444496,
      "step": 15875
    },
    {
      "epoch": 176.44444444444446,
      "grad_norm": 0.006465336307883263,
      "learning_rate": 3.2951027415025806e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4445888,
      "step": 15880
    },
    {
      "epoch": 176.5,
      "grad_norm": 1.443638801574707,
      "learning_rate": 3.294171907871849e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 4447328,
      "step": 15885
    },
    {
      "epoch": 176.55555555555554,
      "grad_norm": 0.10412625968456268,
      "learning_rate": 3.293240951769828e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4448704,
      "step": 15890
    },
    {
      "epoch": 176.61111111111111,
      "grad_norm": 0.6453332901000977,
      "learning_rate": 3.2923098733400846e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 4450128,
      "step": 15895
    },
    {
      "epoch": 176.66666666666666,
      "grad_norm": 0.015564109198749065,
      "learning_rate": 3.291378672726202e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4451584,
      "step": 15900
    },
    {
      "epoch": 176.72222222222223,
      "grad_norm": 0.0007885622326284647,
      "learning_rate": 3.2904473500717824e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4452944,
      "step": 15905
    },
    {
      "epoch": 176.77777777777777,
      "grad_norm": 0.010901553556323051,
      "learning_rate": 3.289515905520449e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4454320,
      "step": 15910
    },
    {
      "epoch": 176.83333333333334,
      "grad_norm": 0.02084287628531456,
      "learning_rate": 3.288584339215841e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4455664,
      "step": 15915
    },
    {
      "epoch": 176.88888888888889,
      "grad_norm": 0.021830419078469276,
      "learning_rate": 3.287652651301617e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4457056,
      "step": 15920
    },
    {
      "epoch": 176.94444444444446,
      "grad_norm": 0.0009285500855185091,
      "learning_rate": 3.286720841921457e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4458448,
      "step": 15925
    },
    {
      "epoch": 177.0,
      "grad_norm": 0.036762841045856476,
      "learning_rate": 3.285788911219056e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4459856,
      "step": 15930
    },
    {
      "epoch": 177.05555555555554,
      "grad_norm": 0.016602875664830208,
      "learning_rate": 3.284856859338131e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4461264,
      "step": 15935
    },
    {
      "epoch": 177.11111111111111,
      "grad_norm": 0.009258931502699852,
      "learning_rate": 3.283924686422414e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4462656,
      "step": 15940
    },
    {
      "epoch": 177.16666666666666,
      "grad_norm": 0.007914085872471333,
      "learning_rate": 3.282992392615659e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4464080,
      "step": 15945
    },
    {
      "epoch": 177.22222222222223,
      "grad_norm": 0.008699534460902214,
      "learning_rate": 3.282059978061638e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4465520,
      "step": 15950
    },
    {
      "epoch": 177.27777777777777,
      "grad_norm": 0.07624513655900955,
      "learning_rate": 3.28112744290414e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4466896,
      "step": 15955
    },
    {
      "epoch": 177.33333333333334,
      "grad_norm": 0.03713138774037361,
      "learning_rate": 3.280194787286974e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4468336,
      "step": 15960
    },
    {
      "epoch": 177.38888888888889,
      "grad_norm": 0.1772083044052124,
      "learning_rate": 3.2792620113539674e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 4469728,
      "step": 15965
    },
    {
      "epoch": 177.44444444444446,
      "grad_norm": 0.0020296303555369377,
      "learning_rate": 3.278329115248966e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4471104,
      "step": 15970
    },
    {
      "epoch": 177.5,
      "grad_norm": 0.008327949792146683,
      "learning_rate": 3.277396099115834e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4472464,
      "step": 15975
    },
    {
      "epoch": 177.55555555555554,
      "grad_norm": 0.0038693726528435946,
      "learning_rate": 3.276462963098454e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4473872,
      "step": 15980
    },
    {
      "epoch": 177.61111111111111,
      "grad_norm": 0.011271920055150986,
      "learning_rate": 3.275529707340728e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4475264,
      "step": 15985
    },
    {
      "epoch": 177.66666666666666,
      "grad_norm": 0.0049786618910729885,
      "learning_rate": 3.274596331986574e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 4476624,
      "step": 15990
    },
    {
      "epoch": 177.72222222222223,
      "grad_norm": 0.0009868332417681813,
      "learning_rate": 3.273662837179932e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4478032,
      "step": 15995
    },
    {
      "epoch": 177.77777777777777,
      "grad_norm": 0.05217833071947098,
      "learning_rate": 3.272729223064758e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4479456,
      "step": 16000
    },
    {
      "epoch": 177.77777777777777,
      "eval_loss": 0.5544477701187134,
      "eval_runtime": 1.2432,
      "eval_samples_per_second": 32.174,
      "eval_steps_per_second": 16.087,
      "num_input_tokens_seen": 4479456,
      "step": 16000
    },
    {
      "epoch": 177.83333333333334,
      "grad_norm": 0.00929749570786953,
      "learning_rate": 3.2717954897850264e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4480816,
      "step": 16005
    },
    {
      "epoch": 177.88888888888889,
      "grad_norm": 0.0011050151661038399,
      "learning_rate": 3.270861637484733e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4482176,
      "step": 16010
    },
    {
      "epoch": 177.94444444444446,
      "grad_norm": 0.0051920549012720585,
      "learning_rate": 3.2699276663078867e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4483600,
      "step": 16015
    },
    {
      "epoch": 178.0,
      "grad_norm": 0.002421023789793253,
      "learning_rate": 3.268993576398519e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4485008,
      "step": 16020
    },
    {
      "epoch": 178.05555555555554,
      "grad_norm": 0.030697382986545563,
      "learning_rate": 3.268059367900678e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4486416,
      "step": 16025
    },
    {
      "epoch": 178.11111111111111,
      "grad_norm": 0.02743428386747837,
      "learning_rate": 3.26712504095843e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4487840,
      "step": 16030
    },
    {
      "epoch": 178.16666666666666,
      "grad_norm": 0.012606530450284481,
      "learning_rate": 3.2661905957158615e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4489280,
      "step": 16035
    },
    {
      "epoch": 178.22222222222223,
      "grad_norm": 0.002131146378815174,
      "learning_rate": 3.2652560323170734e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4490704,
      "step": 16040
    },
    {
      "epoch": 178.27777777777777,
      "grad_norm": 0.002631360199302435,
      "learning_rate": 3.264321350906189e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4492096,
      "step": 16045
    },
    {
      "epoch": 178.33333333333334,
      "grad_norm": 0.0007791582029312849,
      "learning_rate": 3.263386551627346e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4493504,
      "step": 16050
    },
    {
      "epoch": 178.38888888888889,
      "grad_norm": 0.0204211063683033,
      "learning_rate": 3.2624516346247055e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4494912,
      "step": 16055
    },
    {
      "epoch": 178.44444444444446,
      "grad_norm": 0.0756310224533081,
      "learning_rate": 3.2615166000424404e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4496272,
      "step": 16060
    },
    {
      "epoch": 178.5,
      "grad_norm": 0.003242508741095662,
      "learning_rate": 3.260581448024745e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4497712,
      "step": 16065
    },
    {
      "epoch": 178.55555555555554,
      "grad_norm": 0.012768343091011047,
      "learning_rate": 3.2596461787158335e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4499088,
      "step": 16070
    },
    {
      "epoch": 178.61111111111111,
      "grad_norm": 0.05950317531824112,
      "learning_rate": 3.258710792259934e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4500480,
      "step": 16075
    },
    {
      "epoch": 178.66666666666666,
      "grad_norm": 0.012587239034473896,
      "learning_rate": 3.257775288801296e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4501872,
      "step": 16080
    },
    {
      "epoch": 178.72222222222223,
      "grad_norm": 0.0023310058750212193,
      "learning_rate": 3.256839668484186e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 4503264,
      "step": 16085
    },
    {
      "epoch": 178.77777777777777,
      "grad_norm": 0.006789444945752621,
      "learning_rate": 3.255903931452888e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4504672,
      "step": 16090
    },
    {
      "epoch": 178.83333333333334,
      "grad_norm": 0.0015716010238975286,
      "learning_rate": 3.2549680778517045e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4506016,
      "step": 16095
    },
    {
      "epoch": 178.88888888888889,
      "grad_norm": 0.01922130212187767,
      "learning_rate": 3.2540321078249556e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4507472,
      "step": 16100
    },
    {
      "epoch": 178.94444444444446,
      "grad_norm": 0.22187155485153198,
      "learning_rate": 3.2530960215169795e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 4508848,
      "step": 16105
    },
    {
      "epoch": 179.0,
      "grad_norm": 0.001188879832625389,
      "learning_rate": 3.2521598190721345e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4510240,
      "step": 16110
    },
    {
      "epoch": 179.05555555555554,
      "grad_norm": 0.055479831993579865,
      "learning_rate": 3.251223500634792e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4511600,
      "step": 16115
    },
    {
      "epoch": 179.11111111111111,
      "grad_norm": 0.0032421955838799477,
      "learning_rate": 3.2502870663493445e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4512976,
      "step": 16120
    },
    {
      "epoch": 179.16666666666666,
      "grad_norm": 0.5974472761154175,
      "learning_rate": 3.249350516360203e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 4514368,
      "step": 16125
    },
    {
      "epoch": 179.22222222222223,
      "grad_norm": 0.0091685950756073,
      "learning_rate": 3.248413850811797e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4515760,
      "step": 16130
    },
    {
      "epoch": 179.27777777777777,
      "grad_norm": 0.37126076221466064,
      "learning_rate": 3.2474770698485677e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4517168,
      "step": 16135
    },
    {
      "epoch": 179.33333333333334,
      "grad_norm": 0.05872781202197075,
      "learning_rate": 3.246540173614983e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4518576,
      "step": 16140
    },
    {
      "epoch": 179.38888888888889,
      "grad_norm": 0.012793715111911297,
      "learning_rate": 3.2456031622555197e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4519968,
      "step": 16145
    },
    {
      "epoch": 179.44444444444446,
      "grad_norm": 0.011828719638288021,
      "learning_rate": 3.2446660359146794e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4521392,
      "step": 16150
    },
    {
      "epoch": 179.5,
      "grad_norm": 0.009603745304048061,
      "learning_rate": 3.2437287947369786e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4522800,
      "step": 16155
    },
    {
      "epoch": 179.55555555555554,
      "grad_norm": 0.13279420137405396,
      "learning_rate": 3.2427914388669525e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4524224,
      "step": 16160
    },
    {
      "epoch": 179.61111111111111,
      "grad_norm": 0.019085975363850594,
      "learning_rate": 3.241853968449151e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4525648,
      "step": 16165
    },
    {
      "epoch": 179.66666666666666,
      "grad_norm": 0.00562869431450963,
      "learning_rate": 3.240916383628144e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4527040,
      "step": 16170
    },
    {
      "epoch": 179.72222222222223,
      "grad_norm": 0.0006202934309840202,
      "learning_rate": 3.239978684548521e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4528464,
      "step": 16175
    },
    {
      "epoch": 179.77777777777777,
      "grad_norm": 0.014591815881431103,
      "learning_rate": 3.239040871354885e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4529872,
      "step": 16180
    },
    {
      "epoch": 179.83333333333334,
      "grad_norm": 0.002916720462962985,
      "learning_rate": 3.2381029441918596e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4531248,
      "step": 16185
    },
    {
      "epoch": 179.88888888888889,
      "grad_norm": 0.016846610233187675,
      "learning_rate": 3.2371649032040845e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4532688,
      "step": 16190
    },
    {
      "epoch": 179.94444444444446,
      "grad_norm": 0.0017483832780271769,
      "learning_rate": 3.2362267485362174e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4534112,
      "step": 16195
    },
    {
      "epoch": 180.0,
      "grad_norm": 0.030282247811555862,
      "learning_rate": 3.235288480332934e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4535504,
      "step": 16200
    },
    {
      "epoch": 180.0,
      "eval_loss": 0.5789384245872498,
      "eval_runtime": 1.2358,
      "eval_samples_per_second": 32.367,
      "eval_steps_per_second": 16.184,
      "num_input_tokens_seen": 4535504,
      "step": 16200
    },
    {
      "epoch": 180.05555555555554,
      "grad_norm": 0.021332943812012672,
      "learning_rate": 3.234350098738927e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4536896,
      "step": 16205
    },
    {
      "epoch": 180.11111111111111,
      "grad_norm": 0.020235838368535042,
      "learning_rate": 3.233411603898906e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4538272,
      "step": 16210
    },
    {
      "epoch": 180.16666666666666,
      "grad_norm": 0.009922634810209274,
      "learning_rate": 3.232472995957599e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4539680,
      "step": 16215
    },
    {
      "epoch": 180.22222222222223,
      "grad_norm": 0.06573443859815598,
      "learning_rate": 3.231534275059751e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4541104,
      "step": 16220
    },
    {
      "epoch": 180.27777777777777,
      "grad_norm": 0.05494973063468933,
      "learning_rate": 3.230595441350125e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4542512,
      "step": 16225
    },
    {
      "epoch": 180.33333333333334,
      "grad_norm": 0.0010081750806421041,
      "learning_rate": 3.2296564949735e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4543920,
      "step": 16230
    },
    {
      "epoch": 180.38888888888889,
      "grad_norm": 0.017701582983136177,
      "learning_rate": 3.228717436074675e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4545296,
      "step": 16235
    },
    {
      "epoch": 180.44444444444446,
      "grad_norm": 0.00904941651970148,
      "learning_rate": 3.227778264798463e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4546688,
      "step": 16240
    },
    {
      "epoch": 180.5,
      "grad_norm": 0.002129387343302369,
      "learning_rate": 3.226838981289698e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4548064,
      "step": 16245
    },
    {
      "epoch": 180.55555555555554,
      "grad_norm": 0.006667148321866989,
      "learning_rate": 3.225899585693227e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4549472,
      "step": 16250
    },
    {
      "epoch": 180.61111111111111,
      "grad_norm": 0.00024279369972646236,
      "learning_rate": 3.224960078153918e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4550832,
      "step": 16255
    },
    {
      "epoch": 180.66666666666666,
      "grad_norm": 0.015100457705557346,
      "learning_rate": 3.224020458816655e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4552240,
      "step": 16260
    },
    {
      "epoch": 180.72222222222223,
      "grad_norm": 0.03265073522925377,
      "learning_rate": 3.223080727826337e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4553696,
      "step": 16265
    },
    {
      "epoch": 180.77777777777777,
      "grad_norm": 0.019866498187184334,
      "learning_rate": 3.222140885327885e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4555088,
      "step": 16270
    },
    {
      "epoch": 180.83333333333334,
      "grad_norm": 0.017324259504675865,
      "learning_rate": 3.221200931466234e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4556528,
      "step": 16275
    },
    {
      "epoch": 180.88888888888889,
      "grad_norm": 0.0024923677556216717,
      "learning_rate": 3.220260866386336e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4557920,
      "step": 16280
    },
    {
      "epoch": 180.94444444444446,
      "grad_norm": 0.001834722119383514,
      "learning_rate": 3.21932069023316e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4559312,
      "step": 16285
    },
    {
      "epoch": 181.0,
      "grad_norm": 0.09063945710659027,
      "learning_rate": 3.218380403151695e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4560736,
      "step": 16290
    },
    {
      "epoch": 181.05555555555554,
      "grad_norm": 0.07431849092245102,
      "learning_rate": 3.217440005286943e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4562112,
      "step": 16295
    },
    {
      "epoch": 181.11111111111111,
      "grad_norm": 0.0018445284804329276,
      "learning_rate": 3.216499496783928e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4563536,
      "step": 16300
    },
    {
      "epoch": 181.16666666666666,
      "grad_norm": 0.0012988229282200336,
      "learning_rate": 3.2155588777876856e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4564976,
      "step": 16305
    },
    {
      "epoch": 181.22222222222223,
      "grad_norm": 0.022406714037060738,
      "learning_rate": 3.214618148443273e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 4566352,
      "step": 16310
    },
    {
      "epoch": 181.27777777777777,
      "grad_norm": 0.06720859557390213,
      "learning_rate": 3.2136773088957595e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4567728,
      "step": 16315
    },
    {
      "epoch": 181.33333333333334,
      "grad_norm": 0.001130196382291615,
      "learning_rate": 3.2127363592902374e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4569104,
      "step": 16320
    },
    {
      "epoch": 181.38888888888889,
      "grad_norm": 0.04172699525952339,
      "learning_rate": 3.211795299771812e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4570528,
      "step": 16325
    },
    {
      "epoch": 181.44444444444446,
      "grad_norm": 0.0015319030499085784,
      "learning_rate": 3.210854130485605e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4571936,
      "step": 16330
    },
    {
      "epoch": 181.5,
      "grad_norm": 0.00357463164255023,
      "learning_rate": 3.209912851576759e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4573328,
      "step": 16335
    },
    {
      "epoch": 181.55555555555554,
      "grad_norm": 0.04246892035007477,
      "learning_rate": 3.208971463190431e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4574656,
      "step": 16340
    },
    {
      "epoch": 181.61111111111111,
      "grad_norm": 0.011881303042173386,
      "learning_rate": 3.208029965471793e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4576048,
      "step": 16345
    },
    {
      "epoch": 181.66666666666666,
      "grad_norm": 0.004479553084820509,
      "learning_rate": 3.2070883585660364e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4577456,
      "step": 16350
    },
    {
      "epoch": 181.72222222222223,
      "grad_norm": 0.027172980830073357,
      "learning_rate": 3.20614664261837e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4578896,
      "step": 16355
    },
    {
      "epoch": 181.77777777777777,
      "grad_norm": 0.004087011329829693,
      "learning_rate": 3.205204817774016e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4580320,
      "step": 16360
    },
    {
      "epoch": 181.83333333333334,
      "grad_norm": 0.013905558735132217,
      "learning_rate": 3.204262884178218e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4581696,
      "step": 16365
    },
    {
      "epoch": 181.88888888888889,
      "grad_norm": 0.28611451387405396,
      "learning_rate": 3.2033208419762314e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4583104,
      "step": 16370
    },
    {
      "epoch": 181.94444444444446,
      "grad_norm": 0.01573864370584488,
      "learning_rate": 3.2023786913133344e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4584528,
      "step": 16375
    },
    {
      "epoch": 182.0,
      "grad_norm": 0.03460254147648811,
      "learning_rate": 3.201436432334816e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4585936,
      "step": 16380
    },
    {
      "epoch": 182.05555555555554,
      "grad_norm": 0.13070818781852722,
      "learning_rate": 3.2004940651859844e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4587296,
      "step": 16385
    },
    {
      "epoch": 182.11111111111111,
      "grad_norm": 0.0022592563182115555,
      "learning_rate": 3.1995515900121655e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4588704,
      "step": 16390
    },
    {
      "epoch": 182.16666666666666,
      "grad_norm": 0.006968115922063589,
      "learning_rate": 3.1986090069587e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4590080,
      "step": 16395
    },
    {
      "epoch": 182.22222222222223,
      "grad_norm": 0.005919143557548523,
      "learning_rate": 3.1976663161709466e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4591504,
      "step": 16400
    },
    {
      "epoch": 182.22222222222223,
      "eval_loss": 0.5727646946907043,
      "eval_runtime": 1.2353,
      "eval_samples_per_second": 32.381,
      "eval_steps_per_second": 16.191,
      "num_input_tokens_seen": 4591504,
      "step": 16400
    },
    {
      "epoch": 182.27777777777777,
      "grad_norm": 0.0013806161005049944,
      "learning_rate": 3.196723517794279e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4592912,
      "step": 16405
    },
    {
      "epoch": 182.33333333333334,
      "grad_norm": 0.0016553286695852876,
      "learning_rate": 3.19578061197409e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4594304,
      "step": 16410
    },
    {
      "epoch": 182.38888888888889,
      "grad_norm": 0.009581663645803928,
      "learning_rate": 3.194837598855787e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4595696,
      "step": 16415
    },
    {
      "epoch": 182.44444444444446,
      "grad_norm": 0.005458584986627102,
      "learning_rate": 3.193894478584794e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4597120,
      "step": 16420
    },
    {
      "epoch": 182.5,
      "grad_norm": 0.014121548272669315,
      "learning_rate": 3.192951251306553e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4598512,
      "step": 16425
    },
    {
      "epoch": 182.55555555555554,
      "grad_norm": 0.037659045308828354,
      "learning_rate": 3.192007917166521e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4599920,
      "step": 16430
    },
    {
      "epoch": 182.61111111111111,
      "grad_norm": 0.03359154239296913,
      "learning_rate": 3.191064476310171e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4601312,
      "step": 16435
    },
    {
      "epoch": 182.66666666666666,
      "grad_norm": 0.0019298936240375042,
      "learning_rate": 3.1901209288829944e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4602672,
      "step": 16440
    },
    {
      "epoch": 182.72222222222223,
      "grad_norm": 0.005210071802139282,
      "learning_rate": 3.1891772750304985e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4604064,
      "step": 16445
    },
    {
      "epoch": 182.77777777777777,
      "grad_norm": 0.005684972740709782,
      "learning_rate": 3.188233514898206e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4605424,
      "step": 16450
    },
    {
      "epoch": 182.83333333333334,
      "grad_norm": 0.0007979201036505401,
      "learning_rate": 3.187289648631657e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4606832,
      "step": 16455
    },
    {
      "epoch": 182.88888888888889,
      "grad_norm": 0.05935158208012581,
      "learning_rate": 3.186345676376406e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4608240,
      "step": 16460
    },
    {
      "epoch": 182.94444444444446,
      "grad_norm": 0.004913651384413242,
      "learning_rate": 3.1854015982780275e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4609632,
      "step": 16465
    },
    {
      "epoch": 183.0,
      "grad_norm": 0.011998522095382214,
      "learning_rate": 3.1844574144821084e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4611072,
      "step": 16470
    },
    {
      "epoch": 183.05555555555554,
      "grad_norm": 0.0016006026417016983,
      "learning_rate": 3.1835131251342554e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4612496,
      "step": 16475
    },
    {
      "epoch": 183.11111111111111,
      "grad_norm": 0.030570000410079956,
      "learning_rate": 3.182568730380089e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4613920,
      "step": 16480
    },
    {
      "epoch": 183.16666666666666,
      "grad_norm": 0.23467658460140228,
      "learning_rate": 3.181624230365245e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4615344,
      "step": 16485
    },
    {
      "epoch": 183.22222222222223,
      "grad_norm": 0.014957134611904621,
      "learning_rate": 3.180679625235381e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4616736,
      "step": 16490
    },
    {
      "epoch": 183.27777777777777,
      "grad_norm": 0.0019803266040980816,
      "learning_rate": 3.1797349151361646e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4618160,
      "step": 16495
    },
    {
      "epoch": 183.33333333333334,
      "grad_norm": 0.0626499354839325,
      "learning_rate": 3.178790100213281e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4619568,
      "step": 16500
    },
    {
      "epoch": 183.38888888888889,
      "grad_norm": 0.00679313950240612,
      "learning_rate": 3.1778451806124346e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4620928,
      "step": 16505
    },
    {
      "epoch": 183.44444444444446,
      "grad_norm": 0.06001538410782814,
      "learning_rate": 3.176900156479342e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4622320,
      "step": 16510
    },
    {
      "epoch": 183.5,
      "grad_norm": 0.018454905599355698,
      "learning_rate": 3.17595502795974e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4623712,
      "step": 16515
    },
    {
      "epoch": 183.55555555555554,
      "grad_norm": 0.007644526660442352,
      "learning_rate": 3.175009795199377e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4625152,
      "step": 16520
    },
    {
      "epoch": 183.61111111111111,
      "grad_norm": 0.016737008467316628,
      "learning_rate": 3.1740644583440224e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4626512,
      "step": 16525
    },
    {
      "epoch": 183.66666666666666,
      "grad_norm": 0.030453579500317574,
      "learning_rate": 3.173119017539457e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4627952,
      "step": 16530
    },
    {
      "epoch": 183.72222222222223,
      "grad_norm": 0.007452623452991247,
      "learning_rate": 3.172173472931479e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4629360,
      "step": 16535
    },
    {
      "epoch": 183.77777777777777,
      "grad_norm": 0.042817115783691406,
      "learning_rate": 3.1712278246659055e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4630720,
      "step": 16540
    },
    {
      "epoch": 183.83333333333334,
      "grad_norm": 0.0021156060975044966,
      "learning_rate": 3.170282072888566e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4632112,
      "step": 16545
    },
    {
      "epoch": 183.88888888888889,
      "grad_norm": 0.019480083137750626,
      "learning_rate": 3.169336217745307e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4633504,
      "step": 16550
    },
    {
      "epoch": 183.94444444444446,
      "grad_norm": 0.006677242461591959,
      "learning_rate": 3.1683902593819924e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4634896,
      "step": 16555
    },
    {
      "epoch": 184.0,
      "grad_norm": 0.009887549094855785,
      "learning_rate": 3.1674441979445e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4636288,
      "step": 16560
    },
    {
      "epoch": 184.05555555555554,
      "grad_norm": 0.055614322423934937,
      "learning_rate": 3.166498033578725e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4637696,
      "step": 16565
    },
    {
      "epoch": 184.11111111111111,
      "grad_norm": 0.00924955029040575,
      "learning_rate": 3.165551766430578e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4639072,
      "step": 16570
    },
    {
      "epoch": 184.16666666666666,
      "grad_norm": 0.017836671322584152,
      "learning_rate": 3.164605396645984e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4640496,
      "step": 16575
    },
    {
      "epoch": 184.22222222222223,
      "grad_norm": 0.013996032066643238,
      "learning_rate": 3.163658924370886e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4641872,
      "step": 16580
    },
    {
      "epoch": 184.27777777777777,
      "grad_norm": 1.9286911487579346,
      "learning_rate": 3.1627123497512415e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 4643248,
      "step": 16585
    },
    {
      "epoch": 184.33333333333334,
      "grad_norm": 0.008712583221495152,
      "learning_rate": 3.1617656729330245e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4644608,
      "step": 16590
    },
    {
      "epoch": 184.38888888888889,
      "grad_norm": 0.0016983753303065896,
      "learning_rate": 3.1608188940622255e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4646016,
      "step": 16595
    },
    {
      "epoch": 184.44444444444446,
      "grad_norm": 0.003301706863567233,
      "learning_rate": 3.159872013284847e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4647424,
      "step": 16600
    },
    {
      "epoch": 184.44444444444446,
      "eval_loss": 0.5941494107246399,
      "eval_runtime": 1.2343,
      "eval_samples_per_second": 32.406,
      "eval_steps_per_second": 16.203,
      "num_input_tokens_seen": 4647424,
      "step": 16600
    },
    {
      "epoch": 184.5,
      "grad_norm": 0.029102271422743797,
      "learning_rate": 3.1589250307469134e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4648832,
      "step": 16605
    },
    {
      "epoch": 184.55555555555554,
      "grad_norm": 0.02103406935930252,
      "learning_rate": 3.1579779465944586e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4650240,
      "step": 16610
    },
    {
      "epoch": 184.61111111111111,
      "grad_norm": 0.004924386739730835,
      "learning_rate": 3.1570307609735363e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4651616,
      "step": 16615
    },
    {
      "epoch": 184.66666666666666,
      "grad_norm": 0.000424901838414371,
      "learning_rate": 3.156083474030213e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4653040,
      "step": 16620
    },
    {
      "epoch": 184.72222222222223,
      "grad_norm": 0.001546940766274929,
      "learning_rate": 3.155136085910573e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4654432,
      "step": 16625
    },
    {
      "epoch": 184.77777777777777,
      "grad_norm": 0.010297026485204697,
      "learning_rate": 3.154188596760717e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4655808,
      "step": 16630
    },
    {
      "epoch": 184.83333333333334,
      "grad_norm": 0.15949994325637817,
      "learning_rate": 3.153241006726757e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4657248,
      "step": 16635
    },
    {
      "epoch": 184.88888888888889,
      "grad_norm": 0.017740754410624504,
      "learning_rate": 3.152293315954825e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 4658656,
      "step": 16640
    },
    {
      "epoch": 184.94444444444446,
      "grad_norm": 0.009722050279378891,
      "learning_rate": 3.1513455245910666e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 4660064,
      "step": 16645
    },
    {
      "epoch": 185.0,
      "grad_norm": 0.004846709780395031,
      "learning_rate": 3.150397632781643e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4661424,
      "step": 16650
    },
    {
      "epoch": 185.05555555555554,
      "grad_norm": 0.021776456385850906,
      "learning_rate": 3.149449640672731e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4662800,
      "step": 16655
    },
    {
      "epoch": 185.11111111111111,
      "grad_norm": 0.015173740684986115,
      "learning_rate": 3.148501548410523e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4664208,
      "step": 16660
    },
    {
      "epoch": 185.16666666666666,
      "grad_norm": 0.001930282567627728,
      "learning_rate": 3.1475533561412256e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4665648,
      "step": 16665
    },
    {
      "epoch": 185.22222222222223,
      "grad_norm": 0.011682847514748573,
      "learning_rate": 3.146605064011065e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4667024,
      "step": 16670
    },
    {
      "epoch": 185.27777777777777,
      "grad_norm": 0.007854554802179337,
      "learning_rate": 3.145656672166277e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4668416,
      "step": 16675
    },
    {
      "epoch": 185.33333333333334,
      "grad_norm": 0.005636152811348438,
      "learning_rate": 3.144708180753116e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4669808,
      "step": 16680
    },
    {
      "epoch": 185.38888888888889,
      "grad_norm": 0.002160215750336647,
      "learning_rate": 3.143759589917851e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4671200,
      "step": 16685
    },
    {
      "epoch": 185.44444444444446,
      "grad_norm": 0.051377665251493454,
      "learning_rate": 3.142810899806768e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4672608,
      "step": 16690
    },
    {
      "epoch": 185.5,
      "grad_norm": 0.0010301675647497177,
      "learning_rate": 3.141862110566166e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4674016,
      "step": 16695
    },
    {
      "epoch": 185.55555555555554,
      "grad_norm": 0.021467817947268486,
      "learning_rate": 3.1409132223423606e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4675408,
      "step": 16700
    },
    {
      "epoch": 185.61111111111111,
      "grad_norm": 0.020921723917126656,
      "learning_rate": 3.139964235281682e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4676768,
      "step": 16705
    },
    {
      "epoch": 185.66666666666666,
      "grad_norm": 0.021804803982377052,
      "learning_rate": 3.139015149530476e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4678208,
      "step": 16710
    },
    {
      "epoch": 185.72222222222223,
      "grad_norm": 0.010980159044265747,
      "learning_rate": 3.1380659652351034e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4679616,
      "step": 16715
    },
    {
      "epoch": 185.77777777777777,
      "grad_norm": 0.05759742110967636,
      "learning_rate": 3.137116682541941e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4681056,
      "step": 16720
    },
    {
      "epoch": 185.83333333333334,
      "grad_norm": 0.004634797107428312,
      "learning_rate": 3.136167301597379e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4682464,
      "step": 16725
    },
    {
      "epoch": 185.88888888888889,
      "grad_norm": 0.2957983613014221,
      "learning_rate": 3.1352178225478254e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4683824,
      "step": 16730
    },
    {
      "epoch": 185.94444444444446,
      "grad_norm": 0.007319625001400709,
      "learning_rate": 3.1342682455396996e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4685184,
      "step": 16735
    },
    {
      "epoch": 186.0,
      "grad_norm": 0.005763406865298748,
      "learning_rate": 3.133318570719441e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4686608,
      "step": 16740
    },
    {
      "epoch": 186.05555555555554,
      "grad_norm": 0.002483225194737315,
      "learning_rate": 3.132368798233499e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4687984,
      "step": 16745
    },
    {
      "epoch": 186.11111111111111,
      "grad_norm": 0.06564351916313171,
      "learning_rate": 3.131418928228342e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4689376,
      "step": 16750
    },
    {
      "epoch": 186.16666666666666,
      "grad_norm": 0.0023421156220138073,
      "learning_rate": 3.1304689608504514e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4690800,
      "step": 16755
    },
    {
      "epoch": 186.22222222222223,
      "grad_norm": 0.0996004268527031,
      "learning_rate": 3.129518896246324e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4692192,
      "step": 16760
    },
    {
      "epoch": 186.27777777777777,
      "grad_norm": 0.007784257177263498,
      "learning_rate": 3.128568734562472e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 4693568,
      "step": 16765
    },
    {
      "epoch": 186.33333333333334,
      "grad_norm": 0.034896500408649445,
      "learning_rate": 3.127618475945421e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4694992,
      "step": 16770
    },
    {
      "epoch": 186.38888888888889,
      "grad_norm": 0.0008919035317376256,
      "learning_rate": 3.126668120541715e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4696384,
      "step": 16775
    },
    {
      "epoch": 186.44444444444446,
      "grad_norm": 0.004764375742524862,
      "learning_rate": 3.1257176684979096e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4697760,
      "step": 16780
    },
    {
      "epoch": 186.5,
      "grad_norm": 0.001452246098779142,
      "learning_rate": 3.124767119960576e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4699136,
      "step": 16785
    },
    {
      "epoch": 186.55555555555554,
      "grad_norm": 0.031170716509222984,
      "learning_rate": 3.123816475076301e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4700512,
      "step": 16790
    },
    {
      "epoch": 186.61111111111111,
      "grad_norm": 0.0017602195730432868,
      "learning_rate": 3.122865733991687e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4701936,
      "step": 16795
    },
    {
      "epoch": 186.66666666666666,
      "grad_norm": 0.00224185548722744,
      "learning_rate": 3.1219148968533486e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4703376,
      "step": 16800
    },
    {
      "epoch": 186.66666666666666,
      "eval_loss": 0.6517533659934998,
      "eval_runtime": 1.2393,
      "eval_samples_per_second": 32.277,
      "eval_steps_per_second": 16.139,
      "num_input_tokens_seen": 4703376,
      "step": 16800
    },
    {
      "epoch": 186.72222222222223,
      "grad_norm": 0.022078832611441612,
      "learning_rate": 3.120963963807918e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4704800,
      "step": 16805
    },
    {
      "epoch": 186.77777777777777,
      "grad_norm": 0.014795994386076927,
      "learning_rate": 3.12001293500204e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4706176,
      "step": 16810
    },
    {
      "epoch": 186.83333333333334,
      "grad_norm": 0.012826007790863514,
      "learning_rate": 3.1190618105823765e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4707584,
      "step": 16815
    },
    {
      "epoch": 186.88888888888889,
      "grad_norm": 0.4741598069667816,
      "learning_rate": 3.118110590695603e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4709024,
      "step": 16820
    },
    {
      "epoch": 186.94444444444446,
      "grad_norm": 0.0014029371086508036,
      "learning_rate": 3.117159275488407e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4710432,
      "step": 16825
    },
    {
      "epoch": 187.0,
      "grad_norm": 0.009114419110119343,
      "learning_rate": 3.1162078651074956e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4711808,
      "step": 16830
    },
    {
      "epoch": 187.05555555555554,
      "grad_norm": 0.0205463245511055,
      "learning_rate": 3.1152563596995885e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4713184,
      "step": 16835
    },
    {
      "epoch": 187.11111111111111,
      "grad_norm": 0.001877172035165131,
      "learning_rate": 3.1143047594114186e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4714576,
      "step": 16840
    },
    {
      "epoch": 187.16666666666666,
      "grad_norm": 0.21644560992717743,
      "learning_rate": 3.113353064389734e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4715968,
      "step": 16845
    },
    {
      "epoch": 187.22222222222223,
      "grad_norm": 0.014108221046626568,
      "learning_rate": 3.1124012747812993e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4717344,
      "step": 16850
    },
    {
      "epoch": 187.27777777777777,
      "grad_norm": 0.04615611955523491,
      "learning_rate": 3.1114493907328936e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4718752,
      "step": 16855
    },
    {
      "epoch": 187.33333333333334,
      "grad_norm": 0.0005456352955661714,
      "learning_rate": 3.110497412391306e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4720192,
      "step": 16860
    },
    {
      "epoch": 187.38888888888889,
      "grad_norm": 0.006190570071339607,
      "learning_rate": 3.1095453399033466e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4721616,
      "step": 16865
    },
    {
      "epoch": 187.44444444444446,
      "grad_norm": 0.0012050403747707605,
      "learning_rate": 3.108593173415835e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4723024,
      "step": 16870
    },
    {
      "epoch": 187.5,
      "grad_norm": 0.0010924001690000296,
      "learning_rate": 3.107640913075609e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4724464,
      "step": 16875
    },
    {
      "epoch": 187.55555555555554,
      "grad_norm": 0.028330914676189423,
      "learning_rate": 3.106688559029517e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4725840,
      "step": 16880
    },
    {
      "epoch": 187.61111111111111,
      "grad_norm": 0.0032170978374779224,
      "learning_rate": 3.105736111424425e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4727264,
      "step": 16885
    },
    {
      "epoch": 187.66666666666666,
      "grad_norm": 0.01039684098213911,
      "learning_rate": 3.1047835704072136e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4728656,
      "step": 16890
    },
    {
      "epoch": 187.72222222222223,
      "grad_norm": 0.003412947291508317,
      "learning_rate": 3.103830936124775e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4730032,
      "step": 16895
    },
    {
      "epoch": 187.77777777777777,
      "grad_norm": 0.06061859801411629,
      "learning_rate": 3.102878208724018e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4731392,
      "step": 16900
    },
    {
      "epoch": 187.83333333333334,
      "grad_norm": 0.008149913512170315,
      "learning_rate": 3.101925388351865e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4732816,
      "step": 16905
    },
    {
      "epoch": 187.88888888888889,
      "grad_norm": 0.00966637022793293,
      "learning_rate": 3.1009724751552515e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4734240,
      "step": 16910
    },
    {
      "epoch": 187.94444444444446,
      "grad_norm": 0.012817353010177612,
      "learning_rate": 3.100019469281131e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4735600,
      "step": 16915
    },
    {
      "epoch": 188.0,
      "grad_norm": 0.09685617685317993,
      "learning_rate": 3.0990663708764685e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4737040,
      "step": 16920
    },
    {
      "epoch": 188.05555555555554,
      "grad_norm": 0.01027390081435442,
      "learning_rate": 3.098113180088243e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4738480,
      "step": 16925
    },
    {
      "epoch": 188.11111111111111,
      "grad_norm": 0.004665938671678305,
      "learning_rate": 3.097159897063448e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4739936,
      "step": 16930
    },
    {
      "epoch": 188.16666666666666,
      "grad_norm": 0.0014091555494815111,
      "learning_rate": 3.096206521949094e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4741296,
      "step": 16935
    },
    {
      "epoch": 188.22222222222223,
      "grad_norm": 0.00644405372440815,
      "learning_rate": 3.0952530548922006e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4742720,
      "step": 16940
    },
    {
      "epoch": 188.27777777777777,
      "grad_norm": 0.0010356723796576262,
      "learning_rate": 3.0942994960398064e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 4744128,
      "step": 16945
    },
    {
      "epoch": 188.33333333333334,
      "grad_norm": 0.07023164629936218,
      "learning_rate": 3.093345845538961e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4745600,
      "step": 16950
    },
    {
      "epoch": 188.38888888888889,
      "grad_norm": 0.012611870653927326,
      "learning_rate": 3.09239210353673e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4746960,
      "step": 16955
    },
    {
      "epoch": 188.44444444444446,
      "grad_norm": 0.023673055693507195,
      "learning_rate": 3.0914382701801926e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4748352,
      "step": 16960
    },
    {
      "epoch": 188.5,
      "grad_norm": 0.014621411450207233,
      "learning_rate": 3.090484345616441e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4749728,
      "step": 16965
    },
    {
      "epoch": 188.55555555555554,
      "grad_norm": 0.031097954139113426,
      "learning_rate": 3.0895303299925825e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4751136,
      "step": 16970
    },
    {
      "epoch": 188.61111111111111,
      "grad_norm": 0.0011691799154505134,
      "learning_rate": 3.0885762234557393e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4752528,
      "step": 16975
    },
    {
      "epoch": 188.66666666666666,
      "grad_norm": 0.002768714679405093,
      "learning_rate": 3.087622026153045e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4753968,
      "step": 16980
    },
    {
      "epoch": 188.72222222222223,
      "grad_norm": 0.001955850049853325,
      "learning_rate": 3.086667738231651e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4755344,
      "step": 16985
    },
    {
      "epoch": 188.77777777777777,
      "grad_norm": 0.16390983760356903,
      "learning_rate": 3.085713359838718e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4756752,
      "step": 16990
    },
    {
      "epoch": 188.83333333333334,
      "grad_norm": 0.016401438042521477,
      "learning_rate": 3.084758891121425e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4758144,
      "step": 16995
    },
    {
      "epoch": 188.88888888888889,
      "grad_norm": 0.04493455961346626,
      "learning_rate": 3.083804332226963e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4759552,
      "step": 17000
    },
    {
      "epoch": 188.88888888888889,
      "eval_loss": 0.6518311500549316,
      "eval_runtime": 1.2463,
      "eval_samples_per_second": 32.095,
      "eval_steps_per_second": 16.047,
      "num_input_tokens_seen": 4759552,
      "step": 17000
    },
    {
      "epoch": 188.94444444444446,
      "grad_norm": 0.0005188014474697411,
      "learning_rate": 3.082849683302536e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4760912,
      "step": 17005
    },
    {
      "epoch": 189.0,
      "grad_norm": 0.01200709119439125,
      "learning_rate": 3.081894944495363e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4762336,
      "step": 17010
    },
    {
      "epoch": 189.05555555555554,
      "grad_norm": 0.011803235858678818,
      "learning_rate": 3.080940115952677e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4763760,
      "step": 17015
    },
    {
      "epoch": 189.11111111111111,
      "grad_norm": 0.03090417943894863,
      "learning_rate": 3.0799851978217245e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4765216,
      "step": 17020
    },
    {
      "epoch": 189.16666666666666,
      "grad_norm": 0.0015296897618100047,
      "learning_rate": 3.0790301902497666e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4766640,
      "step": 17025
    },
    {
      "epoch": 189.22222222222223,
      "grad_norm": 0.006057863123714924,
      "learning_rate": 3.078075093384076e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4768048,
      "step": 17030
    },
    {
      "epoch": 189.27777777777777,
      "grad_norm": 0.141195148229599,
      "learning_rate": 3.077119907371942e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4769456,
      "step": 17035
    },
    {
      "epoch": 189.33333333333334,
      "grad_norm": 0.010286223143339157,
      "learning_rate": 3.076164632360666e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4770880,
      "step": 17040
    },
    {
      "epoch": 189.38888888888889,
      "grad_norm": 0.0049092392437160015,
      "learning_rate": 3.075209268497563e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4772304,
      "step": 17045
    },
    {
      "epoch": 189.44444444444446,
      "grad_norm": 0.004418463911861181,
      "learning_rate": 3.074253815929961e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4773680,
      "step": 17050
    },
    {
      "epoch": 189.5,
      "grad_norm": 0.006189430598169565,
      "learning_rate": 3.0732982748052054e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4775072,
      "step": 17055
    },
    {
      "epoch": 189.55555555555554,
      "grad_norm": 0.004085351247340441,
      "learning_rate": 3.072342645270651e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4776512,
      "step": 17060
    },
    {
      "epoch": 189.61111111111111,
      "grad_norm": 0.008468273095786572,
      "learning_rate": 3.071386927473668e-05,
      "loss": 0.0017,
      "num_input_tokens_seen": 4777904,
      "step": 17065
    },
    {
      "epoch": 189.66666666666666,
      "grad_norm": 0.02321915328502655,
      "learning_rate": 3.0704311215616404e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4779296,
      "step": 17070
    },
    {
      "epoch": 189.72222222222223,
      "grad_norm": 0.03488648682832718,
      "learning_rate": 3.0694752276819656e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4780672,
      "step": 17075
    },
    {
      "epoch": 189.77777777777777,
      "grad_norm": 2.0000784397125244,
      "learning_rate": 3.068519245982054e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4782048,
      "step": 17080
    },
    {
      "epoch": 189.83333333333334,
      "grad_norm": 0.008651219308376312,
      "learning_rate": 3.0675631766093304e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4783376,
      "step": 17085
    },
    {
      "epoch": 189.88888888888889,
      "grad_norm": 0.10170318931341171,
      "learning_rate": 3.066607019711232e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 4784800,
      "step": 17090
    },
    {
      "epoch": 189.94444444444446,
      "grad_norm": 0.0009312390466220677,
      "learning_rate": 3.065650775435211e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 4786208,
      "step": 17095
    },
    {
      "epoch": 190.0,
      "grad_norm": 0.015483925119042397,
      "learning_rate": 3.0646944439287326e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4787584,
      "step": 17100
    },
    {
      "epoch": 190.05555555555554,
      "grad_norm": 48.30679702758789,
      "learning_rate": 3.0637380253392736e-05,
      "loss": 0.0442,
      "num_input_tokens_seen": 4788976,
      "step": 17105
    },
    {
      "epoch": 190.11111111111111,
      "grad_norm": 0.015148736536502838,
      "learning_rate": 3.062781519814327e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4790400,
      "step": 17110
    },
    {
      "epoch": 190.16666666666666,
      "grad_norm": 0.042309507727622986,
      "learning_rate": 3.0618249275013985e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4791792,
      "step": 17115
    },
    {
      "epoch": 190.22222222222223,
      "grad_norm": 0.0067236702889204025,
      "learning_rate": 3.060868248548005e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4793152,
      "step": 17120
    },
    {
      "epoch": 190.27777777777777,
      "grad_norm": 0.0074440180324018,
      "learning_rate": 3.0599114831016796e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4794576,
      "step": 17125
    },
    {
      "epoch": 190.33333333333334,
      "grad_norm": 0.00924141239374876,
      "learning_rate": 3.0589546313099666e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4795968,
      "step": 17130
    },
    {
      "epoch": 190.38888888888889,
      "grad_norm": 0.016329003497958183,
      "learning_rate": 3.0579976933204255e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4797408,
      "step": 17135
    },
    {
      "epoch": 190.44444444444446,
      "grad_norm": 0.003651233622804284,
      "learning_rate": 3.0570406692806284e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4798832,
      "step": 17140
    },
    {
      "epoch": 190.5,
      "grad_norm": 0.032571423798799515,
      "learning_rate": 3.05608355933816e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4800240,
      "step": 17145
    },
    {
      "epoch": 190.55555555555554,
      "grad_norm": 0.02536970004439354,
      "learning_rate": 3.055126363640618e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4801632,
      "step": 17150
    },
    {
      "epoch": 190.61111111111111,
      "grad_norm": 0.05565860867500305,
      "learning_rate": 3.0541690823356146e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4803040,
      "step": 17155
    },
    {
      "epoch": 190.66666666666666,
      "grad_norm": 0.00671986723318696,
      "learning_rate": 3.053211715570775e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4804448,
      "step": 17160
    },
    {
      "epoch": 190.72222222222223,
      "grad_norm": 0.014902609400451183,
      "learning_rate": 3.052254263493736e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4805872,
      "step": 17165
    },
    {
      "epoch": 190.77777777777777,
      "grad_norm": 0.02106623537838459,
      "learning_rate": 3.0512967262521498e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4807264,
      "step": 17170
    },
    {
      "epoch": 190.83333333333334,
      "grad_norm": 0.01638749986886978,
      "learning_rate": 3.0503391039936803e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4808656,
      "step": 17175
    },
    {
      "epoch": 190.88888888888889,
      "grad_norm": 0.009174205362796783,
      "learning_rate": 3.0493813968660056e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4810032,
      "step": 17180
    },
    {
      "epoch": 190.94444444444446,
      "grad_norm": 0.015883196145296097,
      "learning_rate": 3.0484236050168153e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4811424,
      "step": 17185
    },
    {
      "epoch": 191.0,
      "grad_norm": 0.0013438821770250797,
      "learning_rate": 3.0474657285938123e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4812784,
      "step": 17190
    },
    {
      "epoch": 191.05555555555554,
      "grad_norm": 0.015392344444990158,
      "learning_rate": 3.046507767744715e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4814160,
      "step": 17195
    },
    {
      "epoch": 191.11111111111111,
      "grad_norm": 0.0018575621070340276,
      "learning_rate": 3.045549722617252e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4815552,
      "step": 17200
    },
    {
      "epoch": 191.11111111111111,
      "eval_loss": 0.6296547651290894,
      "eval_runtime": 1.2436,
      "eval_samples_per_second": 32.165,
      "eval_steps_per_second": 16.083,
      "num_input_tokens_seen": 4815552,
      "step": 17200
    },
    {
      "epoch": 191.16666666666666,
      "grad_norm": 0.0036508836783468723,
      "learning_rate": 3.0445915933591658e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4816912,
      "step": 17205
    },
    {
      "epoch": 191.22222222222223,
      "grad_norm": 0.10133422166109085,
      "learning_rate": 3.0436333801182114e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 4818320,
      "step": 17210
    },
    {
      "epoch": 191.27777777777777,
      "grad_norm": 0.009794370271265507,
      "learning_rate": 3.0426750830421596e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4819728,
      "step": 17215
    },
    {
      "epoch": 191.33333333333334,
      "grad_norm": 0.030211517587304115,
      "learning_rate": 3.0417167022787897e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4821136,
      "step": 17220
    },
    {
      "epoch": 191.38888888888889,
      "grad_norm": 0.0015048850327730179,
      "learning_rate": 3.0407582379758966e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4822544,
      "step": 17225
    },
    {
      "epoch": 191.44444444444446,
      "grad_norm": 0.15427513420581818,
      "learning_rate": 3.039799690281287e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4823936,
      "step": 17230
    },
    {
      "epoch": 191.5,
      "grad_norm": 0.00034711946500465274,
      "learning_rate": 3.0388410593427823e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4825344,
      "step": 17235
    },
    {
      "epoch": 191.55555555555554,
      "grad_norm": 0.001787645509466529,
      "learning_rate": 3.0378823453082146e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4826752,
      "step": 17240
    },
    {
      "epoch": 191.61111111111111,
      "grad_norm": 0.0030188963282853365,
      "learning_rate": 3.03692354832543e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4828144,
      "step": 17245
    },
    {
      "epoch": 191.66666666666666,
      "grad_norm": 0.01772664487361908,
      "learning_rate": 3.0359646685422865e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4829568,
      "step": 17250
    },
    {
      "epoch": 191.72222222222223,
      "grad_norm": 0.007356939371675253,
      "learning_rate": 3.035005706106656e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4830896,
      "step": 17255
    },
    {
      "epoch": 191.77777777777777,
      "grad_norm": 0.0022684645373374224,
      "learning_rate": 3.034046661166422e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 4832304,
      "step": 17260
    },
    {
      "epoch": 191.83333333333334,
      "grad_norm": 0.0032660840079188347,
      "learning_rate": 3.033087533869482e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4833696,
      "step": 17265
    },
    {
      "epoch": 191.88888888888889,
      "grad_norm": 0.008421631529927254,
      "learning_rate": 3.0321283243637444e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4835152,
      "step": 17270
    },
    {
      "epoch": 191.94444444444446,
      "grad_norm": 0.07626216858625412,
      "learning_rate": 3.0311690327971326e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4836512,
      "step": 17275
    },
    {
      "epoch": 192.0,
      "grad_norm": 0.000846868846565485,
      "learning_rate": 3.030209659317581e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4837936,
      "step": 17280
    },
    {
      "epoch": 192.05555555555554,
      "grad_norm": 0.01016622968018055,
      "learning_rate": 3.0292502040730362e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4839312,
      "step": 17285
    },
    {
      "epoch": 192.11111111111111,
      "grad_norm": 0.009517084807157516,
      "learning_rate": 3.0282906672114597e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4840672,
      "step": 17290
    },
    {
      "epoch": 192.16666666666666,
      "grad_norm": 0.003567284205928445,
      "learning_rate": 3.027331048880823e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4842080,
      "step": 17295
    },
    {
      "epoch": 192.22222222222223,
      "grad_norm": 0.006507068406790495,
      "learning_rate": 3.0263713492291123e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4843504,
      "step": 17300
    },
    {
      "epoch": 192.27777777777777,
      "grad_norm": 0.0053911032155156136,
      "learning_rate": 3.0254115684043242e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4844912,
      "step": 17305
    },
    {
      "epoch": 192.33333333333334,
      "grad_norm": 0.0010392135009169579,
      "learning_rate": 3.024451706554469e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4846320,
      "step": 17310
    },
    {
      "epoch": 192.38888888888889,
      "grad_norm": 0.0012181082274764776,
      "learning_rate": 3.0234917638275705e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4847696,
      "step": 17315
    },
    {
      "epoch": 192.44444444444446,
      "grad_norm": 0.01989792473614216,
      "learning_rate": 3.0225317403716635e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 4849088,
      "step": 17320
    },
    {
      "epoch": 192.5,
      "grad_norm": 0.0025569351855665445,
      "learning_rate": 3.0215716363347956e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4850512,
      "step": 17325
    },
    {
      "epoch": 192.55555555555554,
      "grad_norm": 0.009604649618268013,
      "learning_rate": 3.0206114518650275e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4851904,
      "step": 17330
    },
    {
      "epoch": 192.61111111111111,
      "grad_norm": 0.005724533926695585,
      "learning_rate": 3.0196511871104304e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4853344,
      "step": 17335
    },
    {
      "epoch": 192.66666666666666,
      "grad_norm": 0.015658464282751083,
      "learning_rate": 3.01869084221909e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4854768,
      "step": 17340
    },
    {
      "epoch": 192.72222222222223,
      "grad_norm": 0.0004696994728874415,
      "learning_rate": 3.0177304173391037e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4856160,
      "step": 17345
    },
    {
      "epoch": 192.77777777777777,
      "grad_norm": 0.00041490784497000277,
      "learning_rate": 3.01676991261858e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 4857536,
      "step": 17350
    },
    {
      "epoch": 192.83333333333334,
      "grad_norm": 0.0013394664274528623,
      "learning_rate": 3.015809328205642e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4858976,
      "step": 17355
    },
    {
      "epoch": 192.88888888888889,
      "grad_norm": 0.0077406298369169235,
      "learning_rate": 3.0148486642484248e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4860384,
      "step": 17360
    },
    {
      "epoch": 192.94444444444446,
      "grad_norm": 0.009317690506577492,
      "learning_rate": 3.0138879208950722e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4861776,
      "step": 17365
    },
    {
      "epoch": 193.0,
      "grad_norm": 0.015749197453260422,
      "learning_rate": 3.012927098293744e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 4863168,
      "step": 17370
    },
    {
      "epoch": 193.05555555555554,
      "grad_norm": 0.008615336380898952,
      "learning_rate": 3.0119661965926123e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4864544,
      "step": 17375
    },
    {
      "epoch": 193.11111111111111,
      "grad_norm": 0.1815434843301773,
      "learning_rate": 3.0110052159398587e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4865952,
      "step": 17380
    },
    {
      "epoch": 193.16666666666666,
      "grad_norm": 0.00294551532715559,
      "learning_rate": 3.0100441564836802e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4867360,
      "step": 17385
    },
    {
      "epoch": 193.22222222222223,
      "grad_norm": 0.0418185256421566,
      "learning_rate": 3.0090830183722817e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4868784,
      "step": 17390
    },
    {
      "epoch": 193.27777777777777,
      "grad_norm": 0.002055473392829299,
      "learning_rate": 3.0081218017538852e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4870208,
      "step": 17395
    },
    {
      "epoch": 193.33333333333334,
      "grad_norm": 0.0014334403676912189,
      "learning_rate": 3.0071605067767212e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4871600,
      "step": 17400
    },
    {
      "epoch": 193.33333333333334,
      "eval_loss": 0.6059907674789429,
      "eval_runtime": 1.2335,
      "eval_samples_per_second": 32.429,
      "eval_steps_per_second": 16.214,
      "num_input_tokens_seen": 4871600,
      "step": 17400
    },
    {
      "epoch": 193.38888888888889,
      "grad_norm": 0.011176282539963722,
      "learning_rate": 3.006199133589034e-05,
      "loss": 0.0022,
      "num_input_tokens_seen": 4872976,
      "step": 17405
    },
    {
      "epoch": 193.44444444444446,
      "grad_norm": 0.030403461307287216,
      "learning_rate": 3.005237682339079e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 4874400,
      "step": 17410
    },
    {
      "epoch": 193.5,
      "grad_norm": 0.0029629834461957216,
      "learning_rate": 3.0042761531751228e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4875792,
      "step": 17415
    },
    {
      "epoch": 193.55555555555554,
      "grad_norm": 0.009587016887962818,
      "learning_rate": 3.0033145462454482e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4877200,
      "step": 17420
    },
    {
      "epoch": 193.61111111111111,
      "grad_norm": 0.0022926051169633865,
      "learning_rate": 3.002352861698345e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4878624,
      "step": 17425
    },
    {
      "epoch": 193.66666666666666,
      "grad_norm": 0.05777903273701668,
      "learning_rate": 3.0013910996821178e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4880048,
      "step": 17430
    },
    {
      "epoch": 193.72222222222223,
      "grad_norm": 0.0042074499651789665,
      "learning_rate": 3.0004292603450817e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4881440,
      "step": 17435
    },
    {
      "epoch": 193.77777777777777,
      "grad_norm": 0.01960034854710102,
      "learning_rate": 2.9994673438355653e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4882880,
      "step": 17440
    },
    {
      "epoch": 193.83333333333334,
      "grad_norm": 0.0006670872098766267,
      "learning_rate": 2.9985053503019078e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 4884272,
      "step": 17445
    },
    {
      "epoch": 193.88888888888889,
      "grad_norm": 0.00339036644436419,
      "learning_rate": 2.99754327989246e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4885664,
      "step": 17450
    },
    {
      "epoch": 193.94444444444446,
      "grad_norm": 0.033558815717697144,
      "learning_rate": 2.9965811327555864e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4887072,
      "step": 17455
    },
    {
      "epoch": 194.0,
      "grad_norm": 0.6252483129501343,
      "learning_rate": 2.995618909039662e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4888464,
      "step": 17460
    },
    {
      "epoch": 194.05555555555554,
      "grad_norm": 0.003322400152683258,
      "learning_rate": 2.9946566088930727e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4889856,
      "step": 17465
    },
    {
      "epoch": 194.11111111111111,
      "grad_norm": 0.01032615453004837,
      "learning_rate": 2.9936942324642192e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4891216,
      "step": 17470
    },
    {
      "epoch": 194.16666666666666,
      "grad_norm": 0.03305903449654579,
      "learning_rate": 2.9927317799015097e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4892640,
      "step": 17475
    },
    {
      "epoch": 194.22222222222223,
      "grad_norm": 0.0027365251444280148,
      "learning_rate": 2.9917692513533685e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4894048,
      "step": 17480
    },
    {
      "epoch": 194.27777777777777,
      "grad_norm": 0.0014948127791285515,
      "learning_rate": 2.990806646968229e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 4895472,
      "step": 17485
    },
    {
      "epoch": 194.33333333333334,
      "grad_norm": 0.007928448729217052,
      "learning_rate": 2.989843966894536e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 4896848,
      "step": 17490
    },
    {
      "epoch": 194.38888888888889,
      "grad_norm": 0.012357492931187153,
      "learning_rate": 2.9888812112807472e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4898224,
      "step": 17495
    },
    {
      "epoch": 194.44444444444446,
      "grad_norm": 0.0018755454802885652,
      "learning_rate": 2.987918380275333e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4899616,
      "step": 17500
    },
    {
      "epoch": 194.5,
      "grad_norm": 0.0042121210135519505,
      "learning_rate": 2.9869554740267724e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4901024,
      "step": 17505
    },
    {
      "epoch": 194.55555555555554,
      "grad_norm": 0.020477665588259697,
      "learning_rate": 2.9859924926835585e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4902432,
      "step": 17510
    },
    {
      "epoch": 194.61111111111111,
      "grad_norm": 0.008911116048693657,
      "learning_rate": 2.9850294363941944e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4903872,
      "step": 17515
    },
    {
      "epoch": 194.66666666666666,
      "grad_norm": 0.0006789251347072423,
      "learning_rate": 2.9840663053071967e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4905248,
      "step": 17520
    },
    {
      "epoch": 194.72222222222223,
      "grad_norm": 0.02423139102756977,
      "learning_rate": 2.983103099571091e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4906656,
      "step": 17525
    },
    {
      "epoch": 194.77777777777777,
      "grad_norm": 0.0046044765040278435,
      "learning_rate": 2.9821398193344164e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4908080,
      "step": 17530
    },
    {
      "epoch": 194.83333333333334,
      "grad_norm": 0.0007262961007654667,
      "learning_rate": 2.9811764647457226e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4909488,
      "step": 17535
    },
    {
      "epoch": 194.88888888888889,
      "grad_norm": 0.01897306740283966,
      "learning_rate": 2.9802130359535714e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4910880,
      "step": 17540
    },
    {
      "epoch": 194.94444444444446,
      "grad_norm": 0.009485314600169659,
      "learning_rate": 2.979249533106535e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4912224,
      "step": 17545
    },
    {
      "epoch": 195.0,
      "grad_norm": 0.010736143216490746,
      "learning_rate": 2.9782859563531986e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4913616,
      "step": 17550
    },
    {
      "epoch": 195.05555555555554,
      "grad_norm": 0.0024703724775463343,
      "learning_rate": 2.977322305842156e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4915024,
      "step": 17555
    },
    {
      "epoch": 195.11111111111111,
      "grad_norm": 0.004089244641363621,
      "learning_rate": 2.9763585817220162e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4916496,
      "step": 17560
    },
    {
      "epoch": 195.16666666666666,
      "grad_norm": 0.0009365016012452543,
      "learning_rate": 2.975394784141397e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4917936,
      "step": 17565
    },
    {
      "epoch": 195.22222222222223,
      "grad_norm": 0.033358559012413025,
      "learning_rate": 2.974430913248928e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 4919392,
      "step": 17570
    },
    {
      "epoch": 195.27777777777777,
      "grad_norm": 0.0002771485596895218,
      "learning_rate": 2.9734669691932497e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4920752,
      "step": 17575
    },
    {
      "epoch": 195.33333333333334,
      "grad_norm": 0.0020003407262265682,
      "learning_rate": 2.9725029521230147e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4922144,
      "step": 17580
    },
    {
      "epoch": 195.38888888888889,
      "grad_norm": 0.0005164783797226846,
      "learning_rate": 2.9715388621868873e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4923536,
      "step": 17585
    },
    {
      "epoch": 195.44444444444446,
      "grad_norm": 0.0008330447017215192,
      "learning_rate": 2.970574699533541e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4924928,
      "step": 17590
    },
    {
      "epoch": 195.5,
      "grad_norm": 0.010417104698717594,
      "learning_rate": 2.969610464311662e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4926288,
      "step": 17595
    },
    {
      "epoch": 195.55555555555554,
      "grad_norm": 0.0029728154186159372,
      "learning_rate": 2.9686461566699487e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4927696,
      "step": 17600
    },
    {
      "epoch": 195.55555555555554,
      "eval_loss": 0.6692536473274231,
      "eval_runtime": 1.2374,
      "eval_samples_per_second": 32.325,
      "eval_steps_per_second": 16.162,
      "num_input_tokens_seen": 4927696,
      "step": 17600
    },
    {
      "epoch": 195.61111111111111,
      "grad_norm": 0.0012054787948727608,
      "learning_rate": 2.9676817767571086e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4929040,
      "step": 17605
    },
    {
      "epoch": 195.66666666666666,
      "grad_norm": 0.008541454561054707,
      "learning_rate": 2.966717324721861e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4930432,
      "step": 17610
    },
    {
      "epoch": 195.72222222222223,
      "grad_norm": 0.0008024678681977093,
      "learning_rate": 2.9657528007129366e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4931840,
      "step": 17615
    },
    {
      "epoch": 195.77777777777777,
      "grad_norm": 0.0031168474815785885,
      "learning_rate": 2.9647882048790777e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4933216,
      "step": 17620
    },
    {
      "epoch": 195.83333333333334,
      "grad_norm": 0.0003132623096462339,
      "learning_rate": 2.963823537369037e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4934576,
      "step": 17625
    },
    {
      "epoch": 195.88888888888889,
      "grad_norm": 0.0005395126645453274,
      "learning_rate": 2.9628587983315775e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 4936016,
      "step": 17630
    },
    {
      "epoch": 195.94444444444446,
      "grad_norm": 0.0017081745900213718,
      "learning_rate": 2.9618939879154746e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4937424,
      "step": 17635
    },
    {
      "epoch": 196.0,
      "grad_norm": 0.008828825317323208,
      "learning_rate": 2.9609291062695143e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4938816,
      "step": 17640
    },
    {
      "epoch": 196.05555555555554,
      "grad_norm": 0.011091477237641811,
      "learning_rate": 2.9599641535424938e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4940224,
      "step": 17645
    },
    {
      "epoch": 196.11111111111111,
      "grad_norm": 0.012488470412790775,
      "learning_rate": 2.9589991298832202e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4941584,
      "step": 17650
    },
    {
      "epoch": 196.16666666666666,
      "grad_norm": 0.0008265786455012858,
      "learning_rate": 2.958034035440513e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4942944,
      "step": 17655
    },
    {
      "epoch": 196.22222222222223,
      "grad_norm": 0.0004136937204748392,
      "learning_rate": 2.957068870363201e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4944384,
      "step": 17660
    },
    {
      "epoch": 196.27777777777777,
      "grad_norm": 0.012245829217135906,
      "learning_rate": 2.956103634800126e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4945792,
      "step": 17665
    },
    {
      "epoch": 196.33333333333334,
      "grad_norm": 0.0006817406974732876,
      "learning_rate": 2.9551383289001384e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 4947200,
      "step": 17670
    },
    {
      "epoch": 196.38888888888889,
      "grad_norm": 0.024983759969472885,
      "learning_rate": 2.9541729528121005e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4948608,
      "step": 17675
    },
    {
      "epoch": 196.44444444444446,
      "grad_norm": 0.0006883350433781743,
      "learning_rate": 2.9532075066848856e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4949984,
      "step": 17680
    },
    {
      "epoch": 196.5,
      "grad_norm": 0.003624364035204053,
      "learning_rate": 2.9522419906673786e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4951328,
      "step": 17685
    },
    {
      "epoch": 196.55555555555554,
      "grad_norm": 0.0018505678744986653,
      "learning_rate": 2.951276404908474e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4952720,
      "step": 17690
    },
    {
      "epoch": 196.61111111111111,
      "grad_norm": 0.009058446623384953,
      "learning_rate": 2.9503107495570752e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4954160,
      "step": 17695
    },
    {
      "epoch": 196.66666666666666,
      "grad_norm": 0.0022728932090103626,
      "learning_rate": 2.9493450247621003e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4955568,
      "step": 17700
    },
    {
      "epoch": 196.72222222222223,
      "grad_norm": 0.04796449467539787,
      "learning_rate": 2.948379230672476e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4956944,
      "step": 17705
    },
    {
      "epoch": 196.77777777777777,
      "grad_norm": 0.3959101438522339,
      "learning_rate": 2.9474133674371396e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4958288,
      "step": 17710
    },
    {
      "epoch": 196.83333333333334,
      "grad_norm": 0.235774964094162,
      "learning_rate": 2.9464474352050387e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 4959616,
      "step": 17715
    },
    {
      "epoch": 196.88888888888889,
      "grad_norm": 0.004563373979181051,
      "learning_rate": 2.9454814341251336e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4961024,
      "step": 17720
    },
    {
      "epoch": 196.94444444444446,
      "grad_norm": 0.020605193451046944,
      "learning_rate": 2.9445153643463942e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4962448,
      "step": 17725
    },
    {
      "epoch": 197.0,
      "grad_norm": 0.21941669285297394,
      "learning_rate": 2.943549226017798e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4963872,
      "step": 17730
    },
    {
      "epoch": 197.05555555555554,
      "grad_norm": 0.029919195920228958,
      "learning_rate": 2.942583019288337e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 4965280,
      "step": 17735
    },
    {
      "epoch": 197.11111111111111,
      "grad_norm": 0.002114388160407543,
      "learning_rate": 2.9416167443070132e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4966688,
      "step": 17740
    },
    {
      "epoch": 197.16666666666666,
      "grad_norm": 0.013762388378381729,
      "learning_rate": 2.9406504012228375e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4968064,
      "step": 17745
    },
    {
      "epoch": 197.22222222222223,
      "grad_norm": 0.0005309412372298539,
      "learning_rate": 2.939683990184832e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4969408,
      "step": 17750
    },
    {
      "epoch": 197.27777777777777,
      "grad_norm": 0.022517520934343338,
      "learning_rate": 2.93871751134203e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4970752,
      "step": 17755
    },
    {
      "epoch": 197.33333333333334,
      "grad_norm": 0.0010879721958190203,
      "learning_rate": 2.9377509648434752e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4972176,
      "step": 17760
    },
    {
      "epoch": 197.38888888888889,
      "grad_norm": 0.0008269868558272719,
      "learning_rate": 2.9367843508382203e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4973600,
      "step": 17765
    },
    {
      "epoch": 197.44444444444446,
      "grad_norm": 0.015279220417141914,
      "learning_rate": 2.9358176694753293e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4974992,
      "step": 17770
    },
    {
      "epoch": 197.5,
      "grad_norm": 0.00831028912216425,
      "learning_rate": 2.9348509209038766e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 4976384,
      "step": 17775
    },
    {
      "epoch": 197.55555555555554,
      "grad_norm": 0.0017769744154065847,
      "learning_rate": 2.933884105272947e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4977808,
      "step": 17780
    },
    {
      "epoch": 197.61111111111111,
      "grad_norm": 0.0011706138029694557,
      "learning_rate": 2.9329172227316366e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4979216,
      "step": 17785
    },
    {
      "epoch": 197.66666666666666,
      "grad_norm": 0.003183252876624465,
      "learning_rate": 2.93195027342905e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4980592,
      "step": 17790
    },
    {
      "epoch": 197.72222222222223,
      "grad_norm": 0.00706462049856782,
      "learning_rate": 2.9309832575143024e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4982016,
      "step": 17795
    },
    {
      "epoch": 197.77777777777777,
      "grad_norm": 0.05480446666479111,
      "learning_rate": 2.930016175136521e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4983424,
      "step": 17800
    },
    {
      "epoch": 197.77777777777777,
      "eval_loss": 0.6280994415283203,
      "eval_runtime": 1.2356,
      "eval_samples_per_second": 32.373,
      "eval_steps_per_second": 16.187,
      "num_input_tokens_seen": 4983424,
      "step": 17800
    },
    {
      "epoch": 197.83333333333334,
      "grad_norm": 0.006793640088289976,
      "learning_rate": 2.9290490264448412e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4984816,
      "step": 17805
    },
    {
      "epoch": 197.88888888888889,
      "grad_norm": 0.008122761733829975,
      "learning_rate": 2.9280818115884094e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4986240,
      "step": 17810
    },
    {
      "epoch": 197.94444444444446,
      "grad_norm": 0.002008171286433935,
      "learning_rate": 2.9271145307163828e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 4987664,
      "step": 17815
    },
    {
      "epoch": 198.0,
      "grad_norm": 0.0007080535287968814,
      "learning_rate": 2.9261471839779287e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4989088,
      "step": 17820
    },
    {
      "epoch": 198.05555555555554,
      "grad_norm": 0.052896060049533844,
      "learning_rate": 2.925179771522223e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4990496,
      "step": 17825
    },
    {
      "epoch": 198.11111111111111,
      "grad_norm": 0.021151285618543625,
      "learning_rate": 2.9242122934984535e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 4991888,
      "step": 17830
    },
    {
      "epoch": 198.16666666666666,
      "grad_norm": 0.013008644804358482,
      "learning_rate": 2.9232447500558176e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4993280,
      "step": 17835
    },
    {
      "epoch": 198.22222222222223,
      "grad_norm": 0.0005825664848089218,
      "learning_rate": 2.9222771413435225e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4994688,
      "step": 17840
    },
    {
      "epoch": 198.27777777777777,
      "grad_norm": 0.0006793682114221156,
      "learning_rate": 2.9213094675107848e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4996112,
      "step": 17845
    },
    {
      "epoch": 198.33333333333334,
      "grad_norm": 0.0012154027353972197,
      "learning_rate": 2.9203417287068335e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 4997520,
      "step": 17850
    },
    {
      "epoch": 198.38888888888889,
      "grad_norm": 0.0012099179439246655,
      "learning_rate": 2.9193739250809042e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 4998928,
      "step": 17855
    },
    {
      "epoch": 198.44444444444446,
      "grad_norm": 0.06409447640180588,
      "learning_rate": 2.9184060567822463e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5000288,
      "step": 17860
    },
    {
      "epoch": 198.5,
      "grad_norm": 0.005899438168853521,
      "learning_rate": 2.9174381239601166e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 5001712,
      "step": 17865
    },
    {
      "epoch": 198.55555555555554,
      "grad_norm": 0.0016274627996608615,
      "learning_rate": 2.916470126763783e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5003104,
      "step": 17870
    },
    {
      "epoch": 198.61111111111111,
      "grad_norm": 0.0008924803696572781,
      "learning_rate": 2.9155020653425203e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5004496,
      "step": 17875
    },
    {
      "epoch": 198.66666666666666,
      "grad_norm": 0.0006649914430454373,
      "learning_rate": 2.9145339398456184e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5005936,
      "step": 17880
    },
    {
      "epoch": 198.72222222222223,
      "grad_norm": 0.010771028697490692,
      "learning_rate": 2.913565750422374e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5007328,
      "step": 17885
    },
    {
      "epoch": 198.77777777777777,
      "grad_norm": 0.0036351503804326057,
      "learning_rate": 2.9125974972220938e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5008720,
      "step": 17890
    },
    {
      "epoch": 198.83333333333334,
      "grad_norm": 0.003074594773352146,
      "learning_rate": 2.9116291803940932e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5010112,
      "step": 17895
    },
    {
      "epoch": 198.88888888888889,
      "grad_norm": 0.00288903946056962,
      "learning_rate": 2.910660800087701e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5011520,
      "step": 17900
    },
    {
      "epoch": 198.94444444444446,
      "grad_norm": 0.01080932468175888,
      "learning_rate": 2.909692356452254e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5012960,
      "step": 17905
    },
    {
      "epoch": 199.0,
      "grad_norm": 0.018417468294501305,
      "learning_rate": 2.9087238496370962e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5014352,
      "step": 17910
    },
    {
      "epoch": 199.05555555555554,
      "grad_norm": 0.019822843372821808,
      "learning_rate": 2.907755279791583e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 5015728,
      "step": 17915
    },
    {
      "epoch": 199.11111111111111,
      "grad_norm": 0.0017486449796706438,
      "learning_rate": 2.906786647065083e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5017152,
      "step": 17920
    },
    {
      "epoch": 199.16666666666666,
      "grad_norm": 0.005522097926586866,
      "learning_rate": 2.9058179516069695e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5018560,
      "step": 17925
    },
    {
      "epoch": 199.22222222222223,
      "grad_norm": 0.02487112022936344,
      "learning_rate": 2.9048491935666282e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5019936,
      "step": 17930
    },
    {
      "epoch": 199.27777777777777,
      "grad_norm": 0.006771773099899292,
      "learning_rate": 2.9038803730934534e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5021344,
      "step": 17935
    },
    {
      "epoch": 199.33333333333334,
      "grad_norm": 0.011191973462700844,
      "learning_rate": 2.9029114903368503e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 5022704,
      "step": 17940
    },
    {
      "epoch": 199.38888888888889,
      "grad_norm": 0.0014036801876500249,
      "learning_rate": 2.9019425454462318e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5024112,
      "step": 17945
    },
    {
      "epoch": 199.44444444444446,
      "grad_norm": 0.00033177912700921297,
      "learning_rate": 2.9009735385710212e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5025504,
      "step": 17950
    },
    {
      "epoch": 199.5,
      "grad_norm": 0.0004240658017806709,
      "learning_rate": 2.900004469860652e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5026912,
      "step": 17955
    },
    {
      "epoch": 199.55555555555554,
      "grad_norm": 1.4772366285324097,
      "learning_rate": 2.8990353394645668e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 5028304,
      "step": 17960
    },
    {
      "epoch": 199.61111111111111,
      "grad_norm": 0.0031349665950983763,
      "learning_rate": 2.8980661475322186e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5029728,
      "step": 17965
    },
    {
      "epoch": 199.66666666666666,
      "grad_norm": 0.0025368407368659973,
      "learning_rate": 2.897096894213067e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5031120,
      "step": 17970
    },
    {
      "epoch": 199.72222222222223,
      "grad_norm": 0.045496921986341476,
      "learning_rate": 2.8961275796565845e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5032512,
      "step": 17975
    },
    {
      "epoch": 199.77777777777777,
      "grad_norm": 0.0015880228020250797,
      "learning_rate": 2.8951582040122517e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5033920,
      "step": 17980
    },
    {
      "epoch": 199.83333333333334,
      "grad_norm": 0.09198308736085892,
      "learning_rate": 2.894188767429557e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5035312,
      "step": 17985
    },
    {
      "epoch": 199.88888888888889,
      "grad_norm": 0.005021082237362862,
      "learning_rate": 2.8932192700580014e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5036720,
      "step": 17990
    },
    {
      "epoch": 199.94444444444446,
      "grad_norm": 0.00446079857647419,
      "learning_rate": 2.8922497120470916e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5038112,
      "step": 17995
    },
    {
      "epoch": 200.0,
      "grad_norm": 0.001583451172336936,
      "learning_rate": 2.891280093546348e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5039536,
      "step": 18000
    },
    {
      "epoch": 200.0,
      "eval_loss": 0.6263655424118042,
      "eval_runtime": 1.2375,
      "eval_samples_per_second": 32.324,
      "eval_steps_per_second": 16.162,
      "num_input_tokens_seen": 5039536,
      "step": 18000
    },
    {
      "epoch": 200.05555555555554,
      "grad_norm": 0.0009574030409567058,
      "learning_rate": 2.890310414705297e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5040944,
      "step": 18005
    },
    {
      "epoch": 200.11111111111111,
      "grad_norm": 0.0003786046872846782,
      "learning_rate": 2.8893406756734742e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5042320,
      "step": 18010
    },
    {
      "epoch": 200.16666666666666,
      "grad_norm": 0.002492388477548957,
      "learning_rate": 2.888370876600427e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5043728,
      "step": 18015
    },
    {
      "epoch": 200.22222222222223,
      "grad_norm": 0.015256421640515327,
      "learning_rate": 2.8874010176357104e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5045104,
      "step": 18020
    },
    {
      "epoch": 200.27777777777777,
      "grad_norm": 0.02849498949944973,
      "learning_rate": 2.886431098928888e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5046512,
      "step": 18025
    },
    {
      "epoch": 200.33333333333334,
      "grad_norm": 0.0009398062247782946,
      "learning_rate": 2.885461120629534e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5047920,
      "step": 18030
    },
    {
      "epoch": 200.38888888888889,
      "grad_norm": 0.0008240463794209063,
      "learning_rate": 2.8844910828872317e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5049328,
      "step": 18035
    },
    {
      "epoch": 200.44444444444446,
      "grad_norm": 0.0020801422651857138,
      "learning_rate": 2.8835209858515715e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5050688,
      "step": 18040
    },
    {
      "epoch": 200.5,
      "grad_norm": 0.011464489623904228,
      "learning_rate": 2.8825508296721566e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5052080,
      "step": 18045
    },
    {
      "epoch": 200.55555555555554,
      "grad_norm": 0.15411420166492462,
      "learning_rate": 2.881580614498596e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5053504,
      "step": 18050
    },
    {
      "epoch": 200.61111111111111,
      "grad_norm": 0.0007736906409263611,
      "learning_rate": 2.8806103404805103e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5054864,
      "step": 18055
    },
    {
      "epoch": 200.66666666666666,
      "grad_norm": 0.007753114681690931,
      "learning_rate": 2.8796400077675257e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5056224,
      "step": 18060
    },
    {
      "epoch": 200.72222222222223,
      "grad_norm": 0.01204466912895441,
      "learning_rate": 2.8786696165092812e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5057600,
      "step": 18065
    },
    {
      "epoch": 200.77777777777777,
      "grad_norm": 0.0010103709064424038,
      "learning_rate": 2.8776991668554236e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5058992,
      "step": 18070
    },
    {
      "epoch": 200.83333333333334,
      "grad_norm": 0.00036953308153897524,
      "learning_rate": 2.876728658955608e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5060432,
      "step": 18075
    },
    {
      "epoch": 200.88888888888889,
      "grad_norm": 0.007732648868113756,
      "learning_rate": 2.8757580929594986e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5061872,
      "step": 18080
    },
    {
      "epoch": 200.94444444444446,
      "grad_norm": 0.006935234647244215,
      "learning_rate": 2.87478746901677e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 5063264,
      "step": 18085
    },
    {
      "epoch": 201.0,
      "grad_norm": 0.0028261649422347546,
      "learning_rate": 2.873816787277103e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5064672,
      "step": 18090
    },
    {
      "epoch": 201.05555555555554,
      "grad_norm": 0.010720157995820045,
      "learning_rate": 2.8728460478901903e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5066080,
      "step": 18095
    },
    {
      "epoch": 201.11111111111111,
      "grad_norm": 0.5185559988021851,
      "learning_rate": 2.8718752510057307e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5067424,
      "step": 18100
    },
    {
      "epoch": 201.16666666666666,
      "grad_norm": 0.002573711099103093,
      "learning_rate": 2.870904396773435e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5068816,
      "step": 18105
    },
    {
      "epoch": 201.22222222222223,
      "grad_norm": 0.17235499620437622,
      "learning_rate": 2.86993348534302e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 5070192,
      "step": 18110
    },
    {
      "epoch": 201.27777777777777,
      "grad_norm": 0.013529950752854347,
      "learning_rate": 2.868962516864212e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5071600,
      "step": 18115
    },
    {
      "epoch": 201.33333333333334,
      "grad_norm": 0.4194115102291107,
      "learning_rate": 2.8679914914867477e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5072992,
      "step": 18120
    },
    {
      "epoch": 201.38888888888889,
      "grad_norm": 0.0008040106622502208,
      "learning_rate": 2.8670204093603713e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5074368,
      "step": 18125
    },
    {
      "epoch": 201.44444444444446,
      "grad_norm": 0.001818204764276743,
      "learning_rate": 2.8660492706348357e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5075744,
      "step": 18130
    },
    {
      "epoch": 201.5,
      "grad_norm": 0.002959734294563532,
      "learning_rate": 2.8650780754599022e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5077168,
      "step": 18135
    },
    {
      "epoch": 201.55555555555554,
      "grad_norm": 0.0021095164120197296,
      "learning_rate": 2.8641068239853407e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5078560,
      "step": 18140
    },
    {
      "epoch": 201.61111111111111,
      "grad_norm": 0.0014778096228837967,
      "learning_rate": 2.863135516360932e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5079968,
      "step": 18145
    },
    {
      "epoch": 201.66666666666666,
      "grad_norm": 0.0016323307063430548,
      "learning_rate": 2.8621641527364633e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5081392,
      "step": 18150
    },
    {
      "epoch": 201.72222222222223,
      "grad_norm": 0.031659118831157684,
      "learning_rate": 2.8611927332617313e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5082784,
      "step": 18155
    },
    {
      "epoch": 201.77777777777777,
      "grad_norm": 0.0034357192926108837,
      "learning_rate": 2.8602212580865405e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5084192,
      "step": 18160
    },
    {
      "epoch": 201.83333333333334,
      "grad_norm": 0.0004938651691190898,
      "learning_rate": 2.859249727360705e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5085584,
      "step": 18165
    },
    {
      "epoch": 201.88888888888889,
      "grad_norm": 0.006087636109441519,
      "learning_rate": 2.8582781412340465e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5086992,
      "step": 18170
    },
    {
      "epoch": 201.94444444444446,
      "grad_norm": 0.001812625559978187,
      "learning_rate": 2.857306499856397e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5088448,
      "step": 18175
    },
    {
      "epoch": 202.0,
      "grad_norm": 0.0018244351958855987,
      "learning_rate": 2.856334803377594e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5089808,
      "step": 18180
    },
    {
      "epoch": 202.05555555555554,
      "grad_norm": 0.01839066669344902,
      "learning_rate": 2.8553630519474867e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5091152,
      "step": 18185
    },
    {
      "epoch": 202.11111111111111,
      "grad_norm": 0.00039451144402846694,
      "learning_rate": 2.8543912457159317e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5092544,
      "step": 18190
    },
    {
      "epoch": 202.16666666666666,
      "grad_norm": 0.004601118620485067,
      "learning_rate": 2.853419384832792e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5093968,
      "step": 18195
    },
    {
      "epoch": 202.22222222222223,
      "grad_norm": 0.025015516206622124,
      "learning_rate": 2.8524474694479423e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5095376,
      "step": 18200
    },
    {
      "epoch": 202.22222222222223,
      "eval_loss": 0.6402644515037537,
      "eval_runtime": 1.2434,
      "eval_samples_per_second": 32.17,
      "eval_steps_per_second": 16.085,
      "num_input_tokens_seen": 5095376,
      "step": 18200
    },
    {
      "epoch": 202.27777777777777,
      "grad_norm": 0.003412172431126237,
      "learning_rate": 2.851475499711264e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5096784,
      "step": 18205
    },
    {
      "epoch": 202.33333333333334,
      "grad_norm": 0.0003656020271591842,
      "learning_rate": 2.8505034757726468e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5098160,
      "step": 18210
    },
    {
      "epoch": 202.38888888888889,
      "grad_norm": 0.0009504305198788643,
      "learning_rate": 2.8495313977819886e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5099600,
      "step": 18215
    },
    {
      "epoch": 202.44444444444446,
      "grad_norm": 0.005396148189902306,
      "learning_rate": 2.8485592658891956e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5101008,
      "step": 18220
    },
    {
      "epoch": 202.5,
      "grad_norm": 0.017163896933197975,
      "learning_rate": 2.8475870802441844e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5102464,
      "step": 18225
    },
    {
      "epoch": 202.55555555555554,
      "grad_norm": 0.006014075130224228,
      "learning_rate": 2.8466148409968774e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5103904,
      "step": 18230
    },
    {
      "epoch": 202.61111111111111,
      "grad_norm": 0.001113795442506671,
      "learning_rate": 2.8456425482972067e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5105312,
      "step": 18235
    },
    {
      "epoch": 202.66666666666666,
      "grad_norm": 0.00790405459702015,
      "learning_rate": 2.84467020229511e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5106704,
      "step": 18240
    },
    {
      "epoch": 202.72222222222223,
      "grad_norm": 0.04096580669283867,
      "learning_rate": 2.8436978031405375e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5108064,
      "step": 18245
    },
    {
      "epoch": 202.77777777777777,
      "grad_norm": 0.0010573806939646602,
      "learning_rate": 2.842725350983445e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5109472,
      "step": 18250
    },
    {
      "epoch": 202.83333333333334,
      "grad_norm": 0.015302878804504871,
      "learning_rate": 2.8417528459737957e-05,
      "loss": 0.0009,
      "num_input_tokens_seen": 5110864,
      "step": 18255
    },
    {
      "epoch": 202.88888888888889,
      "grad_norm": 0.006949119735509157,
      "learning_rate": 2.8407802882615624e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5112224,
      "step": 18260
    },
    {
      "epoch": 202.94444444444446,
      "grad_norm": 0.18251284956932068,
      "learning_rate": 2.8398076779967277e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5113568,
      "step": 18265
    },
    {
      "epoch": 203.0,
      "grad_norm": 0.001961400965228677,
      "learning_rate": 2.8388350153292774e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5114960,
      "step": 18270
    },
    {
      "epoch": 203.05555555555554,
      "grad_norm": 0.0033372591715306044,
      "learning_rate": 2.8378623004092103e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5116320,
      "step": 18275
    },
    {
      "epoch": 203.11111111111111,
      "grad_norm": 0.0003846223989967257,
      "learning_rate": 2.8368895333865302e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5117728,
      "step": 18280
    },
    {
      "epoch": 203.16666666666666,
      "grad_norm": 0.0005424197297543287,
      "learning_rate": 2.835916714411251e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5119168,
      "step": 18285
    },
    {
      "epoch": 203.22222222222223,
      "grad_norm": 0.001958522479981184,
      "learning_rate": 2.8349438436333926e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5120608,
      "step": 18290
    },
    {
      "epoch": 203.27777777777777,
      "grad_norm": 0.0011852108873426914,
      "learning_rate": 2.833970921202984e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5122000,
      "step": 18295
    },
    {
      "epoch": 203.33333333333334,
      "grad_norm": 0.001074046827852726,
      "learning_rate": 2.8329979472700628e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5123424,
      "step": 18300
    },
    {
      "epoch": 203.38888888888889,
      "grad_norm": 0.0002619110746309161,
      "learning_rate": 2.832024921984674e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5124864,
      "step": 18305
    },
    {
      "epoch": 203.44444444444446,
      "grad_norm": 0.0009242625092156231,
      "learning_rate": 2.8310518454968693e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5126272,
      "step": 18310
    },
    {
      "epoch": 203.5,
      "grad_norm": 0.017728235572576523,
      "learning_rate": 2.8300787179567095e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5127680,
      "step": 18315
    },
    {
      "epoch": 203.55555555555554,
      "grad_norm": 0.020962031558156013,
      "learning_rate": 2.8291055395142636e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5129104,
      "step": 18320
    },
    {
      "epoch": 203.61111111111111,
      "grad_norm": 0.31626951694488525,
      "learning_rate": 2.8281323103196073e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 5130496,
      "step": 18325
    },
    {
      "epoch": 203.66666666666666,
      "grad_norm": 0.001371819875203073,
      "learning_rate": 2.8271590305228256e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5131904,
      "step": 18330
    },
    {
      "epoch": 203.72222222222223,
      "grad_norm": 0.0017916315700858831,
      "learning_rate": 2.82618570027401e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5133312,
      "step": 18335
    },
    {
      "epoch": 203.77777777777777,
      "grad_norm": 0.007269116584211588,
      "learning_rate": 2.8252123197232604e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5134704,
      "step": 18340
    },
    {
      "epoch": 203.83333333333334,
      "grad_norm": 0.0019243641290813684,
      "learning_rate": 2.8242388890206843e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 5136096,
      "step": 18345
    },
    {
      "epoch": 203.88888888888889,
      "grad_norm": 0.0034110089763998985,
      "learning_rate": 2.8232654083163967e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5137472,
      "step": 18350
    },
    {
      "epoch": 203.94444444444446,
      "grad_norm": 0.0007355144480243325,
      "learning_rate": 2.822291877760521e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5138896,
      "step": 18355
    },
    {
      "epoch": 204.0,
      "grad_norm": 0.003370251040905714,
      "learning_rate": 2.8213182975031864e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5140272,
      "step": 18360
    },
    {
      "epoch": 204.05555555555554,
      "grad_norm": 0.005067979916930199,
      "learning_rate": 2.8203446676945337e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5141616,
      "step": 18365
    },
    {
      "epoch": 204.11111111111111,
      "grad_norm": 0.18643489480018616,
      "learning_rate": 2.8193709884847075e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 5142976,
      "step": 18370
    },
    {
      "epoch": 204.16666666666666,
      "grad_norm": 0.005866559688001871,
      "learning_rate": 2.8183972600238605e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5144400,
      "step": 18375
    },
    {
      "epoch": 204.22222222222223,
      "grad_norm": 0.0020323130302131176,
      "learning_rate": 2.817423482462156e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5145808,
      "step": 18380
    },
    {
      "epoch": 204.27777777777777,
      "grad_norm": 0.020987408235669136,
      "learning_rate": 2.8164496559497605e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5147216,
      "step": 18385
    },
    {
      "epoch": 204.33333333333334,
      "grad_norm": 0.008186676539480686,
      "learning_rate": 2.815475780636852e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5148624,
      "step": 18390
    },
    {
      "epoch": 204.38888888888889,
      "grad_norm": 0.002305281348526478,
      "learning_rate": 2.814501856673613e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5150032,
      "step": 18395
    },
    {
      "epoch": 204.44444444444446,
      "grad_norm": 0.0007336735725402832,
      "learning_rate": 2.8135278842102353e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5151440,
      "step": 18400
    },
    {
      "epoch": 204.44444444444446,
      "eval_loss": 0.6769424676895142,
      "eval_runtime": 1.2387,
      "eval_samples_per_second": 32.292,
      "eval_steps_per_second": 16.146,
      "num_input_tokens_seen": 5151440,
      "step": 18400
    },
    {
      "epoch": 204.5,
      "grad_norm": 0.00037729216273874044,
      "learning_rate": 2.8125538633969183e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5152864,
      "step": 18405
    },
    {
      "epoch": 204.55555555555554,
      "grad_norm": 0.01430877298116684,
      "learning_rate": 2.8115797943838677e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5154288,
      "step": 18410
    },
    {
      "epoch": 204.61111111111111,
      "grad_norm": 0.0004765752237290144,
      "learning_rate": 2.810605677321298e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5155664,
      "step": 18415
    },
    {
      "epoch": 204.66666666666666,
      "grad_norm": 0.00016226473962888122,
      "learning_rate": 2.809631512359428e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5157040,
      "step": 18420
    },
    {
      "epoch": 204.72222222222223,
      "grad_norm": 0.004809063859283924,
      "learning_rate": 2.8086572996484884e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5158464,
      "step": 18425
    },
    {
      "epoch": 204.77777777777777,
      "grad_norm": 0.005845131352543831,
      "learning_rate": 2.8076830393387143e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5159872,
      "step": 18430
    },
    {
      "epoch": 204.83333333333334,
      "grad_norm": 0.002131752436980605,
      "learning_rate": 2.8067087315803497e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5161280,
      "step": 18435
    },
    {
      "epoch": 204.88888888888889,
      "grad_norm": 0.0003574448055587709,
      "learning_rate": 2.8057343765236433e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5162656,
      "step": 18440
    },
    {
      "epoch": 204.94444444444446,
      "grad_norm": 0.004024159163236618,
      "learning_rate": 2.804759974318854e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5164064,
      "step": 18445
    },
    {
      "epoch": 205.0,
      "grad_norm": 0.0006895678816363215,
      "learning_rate": 2.8037855251162482e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5165456,
      "step": 18450
    },
    {
      "epoch": 205.05555555555554,
      "grad_norm": 0.006115995813161135,
      "learning_rate": 2.802811029066096e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5166880,
      "step": 18455
    },
    {
      "epoch": 205.11111111111111,
      "grad_norm": 0.00044963633990846574,
      "learning_rate": 2.8018364863186764e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5168224,
      "step": 18460
    },
    {
      "epoch": 205.16666666666666,
      "grad_norm": 0.0033260739874094725,
      "learning_rate": 2.800861897024279e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5169664,
      "step": 18465
    },
    {
      "epoch": 205.22222222222223,
      "grad_norm": 0.006484187673777342,
      "learning_rate": 2.799887261333196e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5171136,
      "step": 18470
    },
    {
      "epoch": 205.27777777777777,
      "grad_norm": 0.00210802280344069,
      "learning_rate": 2.798912579395728e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5172544,
      "step": 18475
    },
    {
      "epoch": 205.33333333333334,
      "grad_norm": 0.0013227264862507582,
      "learning_rate": 2.797937851362185e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5173920,
      "step": 18480
    },
    {
      "epoch": 205.38888888888889,
      "grad_norm": 0.011144885793328285,
      "learning_rate": 2.7969630773828802e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5175312,
      "step": 18485
    },
    {
      "epoch": 205.44444444444446,
      "grad_norm": 0.0020640548318624496,
      "learning_rate": 2.7959882576081382e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5176736,
      "step": 18490
    },
    {
      "epoch": 205.5,
      "grad_norm": 0.0024939514696598053,
      "learning_rate": 2.795013392188286e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 5178128,
      "step": 18495
    },
    {
      "epoch": 205.55555555555554,
      "grad_norm": 0.021290766075253487,
      "learning_rate": 2.7940384812736614e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5179520,
      "step": 18500
    },
    {
      "epoch": 205.61111111111111,
      "grad_norm": 0.01199457049369812,
      "learning_rate": 2.7930635250146087e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5180928,
      "step": 18505
    },
    {
      "epoch": 205.66666666666666,
      "grad_norm": 0.004877686500549316,
      "learning_rate": 2.792088523561477e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5182272,
      "step": 18510
    },
    {
      "epoch": 205.72222222222223,
      "grad_norm": 0.001601176569238305,
      "learning_rate": 2.7911134770646246e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5183680,
      "step": 18515
    },
    {
      "epoch": 205.77777777777777,
      "grad_norm": 0.0047949873842298985,
      "learning_rate": 2.7901383856744157e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5185088,
      "step": 18520
    },
    {
      "epoch": 205.83333333333334,
      "grad_norm": 0.0016275065718218684,
      "learning_rate": 2.7891632495412217e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5186464,
      "step": 18525
    },
    {
      "epoch": 205.88888888888889,
      "grad_norm": 0.010919715277850628,
      "learning_rate": 2.7881880688154205e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5187840,
      "step": 18530
    },
    {
      "epoch": 205.94444444444446,
      "grad_norm": 0.009458241052925587,
      "learning_rate": 2.7872128436473977e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5189248,
      "step": 18535
    },
    {
      "epoch": 206.0,
      "grad_norm": 0.007887046784162521,
      "learning_rate": 2.7862375741875448e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5190656,
      "step": 18540
    },
    {
      "epoch": 206.05555555555554,
      "grad_norm": 0.3920457661151886,
      "learning_rate": 2.785262260586261e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 5192048,
      "step": 18545
    },
    {
      "epoch": 206.11111111111111,
      "grad_norm": 0.01992938667535782,
      "learning_rate": 2.7842869029939517e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5193440,
      "step": 18550
    },
    {
      "epoch": 206.16666666666666,
      "grad_norm": 0.01529917772859335,
      "learning_rate": 2.7833115015610296e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5194848,
      "step": 18555
    },
    {
      "epoch": 206.22222222222223,
      "grad_norm": 0.08143259584903717,
      "learning_rate": 2.7823360564379136e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5196240,
      "step": 18560
    },
    {
      "epoch": 206.27777777777777,
      "grad_norm": 0.00012749053712468594,
      "learning_rate": 2.7813605677750297e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5197664,
      "step": 18565
    },
    {
      "epoch": 206.33333333333334,
      "grad_norm": 0.004145378712564707,
      "learning_rate": 2.7803850357228102e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5199056,
      "step": 18570
    },
    {
      "epoch": 206.38888888888889,
      "grad_norm": 0.0011411127634346485,
      "learning_rate": 2.779409460431695e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5200480,
      "step": 18575
    },
    {
      "epoch": 206.44444444444446,
      "grad_norm": 0.06931652128696442,
      "learning_rate": 2.778433842052129e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5201856,
      "step": 18580
    },
    {
      "epoch": 206.5,
      "grad_norm": 0.0047182426787912846,
      "learning_rate": 2.7774581807345664e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5203264,
      "step": 18585
    },
    {
      "epoch": 206.55555555555554,
      "grad_norm": 0.0033846807200461626,
      "learning_rate": 2.776482476629465e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5204672,
      "step": 18590
    },
    {
      "epoch": 206.61111111111111,
      "grad_norm": 0.0035316189751029015,
      "learning_rate": 2.7755067298872924e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5206096,
      "step": 18595
    },
    {
      "epoch": 206.66666666666666,
      "grad_norm": 0.003996088169515133,
      "learning_rate": 2.774530940658518e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5207488,
      "step": 18600
    },
    {
      "epoch": 206.66666666666666,
      "eval_loss": 0.6296755075454712,
      "eval_runtime": 1.2358,
      "eval_samples_per_second": 32.369,
      "eval_steps_per_second": 16.184,
      "num_input_tokens_seen": 5207488,
      "step": 18600
    },
    {
      "epoch": 206.72222222222223,
      "grad_norm": 0.0015808386960998178,
      "learning_rate": 2.7735551090936236e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5208864,
      "step": 18605
    },
    {
      "epoch": 206.77777777777777,
      "grad_norm": 0.0016487984685227275,
      "learning_rate": 2.7725792353430934e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5210288,
      "step": 18610
    },
    {
      "epoch": 206.83333333333334,
      "grad_norm": 0.0034803987946361303,
      "learning_rate": 2.77160331955742e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5211680,
      "step": 18615
    },
    {
      "epoch": 206.88888888888889,
      "grad_norm": 0.0003224753600079566,
      "learning_rate": 2.7706273618871008e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5213072,
      "step": 18620
    },
    {
      "epoch": 206.94444444444446,
      "grad_norm": 0.019706713035702705,
      "learning_rate": 2.769651362482642e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5214464,
      "step": 18625
    },
    {
      "epoch": 207.0,
      "grad_norm": 0.012039609253406525,
      "learning_rate": 2.768675321494555e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5215840,
      "step": 18630
    },
    {
      "epoch": 207.05555555555554,
      "grad_norm": 0.004539318848401308,
      "learning_rate": 2.7676992390733565e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5217232,
      "step": 18635
    },
    {
      "epoch": 207.11111111111111,
      "grad_norm": 0.0036730028223246336,
      "learning_rate": 2.766723115369571e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5218656,
      "step": 18640
    },
    {
      "epoch": 207.16666666666666,
      "grad_norm": 0.00832141749560833,
      "learning_rate": 2.765746950533729e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5220032,
      "step": 18645
    },
    {
      "epoch": 207.22222222222223,
      "grad_norm": 0.00024892642977647483,
      "learning_rate": 2.7647707447163684e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5221424,
      "step": 18650
    },
    {
      "epoch": 207.27777777777777,
      "grad_norm": 0.0003199031634721905,
      "learning_rate": 2.7637944980680315e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5222880,
      "step": 18655
    },
    {
      "epoch": 207.33333333333334,
      "grad_norm": 0.0004773080290760845,
      "learning_rate": 2.762818210739268e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5224256,
      "step": 18660
    },
    {
      "epoch": 207.38888888888889,
      "grad_norm": 0.0006600763881579041,
      "learning_rate": 2.7618418828806332e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5225664,
      "step": 18665
    },
    {
      "epoch": 207.44444444444446,
      "grad_norm": 0.000590238138101995,
      "learning_rate": 2.76086551464269e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5227040,
      "step": 18670
    },
    {
      "epoch": 207.5,
      "grad_norm": 0.0020987235475331545,
      "learning_rate": 2.759889106176006e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5228480,
      "step": 18675
    },
    {
      "epoch": 207.55555555555554,
      "grad_norm": 0.0019797387067228556,
      "learning_rate": 2.758912657631156e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5229888,
      "step": 18680
    },
    {
      "epoch": 207.61111111111111,
      "grad_norm": 0.0012343795970082283,
      "learning_rate": 2.7579361691587198e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5231296,
      "step": 18685
    },
    {
      "epoch": 207.66666666666666,
      "grad_norm": 0.00026556261582300067,
      "learning_rate": 2.756959640909285e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5232688,
      "step": 18690
    },
    {
      "epoch": 207.72222222222223,
      "grad_norm": 0.009100419469177723,
      "learning_rate": 2.7559830730334452e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5234096,
      "step": 18695
    },
    {
      "epoch": 207.77777777777777,
      "grad_norm": 0.008999533019959927,
      "learning_rate": 2.7550064656817988e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5235520,
      "step": 18700
    },
    {
      "epoch": 207.83333333333334,
      "grad_norm": 0.006922734901309013,
      "learning_rate": 2.7540298190049503e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5236928,
      "step": 18705
    },
    {
      "epoch": 207.88888888888889,
      "grad_norm": 0.017758801579475403,
      "learning_rate": 2.7530531331535107e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5238272,
      "step": 18710
    },
    {
      "epoch": 207.94444444444446,
      "grad_norm": 0.024730388075113297,
      "learning_rate": 2.752076408278099e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5239680,
      "step": 18715
    },
    {
      "epoch": 208.0,
      "grad_norm": 0.004606630187481642,
      "learning_rate": 2.751099644529337e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5241056,
      "step": 18720
    },
    {
      "epoch": 208.05555555555554,
      "grad_norm": 0.005506272427737713,
      "learning_rate": 2.7501228420578533e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5242432,
      "step": 18725
    },
    {
      "epoch": 208.11111111111111,
      "grad_norm": 0.006399479694664478,
      "learning_rate": 2.7491460010142857e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5243808,
      "step": 18730
    },
    {
      "epoch": 208.16666666666666,
      "grad_norm": 0.02017606981098652,
      "learning_rate": 2.7481691215492727e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5245152,
      "step": 18735
    },
    {
      "epoch": 208.22222222222223,
      "grad_norm": 0.02468286082148552,
      "learning_rate": 2.747192203813463e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5246544,
      "step": 18740
    },
    {
      "epoch": 208.27777777777777,
      "grad_norm": 0.0019037146121263504,
      "learning_rate": 2.7462152479575087e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5247936,
      "step": 18745
    },
    {
      "epoch": 208.33333333333334,
      "grad_norm": 0.0007145790150389075,
      "learning_rate": 2.7452382541320697e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5249328,
      "step": 18750
    },
    {
      "epoch": 208.38888888888889,
      "grad_norm": 0.003988669253885746,
      "learning_rate": 2.7442612224878096e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5250784,
      "step": 18755
    },
    {
      "epoch": 208.44444444444446,
      "grad_norm": 0.003580026561394334,
      "learning_rate": 2.7432841531753994e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5252192,
      "step": 18760
    },
    {
      "epoch": 208.5,
      "grad_norm": 0.001066897064447403,
      "learning_rate": 2.7423070463455147e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5253616,
      "step": 18765
    },
    {
      "epoch": 208.55555555555554,
      "grad_norm": 0.11214052140712738,
      "learning_rate": 2.7413299021488397e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5254992,
      "step": 18770
    },
    {
      "epoch": 208.61111111111111,
      "grad_norm": 0.013539789244532585,
      "learning_rate": 2.7403527207360615e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5256368,
      "step": 18775
    },
    {
      "epoch": 208.66666666666666,
      "grad_norm": 0.004973128437995911,
      "learning_rate": 2.7393755022578722e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5257808,
      "step": 18780
    },
    {
      "epoch": 208.72222222222223,
      "grad_norm": 0.0016278678085654974,
      "learning_rate": 2.7383982468649714e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5259184,
      "step": 18785
    },
    {
      "epoch": 208.77777777777777,
      "grad_norm": 0.006443746853619814,
      "learning_rate": 2.7374209547080665e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5260592,
      "step": 18790
    },
    {
      "epoch": 208.83333333333334,
      "grad_norm": 0.0011358045740053058,
      "learning_rate": 2.7364436259378663e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5261952,
      "step": 18795
    },
    {
      "epoch": 208.88888888888889,
      "grad_norm": 0.0011134888045489788,
      "learning_rate": 2.735466260705088e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5263360,
      "step": 18800
    },
    {
      "epoch": 208.88888888888889,
      "eval_loss": 0.6870362758636475,
      "eval_runtime": 1.236,
      "eval_samples_per_second": 32.363,
      "eval_steps_per_second": 16.181,
      "num_input_tokens_seen": 5263360,
      "step": 18800
    },
    {
      "epoch": 208.94444444444446,
      "grad_norm": 0.000691875524353236,
      "learning_rate": 2.7344888591604524e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5264800,
      "step": 18805
    },
    {
      "epoch": 209.0,
      "grad_norm": 0.015397688373923302,
      "learning_rate": 2.7335114214546893e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5266240,
      "step": 18810
    },
    {
      "epoch": 209.05555555555554,
      "grad_norm": 0.0010028346441686153,
      "learning_rate": 2.7325339477385293e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5267680,
      "step": 18815
    },
    {
      "epoch": 209.11111111111111,
      "grad_norm": 0.1707204282283783,
      "learning_rate": 2.7315564381627128e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5269088,
      "step": 18820
    },
    {
      "epoch": 209.16666666666666,
      "grad_norm": 0.000368459994206205,
      "learning_rate": 2.7305788928779835e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5270464,
      "step": 18825
    },
    {
      "epoch": 209.22222222222223,
      "grad_norm": 0.0026626328472048044,
      "learning_rate": 2.729601312035091e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5271856,
      "step": 18830
    },
    {
      "epoch": 209.27777777777777,
      "grad_norm": 0.011672831140458584,
      "learning_rate": 2.7286236957847915e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5273280,
      "step": 18835
    },
    {
      "epoch": 209.33333333333334,
      "grad_norm": 0.005192877724766731,
      "learning_rate": 2.7276460442778446e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5274624,
      "step": 18840
    },
    {
      "epoch": 209.38888888888889,
      "grad_norm": 0.0008817364578135312,
      "learning_rate": 2.726668357665017e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5276016,
      "step": 18845
    },
    {
      "epoch": 209.44444444444446,
      "grad_norm": 0.002411074237897992,
      "learning_rate": 2.7256906360970808e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5277424,
      "step": 18850
    },
    {
      "epoch": 209.5,
      "grad_norm": 0.00047497969353571534,
      "learning_rate": 2.7247128797248117e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 5278864,
      "step": 18855
    },
    {
      "epoch": 209.55555555555554,
      "grad_norm": 0.00240908982232213,
      "learning_rate": 2.7237350886989925e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5280240,
      "step": 18860
    },
    {
      "epoch": 209.61111111111111,
      "grad_norm": 0.0023590652272105217,
      "learning_rate": 2.7227572631704107e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5281632,
      "step": 18865
    },
    {
      "epoch": 209.66666666666666,
      "grad_norm": 0.002305972157046199,
      "learning_rate": 2.7217794032898596e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5283040,
      "step": 18870
    },
    {
      "epoch": 209.72222222222223,
      "grad_norm": 0.008171000517904758,
      "learning_rate": 2.7208015092081384e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5284464,
      "step": 18875
    },
    {
      "epoch": 209.77777777777777,
      "grad_norm": 0.3552607595920563,
      "learning_rate": 2.719823581076049e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5285824,
      "step": 18880
    },
    {
      "epoch": 209.83333333333334,
      "grad_norm": 0.001556744915433228,
      "learning_rate": 2.718845619044401e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5287216,
      "step": 18885
    },
    {
      "epoch": 209.88888888888889,
      "grad_norm": 0.0017997295362874866,
      "learning_rate": 2.7178676232640088e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5288576,
      "step": 18890
    },
    {
      "epoch": 209.94444444444446,
      "grad_norm": 0.2712509036064148,
      "learning_rate": 2.716889593885691e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 5290000,
      "step": 18895
    },
    {
      "epoch": 210.0,
      "grad_norm": 0.00079802778782323,
      "learning_rate": 2.7159115310602716e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5291424,
      "step": 18900
    },
    {
      "epoch": 210.05555555555554,
      "grad_norm": 0.0077384039759635925,
      "learning_rate": 2.7149334349385814e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 5292832,
      "step": 18905
    },
    {
      "epoch": 210.11111111111111,
      "grad_norm": 0.00015461959992535412,
      "learning_rate": 2.713955305671454e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5294240,
      "step": 18910
    },
    {
      "epoch": 210.16666666666666,
      "grad_norm": 0.002063469961285591,
      "learning_rate": 2.71297714340973e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5295632,
      "step": 18915
    },
    {
      "epoch": 210.22222222222223,
      "grad_norm": 0.003593848552554846,
      "learning_rate": 2.7119989483042545e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5297136,
      "step": 18920
    },
    {
      "epoch": 210.27777777777777,
      "grad_norm": 0.007934455759823322,
      "learning_rate": 2.7110207205058768e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5298528,
      "step": 18925
    },
    {
      "epoch": 210.33333333333334,
      "grad_norm": 0.004264259245246649,
      "learning_rate": 2.7100424601654517e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5299920,
      "step": 18930
    },
    {
      "epoch": 210.38888888888889,
      "grad_norm": 0.0007560940575785935,
      "learning_rate": 2.7090641674338403e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5301328,
      "step": 18935
    },
    {
      "epoch": 210.44444444444446,
      "grad_norm": 0.023967988789081573,
      "learning_rate": 2.7080858424619072e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5302688,
      "step": 18940
    },
    {
      "epoch": 210.5,
      "grad_norm": 0.0006330738542601466,
      "learning_rate": 2.707107485400521e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5304064,
      "step": 18945
    },
    {
      "epoch": 210.55555555555554,
      "grad_norm": 0.0009235520847141743,
      "learning_rate": 2.7061290964005586e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5305440,
      "step": 18950
    },
    {
      "epoch": 210.61111111111111,
      "grad_norm": 0.0019444363424554467,
      "learning_rate": 2.7051506756129e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5306800,
      "step": 18955
    },
    {
      "epoch": 210.66666666666666,
      "grad_norm": 0.0082705058157444,
      "learning_rate": 2.704172223188428e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5308192,
      "step": 18960
    },
    {
      "epoch": 210.72222222222223,
      "grad_norm": 0.016855286434292793,
      "learning_rate": 2.7031937392780334e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5309600,
      "step": 18965
    },
    {
      "epoch": 210.77777777777777,
      "grad_norm": 0.00274306139908731,
      "learning_rate": 2.702215224032611e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5310960,
      "step": 18970
    },
    {
      "epoch": 210.83333333333334,
      "grad_norm": 0.006607595831155777,
      "learning_rate": 2.70123667760306e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5312352,
      "step": 18975
    },
    {
      "epoch": 210.88888888888889,
      "grad_norm": 0.0008605016628280282,
      "learning_rate": 2.7002581001402845e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5313776,
      "step": 18980
    },
    {
      "epoch": 210.94444444444446,
      "grad_norm": 0.012738210149109364,
      "learning_rate": 2.6992794917951923e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5315136,
      "step": 18985
    },
    {
      "epoch": 211.0,
      "grad_norm": 0.0004049587878398597,
      "learning_rate": 2.6983008527187e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5316560,
      "step": 18990
    },
    {
      "epoch": 211.05555555555554,
      "grad_norm": 0.0007220018887892365,
      "learning_rate": 2.697322183061723e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5317952,
      "step": 18995
    },
    {
      "epoch": 211.11111111111111,
      "grad_norm": 0.008793555200099945,
      "learning_rate": 2.696343482975186e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5319344,
      "step": 19000
    },
    {
      "epoch": 211.11111111111111,
      "eval_loss": 0.6698595285415649,
      "eval_runtime": 1.2344,
      "eval_samples_per_second": 32.403,
      "eval_steps_per_second": 16.202,
      "num_input_tokens_seen": 5319344,
      "step": 19000
    },
    {
      "epoch": 211.16666666666666,
      "grad_norm": 0.036329302936792374,
      "learning_rate": 2.695364752610016e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5320736,
      "step": 19005
    },
    {
      "epoch": 211.22222222222223,
      "grad_norm": 0.00045423279516398907,
      "learning_rate": 2.6943859921171467e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5322176,
      "step": 19010
    },
    {
      "epoch": 211.27777777777777,
      "grad_norm": 0.0031093910802155733,
      "learning_rate": 2.6934072016475143e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5323552,
      "step": 19015
    },
    {
      "epoch": 211.33333333333334,
      "grad_norm": 0.003100471803918481,
      "learning_rate": 2.6924283813520606e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5324992,
      "step": 19020
    },
    {
      "epoch": 211.38888888888889,
      "grad_norm": 0.009342623874545097,
      "learning_rate": 2.691449531381733e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5326416,
      "step": 19025
    },
    {
      "epoch": 211.44444444444446,
      "grad_norm": 0.0041899848729372025,
      "learning_rate": 2.6904706518874816e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5327808,
      "step": 19030
    },
    {
      "epoch": 211.5,
      "grad_norm": 0.0033640332985669374,
      "learning_rate": 2.6894917430202615e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5329216,
      "step": 19035
    },
    {
      "epoch": 211.55555555555554,
      "grad_norm": 0.002021713647991419,
      "learning_rate": 2.6885128049310343e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5330624,
      "step": 19040
    },
    {
      "epoch": 211.61111111111111,
      "grad_norm": 0.008253009989857674,
      "learning_rate": 2.687533837770762e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5332016,
      "step": 19045
    },
    {
      "epoch": 211.66666666666666,
      "grad_norm": 0.006206140387803316,
      "learning_rate": 2.6865548416904162e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5333424,
      "step": 19050
    },
    {
      "epoch": 211.72222222222223,
      "grad_norm": 0.0002617515274323523,
      "learning_rate": 2.68557581684097e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5334832,
      "step": 19055
    },
    {
      "epoch": 211.77777777777777,
      "grad_norm": 0.005429750308394432,
      "learning_rate": 2.6845967633733998e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5336208,
      "step": 19060
    },
    {
      "epoch": 211.83333333333334,
      "grad_norm": 0.008093767799437046,
      "learning_rate": 2.683617681438689e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5337584,
      "step": 19065
    },
    {
      "epoch": 211.88888888888889,
      "grad_norm": 0.07269459217786789,
      "learning_rate": 2.682638571187825e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5338944,
      "step": 19070
    },
    {
      "epoch": 211.94444444444446,
      "grad_norm": 0.0002467484155204147,
      "learning_rate": 2.6816594327717976e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5340368,
      "step": 19075
    },
    {
      "epoch": 212.0,
      "grad_norm": 0.0016316493274644017,
      "learning_rate": 2.680680266341603e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5341776,
      "step": 19080
    },
    {
      "epoch": 212.05555555555554,
      "grad_norm": 0.0025658716913312674,
      "learning_rate": 2.67970107204824e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5343120,
      "step": 19085
    },
    {
      "epoch": 212.11111111111111,
      "grad_norm": 0.0804070308804512,
      "learning_rate": 2.6787218500427142e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5344496,
      "step": 19090
    },
    {
      "epoch": 212.16666666666666,
      "grad_norm": 0.002394438022747636,
      "learning_rate": 2.6777426004760332e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5345920,
      "step": 19095
    },
    {
      "epoch": 212.22222222222223,
      "grad_norm": 0.17734353244304657,
      "learning_rate": 2.6767633234992094e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5347312,
      "step": 19100
    },
    {
      "epoch": 212.27777777777777,
      "grad_norm": 0.00019388084183447063,
      "learning_rate": 2.6757840192632598e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5348768,
      "step": 19105
    },
    {
      "epoch": 212.33333333333334,
      "grad_norm": 0.00308228749781847,
      "learning_rate": 2.6748046879192052e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5350192,
      "step": 19110
    },
    {
      "epoch": 212.38888888888889,
      "grad_norm": 0.0012837642570957541,
      "learning_rate": 2.673825329618071e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5351600,
      "step": 19115
    },
    {
      "epoch": 212.44444444444446,
      "grad_norm": 0.034200847148895264,
      "learning_rate": 2.6728459445108866e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5353040,
      "step": 19120
    },
    {
      "epoch": 212.5,
      "grad_norm": 0.0034910524263978004,
      "learning_rate": 2.6718665327486854e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5354416,
      "step": 19125
    },
    {
      "epoch": 212.55555555555554,
      "grad_norm": 0.00024852470960468054,
      "learning_rate": 2.6708870944825048e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5355808,
      "step": 19130
    },
    {
      "epoch": 212.61111111111111,
      "grad_norm": 0.0008710575057193637,
      "learning_rate": 2.6699076298633874e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5357248,
      "step": 19135
    },
    {
      "epoch": 212.66666666666666,
      "grad_norm": 0.00507877254858613,
      "learning_rate": 2.6689281390423788e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5358656,
      "step": 19140
    },
    {
      "epoch": 212.72222222222223,
      "grad_norm": 0.007229275070130825,
      "learning_rate": 2.667948622170527e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5360016,
      "step": 19145
    },
    {
      "epoch": 212.77777777777777,
      "grad_norm": 0.013118011876940727,
      "learning_rate": 2.6669690793988873e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5361376,
      "step": 19150
    },
    {
      "epoch": 212.83333333333334,
      "grad_norm": 0.00031091555138118565,
      "learning_rate": 2.665989510878518e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5362784,
      "step": 19155
    },
    {
      "epoch": 212.88888888888889,
      "grad_norm": 0.0023389735724776983,
      "learning_rate": 2.6650099167604793e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5364176,
      "step": 19160
    },
    {
      "epoch": 212.94444444444446,
      "grad_norm": 0.0013293421361595392,
      "learning_rate": 2.6640302971958376e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5365584,
      "step": 19165
    },
    {
      "epoch": 213.0,
      "grad_norm": 0.0049838232807815075,
      "learning_rate": 2.6630506523356635e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5366960,
      "step": 19170
    },
    {
      "epoch": 213.05555555555554,
      "grad_norm": 0.001437244936823845,
      "learning_rate": 2.6620709823310297e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5368320,
      "step": 19175
    },
    {
      "epoch": 213.11111111111111,
      "grad_norm": 0.007783041335642338,
      "learning_rate": 2.661091287333014e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5369760,
      "step": 19180
    },
    {
      "epoch": 213.16666666666666,
      "grad_norm": 0.005224299617111683,
      "learning_rate": 2.660111567492696e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5371136,
      "step": 19185
    },
    {
      "epoch": 213.22222222222223,
      "grad_norm": 0.001999989617615938,
      "learning_rate": 2.6591318229611635e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5372528,
      "step": 19190
    },
    {
      "epoch": 213.27777777777777,
      "grad_norm": 0.004012868739664555,
      "learning_rate": 2.6581520538895037e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5373888,
      "step": 19195
    },
    {
      "epoch": 213.33333333333334,
      "grad_norm": 0.002283406211063266,
      "learning_rate": 2.6571722604288102e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5375280,
      "step": 19200
    },
    {
      "epoch": 213.33333333333334,
      "eval_loss": 0.6956033706665039,
      "eval_runtime": 1.2349,
      "eval_samples_per_second": 32.391,
      "eval_steps_per_second": 16.196,
      "num_input_tokens_seen": 5375280,
      "step": 19200
    },
    {
      "epoch": 213.38888888888889,
      "grad_norm": 0.00474558025598526,
      "learning_rate": 2.656192442730179e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5376704,
      "step": 19205
    },
    {
      "epoch": 213.44444444444446,
      "grad_norm": 0.00738050602376461,
      "learning_rate": 2.6552126009447098e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5378080,
      "step": 19210
    },
    {
      "epoch": 213.5,
      "grad_norm": 0.00030172584229148924,
      "learning_rate": 2.654232735223507e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5379504,
      "step": 19215
    },
    {
      "epoch": 213.55555555555554,
      "grad_norm": 0.0039890524931252,
      "learning_rate": 2.6532528457176787e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5380896,
      "step": 19220
    },
    {
      "epoch": 213.61111111111111,
      "grad_norm": 0.0005801186780445278,
      "learning_rate": 2.6522729325783348e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5382304,
      "step": 19225
    },
    {
      "epoch": 213.66666666666666,
      "grad_norm": 0.0010160926030948758,
      "learning_rate": 2.6512929959565914e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5383664,
      "step": 19230
    },
    {
      "epoch": 213.72222222222223,
      "grad_norm": 0.013006860390305519,
      "learning_rate": 2.6503130360035673e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5385088,
      "step": 19235
    },
    {
      "epoch": 213.77777777777777,
      "grad_norm": 0.0030064519960433245,
      "learning_rate": 2.6493330528703835e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5386512,
      "step": 19240
    },
    {
      "epoch": 213.83333333333334,
      "grad_norm": 0.0009583337232470512,
      "learning_rate": 2.648353046708167e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5387968,
      "step": 19245
    },
    {
      "epoch": 213.88888888888889,
      "grad_norm": 0.0009014995885081589,
      "learning_rate": 2.647373017668046e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5389392,
      "step": 19250
    },
    {
      "epoch": 213.94444444444446,
      "grad_norm": 0.003017202252522111,
      "learning_rate": 2.6463929659011537e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5390800,
      "step": 19255
    },
    {
      "epoch": 214.0,
      "grad_norm": 0.04175446182489395,
      "learning_rate": 2.6454128915586262e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5392192,
      "step": 19260
    },
    {
      "epoch": 214.05555555555554,
      "grad_norm": 0.00027273318846710026,
      "learning_rate": 2.6444327947916036e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5393568,
      "step": 19265
    },
    {
      "epoch": 214.11111111111111,
      "grad_norm": 0.0031424658372998238,
      "learning_rate": 2.6434526757512292e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5395008,
      "step": 19270
    },
    {
      "epoch": 214.16666666666666,
      "grad_norm": 0.0010212355991825461,
      "learning_rate": 2.6424725345886486e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5396384,
      "step": 19275
    },
    {
      "epoch": 214.22222222222223,
      "grad_norm": 0.00024267328262794763,
      "learning_rate": 2.641492371455014e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5397744,
      "step": 19280
    },
    {
      "epoch": 214.27777777777777,
      "grad_norm": 0.005459020845592022,
      "learning_rate": 2.640512186501477e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5399168,
      "step": 19285
    },
    {
      "epoch": 214.33333333333334,
      "grad_norm": 0.017112674191594124,
      "learning_rate": 2.639531979879195e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5400560,
      "step": 19290
    },
    {
      "epoch": 214.38888888888889,
      "grad_norm": 0.09359753876924515,
      "learning_rate": 2.638551751739328e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5401984,
      "step": 19295
    },
    {
      "epoch": 214.44444444444446,
      "grad_norm": 0.0028370791114866734,
      "learning_rate": 2.6375715022330404e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5403392,
      "step": 19300
    },
    {
      "epoch": 214.5,
      "grad_norm": 0.00034261125256307423,
      "learning_rate": 2.6365912315114976e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5404800,
      "step": 19305
    },
    {
      "epoch": 214.55555555555554,
      "grad_norm": 0.0001545231934869662,
      "learning_rate": 2.6356109397258704e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5406208,
      "step": 19310
    },
    {
      "epoch": 214.61111111111111,
      "grad_norm": 0.0012805759906768799,
      "learning_rate": 2.6346306270273325e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5407616,
      "step": 19315
    },
    {
      "epoch": 214.66666666666666,
      "grad_norm": 0.0013492514844983816,
      "learning_rate": 2.6336502935670608e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5409040,
      "step": 19320
    },
    {
      "epoch": 214.72222222222223,
      "grad_norm": 0.006651722826063633,
      "learning_rate": 2.6326699394962333e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5410448,
      "step": 19325
    },
    {
      "epoch": 214.77777777777777,
      "grad_norm": 0.009999210014939308,
      "learning_rate": 2.6316895649660334e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5411808,
      "step": 19330
    },
    {
      "epoch": 214.83333333333334,
      "grad_norm": 0.002193124732002616,
      "learning_rate": 2.6307091701276486e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5413232,
      "step": 19335
    },
    {
      "epoch": 214.88888888888889,
      "grad_norm": 0.016242273151874542,
      "learning_rate": 2.629728755132267e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5414656,
      "step": 19340
    },
    {
      "epoch": 214.94444444444446,
      "grad_norm": 0.0014452234609052539,
      "learning_rate": 2.628748320131081e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5416080,
      "step": 19345
    },
    {
      "epoch": 215.0,
      "grad_norm": 0.005534008145332336,
      "learning_rate": 2.6277678652752856e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5417472,
      "step": 19350
    },
    {
      "epoch": 215.05555555555554,
      "grad_norm": 0.00019919613259844482,
      "learning_rate": 2.6267873907160807e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5418880,
      "step": 19355
    },
    {
      "epoch": 215.11111111111111,
      "grad_norm": 0.0023930673487484455,
      "learning_rate": 2.6258068966046668e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5420272,
      "step": 19360
    },
    {
      "epoch": 215.16666666666666,
      "grad_norm": 0.00014251100947149098,
      "learning_rate": 2.6248263830922475e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5421664,
      "step": 19365
    },
    {
      "epoch": 215.22222222222223,
      "grad_norm": 0.04306882619857788,
      "learning_rate": 2.6238458503300318e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5423072,
      "step": 19370
    },
    {
      "epoch": 215.27777777777777,
      "grad_norm": 0.00094628642546013,
      "learning_rate": 2.6228652984692292e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5424496,
      "step": 19375
    },
    {
      "epoch": 215.33333333333334,
      "grad_norm": 0.0007458815234713256,
      "learning_rate": 2.621884727661054e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5425904,
      "step": 19380
    },
    {
      "epoch": 215.38888888888889,
      "grad_norm": 0.008212692104279995,
      "learning_rate": 2.6209041380567222e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5427296,
      "step": 19385
    },
    {
      "epoch": 215.44444444444446,
      "grad_norm": 0.0006181761273182929,
      "learning_rate": 2.6199235298074527e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5428688,
      "step": 19390
    },
    {
      "epoch": 215.5,
      "grad_norm": 0.0029916707426309586,
      "learning_rate": 2.618942903064468e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5430080,
      "step": 19395
    },
    {
      "epoch": 215.55555555555554,
      "grad_norm": 0.01004949864000082,
      "learning_rate": 2.6179622579789932e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5431520,
      "step": 19400
    },
    {
      "epoch": 215.55555555555554,
      "eval_loss": 0.7282935976982117,
      "eval_runtime": 1.2369,
      "eval_samples_per_second": 32.339,
      "eval_steps_per_second": 16.17,
      "num_input_tokens_seen": 5431520,
      "step": 19400
    },
    {
      "epoch": 215.61111111111111,
      "grad_norm": 0.00701445247977972,
      "learning_rate": 2.6169815947022553e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5432960,
      "step": 19405
    },
    {
      "epoch": 215.66666666666666,
      "grad_norm": 0.002945417771115899,
      "learning_rate": 2.6160009133854853e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5434352,
      "step": 19410
    },
    {
      "epoch": 215.72222222222223,
      "grad_norm": 0.0015321287792176008,
      "learning_rate": 2.6150202141799168e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5435760,
      "step": 19415
    },
    {
      "epoch": 215.77777777777777,
      "grad_norm": 0.004911166150122881,
      "learning_rate": 2.614039497236786e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5437168,
      "step": 19420
    },
    {
      "epoch": 215.83333333333334,
      "grad_norm": 0.002216619672253728,
      "learning_rate": 2.6130587627073315e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5438544,
      "step": 19425
    },
    {
      "epoch": 215.88888888888889,
      "grad_norm": 0.009438734501600266,
      "learning_rate": 2.6120780107427956e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5439904,
      "step": 19430
    },
    {
      "epoch": 215.94444444444446,
      "grad_norm": 0.004374843090772629,
      "learning_rate": 2.6110972414944214e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5441360,
      "step": 19435
    },
    {
      "epoch": 216.0,
      "grad_norm": 0.0002877140650525689,
      "learning_rate": 2.6101164551134565e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5442752,
      "step": 19440
    },
    {
      "epoch": 216.05555555555554,
      "grad_norm": 0.0015668122796341777,
      "learning_rate": 2.6091356517511505e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5444176,
      "step": 19445
    },
    {
      "epoch": 216.11111111111111,
      "grad_norm": 0.00037309827166609466,
      "learning_rate": 2.608154831558755e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5445584,
      "step": 19450
    },
    {
      "epoch": 216.16666666666666,
      "grad_norm": 9.21193859539926e-05,
      "learning_rate": 2.607173994687526e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5446976,
      "step": 19455
    },
    {
      "epoch": 216.22222222222223,
      "grad_norm": 0.004722981248050928,
      "learning_rate": 2.6061931412887196e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5448384,
      "step": 19460
    },
    {
      "epoch": 216.27777777777777,
      "grad_norm": 0.13040518760681152,
      "learning_rate": 2.6052122715135973e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5449808,
      "step": 19465
    },
    {
      "epoch": 216.33333333333334,
      "grad_norm": 0.0005271522677503526,
      "learning_rate": 2.60423138551342e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5451232,
      "step": 19470
    },
    {
      "epoch": 216.38888888888889,
      "grad_norm": 0.001754550845362246,
      "learning_rate": 2.6032504834394527e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5452624,
      "step": 19475
    },
    {
      "epoch": 216.44444444444446,
      "grad_norm": 0.004653278272598982,
      "learning_rate": 2.602269565442964e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5453968,
      "step": 19480
    },
    {
      "epoch": 216.5,
      "grad_norm": 0.0005853786133229733,
      "learning_rate": 2.6012886316752227e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5455392,
      "step": 19485
    },
    {
      "epoch": 216.55555555555554,
      "grad_norm": 0.002279987558722496,
      "learning_rate": 2.6003076822875018e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5456800,
      "step": 19490
    },
    {
      "epoch": 216.61111111111111,
      "grad_norm": 0.003286063903942704,
      "learning_rate": 2.5993267174310755e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5458192,
      "step": 19495
    },
    {
      "epoch": 216.66666666666666,
      "grad_norm": 0.012572264298796654,
      "learning_rate": 2.5983457372572218e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5459568,
      "step": 19500
    },
    {
      "epoch": 216.72222222222223,
      "grad_norm": 0.005131459329277277,
      "learning_rate": 2.597364741917219e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5460960,
      "step": 19505
    },
    {
      "epoch": 216.77777777777777,
      "grad_norm": 0.00036715020542033017,
      "learning_rate": 2.5963837315623492e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5462368,
      "step": 19510
    },
    {
      "epoch": 216.83333333333334,
      "grad_norm": 0.0028951987624168396,
      "learning_rate": 2.595402706343897e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5463792,
      "step": 19515
    },
    {
      "epoch": 216.88888888888889,
      "grad_norm": 0.13571612536907196,
      "learning_rate": 2.594421666413148e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5465232,
      "step": 19520
    },
    {
      "epoch": 216.94444444444446,
      "grad_norm": 0.004599777050316334,
      "learning_rate": 2.5934406119213928e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5466640,
      "step": 19525
    },
    {
      "epoch": 217.0,
      "grad_norm": 0.011180393397808075,
      "learning_rate": 2.5924595430199193e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5467984,
      "step": 19530
    },
    {
      "epoch": 217.05555555555554,
      "grad_norm": 0.0006701169768348336,
      "learning_rate": 2.5914784598600238e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5469376,
      "step": 19535
    },
    {
      "epoch": 217.11111111111111,
      "grad_norm": 0.0010564594995230436,
      "learning_rate": 2.5904973625930002e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5470720,
      "step": 19540
    },
    {
      "epoch": 217.16666666666666,
      "grad_norm": 0.0004452488792594522,
      "learning_rate": 2.5895162513701456e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5472112,
      "step": 19545
    },
    {
      "epoch": 217.22222222222223,
      "grad_norm": 0.0038704259786754847,
      "learning_rate": 2.5885351263427593e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5473520,
      "step": 19550
    },
    {
      "epoch": 217.27777777777777,
      "grad_norm": 0.0004953907919116318,
      "learning_rate": 2.5875539876621448e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5474912,
      "step": 19555
    },
    {
      "epoch": 217.33333333333334,
      "grad_norm": 0.000853539037052542,
      "learning_rate": 2.586572835479605e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5476288,
      "step": 19560
    },
    {
      "epoch": 217.38888888888889,
      "grad_norm": 0.00757192587479949,
      "learning_rate": 2.585591669946446e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5477632,
      "step": 19565
    },
    {
      "epoch": 217.44444444444446,
      "grad_norm": 0.00047957574133761227,
      "learning_rate": 2.5846104912139756e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5479056,
      "step": 19570
    },
    {
      "epoch": 217.5,
      "grad_norm": 0.10632050782442093,
      "learning_rate": 2.583629299433505e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5480480,
      "step": 19575
    },
    {
      "epoch": 217.55555555555554,
      "grad_norm": 0.010511273518204689,
      "learning_rate": 2.582648094756345e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5481904,
      "step": 19580
    },
    {
      "epoch": 217.61111111111111,
      "grad_norm": 0.00019611854804679751,
      "learning_rate": 2.5816668773338098e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5483280,
      "step": 19585
    },
    {
      "epoch": 217.66666666666666,
      "grad_norm": 0.00611312547698617,
      "learning_rate": 2.580685647317216e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5484688,
      "step": 19590
    },
    {
      "epoch": 217.72222222222223,
      "grad_norm": 0.0024877754040062428,
      "learning_rate": 2.5797044048578818e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5486064,
      "step": 19595
    },
    {
      "epoch": 217.77777777777777,
      "grad_norm": 0.0002242909831693396,
      "learning_rate": 2.5787231501071262e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5487472,
      "step": 19600
    },
    {
      "epoch": 217.77777777777777,
      "eval_loss": 0.7154077291488647,
      "eval_runtime": 1.2442,
      "eval_samples_per_second": 32.15,
      "eval_steps_per_second": 16.075,
      "num_input_tokens_seen": 5487472,
      "step": 19600
    },
    {
      "epoch": 217.83333333333334,
      "grad_norm": 0.003620681120082736,
      "learning_rate": 2.577741883216272e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5488880,
      "step": 19605
    },
    {
      "epoch": 217.88888888888889,
      "grad_norm": 0.00041727564530447125,
      "learning_rate": 2.576760604336642e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5490304,
      "step": 19610
    },
    {
      "epoch": 217.94444444444446,
      "grad_norm": 0.007721369154751301,
      "learning_rate": 2.575779313619563e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5491712,
      "step": 19615
    },
    {
      "epoch": 218.0,
      "grad_norm": 0.00402510492131114,
      "learning_rate": 2.5747980112163605e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5493136,
      "step": 19620
    },
    {
      "epoch": 218.05555555555554,
      "grad_norm": 0.025945942848920822,
      "learning_rate": 2.5738166972783656e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5494528,
      "step": 19625
    },
    {
      "epoch": 218.11111111111111,
      "grad_norm": 7.888002437539399e-05,
      "learning_rate": 2.5728353719569075e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5495920,
      "step": 19630
    },
    {
      "epoch": 218.16666666666666,
      "grad_norm": 0.000701736775226891,
      "learning_rate": 2.57185403540332e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5497296,
      "step": 19635
    },
    {
      "epoch": 218.22222222222223,
      "grad_norm": 0.036883458495140076,
      "learning_rate": 2.5708726877689375e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5498688,
      "step": 19640
    },
    {
      "epoch": 218.27777777777777,
      "grad_norm": 0.0005596756236627698,
      "learning_rate": 2.5698913292050964e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5500096,
      "step": 19645
    },
    {
      "epoch": 218.33333333333334,
      "grad_norm": 0.017529066652059555,
      "learning_rate": 2.568909959863133e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5501520,
      "step": 19650
    },
    {
      "epoch": 218.38888888888889,
      "grad_norm": 0.0029092179611325264,
      "learning_rate": 2.5679285798943887e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5502928,
      "step": 19655
    },
    {
      "epoch": 218.44444444444446,
      "grad_norm": 0.003232925897464156,
      "learning_rate": 2.5669471894502035e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5504304,
      "step": 19660
    },
    {
      "epoch": 218.5,
      "grad_norm": 0.0015763393603265285,
      "learning_rate": 2.56596578868192e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5505664,
      "step": 19665
    },
    {
      "epoch": 218.55555555555554,
      "grad_norm": 0.0014231560053303838,
      "learning_rate": 2.564984377740883e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5507088,
      "step": 19670
    },
    {
      "epoch": 218.61111111111111,
      "grad_norm": 0.0009774237405508757,
      "learning_rate": 2.564002956778438e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5508512,
      "step": 19675
    },
    {
      "epoch": 218.66666666666666,
      "grad_norm": 0.0007751050870865583,
      "learning_rate": 2.563021525945934e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5509952,
      "step": 19680
    },
    {
      "epoch": 218.72222222222223,
      "grad_norm": 0.00024754140758886933,
      "learning_rate": 2.562040085394718e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5511344,
      "step": 19685
    },
    {
      "epoch": 218.77777777777777,
      "grad_norm": 0.0003431342775002122,
      "learning_rate": 2.56105863527614e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5512720,
      "step": 19690
    },
    {
      "epoch": 218.83333333333334,
      "grad_norm": 0.006441803183406591,
      "learning_rate": 2.5600771757415548e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5514112,
      "step": 19695
    },
    {
      "epoch": 218.88888888888889,
      "grad_norm": 0.00036616806755773723,
      "learning_rate": 2.5590957069423134e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5515568,
      "step": 19700
    },
    {
      "epoch": 218.94444444444446,
      "grad_norm": 0.0017557453829795122,
      "learning_rate": 2.5581142290297716e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5516992,
      "step": 19705
    },
    {
      "epoch": 219.0,
      "grad_norm": 0.0009431534563191235,
      "learning_rate": 2.557132742155285e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5518368,
      "step": 19710
    },
    {
      "epoch": 219.05555555555554,
      "grad_norm": 0.0017399274511262774,
      "learning_rate": 2.556151246470212e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5519776,
      "step": 19715
    },
    {
      "epoch": 219.11111111111111,
      "grad_norm": 0.0018469748320057988,
      "learning_rate": 2.5551697421259114e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5521216,
      "step": 19720
    },
    {
      "epoch": 219.16666666666666,
      "grad_norm": 0.021235428750514984,
      "learning_rate": 2.554188229273743e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5522608,
      "step": 19725
    },
    {
      "epoch": 219.22222222222223,
      "grad_norm": 0.0017678450094535947,
      "learning_rate": 2.5532067080650678e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5524016,
      "step": 19730
    },
    {
      "epoch": 219.27777777777777,
      "grad_norm": 0.0005455154459923506,
      "learning_rate": 2.55222517865125e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5525424,
      "step": 19735
    },
    {
      "epoch": 219.33333333333334,
      "grad_norm": 0.0007837860030122101,
      "learning_rate": 2.5512436411836538e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5526816,
      "step": 19740
    },
    {
      "epoch": 219.38888888888889,
      "grad_norm": 0.0014398441417142749,
      "learning_rate": 2.5502620958136443e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5528192,
      "step": 19745
    },
    {
      "epoch": 219.44444444444446,
      "grad_norm": 0.0011789303971454501,
      "learning_rate": 2.5492805426925874e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5529616,
      "step": 19750
    },
    {
      "epoch": 219.5,
      "grad_norm": 0.005147889256477356,
      "learning_rate": 2.5482989819718523e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5531008,
      "step": 19755
    },
    {
      "epoch": 219.55555555555554,
      "grad_norm": 0.0002831345482263714,
      "learning_rate": 2.5473174138028065e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5532384,
      "step": 19760
    },
    {
      "epoch": 219.61111111111111,
      "grad_norm": 0.004809612408280373,
      "learning_rate": 2.5463358383368212e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5533728,
      "step": 19765
    },
    {
      "epoch": 219.66666666666666,
      "grad_norm": 0.0002769332495518029,
      "learning_rate": 2.545354255725267e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5535152,
      "step": 19770
    },
    {
      "epoch": 219.72222222222223,
      "grad_norm": 0.0022134738974273205,
      "learning_rate": 2.5443726661195165e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5536560,
      "step": 19775
    },
    {
      "epoch": 219.77777777777777,
      "grad_norm": 0.0014782652724534273,
      "learning_rate": 2.543391069670944e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5537936,
      "step": 19780
    },
    {
      "epoch": 219.83333333333334,
      "grad_norm": 0.0006597347091883421,
      "learning_rate": 2.5424094665309228e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5539360,
      "step": 19785
    },
    {
      "epoch": 219.88888888888889,
      "grad_norm": 0.004858363885432482,
      "learning_rate": 2.5414278568508292e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5540784,
      "step": 19790
    },
    {
      "epoch": 219.94444444444446,
      "grad_norm": 0.002755743684247136,
      "learning_rate": 2.540446240782039e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5542160,
      "step": 19795
    },
    {
      "epoch": 220.0,
      "grad_norm": 0.0001539865625090897,
      "learning_rate": 2.5394646184759307e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5543504,
      "step": 19800
    },
    {
      "epoch": 220.0,
      "eval_loss": 0.7333426475524902,
      "eval_runtime": 1.2357,
      "eval_samples_per_second": 32.371,
      "eval_steps_per_second": 16.186,
      "num_input_tokens_seen": 5543504,
      "step": 19800
    },
    {
      "epoch": 220.05555555555554,
      "grad_norm": 0.0076450929045677185,
      "learning_rate": 2.538482990083882e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5544928,
      "step": 19805
    },
    {
      "epoch": 220.11111111111111,
      "grad_norm": 0.0007812681724317372,
      "learning_rate": 2.5375013557572725e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5546352,
      "step": 19810
    },
    {
      "epoch": 220.16666666666666,
      "grad_norm": 0.011557623744010925,
      "learning_rate": 2.536519715647483e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5547712,
      "step": 19815
    },
    {
      "epoch": 220.22222222222223,
      "grad_norm": 0.0009070027736015618,
      "learning_rate": 2.535538069905894e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5549168,
      "step": 19820
    },
    {
      "epoch": 220.27777777777777,
      "grad_norm": 0.00716383708640933,
      "learning_rate": 2.534556418683888e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5550576,
      "step": 19825
    },
    {
      "epoch": 220.33333333333334,
      "grad_norm": 0.00122284225653857,
      "learning_rate": 2.5335747621328486e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5551968,
      "step": 19830
    },
    {
      "epoch": 220.38888888888889,
      "grad_norm": 0.0014776497846469283,
      "learning_rate": 2.5325931004041586e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5553376,
      "step": 19835
    },
    {
      "epoch": 220.44444444444446,
      "grad_norm": 0.005764591507613659,
      "learning_rate": 2.5316114336492032e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5554736,
      "step": 19840
    },
    {
      "epoch": 220.5,
      "grad_norm": 0.002513699233531952,
      "learning_rate": 2.530629762019367e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5556112,
      "step": 19845
    },
    {
      "epoch": 220.55555555555554,
      "grad_norm": 0.00022575710318051279,
      "learning_rate": 2.5296480856660364e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5557520,
      "step": 19850
    },
    {
      "epoch": 220.61111111111111,
      "grad_norm": 0.00010080587526317686,
      "learning_rate": 2.528666404740599e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5558944,
      "step": 19855
    },
    {
      "epoch": 220.66666666666666,
      "grad_norm": 0.0057403999380767345,
      "learning_rate": 2.527684719394442e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5560352,
      "step": 19860
    },
    {
      "epoch": 220.72222222222223,
      "grad_norm": 0.13213077187538147,
      "learning_rate": 2.526703029778953e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 5561776,
      "step": 19865
    },
    {
      "epoch": 220.77777777777777,
      "grad_norm": 0.010376871563494205,
      "learning_rate": 2.5257213360455208e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5563136,
      "step": 19870
    },
    {
      "epoch": 220.83333333333334,
      "grad_norm": 0.002284600166603923,
      "learning_rate": 2.5247396383455353e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5564528,
      "step": 19875
    },
    {
      "epoch": 220.88888888888889,
      "grad_norm": 0.015126712620258331,
      "learning_rate": 2.523757936830387e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5565888,
      "step": 19880
    },
    {
      "epoch": 220.94444444444446,
      "grad_norm": 0.0030041448771953583,
      "learning_rate": 2.5227762316514662e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5567312,
      "step": 19885
    },
    {
      "epoch": 221.0,
      "grad_norm": 0.00023891235468909144,
      "learning_rate": 2.5217945229601648e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5568672,
      "step": 19890
    },
    {
      "epoch": 221.05555555555554,
      "grad_norm": 0.002823425689712167,
      "learning_rate": 2.5208128109078738e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5570096,
      "step": 19895
    },
    {
      "epoch": 221.11111111111111,
      "grad_norm": 0.0002438022056594491,
      "learning_rate": 2.5198310956459853e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5571456,
      "step": 19900
    },
    {
      "epoch": 221.16666666666666,
      "grad_norm": 0.0004799234156962484,
      "learning_rate": 2.518849377325893e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5572848,
      "step": 19905
    },
    {
      "epoch": 221.22222222222223,
      "grad_norm": 0.0029586548916995525,
      "learning_rate": 2.51786765609899e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5574240,
      "step": 19910
    },
    {
      "epoch": 221.27777777777777,
      "grad_norm": 0.0010623052949085832,
      "learning_rate": 2.5168859321166694e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5575616,
      "step": 19915
    },
    {
      "epoch": 221.33333333333334,
      "grad_norm": 0.00025456302682869136,
      "learning_rate": 2.515904205530326e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5577040,
      "step": 19920
    },
    {
      "epoch": 221.38888888888889,
      "grad_norm": 0.006686328910291195,
      "learning_rate": 2.514922476491355e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5578432,
      "step": 19925
    },
    {
      "epoch": 221.44444444444446,
      "grad_norm": 0.0014559815172106028,
      "learning_rate": 2.51394074515115e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5579856,
      "step": 19930
    },
    {
      "epoch": 221.5,
      "grad_norm": 0.004007631912827492,
      "learning_rate": 2.5129590116611067e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5581216,
      "step": 19935
    },
    {
      "epoch": 221.55555555555554,
      "grad_norm": 0.00314637110568583,
      "learning_rate": 2.5119772761726212e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5582656,
      "step": 19940
    },
    {
      "epoch": 221.61111111111111,
      "grad_norm": 0.0014572396175935864,
      "learning_rate": 2.5109955388370893e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5584032,
      "step": 19945
    },
    {
      "epoch": 221.66666666666666,
      "grad_norm": 0.02552792616188526,
      "learning_rate": 2.510013799805907e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5585408,
      "step": 19950
    },
    {
      "epoch": 221.72222222222223,
      "grad_norm": 0.0004059635102748871,
      "learning_rate": 2.5090320592304706e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5586848,
      "step": 19955
    },
    {
      "epoch": 221.77777777777777,
      "grad_norm": 0.14008809626102448,
      "learning_rate": 2.5080503172621777e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5588224,
      "step": 19960
    },
    {
      "epoch": 221.83333333333334,
      "grad_norm": 0.00026557070668786764,
      "learning_rate": 2.5070685740524246e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5589664,
      "step": 19965
    },
    {
      "epoch": 221.88888888888889,
      "grad_norm": 0.019760003313422203,
      "learning_rate": 2.5060868297526084e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5591056,
      "step": 19970
    },
    {
      "epoch": 221.94444444444446,
      "grad_norm": 0.000770705402828753,
      "learning_rate": 2.5051050845141267e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5592432,
      "step": 19975
    },
    {
      "epoch": 222.0,
      "grad_norm": 0.002022247761487961,
      "learning_rate": 2.5041233384883765e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5593824,
      "step": 19980
    },
    {
      "epoch": 222.05555555555554,
      "grad_norm": 0.0017993610817939043,
      "learning_rate": 2.5031415918267564e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5595264,
      "step": 19985
    },
    {
      "epoch": 222.11111111111111,
      "grad_norm": 0.0008889834862202406,
      "learning_rate": 2.5021598446806626e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5596688,
      "step": 19990
    },
    {
      "epoch": 222.16666666666666,
      "grad_norm": 0.0002443735138513148,
      "learning_rate": 2.5011780972014937e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5598080,
      "step": 19995
    },
    {
      "epoch": 222.22222222222223,
      "grad_norm": 0.004774985369294882,
      "learning_rate": 2.5001963495406478e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5599440,
      "step": 20000
    },
    {
      "epoch": 222.22222222222223,
      "eval_loss": 0.7466219663619995,
      "eval_runtime": 1.2326,
      "eval_samples_per_second": 32.451,
      "eval_steps_per_second": 16.226,
      "num_input_tokens_seen": 5599440,
      "step": 20000
    },
    {
      "epoch": 222.27777777777777,
      "grad_norm": 0.00418256688863039,
      "learning_rate": 2.499214601849522e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5600848,
      "step": 20005
    },
    {
      "epoch": 222.33333333333334,
      "grad_norm": 0.002264811424538493,
      "learning_rate": 2.4982328542795148e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5602240,
      "step": 20010
    },
    {
      "epoch": 222.38888888888889,
      "grad_norm": 0.0004665813175961375,
      "learning_rate": 2.497251106982024e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5603600,
      "step": 20015
    },
    {
      "epoch": 222.44444444444446,
      "grad_norm": 0.011342580430209637,
      "learning_rate": 2.4962693601084458e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5604992,
      "step": 20020
    },
    {
      "epoch": 222.5,
      "grad_norm": 0.0038666308391839266,
      "learning_rate": 2.4952876138101794e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5606384,
      "step": 20025
    },
    {
      "epoch": 222.55555555555554,
      "grad_norm": 0.0006547505035996437,
      "learning_rate": 2.4943058682386233e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5607792,
      "step": 20030
    },
    {
      "epoch": 222.61111111111111,
      "grad_norm": 0.01017184741795063,
      "learning_rate": 2.493324123545173e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5609216,
      "step": 20035
    },
    {
      "epoch": 222.66666666666666,
      "grad_norm": 0.00034825733746401966,
      "learning_rate": 2.4923423798812272e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5610656,
      "step": 20040
    },
    {
      "epoch": 222.72222222222223,
      "grad_norm": 0.0008934308425523341,
      "learning_rate": 2.4913606373981825e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5612064,
      "step": 20045
    },
    {
      "epoch": 222.77777777777777,
      "grad_norm": 0.0015530198579654098,
      "learning_rate": 2.4903788962474357e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5613472,
      "step": 20050
    },
    {
      "epoch": 222.83333333333334,
      "grad_norm": 0.0006859037093818188,
      "learning_rate": 2.489397156580385e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5614816,
      "step": 20055
    },
    {
      "epoch": 222.88888888888889,
      "grad_norm": 0.0032898138742893934,
      "learning_rate": 2.4884154185484246e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5616208,
      "step": 20060
    },
    {
      "epoch": 222.94444444444446,
      "grad_norm": 0.000244922615820542,
      "learning_rate": 2.4874336823029526e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5617632,
      "step": 20065
    },
    {
      "epoch": 223.0,
      "grad_norm": 0.015488671138882637,
      "learning_rate": 2.4864519479953656e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5619024,
      "step": 20070
    },
    {
      "epoch": 223.05555555555554,
      "grad_norm": 0.0006168658728711307,
      "learning_rate": 2.485470215777058e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5620448,
      "step": 20075
    },
    {
      "epoch": 223.11111111111111,
      "grad_norm": 0.0010577136417850852,
      "learning_rate": 2.4844884857994258e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5621840,
      "step": 20080
    },
    {
      "epoch": 223.16666666666666,
      "grad_norm": 0.0003592760185711086,
      "learning_rate": 2.4835067582138638e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5623280,
      "step": 20085
    },
    {
      "epoch": 223.22222222222223,
      "grad_norm": 0.0002019365638261661,
      "learning_rate": 2.4825250331717666e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5624672,
      "step": 20090
    },
    {
      "epoch": 223.27777777777777,
      "grad_norm": 0.0010637444211170077,
      "learning_rate": 2.4815433108245298e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5626112,
      "step": 20095
    },
    {
      "epoch": 223.33333333333334,
      "grad_norm": 0.0008902890258468688,
      "learning_rate": 2.4805615913235456e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5627552,
      "step": 20100
    },
    {
      "epoch": 223.38888888888889,
      "grad_norm": 0.001506911008618772,
      "learning_rate": 2.479579874820208e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5628944,
      "step": 20105
    },
    {
      "epoch": 223.44444444444446,
      "grad_norm": 0.004840063396841288,
      "learning_rate": 2.4785981614659115e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5630320,
      "step": 20110
    },
    {
      "epoch": 223.5,
      "grad_norm": 0.0004646486195269972,
      "learning_rate": 2.477616451412047e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5631664,
      "step": 20115
    },
    {
      "epoch": 223.55555555555554,
      "grad_norm": 0.0003245874831918627,
      "learning_rate": 2.476634744810007e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5633056,
      "step": 20120
    },
    {
      "epoch": 223.61111111111111,
      "grad_norm": 0.0016114027239382267,
      "learning_rate": 2.475653041811183e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5634448,
      "step": 20125
    },
    {
      "epoch": 223.66666666666666,
      "grad_norm": 0.00013655539078172296,
      "learning_rate": 2.4746713425669652e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5635824,
      "step": 20130
    },
    {
      "epoch": 223.72222222222223,
      "grad_norm": 0.007564155384898186,
      "learning_rate": 2.4736896472287458e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5637232,
      "step": 20135
    },
    {
      "epoch": 223.77777777777777,
      "grad_norm": 0.0038199329283088446,
      "learning_rate": 2.4727079559479124e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5638640,
      "step": 20140
    },
    {
      "epoch": 223.83333333333334,
      "grad_norm": 0.0005157505511306226,
      "learning_rate": 2.4717262688758557e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5640048,
      "step": 20145
    },
    {
      "epoch": 223.88888888888889,
      "grad_norm": 0.00045148003846406937,
      "learning_rate": 2.4707445861639637e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5641472,
      "step": 20150
    },
    {
      "epoch": 223.94444444444446,
      "grad_norm": 0.0020378355402499437,
      "learning_rate": 2.4697629079636244e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5642880,
      "step": 20155
    },
    {
      "epoch": 224.0,
      "grad_norm": 0.0018398732645437121,
      "learning_rate": 2.4687812344262244e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5644224,
      "step": 20160
    },
    {
      "epoch": 224.05555555555554,
      "grad_norm": 0.0004016743041574955,
      "learning_rate": 2.46779956570315e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5645600,
      "step": 20165
    },
    {
      "epoch": 224.11111111111111,
      "grad_norm": 0.0015042162267491221,
      "learning_rate": 2.466817901945787e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5647024,
      "step": 20170
    },
    {
      "epoch": 224.16666666666666,
      "grad_norm": 0.18290311098098755,
      "learning_rate": 2.4658362433055217e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5648448,
      "step": 20175
    },
    {
      "epoch": 224.22222222222223,
      "grad_norm": 0.0024623044300824404,
      "learning_rate": 2.4648545899337356e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5649856,
      "step": 20180
    },
    {
      "epoch": 224.27777777777777,
      "grad_norm": 0.1749204844236374,
      "learning_rate": 2.4638729419818143e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 5651216,
      "step": 20185
    },
    {
      "epoch": 224.33333333333334,
      "grad_norm": 0.05198056623339653,
      "learning_rate": 2.46289129960114e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5652624,
      "step": 20190
    },
    {
      "epoch": 224.38888888888889,
      "grad_norm": 0.015053823590278625,
      "learning_rate": 2.4619096629430924e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5654016,
      "step": 20195
    },
    {
      "epoch": 224.44444444444446,
      "grad_norm": 0.015049783512949944,
      "learning_rate": 2.4609280321590543e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5655424,
      "step": 20200
    },
    {
      "epoch": 224.44444444444446,
      "eval_loss": 0.6736544370651245,
      "eval_runtime": 1.2515,
      "eval_samples_per_second": 31.961,
      "eval_steps_per_second": 15.981,
      "num_input_tokens_seen": 5655424,
      "step": 20200
    },
    {
      "epoch": 224.5,
      "grad_norm": 0.002839656313881278,
      "learning_rate": 2.4599464074004037e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5656832,
      "step": 20205
    },
    {
      "epoch": 224.55555555555554,
      "grad_norm": 0.001531117013655603,
      "learning_rate": 2.4589647888185204e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5658208,
      "step": 20210
    },
    {
      "epoch": 224.61111111111111,
      "grad_norm": 0.00163594342302531,
      "learning_rate": 2.4579831765647836e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5659568,
      "step": 20215
    },
    {
      "epoch": 224.66666666666666,
      "grad_norm": 0.00010060337808681652,
      "learning_rate": 2.4570015707905676e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5661024,
      "step": 20220
    },
    {
      "epoch": 224.72222222222223,
      "grad_norm": 0.001997609855607152,
      "learning_rate": 2.4560199716472508e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5662416,
      "step": 20225
    },
    {
      "epoch": 224.77777777777777,
      "grad_norm": 0.010642443783581257,
      "learning_rate": 2.455038379286207e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5663840,
      "step": 20230
    },
    {
      "epoch": 224.83333333333334,
      "grad_norm": 0.0020829045679420233,
      "learning_rate": 2.4540567938588095e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5665264,
      "step": 20235
    },
    {
      "epoch": 224.88888888888889,
      "grad_norm": 0.00017531994672026485,
      "learning_rate": 2.4530752155164328e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5666656,
      "step": 20240
    },
    {
      "epoch": 224.94444444444446,
      "grad_norm": 0.00046504594502039254,
      "learning_rate": 2.4520936444104463e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5668064,
      "step": 20245
    },
    {
      "epoch": 225.0,
      "grad_norm": 0.0019951944705098867,
      "learning_rate": 2.4511120806922218e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5669456,
      "step": 20250
    },
    {
      "epoch": 225.05555555555554,
      "grad_norm": 0.00510542094707489,
      "learning_rate": 2.45013052451313e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5670800,
      "step": 20255
    },
    {
      "epoch": 225.11111111111111,
      "grad_norm": 0.0004848548851441592,
      "learning_rate": 2.4491489760245376e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5672160,
      "step": 20260
    },
    {
      "epoch": 225.16666666666666,
      "grad_norm": 0.000946118903812021,
      "learning_rate": 2.4481674353778115e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5673568,
      "step": 20265
    },
    {
      "epoch": 225.22222222222223,
      "grad_norm": 0.002383957151323557,
      "learning_rate": 2.447185902724319e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5674960,
      "step": 20270
    },
    {
      "epoch": 225.27777777777777,
      "grad_norm": 0.0029119013343006372,
      "learning_rate": 2.4462043782154233e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5676352,
      "step": 20275
    },
    {
      "epoch": 225.33333333333334,
      "grad_norm": 0.0003957401495426893,
      "learning_rate": 2.4452228620024895e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5677776,
      "step": 20280
    },
    {
      "epoch": 225.38888888888889,
      "grad_norm": 0.002162825083360076,
      "learning_rate": 2.4442413542368776e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5679216,
      "step": 20285
    },
    {
      "epoch": 225.44444444444446,
      "grad_norm": 0.004529166966676712,
      "learning_rate": 2.4432598550699502e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5680640,
      "step": 20290
    },
    {
      "epoch": 225.5,
      "grad_norm": 0.00011410571460146457,
      "learning_rate": 2.4422783646530663e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5682048,
      "step": 20295
    },
    {
      "epoch": 225.55555555555554,
      "grad_norm": 0.007560293190181255,
      "learning_rate": 2.441296883137584e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5683440,
      "step": 20300
    },
    {
      "epoch": 225.61111111111111,
      "grad_norm": 0.002714507980272174,
      "learning_rate": 2.4403154106748592e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5684816,
      "step": 20305
    },
    {
      "epoch": 225.66666666666666,
      "grad_norm": 0.00020149392366874963,
      "learning_rate": 2.4393339474162494e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5686208,
      "step": 20310
    },
    {
      "epoch": 225.72222222222223,
      "grad_norm": 0.0035374287981539965,
      "learning_rate": 2.4383524935131062e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5687600,
      "step": 20315
    },
    {
      "epoch": 225.77777777777777,
      "grad_norm": 0.0003783237771131098,
      "learning_rate": 2.437371049116784e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5689008,
      "step": 20320
    },
    {
      "epoch": 225.83333333333334,
      "grad_norm": 0.002999288495630026,
      "learning_rate": 2.436389614378632e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5690416,
      "step": 20325
    },
    {
      "epoch": 225.88888888888889,
      "grad_norm": 0.0003912140382453799,
      "learning_rate": 2.435408189450002e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5691792,
      "step": 20330
    },
    {
      "epoch": 225.94444444444446,
      "grad_norm": 0.003992462530732155,
      "learning_rate": 2.4344267744822406e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5693200,
      "step": 20335
    },
    {
      "epoch": 226.0,
      "grad_norm": 0.0016882536001503468,
      "learning_rate": 2.4334453696266944e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5694624,
      "step": 20340
    },
    {
      "epoch": 226.05555555555554,
      "grad_norm": 0.003167676040902734,
      "learning_rate": 2.432463975034708e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5695984,
      "step": 20345
    },
    {
      "epoch": 226.11111111111111,
      "grad_norm": 0.0038761685136705637,
      "learning_rate": 2.4314825908576265e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5697344,
      "step": 20350
    },
    {
      "epoch": 226.16666666666666,
      "grad_norm": 0.00247882679104805,
      "learning_rate": 2.4305012172467897e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5698704,
      "step": 20355
    },
    {
      "epoch": 226.22222222222223,
      "grad_norm": 0.0006453740643337369,
      "learning_rate": 2.4295198543535393e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5700096,
      "step": 20360
    },
    {
      "epoch": 226.27777777777777,
      "grad_norm": 0.001984697999432683,
      "learning_rate": 2.4285385023292124e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5701504,
      "step": 20365
    },
    {
      "epoch": 226.33333333333334,
      "grad_norm": 0.0007347746286541224,
      "learning_rate": 2.427557161325147e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5702864,
      "step": 20370
    },
    {
      "epoch": 226.38888888888889,
      "grad_norm": 0.002518095076084137,
      "learning_rate": 2.4265758314926778e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5704304,
      "step": 20375
    },
    {
      "epoch": 226.44444444444446,
      "grad_norm": 0.0002868259616661817,
      "learning_rate": 2.4255945129831373e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5705760,
      "step": 20380
    },
    {
      "epoch": 226.5,
      "grad_norm": 0.0007307397900149226,
      "learning_rate": 2.4246132059478578e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5707168,
      "step": 20385
    },
    {
      "epoch": 226.55555555555554,
      "grad_norm": 0.0003357003442943096,
      "learning_rate": 2.4236319105381706e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5708528,
      "step": 20390
    },
    {
      "epoch": 226.61111111111111,
      "grad_norm": 0.005860127974301577,
      "learning_rate": 2.422650626905401e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5709920,
      "step": 20395
    },
    {
      "epoch": 226.66666666666666,
      "grad_norm": 0.0014151382492855191,
      "learning_rate": 2.4216693552008785e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5711344,
      "step": 20400
    },
    {
      "epoch": 226.66666666666666,
      "eval_loss": 0.7212114930152893,
      "eval_runtime": 1.2363,
      "eval_samples_per_second": 32.353,
      "eval_steps_per_second": 16.177,
      "num_input_tokens_seen": 5711344,
      "step": 20400
    },
    {
      "epoch": 226.72222222222223,
      "grad_norm": 0.0017602695152163506,
      "learning_rate": 2.4206880955759247e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5712688,
      "step": 20405
    },
    {
      "epoch": 226.77777777777777,
      "grad_norm": 0.0009254604810848832,
      "learning_rate": 2.419706848181863e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5714128,
      "step": 20410
    },
    {
      "epoch": 226.83333333333334,
      "grad_norm": 0.002854111837223172,
      "learning_rate": 2.4187256131700153e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5715520,
      "step": 20415
    },
    {
      "epoch": 226.88888888888889,
      "grad_norm": 0.0001906455581774935,
      "learning_rate": 2.4177443906916985e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5716992,
      "step": 20420
    },
    {
      "epoch": 226.94444444444446,
      "grad_norm": 0.0007874320726841688,
      "learning_rate": 2.4167631808982303e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5718416,
      "step": 20425
    },
    {
      "epoch": 227.0,
      "grad_norm": 0.0027836342342197895,
      "learning_rate": 2.4157819839409264e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5719840,
      "step": 20430
    },
    {
      "epoch": 227.05555555555554,
      "grad_norm": 0.01775851845741272,
      "learning_rate": 2.414800799971098e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5721232,
      "step": 20435
    },
    {
      "epoch": 227.11111111111111,
      "grad_norm": 0.0019144449615851045,
      "learning_rate": 2.4138196291400582e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5722688,
      "step": 20440
    },
    {
      "epoch": 227.16666666666666,
      "grad_norm": 0.00040763505967333913,
      "learning_rate": 2.412838471599114e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5724080,
      "step": 20445
    },
    {
      "epoch": 227.22222222222223,
      "grad_norm": 0.00011591536895139143,
      "learning_rate": 2.411857327499572e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5725504,
      "step": 20450
    },
    {
      "epoch": 227.27777777777777,
      "grad_norm": 0.0018772375769913197,
      "learning_rate": 2.410876196992739e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5726896,
      "step": 20455
    },
    {
      "epoch": 227.33333333333334,
      "grad_norm": 0.02366393618285656,
      "learning_rate": 2.4098950802299156e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5728240,
      "step": 20460
    },
    {
      "epoch": 227.38888888888889,
      "grad_norm": 0.00046645026304759085,
      "learning_rate": 2.4089139773624027e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5729632,
      "step": 20465
    },
    {
      "epoch": 227.44444444444446,
      "grad_norm": 0.0004517661000136286,
      "learning_rate": 2.4079328885415007e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5731040,
      "step": 20470
    },
    {
      "epoch": 227.5,
      "grad_norm": 0.002851464319974184,
      "learning_rate": 2.4069518139185036e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5732464,
      "step": 20475
    },
    {
      "epoch": 227.55555555555554,
      "grad_norm": 0.0035426733084023,
      "learning_rate": 2.405970753644706e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5733824,
      "step": 20480
    },
    {
      "epoch": 227.61111111111111,
      "grad_norm": 0.0013435344444587827,
      "learning_rate": 2.4049897078714e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5735232,
      "step": 20485
    },
    {
      "epoch": 227.66666666666666,
      "grad_norm": 0.0014604725874960423,
      "learning_rate": 2.404008676749874e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5736608,
      "step": 20490
    },
    {
      "epoch": 227.72222222222223,
      "grad_norm": 0.057991404086351395,
      "learning_rate": 2.403027660431418e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5738032,
      "step": 20495
    },
    {
      "epoch": 227.77777777777777,
      "grad_norm": 0.0010146030690521002,
      "learning_rate": 2.402046659067314e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5739424,
      "step": 20500
    },
    {
      "epoch": 227.83333333333334,
      "grad_norm": 0.000716249574907124,
      "learning_rate": 2.401065672808847e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5740800,
      "step": 20505
    },
    {
      "epoch": 227.88888888888889,
      "grad_norm": 0.002371674869209528,
      "learning_rate": 2.400084701807296e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5742240,
      "step": 20510
    },
    {
      "epoch": 227.94444444444446,
      "grad_norm": 0.0005617579445242882,
      "learning_rate": 2.39910374621394e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5743616,
      "step": 20515
    },
    {
      "epoch": 228.0,
      "grad_norm": 0.05637703463435173,
      "learning_rate": 2.3981228061800544e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5745024,
      "step": 20520
    },
    {
      "epoch": 228.05555555555554,
      "grad_norm": 0.0014090943150222301,
      "learning_rate": 2.3971418818569115e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5746448,
      "step": 20525
    },
    {
      "epoch": 228.11111111111111,
      "grad_norm": 0.000704960897564888,
      "learning_rate": 2.3961609733957832e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5747872,
      "step": 20530
    },
    {
      "epoch": 228.16666666666666,
      "grad_norm": 0.0010998508660122752,
      "learning_rate": 2.395180080947939e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5749264,
      "step": 20535
    },
    {
      "epoch": 228.22222222222223,
      "grad_norm": 0.0009674277389422059,
      "learning_rate": 2.394199204664642e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5750704,
      "step": 20540
    },
    {
      "epoch": 228.27777777777777,
      "grad_norm": 0.0007467707036994398,
      "learning_rate": 2.3932183446971583e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5752080,
      "step": 20545
    },
    {
      "epoch": 228.33333333333334,
      "grad_norm": 0.02736949920654297,
      "learning_rate": 2.3922375011967473e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5753472,
      "step": 20550
    },
    {
      "epoch": 228.38888888888889,
      "grad_norm": 0.00032819833722896874,
      "learning_rate": 2.3912566743146676e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5754880,
      "step": 20555
    },
    {
      "epoch": 228.44444444444446,
      "grad_norm": 0.001383372349664569,
      "learning_rate": 2.390275864202176e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5756272,
      "step": 20560
    },
    {
      "epoch": 228.5,
      "grad_norm": 0.0003688345314003527,
      "learning_rate": 2.3892950710105243e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5757632,
      "step": 20565
    },
    {
      "epoch": 228.55555555555554,
      "grad_norm": 0.00021442797151394188,
      "learning_rate": 2.3883142948909635e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5758976,
      "step": 20570
    },
    {
      "epoch": 228.61111111111111,
      "grad_norm": 0.0007414563442580402,
      "learning_rate": 2.3873335359947433e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5760384,
      "step": 20575
    },
    {
      "epoch": 228.66666666666666,
      "grad_norm": 0.0004348534857854247,
      "learning_rate": 2.3863527944731066e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5761792,
      "step": 20580
    },
    {
      "epoch": 228.72222222222223,
      "grad_norm": 0.013642451725900173,
      "learning_rate": 2.385372070477298e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5763216,
      "step": 20585
    },
    {
      "epoch": 228.77777777777777,
      "grad_norm": 0.00032532456680200994,
      "learning_rate": 2.384391364158556e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5764576,
      "step": 20590
    },
    {
      "epoch": 228.83333333333334,
      "grad_norm": 0.018917705863714218,
      "learning_rate": 2.3834106756681185e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5765968,
      "step": 20595
    },
    {
      "epoch": 228.88888888888889,
      "grad_norm": 0.0028624841943383217,
      "learning_rate": 2.3824300051572206e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5767376,
      "step": 20600
    },
    {
      "epoch": 228.88888888888889,
      "eval_loss": 0.7255113124847412,
      "eval_runtime": 1.2382,
      "eval_samples_per_second": 32.304,
      "eval_steps_per_second": 16.152,
      "num_input_tokens_seen": 5767376,
      "step": 20600
    },
    {
      "epoch": 228.94444444444446,
      "grad_norm": 0.003460530424490571,
      "learning_rate": 2.3814493527770923e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5768832,
      "step": 20605
    },
    {
      "epoch": 229.0,
      "grad_norm": 0.00013375995331443846,
      "learning_rate": 2.3804687186789637e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5770256,
      "step": 20610
    },
    {
      "epoch": 229.05555555555554,
      "grad_norm": 0.0010316699044778943,
      "learning_rate": 2.379488103014062e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5771632,
      "step": 20615
    },
    {
      "epoch": 229.11111111111111,
      "grad_norm": 0.00013584535918198526,
      "learning_rate": 2.3785075059336086e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5773024,
      "step": 20620
    },
    {
      "epoch": 229.16666666666666,
      "grad_norm": 0.00025520139024592936,
      "learning_rate": 2.3775269275888248e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5774464,
      "step": 20625
    },
    {
      "epoch": 229.22222222222223,
      "grad_norm": 0.0005701640038751066,
      "learning_rate": 2.3765463681309274e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5775824,
      "step": 20630
    },
    {
      "epoch": 229.27777777777777,
      "grad_norm": 0.004068506881594658,
      "learning_rate": 2.3755658277111313e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5777232,
      "step": 20635
    },
    {
      "epoch": 229.33333333333334,
      "grad_norm": 0.000523449678439647,
      "learning_rate": 2.374585306480649e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5778592,
      "step": 20640
    },
    {
      "epoch": 229.38888888888889,
      "grad_norm": 0.0005864353151991963,
      "learning_rate": 2.3736048045906877e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5779984,
      "step": 20645
    },
    {
      "epoch": 229.44444444444446,
      "grad_norm": 0.003829934634268284,
      "learning_rate": 2.372624322192454e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5781376,
      "step": 20650
    },
    {
      "epoch": 229.5,
      "grad_norm": 0.0009692417224869132,
      "learning_rate": 2.3716438594371516e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5782752,
      "step": 20655
    },
    {
      "epoch": 229.55555555555554,
      "grad_norm": 0.0007199915708042681,
      "learning_rate": 2.3706634164759784e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5784176,
      "step": 20660
    },
    {
      "epoch": 229.61111111111111,
      "grad_norm": 0.0004150182066950947,
      "learning_rate": 2.3696829934601323e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5785552,
      "step": 20665
    },
    {
      "epoch": 229.66666666666666,
      "grad_norm": 0.0011126712197437882,
      "learning_rate": 2.3687025905408053e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5786960,
      "step": 20670
    },
    {
      "epoch": 229.72222222222223,
      "grad_norm": 0.00011156131222378463,
      "learning_rate": 2.3677222078691886e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5788352,
      "step": 20675
    },
    {
      "epoch": 229.77777777777777,
      "grad_norm": 0.0002346731926081702,
      "learning_rate": 2.366741845596471e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5789696,
      "step": 20680
    },
    {
      "epoch": 229.83333333333334,
      "grad_norm": 0.00030767955468036234,
      "learning_rate": 2.3657615038738343e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5791072,
      "step": 20685
    },
    {
      "epoch": 229.88888888888889,
      "grad_norm": 0.00025379142607562244,
      "learning_rate": 2.3647811828524614e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5792528,
      "step": 20690
    },
    {
      "epoch": 229.94444444444446,
      "grad_norm": 0.007639603689312935,
      "learning_rate": 2.363800882683529e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5793920,
      "step": 20695
    },
    {
      "epoch": 230.0,
      "grad_norm": 0.004349290858954191,
      "learning_rate": 2.3628206035182125e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5795360,
      "step": 20700
    },
    {
      "epoch": 230.05555555555554,
      "grad_norm": 0.0001910125429276377,
      "learning_rate": 2.361840345507683e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5796736,
      "step": 20705
    },
    {
      "epoch": 230.11111111111111,
      "grad_norm": 0.000889265036676079,
      "learning_rate": 2.3608601088031073e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5798144,
      "step": 20710
    },
    {
      "epoch": 230.16666666666666,
      "grad_norm": 0.0030657600145787,
      "learning_rate": 2.3598798935556516e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5799520,
      "step": 20715
    },
    {
      "epoch": 230.22222222222223,
      "grad_norm": 0.0011600542347878218,
      "learning_rate": 2.3588996999164784e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5800896,
      "step": 20720
    },
    {
      "epoch": 230.27777777777777,
      "grad_norm": 0.1517433226108551,
      "learning_rate": 2.3579195280367434e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5802256,
      "step": 20725
    },
    {
      "epoch": 230.33333333333334,
      "grad_norm": 0.00047028117114678025,
      "learning_rate": 2.356939378067603e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5803616,
      "step": 20730
    },
    {
      "epoch": 230.38888888888889,
      "grad_norm": 5.1916245865868405e-05,
      "learning_rate": 2.3559592501602092e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5804960,
      "step": 20735
    },
    {
      "epoch": 230.44444444444446,
      "grad_norm": 0.0005077195237390697,
      "learning_rate": 2.3549791444657076e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5806368,
      "step": 20740
    },
    {
      "epoch": 230.5,
      "grad_norm": 0.1127915158867836,
      "learning_rate": 2.353999061135246e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5807712,
      "step": 20745
    },
    {
      "epoch": 230.55555555555554,
      "grad_norm": 9.741583926370367e-05,
      "learning_rate": 2.3530190003199626e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5809072,
      "step": 20750
    },
    {
      "epoch": 230.61111111111111,
      "grad_norm": 0.00271789962425828,
      "learning_rate": 2.3520389621709965e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5810480,
      "step": 20755
    },
    {
      "epoch": 230.66666666666666,
      "grad_norm": 0.00035625023883767426,
      "learning_rate": 2.351058946839483e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5811888,
      "step": 20760
    },
    {
      "epoch": 230.72222222222223,
      "grad_norm": 0.0006307157455012202,
      "learning_rate": 2.350078954476551e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5813280,
      "step": 20765
    },
    {
      "epoch": 230.77777777777777,
      "grad_norm": 0.000331040529999882,
      "learning_rate": 2.3490989852333272e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5814704,
      "step": 20770
    },
    {
      "epoch": 230.83333333333334,
      "grad_norm": 0.0008514536893926561,
      "learning_rate": 2.3481190392609377e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5816144,
      "step": 20775
    },
    {
      "epoch": 230.88888888888889,
      "grad_norm": 0.0005160507280379534,
      "learning_rate": 2.3471391167105e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5817584,
      "step": 20780
    },
    {
      "epoch": 230.94444444444446,
      "grad_norm": 0.0007008324027992785,
      "learning_rate": 2.3461592177331325e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5819008,
      "step": 20785
    },
    {
      "epoch": 231.0,
      "grad_norm": 0.000677600153721869,
      "learning_rate": 2.345179342479946e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5820432,
      "step": 20790
    },
    {
      "epoch": 231.05555555555554,
      "grad_norm": 0.00112150888890028,
      "learning_rate": 2.3441994911020503e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5821872,
      "step": 20795
    },
    {
      "epoch": 231.11111111111111,
      "grad_norm": 0.005888773128390312,
      "learning_rate": 2.3432196637505522e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5823264,
      "step": 20800
    },
    {
      "epoch": 231.11111111111111,
      "eval_loss": 0.7554715871810913,
      "eval_runtime": 1.2438,
      "eval_samples_per_second": 32.158,
      "eval_steps_per_second": 16.079,
      "num_input_tokens_seen": 5823264,
      "step": 20800
    },
    {
      "epoch": 231.16666666666666,
      "grad_norm": 0.00020976732776034623,
      "learning_rate": 2.3422398605765515e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5824640,
      "step": 20805
    },
    {
      "epoch": 231.22222222222223,
      "grad_norm": 0.005016999319195747,
      "learning_rate": 2.3412600817311462e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5826000,
      "step": 20810
    },
    {
      "epoch": 231.27777777777777,
      "grad_norm": 0.003038015915080905,
      "learning_rate": 2.3402803273654326e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5827392,
      "step": 20815
    },
    {
      "epoch": 231.33333333333334,
      "grad_norm": 0.0046798367984592915,
      "learning_rate": 2.3393005976304983e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5828768,
      "step": 20820
    },
    {
      "epoch": 231.38888888888889,
      "grad_norm": 0.0030180790927261114,
      "learning_rate": 2.338320892677432e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5830160,
      "step": 20825
    },
    {
      "epoch": 231.44444444444446,
      "grad_norm": 0.0031906384974718094,
      "learning_rate": 2.3373412126573155e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5831600,
      "step": 20830
    },
    {
      "epoch": 231.5,
      "grad_norm": 0.0007507085683755577,
      "learning_rate": 2.3363615577212285e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5832992,
      "step": 20835
    },
    {
      "epoch": 231.55555555555554,
      "grad_norm": 0.003066964913159609,
      "learning_rate": 2.3353819280202455e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5834384,
      "step": 20840
    },
    {
      "epoch": 231.61111111111111,
      "grad_norm": 0.002055646851658821,
      "learning_rate": 2.334402323705438e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5835776,
      "step": 20845
    },
    {
      "epoch": 231.66666666666666,
      "grad_norm": 0.0018859306583181024,
      "learning_rate": 2.3334227449278725e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5837184,
      "step": 20850
    },
    {
      "epoch": 231.72222222222223,
      "grad_norm": 0.0013519146014004946,
      "learning_rate": 2.3324431918386143e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5838592,
      "step": 20855
    },
    {
      "epoch": 231.77777777777777,
      "grad_norm": 0.0006540620233863592,
      "learning_rate": 2.3314636645887207e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5840000,
      "step": 20860
    },
    {
      "epoch": 231.83333333333334,
      "grad_norm": 0.0025097625330090523,
      "learning_rate": 2.3304841633292487e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5841424,
      "step": 20865
    },
    {
      "epoch": 231.88888888888889,
      "grad_norm": 0.0005825088592246175,
      "learning_rate": 2.329504688211248e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5842816,
      "step": 20870
    },
    {
      "epoch": 231.94444444444446,
      "grad_norm": 0.001947994576767087,
      "learning_rate": 2.3285252393857677e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5844224,
      "step": 20875
    },
    {
      "epoch": 232.0,
      "grad_norm": 0.0008009826997295022,
      "learning_rate": 2.327545817003851e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5845632,
      "step": 20880
    },
    {
      "epoch": 232.05555555555554,
      "grad_norm": 0.002646626206114888,
      "learning_rate": 2.326566421216535e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5847040,
      "step": 20885
    },
    {
      "epoch": 232.11111111111111,
      "grad_norm": 0.011152997612953186,
      "learning_rate": 2.3255870521748565e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5848448,
      "step": 20890
    },
    {
      "epoch": 232.16666666666666,
      "grad_norm": 0.0075728390365839005,
      "learning_rate": 2.3246077100298474e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5849824,
      "step": 20895
    },
    {
      "epoch": 232.22222222222223,
      "grad_norm": 0.005118212662637234,
      "learning_rate": 2.3236283949325328e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5851200,
      "step": 20900
    },
    {
      "epoch": 232.27777777777777,
      "grad_norm": 0.003165455302223563,
      "learning_rate": 2.3226491070339368e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5852624,
      "step": 20905
    },
    {
      "epoch": 232.33333333333334,
      "grad_norm": 0.00089990597916767,
      "learning_rate": 2.3216698464850762e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5854032,
      "step": 20910
    },
    {
      "epoch": 232.38888888888889,
      "grad_norm": 0.0009931010426953435,
      "learning_rate": 2.320690613436967e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5855456,
      "step": 20915
    },
    {
      "epoch": 232.44444444444446,
      "grad_norm": 0.00021781631221529096,
      "learning_rate": 2.3197114080406192e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5856880,
      "step": 20920
    },
    {
      "epoch": 232.5,
      "grad_norm": 0.0004580788081511855,
      "learning_rate": 2.3187322304470365e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5858256,
      "step": 20925
    },
    {
      "epoch": 232.55555555555554,
      "grad_norm": 0.0006061866879463196,
      "learning_rate": 2.3177530808072222e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5859680,
      "step": 20930
    },
    {
      "epoch": 232.61111111111111,
      "grad_norm": 0.0005771088181063533,
      "learning_rate": 2.316773959272174e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5861040,
      "step": 20935
    },
    {
      "epoch": 232.66666666666666,
      "grad_norm": 0.0007574463961645961,
      "learning_rate": 2.3157948659928823e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5862496,
      "step": 20940
    },
    {
      "epoch": 232.72222222222223,
      "grad_norm": 0.0005946329329162836,
      "learning_rate": 2.3148158011203388e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5863872,
      "step": 20945
    },
    {
      "epoch": 232.77777777777777,
      "grad_norm": 0.001873006229288876,
      "learning_rate": 2.3138367648055253e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5865328,
      "step": 20950
    },
    {
      "epoch": 232.83333333333334,
      "grad_norm": 0.005711598787456751,
      "learning_rate": 2.312857757199422e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 5866720,
      "step": 20955
    },
    {
      "epoch": 232.88888888888889,
      "grad_norm": 0.007503168657422066,
      "learning_rate": 2.3118787784530048e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5868096,
      "step": 20960
    },
    {
      "epoch": 232.94444444444446,
      "grad_norm": 0.00011594441457418725,
      "learning_rate": 2.310899828717243e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5869488,
      "step": 20965
    },
    {
      "epoch": 233.0,
      "grad_norm": 0.00014526846644002944,
      "learning_rate": 2.309920908143104e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5870896,
      "step": 20970
    },
    {
      "epoch": 233.05555555555554,
      "grad_norm": 0.0019214852945879102,
      "learning_rate": 2.308942016881551e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5872272,
      "step": 20975
    },
    {
      "epoch": 233.11111111111111,
      "grad_norm": 0.0041165463626384735,
      "learning_rate": 2.307963155083539e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5873632,
      "step": 20980
    },
    {
      "epoch": 233.16666666666666,
      "grad_norm": 0.0022857736330479383,
      "learning_rate": 2.306984322900022e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5874992,
      "step": 20985
    },
    {
      "epoch": 233.22222222222223,
      "grad_norm": 0.00012548196536954492,
      "learning_rate": 2.3060055204819482e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5876368,
      "step": 20990
    },
    {
      "epoch": 233.27777777777777,
      "grad_norm": 0.000627894711215049,
      "learning_rate": 2.3050267479802604e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5877824,
      "step": 20995
    },
    {
      "epoch": 233.33333333333334,
      "grad_norm": 0.00300599099136889,
      "learning_rate": 2.304048005545899e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5879248,
      "step": 21000
    },
    {
      "epoch": 233.33333333333334,
      "eval_loss": 0.7266243100166321,
      "eval_runtime": 1.2432,
      "eval_samples_per_second": 32.174,
      "eval_steps_per_second": 16.087,
      "num_input_tokens_seen": 5879248,
      "step": 21000
    },
    {
      "epoch": 233.38888888888889,
      "grad_norm": 0.0003690715238917619,
      "learning_rate": 2.3030692933297972e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5880688,
      "step": 21005
    },
    {
      "epoch": 233.44444444444446,
      "grad_norm": 0.0008754149894230068,
      "learning_rate": 2.3020906114828843e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5882080,
      "step": 21010
    },
    {
      "epoch": 233.5,
      "grad_norm": 0.004159186966717243,
      "learning_rate": 2.301111960156088e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5883472,
      "step": 21015
    },
    {
      "epoch": 233.55555555555554,
      "grad_norm": 0.00020202892483212054,
      "learning_rate": 2.300133339500326e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5884912,
      "step": 21020
    },
    {
      "epoch": 233.61111111111111,
      "grad_norm": 0.002664302010089159,
      "learning_rate": 2.2991547496665148e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5886288,
      "step": 21025
    },
    {
      "epoch": 233.66666666666666,
      "grad_norm": 0.003940335474908352,
      "learning_rate": 2.298176190805565e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5887712,
      "step": 21030
    },
    {
      "epoch": 233.72222222222223,
      "grad_norm": 0.00012228562263771892,
      "learning_rate": 2.2971976630683826e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5889088,
      "step": 21035
    },
    {
      "epoch": 233.77777777777777,
      "grad_norm": 0.0009134980500675738,
      "learning_rate": 2.29621916660587e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5890512,
      "step": 21040
    },
    {
      "epoch": 233.83333333333334,
      "grad_norm": 0.0001370518293697387,
      "learning_rate": 2.295240701568922e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5891872,
      "step": 21045
    },
    {
      "epoch": 233.88888888888889,
      "grad_norm": 0.0003335162182338536,
      "learning_rate": 2.2942622681084312e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5893264,
      "step": 21050
    },
    {
      "epoch": 233.94444444444446,
      "grad_norm": 0.022168990224599838,
      "learning_rate": 2.293283866375284e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5894672,
      "step": 21055
    },
    {
      "epoch": 234.0,
      "grad_norm": 0.001616974943317473,
      "learning_rate": 2.2923054965203627e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5896096,
      "step": 21060
    },
    {
      "epoch": 234.05555555555554,
      "grad_norm": 0.0003439433639869094,
      "learning_rate": 2.2913271586945443e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5897440,
      "step": 21065
    },
    {
      "epoch": 234.11111111111111,
      "grad_norm": 0.00039538013515993953,
      "learning_rate": 2.290348853048699e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5898848,
      "step": 21070
    },
    {
      "epoch": 234.16666666666666,
      "grad_norm": 0.01118369959294796,
      "learning_rate": 2.2893705797336956e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5900256,
      "step": 21075
    },
    {
      "epoch": 234.22222222222223,
      "grad_norm": 6.868342461530119e-05,
      "learning_rate": 2.288392338900397e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5901664,
      "step": 21080
    },
    {
      "epoch": 234.27777777777777,
      "grad_norm": 0.001161991385743022,
      "learning_rate": 2.2874141306996576e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5903008,
      "step": 21085
    },
    {
      "epoch": 234.33333333333334,
      "grad_norm": 0.00013627692533191293,
      "learning_rate": 2.2864359552823312e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5904432,
      "step": 21090
    },
    {
      "epoch": 234.38888888888889,
      "grad_norm": 0.006582265254110098,
      "learning_rate": 2.2854578127992648e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5905840,
      "step": 21095
    },
    {
      "epoch": 234.44444444444446,
      "grad_norm": 0.0002698028110899031,
      "learning_rate": 2.2844797034012988e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5907264,
      "step": 21100
    },
    {
      "epoch": 234.5,
      "grad_norm": 0.00037913487176410854,
      "learning_rate": 2.2835016272392722e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5908640,
      "step": 21105
    },
    {
      "epoch": 234.55555555555554,
      "grad_norm": 0.00018694743630476296,
      "learning_rate": 2.2825235844640142e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5910032,
      "step": 21110
    },
    {
      "epoch": 234.61111111111111,
      "grad_norm": 0.0023195783142000437,
      "learning_rate": 2.2815455752263522e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5911376,
      "step": 21115
    },
    {
      "epoch": 234.66666666666666,
      "grad_norm": 0.00021994111011736095,
      "learning_rate": 2.2805675996771092e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5912800,
      "step": 21120
    },
    {
      "epoch": 234.72222222222223,
      "grad_norm": 0.00027125125052407384,
      "learning_rate": 2.2795896579670987e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5914192,
      "step": 21125
    },
    {
      "epoch": 234.77777777777777,
      "grad_norm": 0.0003167249960824847,
      "learning_rate": 2.2786117502471337e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5915600,
      "step": 21130
    },
    {
      "epoch": 234.83333333333334,
      "grad_norm": 0.00029821140924468637,
      "learning_rate": 2.2776338766680185e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5916928,
      "step": 21135
    },
    {
      "epoch": 234.88888888888889,
      "grad_norm": 0.0035832817666232586,
      "learning_rate": 2.2766560373805533e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5918368,
      "step": 21140
    },
    {
      "epoch": 234.94444444444446,
      "grad_norm": 0.00021079865109641105,
      "learning_rate": 2.2756782325355353e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5919840,
      "step": 21145
    },
    {
      "epoch": 235.0,
      "grad_norm": 0.00016777431301306933,
      "learning_rate": 2.2747004622837514e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5921216,
      "step": 21150
    },
    {
      "epoch": 235.05555555555554,
      "grad_norm": 0.0002280806947965175,
      "learning_rate": 2.2737227267759878e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5922608,
      "step": 21155
    },
    {
      "epoch": 235.11111111111111,
      "grad_norm": 0.0028204612899571657,
      "learning_rate": 2.272745026163024e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5924000,
      "step": 21160
    },
    {
      "epoch": 235.16666666666666,
      "grad_norm": 0.0008874436607584357,
      "learning_rate": 2.271767360595633e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5925376,
      "step": 21165
    },
    {
      "epoch": 235.22222222222223,
      "grad_norm": 0.0017236542189493775,
      "learning_rate": 2.270789730224583e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5926784,
      "step": 21170
    },
    {
      "epoch": 235.27777777777777,
      "grad_norm": 7.882290810812265e-05,
      "learning_rate": 2.2698121352006367e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5928240,
      "step": 21175
    },
    {
      "epoch": 235.33333333333334,
      "grad_norm": 0.0016718818806111813,
      "learning_rate": 2.2688345756745517e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5929600,
      "step": 21180
    },
    {
      "epoch": 235.38888888888889,
      "grad_norm": 0.0010163148399442434,
      "learning_rate": 2.267857051797081e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5930992,
      "step": 21185
    },
    {
      "epoch": 235.44444444444446,
      "grad_norm": 0.00022368023928720504,
      "learning_rate": 2.2668795637189695e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5932368,
      "step": 21190
    },
    {
      "epoch": 235.5,
      "grad_norm": 0.000657191383652389,
      "learning_rate": 2.2659021115909586e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5933744,
      "step": 21195
    },
    {
      "epoch": 235.55555555555554,
      "grad_norm": 0.00027443544240668416,
      "learning_rate": 2.2649246955637847e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5935168,
      "step": 21200
    },
    {
      "epoch": 235.55555555555554,
      "eval_loss": 0.7732191681861877,
      "eval_runtime": 1.2376,
      "eval_samples_per_second": 32.32,
      "eval_steps_per_second": 16.16,
      "num_input_tokens_seen": 5935168,
      "step": 21200
    },
    {
      "epoch": 235.61111111111111,
      "grad_norm": 0.0017644542967900634,
      "learning_rate": 2.2639473157881766e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5936544,
      "step": 21205
    },
    {
      "epoch": 235.66666666666666,
      "grad_norm": 0.0004947429406456649,
      "learning_rate": 2.2629699724148594e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5937904,
      "step": 21210
    },
    {
      "epoch": 235.72222222222223,
      "grad_norm": 0.00051125418394804,
      "learning_rate": 2.26199266559455e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5939296,
      "step": 21215
    },
    {
      "epoch": 235.77777777777777,
      "grad_norm": 0.0013317973352968693,
      "learning_rate": 2.2610153954779625e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5940704,
      "step": 21220
    },
    {
      "epoch": 235.83333333333334,
      "grad_norm": 0.0004946655826643109,
      "learning_rate": 2.2600381622158056e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5942080,
      "step": 21225
    },
    {
      "epoch": 235.88888888888889,
      "grad_norm": 0.0001506832049926743,
      "learning_rate": 2.2590609659587783e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5943536,
      "step": 21230
    },
    {
      "epoch": 235.94444444444446,
      "grad_norm": 0.0013774980325251818,
      "learning_rate": 2.2580838068575787e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5944912,
      "step": 21235
    },
    {
      "epoch": 236.0,
      "grad_norm": 0.0003489530645310879,
      "learning_rate": 2.257106685062896e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5946352,
      "step": 21240
    },
    {
      "epoch": 236.05555555555554,
      "grad_norm": 0.000307282229186967,
      "learning_rate": 2.256129600725415e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5947744,
      "step": 21245
    },
    {
      "epoch": 236.11111111111111,
      "grad_norm": 0.0005309681873768568,
      "learning_rate": 2.2551525539958145e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5949136,
      "step": 21250
    },
    {
      "epoch": 236.16666666666666,
      "grad_norm": 0.0032207826152443886,
      "learning_rate": 2.2541755450247663e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5950544,
      "step": 21255
    },
    {
      "epoch": 236.22222222222223,
      "grad_norm": 0.0001347365032415837,
      "learning_rate": 2.2531985739629382e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5951968,
      "step": 21260
    },
    {
      "epoch": 236.27777777777777,
      "grad_norm": 0.003419547574594617,
      "learning_rate": 2.2522216409609924e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5953376,
      "step": 21265
    },
    {
      "epoch": 236.33333333333334,
      "grad_norm": 0.0070363678969442844,
      "learning_rate": 2.2512447461695826e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5954816,
      "step": 21270
    },
    {
      "epoch": 236.38888888888889,
      "grad_norm": 0.00020558889082167298,
      "learning_rate": 2.2502678897393593e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5956208,
      "step": 21275
    },
    {
      "epoch": 236.44444444444446,
      "grad_norm": 0.0001570586464367807,
      "learning_rate": 2.2492910718209665e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5957616,
      "step": 21280
    },
    {
      "epoch": 236.5,
      "grad_norm": 0.0007860290352255106,
      "learning_rate": 2.2483142925650398e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5959024,
      "step": 21285
    },
    {
      "epoch": 236.55555555555554,
      "grad_norm": 0.006108334753662348,
      "learning_rate": 2.247337552122213e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5960400,
      "step": 21290
    },
    {
      "epoch": 236.61111111111111,
      "grad_norm": 0.0003261260862927884,
      "learning_rate": 2.24636085064311e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5961840,
      "step": 21295
    },
    {
      "epoch": 236.66666666666666,
      "grad_norm": 0.0022011438850313425,
      "learning_rate": 2.245384188278351e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5963232,
      "step": 21300
    },
    {
      "epoch": 236.72222222222223,
      "grad_norm": 0.0015063106548041105,
      "learning_rate": 2.2444075651785513e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5964656,
      "step": 21305
    },
    {
      "epoch": 236.77777777777777,
      "grad_norm": 0.016242237761616707,
      "learning_rate": 2.243430981494316e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5966080,
      "step": 21310
    },
    {
      "epoch": 236.83333333333334,
      "grad_norm": 0.0021395704243332148,
      "learning_rate": 2.2424544373762475e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5967504,
      "step": 21315
    },
    {
      "epoch": 236.88888888888889,
      "grad_norm": 0.11781240999698639,
      "learning_rate": 2.2414779329749418e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5968896,
      "step": 21320
    },
    {
      "epoch": 236.94444444444446,
      "grad_norm": 0.0015751213068142533,
      "learning_rate": 2.2405014684409873e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5970272,
      "step": 21325
    },
    {
      "epoch": 237.0,
      "grad_norm": 0.002880740910768509,
      "learning_rate": 2.239525043924968e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5971664,
      "step": 21330
    },
    {
      "epoch": 237.05555555555554,
      "grad_norm": 0.0008667429210618138,
      "learning_rate": 2.2385486595774592e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5972992,
      "step": 21335
    },
    {
      "epoch": 237.11111111111111,
      "grad_norm": 0.0029471544548869133,
      "learning_rate": 2.237572315549033e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5974368,
      "step": 21340
    },
    {
      "epoch": 237.16666666666666,
      "grad_norm": 0.00028032076079398394,
      "learning_rate": 2.2365960119902545e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5975728,
      "step": 21345
    },
    {
      "epoch": 237.22222222222223,
      "grad_norm": 0.3634391725063324,
      "learning_rate": 2.2356197490516806e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5977152,
      "step": 21350
    },
    {
      "epoch": 237.27777777777777,
      "grad_norm": 0.00028022192418575287,
      "learning_rate": 2.234643526883863e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5978608,
      "step": 21355
    },
    {
      "epoch": 237.33333333333334,
      "grad_norm": 0.0014463234692811966,
      "learning_rate": 2.2336673456373497e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5979968,
      "step": 21360
    },
    {
      "epoch": 237.38888888888889,
      "grad_norm": 0.0008377303602173924,
      "learning_rate": 2.2326912054626772e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5981376,
      "step": 21365
    },
    {
      "epoch": 237.44444444444446,
      "grad_norm": 0.00014229127555154264,
      "learning_rate": 2.2317151065103813e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5982816,
      "step": 21370
    },
    {
      "epoch": 237.5,
      "grad_norm": 0.0004443724756129086,
      "learning_rate": 2.2307390489309865e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5984208,
      "step": 21375
    },
    {
      "epoch": 237.55555555555554,
      "grad_norm": 0.0007499324274249375,
      "learning_rate": 2.2297630328750146e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5985616,
      "step": 21380
    },
    {
      "epoch": 237.61111111111111,
      "grad_norm": 0.006795480381697416,
      "learning_rate": 2.228787058492979e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5987024,
      "step": 21385
    },
    {
      "epoch": 237.66666666666666,
      "grad_norm": 0.0003962942573707551,
      "learning_rate": 2.2278111259353875e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5988448,
      "step": 21390
    },
    {
      "epoch": 237.72222222222223,
      "grad_norm": 0.004653853829950094,
      "learning_rate": 2.2268352353527395e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5989856,
      "step": 21395
    },
    {
      "epoch": 237.77777777777777,
      "grad_norm": 0.06484623998403549,
      "learning_rate": 2.225859386895533e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5991232,
      "step": 21400
    },
    {
      "epoch": 237.77777777777777,
      "eval_loss": 0.7357824444770813,
      "eval_runtime": 1.2359,
      "eval_samples_per_second": 32.365,
      "eval_steps_per_second": 16.182,
      "num_input_tokens_seen": 5991232,
      "step": 21400
    },
    {
      "epoch": 237.83333333333334,
      "grad_norm": 0.00027487004990689456,
      "learning_rate": 2.2248835807142525e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5992656,
      "step": 21405
    },
    {
      "epoch": 237.88888888888889,
      "grad_norm": 0.05770151689648628,
      "learning_rate": 2.2239078169593826e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 5994032,
      "step": 21410
    },
    {
      "epoch": 237.94444444444446,
      "grad_norm": 0.015806807205080986,
      "learning_rate": 2.222932095781396e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5995456,
      "step": 21415
    },
    {
      "epoch": 238.0,
      "grad_norm": 0.0019664207939058542,
      "learning_rate": 2.221956417330762e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5996896,
      "step": 21420
    },
    {
      "epoch": 238.05555555555554,
      "grad_norm": 0.0004182085976935923,
      "learning_rate": 2.2209807817579438e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5998288,
      "step": 21425
    },
    {
      "epoch": 238.11111111111111,
      "grad_norm": 0.0014993990771472454,
      "learning_rate": 2.220005189213394e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 5999712,
      "step": 21430
    },
    {
      "epoch": 238.16666666666666,
      "grad_norm": 0.03632230684161186,
      "learning_rate": 2.2190296398475624e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6001136,
      "step": 21435
    },
    {
      "epoch": 238.22222222222223,
      "grad_norm": 0.0016670881304889917,
      "learning_rate": 2.2180541338108926e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6002512,
      "step": 21440
    },
    {
      "epoch": 238.27777777777777,
      "grad_norm": 0.00013442672207020223,
      "learning_rate": 2.2170786712538176e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6003888,
      "step": 21445
    },
    {
      "epoch": 238.33333333333334,
      "grad_norm": 0.0005539089324884117,
      "learning_rate": 2.216103252326768e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6005280,
      "step": 21450
    },
    {
      "epoch": 238.38888888888889,
      "grad_norm": 0.0001687972981017083,
      "learning_rate": 2.2151278771801635e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6006688,
      "step": 21455
    },
    {
      "epoch": 238.44444444444446,
      "grad_norm": 0.0020868543069809675,
      "learning_rate": 2.21415254596442e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6008096,
      "step": 21460
    },
    {
      "epoch": 238.5,
      "grad_norm": 0.0005066402372904122,
      "learning_rate": 2.213177258829947e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6009568,
      "step": 21465
    },
    {
      "epoch": 238.55555555555554,
      "grad_norm": 9.003953164210543e-05,
      "learning_rate": 2.2122020159271445e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6010992,
      "step": 21470
    },
    {
      "epoch": 238.61111111111111,
      "grad_norm": 0.0008671157411299646,
      "learning_rate": 2.2112268174064075e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6012384,
      "step": 21475
    },
    {
      "epoch": 238.66666666666666,
      "grad_norm": 0.006109389942139387,
      "learning_rate": 2.2102516634181253e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6013776,
      "step": 21480
    },
    {
      "epoch": 238.72222222222223,
      "grad_norm": 0.001790319220162928,
      "learning_rate": 2.209276554112677e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6015200,
      "step": 21485
    },
    {
      "epoch": 238.77777777777777,
      "grad_norm": 0.00023971502378117293,
      "learning_rate": 2.2083014896404384e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6016624,
      "step": 21490
    },
    {
      "epoch": 238.83333333333334,
      "grad_norm": 0.007142891641706228,
      "learning_rate": 2.207326470151775e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6018032,
      "step": 21495
    },
    {
      "epoch": 238.88888888888889,
      "grad_norm": 0.00011377265764167532,
      "learning_rate": 2.2063514957970477e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6019424,
      "step": 21500
    },
    {
      "epoch": 238.94444444444446,
      "grad_norm": 0.0006495742127299309,
      "learning_rate": 2.205376566726611e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6020832,
      "step": 21505
    },
    {
      "epoch": 239.0,
      "grad_norm": 0.0010626354487612844,
      "learning_rate": 2.204401683090809e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6022192,
      "step": 21510
    },
    {
      "epoch": 239.05555555555554,
      "grad_norm": 0.0014033233746886253,
      "learning_rate": 2.203426845039982e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6023616,
      "step": 21515
    },
    {
      "epoch": 239.11111111111111,
      "grad_norm": 0.00032312949770130217,
      "learning_rate": 2.202452052724464e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6025008,
      "step": 21520
    },
    {
      "epoch": 239.16666666666666,
      "grad_norm": 0.0003892479289788753,
      "learning_rate": 2.2014773062945777e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6026368,
      "step": 21525
    },
    {
      "epoch": 239.22222222222223,
      "grad_norm": 0.00038362134364433587,
      "learning_rate": 2.2005026059006427e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6027760,
      "step": 21530
    },
    {
      "epoch": 239.27777777777777,
      "grad_norm": 0.0016086366958916187,
      "learning_rate": 2.1995279516929695e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6029120,
      "step": 21535
    },
    {
      "epoch": 239.33333333333334,
      "grad_norm": 0.002279398264363408,
      "learning_rate": 2.1985533438218613e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6030560,
      "step": 21540
    },
    {
      "epoch": 239.38888888888889,
      "grad_norm": 0.001118828309699893,
      "learning_rate": 2.197578782437617e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6031952,
      "step": 21545
    },
    {
      "epoch": 239.44444444444446,
      "grad_norm": 0.0002888777817133814,
      "learning_rate": 2.196604267690524e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6033376,
      "step": 21550
    },
    {
      "epoch": 239.5,
      "grad_norm": 0.0003340178227517754,
      "learning_rate": 2.195629799730865e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6034768,
      "step": 21555
    },
    {
      "epoch": 239.55555555555554,
      "grad_norm": 0.000872036733198911,
      "learning_rate": 2.1946553787089173e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6036176,
      "step": 21560
    },
    {
      "epoch": 239.61111111111111,
      "grad_norm": 0.0003777380334213376,
      "learning_rate": 2.193681004774947e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6037568,
      "step": 21565
    },
    {
      "epoch": 239.66666666666666,
      "grad_norm": 0.00036818961962126195,
      "learning_rate": 2.1927066780792154e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6038992,
      "step": 21570
    },
    {
      "epoch": 239.72222222222223,
      "grad_norm": 0.0028850154485553503,
      "learning_rate": 2.191732398771975e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6040384,
      "step": 21575
    },
    {
      "epoch": 239.77777777777777,
      "grad_norm": 0.0003492069954518229,
      "learning_rate": 2.1907581670034725e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6041792,
      "step": 21580
    },
    {
      "epoch": 239.83333333333334,
      "grad_norm": 0.0007839987520128489,
      "learning_rate": 2.189783982923948e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6043168,
      "step": 21585
    },
    {
      "epoch": 239.88888888888889,
      "grad_norm": 0.00032084438134916127,
      "learning_rate": 2.1888098466836303e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6044608,
      "step": 21590
    },
    {
      "epoch": 239.94444444444446,
      "grad_norm": 0.00025832673418335617,
      "learning_rate": 2.1878357584327457e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6045984,
      "step": 21595
    },
    {
      "epoch": 240.0,
      "grad_norm": 0.0023369784466922283,
      "learning_rate": 2.1868617183215103e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6047376,
      "step": 21600
    },
    {
      "epoch": 240.0,
      "eval_loss": 0.8028701543807983,
      "eval_runtime": 1.2335,
      "eval_samples_per_second": 32.429,
      "eval_steps_per_second": 16.214,
      "num_input_tokens_seen": 6047376,
      "step": 21600
    },
    {
      "epoch": 240.05555555555554,
      "grad_norm": 0.00020625288016162813,
      "learning_rate": 2.1858877265001327e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6048816,
      "step": 21605
    },
    {
      "epoch": 240.11111111111111,
      "grad_norm": 0.001180768827907741,
      "learning_rate": 2.184913783118816e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6050192,
      "step": 21610
    },
    {
      "epoch": 240.16666666666666,
      "grad_norm": 0.0002595208352431655,
      "learning_rate": 2.1839398883277522e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6051600,
      "step": 21615
    },
    {
      "epoch": 240.22222222222223,
      "grad_norm": 0.0007999226218089461,
      "learning_rate": 2.182966042277129e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6052992,
      "step": 21620
    },
    {
      "epoch": 240.27777777777777,
      "grad_norm": 0.00011349567648721859,
      "learning_rate": 2.181992245117128e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6054384,
      "step": 21625
    },
    {
      "epoch": 240.33333333333334,
      "grad_norm": 0.0003770995244849473,
      "learning_rate": 2.181018496997918e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6055744,
      "step": 21630
    },
    {
      "epoch": 240.38888888888889,
      "grad_norm": 0.0006126409280113876,
      "learning_rate": 2.1800447980696648e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6057184,
      "step": 21635
    },
    {
      "epoch": 240.44444444444446,
      "grad_norm": 0.00023074995260685682,
      "learning_rate": 2.1790711484825248e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6058576,
      "step": 21640
    },
    {
      "epoch": 240.5,
      "grad_norm": 0.0003769552567973733,
      "learning_rate": 2.178097548386646e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6059968,
      "step": 21645
    },
    {
      "epoch": 240.55555555555554,
      "grad_norm": 0.019768182188272476,
      "learning_rate": 2.1771239979321712e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6061424,
      "step": 21650
    },
    {
      "epoch": 240.61111111111111,
      "grad_norm": 0.0003155511221848428,
      "learning_rate": 2.1761504972692327e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6062800,
      "step": 21655
    },
    {
      "epoch": 240.66666666666666,
      "grad_norm": 0.0010528338607400656,
      "learning_rate": 2.1751770465479572e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6064208,
      "step": 21660
    },
    {
      "epoch": 240.72222222222223,
      "grad_norm": 0.01256028562784195,
      "learning_rate": 2.174203645918464e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6065616,
      "step": 21665
    },
    {
      "epoch": 240.77777777777777,
      "grad_norm": 6.851805665064603e-05,
      "learning_rate": 2.1732302955308624e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6067008,
      "step": 21670
    },
    {
      "epoch": 240.83333333333334,
      "grad_norm": 5.577411502599716e-05,
      "learning_rate": 2.172256995535255e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6068400,
      "step": 21675
    },
    {
      "epoch": 240.88888888888889,
      "grad_norm": 0.012524639256298542,
      "learning_rate": 2.171283746081739e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6069760,
      "step": 21680
    },
    {
      "epoch": 240.94444444444446,
      "grad_norm": 0.027839042246341705,
      "learning_rate": 2.1703105473203988e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6071120,
      "step": 21685
    },
    {
      "epoch": 241.0,
      "grad_norm": 0.0005985989118926227,
      "learning_rate": 2.1693373994013168e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6072512,
      "step": 21690
    },
    {
      "epoch": 241.05555555555554,
      "grad_norm": 0.0007040696800686419,
      "learning_rate": 2.168364302474562e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6073872,
      "step": 21695
    },
    {
      "epoch": 241.11111111111111,
      "grad_norm": 0.0003523552732076496,
      "learning_rate": 2.167391256690199e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6075280,
      "step": 21700
    },
    {
      "epoch": 241.16666666666666,
      "grad_norm": 0.0003464978071860969,
      "learning_rate": 2.1664182621982855e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6076704,
      "step": 21705
    },
    {
      "epoch": 241.22222222222223,
      "grad_norm": 0.002573705743998289,
      "learning_rate": 2.1654453191488673e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6078080,
      "step": 21710
    },
    {
      "epoch": 241.27777777777777,
      "grad_norm": 0.0008546922472305596,
      "learning_rate": 2.1644724276919846e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6079504,
      "step": 21715
    },
    {
      "epoch": 241.33333333333334,
      "grad_norm": 0.0009461546433158219,
      "learning_rate": 2.1634995879776715e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6080896,
      "step": 21720
    },
    {
      "epoch": 241.38888888888889,
      "grad_norm": 0.007622463162988424,
      "learning_rate": 2.162526800155949e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6082304,
      "step": 21725
    },
    {
      "epoch": 241.44444444444446,
      "grad_norm": 0.0034643837716430426,
      "learning_rate": 2.1615540643768363e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6083696,
      "step": 21730
    },
    {
      "epoch": 241.5,
      "grad_norm": 0.0229982677847147,
      "learning_rate": 2.160581380790339e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6085072,
      "step": 21735
    },
    {
      "epoch": 241.55555555555554,
      "grad_norm": 0.003978033550083637,
      "learning_rate": 2.1596087495464586e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6086464,
      "step": 21740
    },
    {
      "epoch": 241.61111111111111,
      "grad_norm": 0.002815583487972617,
      "learning_rate": 2.1586361707951866e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6087808,
      "step": 21745
    },
    {
      "epoch": 241.66666666666666,
      "grad_norm": 0.00038937985664233565,
      "learning_rate": 2.157663644686507e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6089216,
      "step": 21750
    },
    {
      "epoch": 241.72222222222223,
      "grad_norm": 8.912024350138381e-05,
      "learning_rate": 2.156691171370396e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6090640,
      "step": 21755
    },
    {
      "epoch": 241.77777777777777,
      "grad_norm": 0.0015231460565701127,
      "learning_rate": 2.1557187509968195e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6092016,
      "step": 21760
    },
    {
      "epoch": 241.83333333333334,
      "grad_norm": 0.0030128054786473513,
      "learning_rate": 2.1547463837157382e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6093456,
      "step": 21765
    },
    {
      "epoch": 241.88888888888889,
      "grad_norm": 0.0020131065975874662,
      "learning_rate": 2.1537740696771045e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6094880,
      "step": 21770
    },
    {
      "epoch": 241.94444444444446,
      "grad_norm": 0.0002807459095492959,
      "learning_rate": 2.1528018090308587e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6096352,
      "step": 21775
    },
    {
      "epoch": 242.0,
      "grad_norm": 0.008619074709713459,
      "learning_rate": 2.151829601926938e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6097712,
      "step": 21780
    },
    {
      "epoch": 242.05555555555554,
      "grad_norm": 0.000108494030428119,
      "learning_rate": 2.1508574485152684e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6099072,
      "step": 21785
    },
    {
      "epoch": 242.11111111111111,
      "grad_norm": 0.00023195327958092093,
      "learning_rate": 2.1498853489457667e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6100496,
      "step": 21790
    },
    {
      "epoch": 242.16666666666666,
      "grad_norm": 0.0007308248314075172,
      "learning_rate": 2.1489133033683455e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6101888,
      "step": 21795
    },
    {
      "epoch": 242.22222222222223,
      "grad_norm": 6.735136412316933e-05,
      "learning_rate": 2.1479413119329038e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6103328,
      "step": 21800
    },
    {
      "epoch": 242.22222222222223,
      "eval_loss": 0.7473093271255493,
      "eval_runtime": 1.236,
      "eval_samples_per_second": 32.362,
      "eval_steps_per_second": 16.181,
      "num_input_tokens_seen": 6103328,
      "step": 21800
    },
    {
      "epoch": 242.27777777777777,
      "grad_norm": 0.003484840737655759,
      "learning_rate": 2.1469693747893355e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6104704,
      "step": 21805
    },
    {
      "epoch": 242.33333333333334,
      "grad_norm": 0.011738155968487263,
      "learning_rate": 2.1459974920875274e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6106144,
      "step": 21810
    },
    {
      "epoch": 242.38888888888889,
      "grad_norm": 0.0003543765633367002,
      "learning_rate": 2.145025663977354e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6107552,
      "step": 21815
    },
    {
      "epoch": 242.44444444444446,
      "grad_norm": 0.00037464601336978376,
      "learning_rate": 2.1440538906086844e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6108912,
      "step": 21820
    },
    {
      "epoch": 242.5,
      "grad_norm": 0.0003524732601363212,
      "learning_rate": 2.1430821721313782e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6110320,
      "step": 21825
    },
    {
      "epoch": 242.55555555555554,
      "grad_norm": 0.00028772602672688663,
      "learning_rate": 2.142110508695286e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6111760,
      "step": 21830
    },
    {
      "epoch": 242.61111111111111,
      "grad_norm": 0.0018252568552270532,
      "learning_rate": 2.1411389004502515e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6113152,
      "step": 21835
    },
    {
      "epoch": 242.66666666666666,
      "grad_norm": 0.0012488211505115032,
      "learning_rate": 2.140167347546107e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6114512,
      "step": 21840
    },
    {
      "epoch": 242.72222222222223,
      "grad_norm": 0.0002760677016340196,
      "learning_rate": 2.1391958501326793e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6115888,
      "step": 21845
    },
    {
      "epoch": 242.77777777777777,
      "grad_norm": 0.0003354648652020842,
      "learning_rate": 2.1382244083597873e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6117296,
      "step": 21850
    },
    {
      "epoch": 242.83333333333334,
      "grad_norm": 6.230926373973489e-05,
      "learning_rate": 2.137253022377237e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6118688,
      "step": 21855
    },
    {
      "epoch": 242.88888888888889,
      "grad_norm": 0.00011737326713046059,
      "learning_rate": 2.136281692334829e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6120096,
      "step": 21860
    },
    {
      "epoch": 242.94444444444446,
      "grad_norm": 0.0004837668384425342,
      "learning_rate": 2.135310418382356e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6121504,
      "step": 21865
    },
    {
      "epoch": 243.0,
      "grad_norm": 0.008514154702425003,
      "learning_rate": 2.134339200669598e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6122896,
      "step": 21870
    },
    {
      "epoch": 243.05555555555554,
      "grad_norm": 0.08911944925785065,
      "learning_rate": 2.133368039346331e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6124304,
      "step": 21875
    },
    {
      "epoch": 243.11111111111111,
      "grad_norm": 0.00015840162814129144,
      "learning_rate": 2.1323969345623195e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6125744,
      "step": 21880
    },
    {
      "epoch": 243.16666666666666,
      "grad_norm": 0.002767866011708975,
      "learning_rate": 2.1314258864673207e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6127168,
      "step": 21885
    },
    {
      "epoch": 243.22222222222223,
      "grad_norm": 0.8894241452217102,
      "learning_rate": 2.130454895211082e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 6128544,
      "step": 21890
    },
    {
      "epoch": 243.27777777777777,
      "grad_norm": 0.00498094130307436,
      "learning_rate": 2.129483960943342e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6129936,
      "step": 21895
    },
    {
      "epoch": 243.33333333333334,
      "grad_norm": 0.0009347967570647597,
      "learning_rate": 2.128513083813831e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 6131376,
      "step": 21900
    },
    {
      "epoch": 243.38888888888889,
      "grad_norm": 0.0024107841309159994,
      "learning_rate": 2.1275422639722724e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 6132784,
      "step": 21905
    },
    {
      "epoch": 243.44444444444446,
      "grad_norm": 0.00023099765530787408,
      "learning_rate": 2.126571501568376e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6134176,
      "step": 21910
    },
    {
      "epoch": 243.5,
      "grad_norm": 0.020306922495365143,
      "learning_rate": 2.1256007967518478e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6135568,
      "step": 21915
    },
    {
      "epoch": 243.55555555555554,
      "grad_norm": 0.00010030933481175452,
      "learning_rate": 2.124630149672381e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 6136960,
      "step": 21920
    },
    {
      "epoch": 243.61111111111111,
      "grad_norm": 0.03159477934241295,
      "learning_rate": 2.1236595604796624e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6138336,
      "step": 21925
    },
    {
      "epoch": 243.66666666666666,
      "grad_norm": 0.0005163029418326914,
      "learning_rate": 2.1226890293233693e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6139744,
      "step": 21930
    },
    {
      "epoch": 243.72222222222223,
      "grad_norm": 1.091316819190979,
      "learning_rate": 2.1217185563531694e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 6141168,
      "step": 21935
    },
    {
      "epoch": 243.77777777777777,
      "grad_norm": 0.0003699126827996224,
      "learning_rate": 2.120748141718721e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6142592,
      "step": 21940
    },
    {
      "epoch": 243.83333333333334,
      "grad_norm": 0.0004727435589302331,
      "learning_rate": 2.1197777855696765e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6143984,
      "step": 21945
    },
    {
      "epoch": 243.88888888888889,
      "grad_norm": 0.015620280988514423,
      "learning_rate": 2.1188074880556746e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6145392,
      "step": 21950
    },
    {
      "epoch": 243.94444444444446,
      "grad_norm": 0.0002238462184322998,
      "learning_rate": 2.1178372493263495e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6146800,
      "step": 21955
    },
    {
      "epoch": 244.0,
      "grad_norm": 0.11674918979406357,
      "learning_rate": 2.116867069531322e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6148176,
      "step": 21960
    },
    {
      "epoch": 244.05555555555554,
      "grad_norm": 0.005939237307757139,
      "learning_rate": 2.1158969488202073e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6149616,
      "step": 21965
    },
    {
      "epoch": 244.11111111111111,
      "grad_norm": 0.0011894013732671738,
      "learning_rate": 2.114926887342611e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6150992,
      "step": 21970
    },
    {
      "epoch": 244.16666666666666,
      "grad_norm": 0.0013917149044573307,
      "learning_rate": 2.113956885248127e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6152384,
      "step": 21975
    },
    {
      "epoch": 244.22222222222223,
      "grad_norm": 0.020237235352396965,
      "learning_rate": 2.112986942686342e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6153792,
      "step": 21980
    },
    {
      "epoch": 244.27777777777777,
      "grad_norm": 0.0023455603513866663,
      "learning_rate": 2.112017059806835e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6155200,
      "step": 21985
    },
    {
      "epoch": 244.33333333333334,
      "grad_norm": 0.0009728614240884781,
      "learning_rate": 2.1110472367591724e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6156576,
      "step": 21990
    },
    {
      "epoch": 244.38888888888889,
      "grad_norm": 0.002661295235157013,
      "learning_rate": 2.1100774736929145e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6158000,
      "step": 21995
    },
    {
      "epoch": 244.44444444444446,
      "grad_norm": 0.014106548391282558,
      "learning_rate": 2.10910777075761e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6159376,
      "step": 22000
    },
    {
      "epoch": 244.44444444444446,
      "eval_loss": 0.7875652313232422,
      "eval_runtime": 1.2477,
      "eval_samples_per_second": 32.06,
      "eval_steps_per_second": 16.03,
      "num_input_tokens_seen": 6159376,
      "step": 22000
    },
    {
      "epoch": 244.5,
      "grad_norm": 0.00041917889029718935,
      "learning_rate": 2.108138128102799e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6160800,
      "step": 22005
    },
    {
      "epoch": 244.55555555555554,
      "grad_norm": 0.00013069169654045254,
      "learning_rate": 2.107168545878014e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6162192,
      "step": 22010
    },
    {
      "epoch": 244.61111111111111,
      "grad_norm": 0.005526286084204912,
      "learning_rate": 2.106199024232775e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6163600,
      "step": 22015
    },
    {
      "epoch": 244.66666666666666,
      "grad_norm": 0.0003067751240450889,
      "learning_rate": 2.105229563316595e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6165024,
      "step": 22020
    },
    {
      "epoch": 244.72222222222223,
      "grad_norm": 0.0004218422982376069,
      "learning_rate": 2.1042601632789784e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6166432,
      "step": 22025
    },
    {
      "epoch": 244.77777777777777,
      "grad_norm": 0.0014909007586538792,
      "learning_rate": 2.103290824269417e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6167824,
      "step": 22030
    },
    {
      "epoch": 244.83333333333334,
      "grad_norm": 0.00014724655193276703,
      "learning_rate": 2.1023215464373965e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6169184,
      "step": 22035
    },
    {
      "epoch": 244.88888888888889,
      "grad_norm": 0.0004448811523616314,
      "learning_rate": 2.1013523299323908e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6170592,
      "step": 22040
    },
    {
      "epoch": 244.94444444444446,
      "grad_norm": 0.009845199063420296,
      "learning_rate": 2.1003831749038654e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6171984,
      "step": 22045
    },
    {
      "epoch": 245.0,
      "grad_norm": 0.007258197758346796,
      "learning_rate": 2.099414081501277e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6173376,
      "step": 22050
    },
    {
      "epoch": 245.05555555555554,
      "grad_norm": 0.0013119925279170275,
      "learning_rate": 2.09844504987407e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6174752,
      "step": 22055
    },
    {
      "epoch": 245.11111111111111,
      "grad_norm": 0.0026470746379345655,
      "learning_rate": 2.097476080171683e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6176128,
      "step": 22060
    },
    {
      "epoch": 245.16666666666666,
      "grad_norm": 0.0003085501375608146,
      "learning_rate": 2.0965071725435436e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6177504,
      "step": 22065
    },
    {
      "epoch": 245.22222222222223,
      "grad_norm": 0.0006096676806919277,
      "learning_rate": 2.0955383271390684e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6178944,
      "step": 22070
    },
    {
      "epoch": 245.27777777777777,
      "grad_norm": 0.007695350795984268,
      "learning_rate": 2.094569544107666e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6180352,
      "step": 22075
    },
    {
      "epoch": 245.33333333333334,
      "grad_norm": 0.0008652419201098382,
      "learning_rate": 2.093600823598735e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6181680,
      "step": 22080
    },
    {
      "epoch": 245.38888888888889,
      "grad_norm": 0.0016675074584782124,
      "learning_rate": 2.092632165761663e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6183088,
      "step": 22085
    },
    {
      "epoch": 245.44444444444446,
      "grad_norm": 0.020046178251504898,
      "learning_rate": 2.091663570745832e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6184464,
      "step": 22090
    },
    {
      "epoch": 245.5,
      "grad_norm": 0.001355786225758493,
      "learning_rate": 2.0906950387006086e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6185808,
      "step": 22095
    },
    {
      "epoch": 245.55555555555554,
      "grad_norm": 0.00623982772231102,
      "learning_rate": 2.0897265697753543e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6187216,
      "step": 22100
    },
    {
      "epoch": 245.61111111111111,
      "grad_norm": 0.0008854445186443627,
      "learning_rate": 2.088758164119419e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6188592,
      "step": 22105
    },
    {
      "epoch": 245.66666666666666,
      "grad_norm": 0.0004466006357688457,
      "learning_rate": 2.0877898218821428e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6190016,
      "step": 22110
    },
    {
      "epoch": 245.72222222222223,
      "grad_norm": 0.0005496328813023865,
      "learning_rate": 2.0868215432128565e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6191408,
      "step": 22115
    },
    {
      "epoch": 245.77777777777777,
      "grad_norm": 0.00010033588478108868,
      "learning_rate": 2.0858533282608796e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6192832,
      "step": 22120
    },
    {
      "epoch": 245.83333333333334,
      "grad_norm": 0.00024442083667963743,
      "learning_rate": 2.084885177175524e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6194272,
      "step": 22125
    },
    {
      "epoch": 245.88888888888889,
      "grad_norm": 0.00029792284476570785,
      "learning_rate": 2.0839170901060917e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6195696,
      "step": 22130
    },
    {
      "epoch": 245.94444444444446,
      "grad_norm": 0.0010604608105495572,
      "learning_rate": 2.082949067201872e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6197104,
      "step": 22135
    },
    {
      "epoch": 246.0,
      "grad_norm": 0.00014138991537038237,
      "learning_rate": 2.0819811086121475e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6198512,
      "step": 22140
    },
    {
      "epoch": 246.05555555555554,
      "grad_norm": 0.005652309861034155,
      "learning_rate": 2.08101321448619e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6199888,
      "step": 22145
    },
    {
      "epoch": 246.11111111111111,
      "grad_norm": 0.0007807909860275686,
      "learning_rate": 2.080045384973259e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6201296,
      "step": 22150
    },
    {
      "epoch": 246.16666666666666,
      "grad_norm": 0.00022988597629591823,
      "learning_rate": 2.0790776202226082e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6202688,
      "step": 22155
    },
    {
      "epoch": 246.22222222222223,
      "grad_norm": 0.0006423001177608967,
      "learning_rate": 2.078109920383477e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6204048,
      "step": 22160
    },
    {
      "epoch": 246.27777777777777,
      "grad_norm": 0.0012303474359214306,
      "learning_rate": 2.0771422856050978e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6205488,
      "step": 22165
    },
    {
      "epoch": 246.33333333333334,
      "grad_norm": 0.00011669018567772582,
      "learning_rate": 2.076174716036693e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6206912,
      "step": 22170
    },
    {
      "epoch": 246.38888888888889,
      "grad_norm": 0.001560780219733715,
      "learning_rate": 2.075207211827472e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6208304,
      "step": 22175
    },
    {
      "epoch": 246.44444444444446,
      "grad_norm": 0.00033557586721144617,
      "learning_rate": 2.074239773126638e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6209728,
      "step": 22180
    },
    {
      "epoch": 246.5,
      "grad_norm": 0.005448055919259787,
      "learning_rate": 2.073272400083382e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6211168,
      "step": 22185
    },
    {
      "epoch": 246.55555555555554,
      "grad_norm": 0.0006780593539588153,
      "learning_rate": 2.072305092846883e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6212560,
      "step": 22190
    },
    {
      "epoch": 246.61111111111111,
      "grad_norm": 0.00020185804169159383,
      "learning_rate": 2.0713378515663152e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6213968,
      "step": 22195
    },
    {
      "epoch": 246.66666666666666,
      "grad_norm": 7.89690893725492e-05,
      "learning_rate": 2.070370676390836e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6215360,
      "step": 22200
    },
    {
      "epoch": 246.66666666666666,
      "eval_loss": 0.802726149559021,
      "eval_runtime": 1.2336,
      "eval_samples_per_second": 32.425,
      "eval_steps_per_second": 16.212,
      "num_input_tokens_seen": 6215360,
      "step": 22200
    },
    {
      "epoch": 246.72222222222223,
      "grad_norm": 0.00012056640116497874,
      "learning_rate": 2.0694035674695974e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6216784,
      "step": 22205
    },
    {
      "epoch": 246.77777777777777,
      "grad_norm": 0.004524414427578449,
      "learning_rate": 2.0684365249517416e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6218192,
      "step": 22210
    },
    {
      "epoch": 246.83333333333334,
      "grad_norm": 0.0008751393179409206,
      "learning_rate": 2.067469548986396e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6219616,
      "step": 22215
    },
    {
      "epoch": 246.88888888888889,
      "grad_norm": 0.0008465036517009139,
      "learning_rate": 2.066502639722681e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6220976,
      "step": 22220
    },
    {
      "epoch": 246.94444444444446,
      "grad_norm": 0.0005188906798139215,
      "learning_rate": 2.065535797309708e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6222352,
      "step": 22225
    },
    {
      "epoch": 247.0,
      "grad_norm": 0.0005821469821967185,
      "learning_rate": 2.0645690218965736e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6223728,
      "step": 22230
    },
    {
      "epoch": 247.05555555555554,
      "grad_norm": 0.0020115869119763374,
      "learning_rate": 2.063602313632369e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6225104,
      "step": 22235
    },
    {
      "epoch": 247.11111111111111,
      "grad_norm": 0.0005818973295390606,
      "learning_rate": 2.0626356726661704e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6226464,
      "step": 22240
    },
    {
      "epoch": 247.16666666666666,
      "grad_norm": 0.0015684099635109305,
      "learning_rate": 2.0616690991470477e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6227840,
      "step": 22245
    },
    {
      "epoch": 247.22222222222223,
      "grad_norm": 0.0006305472343228757,
      "learning_rate": 2.0607025932240595e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6229232,
      "step": 22250
    },
    {
      "epoch": 247.27777777777777,
      "grad_norm": 0.00011363387602614239,
      "learning_rate": 2.059736155046251e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6230624,
      "step": 22255
    },
    {
      "epoch": 247.33333333333334,
      "grad_norm": 0.00015310783055610955,
      "learning_rate": 2.0587697847626603e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6232032,
      "step": 22260
    },
    {
      "epoch": 247.38888888888889,
      "grad_norm": 0.0004827035591006279,
      "learning_rate": 2.057803482522314e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6233424,
      "step": 22265
    },
    {
      "epoch": 247.44444444444446,
      "grad_norm": 0.000142842618515715,
      "learning_rate": 2.056837248474227e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6234800,
      "step": 22270
    },
    {
      "epoch": 247.5,
      "grad_norm": 0.0009167765383608639,
      "learning_rate": 2.0558710827674064e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6236208,
      "step": 22275
    },
    {
      "epoch": 247.55555555555554,
      "grad_norm": 0.04829491302371025,
      "learning_rate": 2.054904985550845e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6237632,
      "step": 22280
    },
    {
      "epoch": 247.61111111111111,
      "grad_norm": 0.00012714818876702338,
      "learning_rate": 2.0539389569735287e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6239072,
      "step": 22285
    },
    {
      "epoch": 247.66666666666666,
      "grad_norm": 0.002196407411247492,
      "learning_rate": 2.052972997184431e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6240464,
      "step": 22290
    },
    {
      "epoch": 247.72222222222223,
      "grad_norm": 0.00010137445497093722,
      "learning_rate": 2.0520071063325146e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6241840,
      "step": 22295
    },
    {
      "epoch": 247.77777777777777,
      "grad_norm": 0.0007277659606188536,
      "learning_rate": 2.051041284566732e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6243280,
      "step": 22300
    },
    {
      "epoch": 247.83333333333334,
      "grad_norm": 0.0009279994992539287,
      "learning_rate": 2.050075532036026e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6244704,
      "step": 22305
    },
    {
      "epoch": 247.88888888888889,
      "grad_norm": 0.000633431423921138,
      "learning_rate": 2.0491098488893264e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6246128,
      "step": 22310
    },
    {
      "epoch": 247.94444444444446,
      "grad_norm": 0.00030119059374555945,
      "learning_rate": 2.0481442352755546e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6247504,
      "step": 22315
    },
    {
      "epoch": 248.0,
      "grad_norm": 0.0005740798660553992,
      "learning_rate": 2.0471786913436198e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6248896,
      "step": 22320
    },
    {
      "epoch": 248.05555555555554,
      "grad_norm": 9.961763862520456e-05,
      "learning_rate": 2.0462132172424218e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6250288,
      "step": 22325
    },
    {
      "epoch": 248.11111111111111,
      "grad_norm": 0.008852814324200153,
      "learning_rate": 2.0452478131208484e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6251696,
      "step": 22330
    },
    {
      "epoch": 248.16666666666666,
      "grad_norm": 0.00027685597888194025,
      "learning_rate": 2.0442824791277765e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6253056,
      "step": 22335
    },
    {
      "epoch": 248.22222222222223,
      "grad_norm": 0.0009252562886103988,
      "learning_rate": 2.0433172154120727e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6254464,
      "step": 22340
    },
    {
      "epoch": 248.27777777777777,
      "grad_norm": 0.0001054987296811305,
      "learning_rate": 2.0423520221225947e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6255872,
      "step": 22345
    },
    {
      "epoch": 248.33333333333334,
      "grad_norm": 0.008695971220731735,
      "learning_rate": 2.0413868994081848e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6257248,
      "step": 22350
    },
    {
      "epoch": 248.38888888888889,
      "grad_norm": 0.005030008964240551,
      "learning_rate": 2.0404218474176795e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6258608,
      "step": 22355
    },
    {
      "epoch": 248.44444444444446,
      "grad_norm": 0.0005653531407006085,
      "learning_rate": 2.0394568662999002e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6260000,
      "step": 22360
    },
    {
      "epoch": 248.5,
      "grad_norm": 0.002148991683498025,
      "learning_rate": 2.0384919562036593e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6261440,
      "step": 22365
    },
    {
      "epoch": 248.55555555555554,
      "grad_norm": 0.0011706982040777802,
      "learning_rate": 2.0375271172777593e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6262896,
      "step": 22370
    },
    {
      "epoch": 248.61111111111111,
      "grad_norm": 0.0006818120018579066,
      "learning_rate": 2.0365623496709885e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6264272,
      "step": 22375
    },
    {
      "epoch": 248.66666666666666,
      "grad_norm": 0.00024801879771985114,
      "learning_rate": 2.0355976535321283e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6265648,
      "step": 22380
    },
    {
      "epoch": 248.72222222222223,
      "grad_norm": 0.00045034216600470245,
      "learning_rate": 2.034633029009945e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6267040,
      "step": 22385
    },
    {
      "epoch": 248.77777777777777,
      "grad_norm": 0.0003194789169356227,
      "learning_rate": 2.0336684762531972e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6268432,
      "step": 22390
    },
    {
      "epoch": 248.83333333333334,
      "grad_norm": 0.0015080212615430355,
      "learning_rate": 2.032703995410631e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6269824,
      "step": 22395
    },
    {
      "epoch": 248.88888888888889,
      "grad_norm": 0.00033105211332440376,
      "learning_rate": 2.031739586630981e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6271232,
      "step": 22400
    },
    {
      "epoch": 248.88888888888889,
      "eval_loss": 0.7957814335823059,
      "eval_runtime": 1.2402,
      "eval_samples_per_second": 32.252,
      "eval_steps_per_second": 16.126,
      "num_input_tokens_seen": 6271232,
      "step": 22400
    },
    {
      "epoch": 248.94444444444446,
      "grad_norm": 0.0006263927207328379,
      "learning_rate": 2.0307752500629707e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6272656,
      "step": 22405
    },
    {
      "epoch": 249.0,
      "grad_norm": 0.010833787731826305,
      "learning_rate": 2.0298109858553144e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6274048,
      "step": 22410
    },
    {
      "epoch": 249.05555555555554,
      "grad_norm": 0.0001508628629380837,
      "learning_rate": 2.028846794156712e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6275504,
      "step": 22415
    },
    {
      "epoch": 249.11111111111111,
      "grad_norm": 0.0002532829239498824,
      "learning_rate": 2.027882675115856e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6276880,
      "step": 22420
    },
    {
      "epoch": 249.16666666666666,
      "grad_norm": 0.00025912950513884425,
      "learning_rate": 2.026918628881423e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6278288,
      "step": 22425
    },
    {
      "epoch": 249.22222222222223,
      "grad_norm": 0.0016188069712370634,
      "learning_rate": 2.0259546556020833e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6279680,
      "step": 22430
    },
    {
      "epoch": 249.27777777777777,
      "grad_norm": 0.00047227588947862387,
      "learning_rate": 2.024990755426493e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6281088,
      "step": 22435
    },
    {
      "epoch": 249.33333333333334,
      "grad_norm": 0.0009197634644806385,
      "learning_rate": 2.0240269285032975e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6282480,
      "step": 22440
    },
    {
      "epoch": 249.38888888888889,
      "grad_norm": 0.0005078706308268011,
      "learning_rate": 2.0230631749811306e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6283872,
      "step": 22445
    },
    {
      "epoch": 249.44444444444446,
      "grad_norm": 0.00020499633683357388,
      "learning_rate": 2.0220994950086162e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6285232,
      "step": 22450
    },
    {
      "epoch": 249.5,
      "grad_norm": 0.00024175707949325442,
      "learning_rate": 2.021135888734365e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6286592,
      "step": 22455
    },
    {
      "epoch": 249.55555555555554,
      "grad_norm": 6.565564399352297e-05,
      "learning_rate": 2.0201723563069783e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6288000,
      "step": 22460
    },
    {
      "epoch": 249.61111111111111,
      "grad_norm": 0.00026394875021651387,
      "learning_rate": 2.0192088978750433e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6289392,
      "step": 22465
    },
    {
      "epoch": 249.66666666666666,
      "grad_norm": 0.00023071446048561484,
      "learning_rate": 2.0182455135871385e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6290800,
      "step": 22470
    },
    {
      "epoch": 249.72222222222223,
      "grad_norm": 0.001057036337442696,
      "learning_rate": 2.0172822035918305e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6292176,
      "step": 22475
    },
    {
      "epoch": 249.77777777777777,
      "grad_norm": 0.004844668321311474,
      "learning_rate": 2.016318968037671e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6293552,
      "step": 22480
    },
    {
      "epoch": 249.83333333333334,
      "grad_norm": 0.0025392130482941866,
      "learning_rate": 2.015355807073206e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6294944,
      "step": 22485
    },
    {
      "epoch": 249.88888888888889,
      "grad_norm": 0.019238445907831192,
      "learning_rate": 2.0143927208469664e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6296384,
      "step": 22490
    },
    {
      "epoch": 249.94444444444446,
      "grad_norm": 5.6770943047013134e-05,
      "learning_rate": 2.0134297095074708e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6297808,
      "step": 22495
    },
    {
      "epoch": 250.0,
      "grad_norm": 4.439561234903522e-05,
      "learning_rate": 2.0124667732032297e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6299216,
      "step": 22500
    },
    {
      "epoch": 250.05555555555554,
      "grad_norm": 7.482501678168774e-05,
      "learning_rate": 2.011503912082738e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6300656,
      "step": 22505
    },
    {
      "epoch": 250.11111111111111,
      "grad_norm": 0.0010432725539430976,
      "learning_rate": 2.0105411262944823e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6302080,
      "step": 22510
    },
    {
      "epoch": 250.16666666666666,
      "grad_norm": 0.0015005022287368774,
      "learning_rate": 2.0095784159869366e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6303456,
      "step": 22515
    },
    {
      "epoch": 250.22222222222223,
      "grad_norm": 0.0002607002097647637,
      "learning_rate": 2.0086157813085608e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6304816,
      "step": 22520
    },
    {
      "epoch": 250.27777777777777,
      "grad_norm": 0.0003564483777154237,
      "learning_rate": 2.0076532224078068e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6306192,
      "step": 22525
    },
    {
      "epoch": 250.33333333333334,
      "grad_norm": 0.0005455950740724802,
      "learning_rate": 2.0066907394331142e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6307568,
      "step": 22530
    },
    {
      "epoch": 250.38888888888889,
      "grad_norm": 0.00024444193695671856,
      "learning_rate": 2.0057283325329077e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6308944,
      "step": 22535
    },
    {
      "epoch": 250.44444444444446,
      "grad_norm": 0.000522799149621278,
      "learning_rate": 2.0047660018556047e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6310352,
      "step": 22540
    },
    {
      "epoch": 250.5,
      "grad_norm": 0.0018539021257311106,
      "learning_rate": 2.0038037475496075e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6311712,
      "step": 22545
    },
    {
      "epoch": 250.55555555555554,
      "grad_norm": 0.0006793913780711591,
      "learning_rate": 2.0028415697633073e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6313184,
      "step": 22550
    },
    {
      "epoch": 250.61111111111111,
      "grad_norm": 0.00040901050670072436,
      "learning_rate": 2.0018794686450858e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6314560,
      "step": 22555
    },
    {
      "epoch": 250.66666666666666,
      "grad_norm": 0.00010216861846856773,
      "learning_rate": 2.0009174443433088e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6315984,
      "step": 22560
    },
    {
      "epoch": 250.72222222222223,
      "grad_norm": 0.010695104487240314,
      "learning_rate": 1.999955497006334e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6317360,
      "step": 22565
    },
    {
      "epoch": 250.77777777777777,
      "grad_norm": 0.003286150749772787,
      "learning_rate": 1.9989936267825067e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6318768,
      "step": 22570
    },
    {
      "epoch": 250.83333333333334,
      "grad_norm": 0.0001294455141760409,
      "learning_rate": 1.9980318338201572e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6320144,
      "step": 22575
    },
    {
      "epoch": 250.88888888888889,
      "grad_norm": 0.030317174270749092,
      "learning_rate": 1.997070118267607e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6321568,
      "step": 22580
    },
    {
      "epoch": 250.94444444444446,
      "grad_norm": 0.005951566621661186,
      "learning_rate": 1.9961084802731654e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6322960,
      "step": 22585
    },
    {
      "epoch": 251.0,
      "grad_norm": 0.00017364397353958338,
      "learning_rate": 1.9951469199851273e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6324320,
      "step": 22590
    },
    {
      "epoch": 251.05555555555554,
      "grad_norm": 0.00016950833378359675,
      "learning_rate": 1.99418543755178e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6325760,
      "step": 22595
    },
    {
      "epoch": 251.11111111111111,
      "grad_norm": 0.016036497429013252,
      "learning_rate": 1.9932240331213936e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6327136,
      "step": 22600
    },
    {
      "epoch": 251.11111111111111,
      "eval_loss": 0.7958273887634277,
      "eval_runtime": 1.2389,
      "eval_samples_per_second": 32.287,
      "eval_steps_per_second": 16.143,
      "num_input_tokens_seen": 6327136,
      "step": 22600
    },
    {
      "epoch": 251.16666666666666,
      "grad_norm": 0.00033177228760905564,
      "learning_rate": 1.9922627068422297e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6328512,
      "step": 22605
    },
    {
      "epoch": 251.22222222222223,
      "grad_norm": 0.0003435334947425872,
      "learning_rate": 1.991301458862538e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6329904,
      "step": 22610
    },
    {
      "epoch": 251.27777777777777,
      "grad_norm": 9.46634536376223e-05,
      "learning_rate": 1.9903402893305536e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6331344,
      "step": 22615
    },
    {
      "epoch": 251.33333333333334,
      "grad_norm": 0.0028096153400838375,
      "learning_rate": 1.9893791983945016e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6332768,
      "step": 22620
    },
    {
      "epoch": 251.38888888888889,
      "grad_norm": 0.0003079261805396527,
      "learning_rate": 1.988418186202594e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6334192,
      "step": 22625
    },
    {
      "epoch": 251.44444444444446,
      "grad_norm": 0.00013613687769975513,
      "learning_rate": 1.98745725290303e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6335568,
      "step": 22630
    },
    {
      "epoch": 251.5,
      "grad_norm": 0.000299863371765241,
      "learning_rate": 1.986496398644e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6336976,
      "step": 22635
    },
    {
      "epoch": 251.55555555555554,
      "grad_norm": 0.0007070168503560126,
      "learning_rate": 1.9855356235736777e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6338384,
      "step": 22640
    },
    {
      "epoch": 251.61111111111111,
      "grad_norm": 0.0005753046134486794,
      "learning_rate": 1.9845749278402277e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6339792,
      "step": 22645
    },
    {
      "epoch": 251.66666666666666,
      "grad_norm": 0.0001406440424034372,
      "learning_rate": 1.9836143115918006e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6341184,
      "step": 22650
    },
    {
      "epoch": 251.72222222222223,
      "grad_norm": 0.001287655672058463,
      "learning_rate": 1.9826537749765367e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6342608,
      "step": 22655
    },
    {
      "epoch": 251.77777777777777,
      "grad_norm": 0.0015251574805006385,
      "learning_rate": 1.9816933181425625e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6344048,
      "step": 22660
    },
    {
      "epoch": 251.83333333333334,
      "grad_norm": 0.002012518933042884,
      "learning_rate": 1.9807329412379903e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6345472,
      "step": 22665
    },
    {
      "epoch": 251.88888888888889,
      "grad_norm": 0.0014739936450496316,
      "learning_rate": 1.9797726444109247e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6346896,
      "step": 22670
    },
    {
      "epoch": 251.94444444444446,
      "grad_norm": 0.0005917931557632983,
      "learning_rate": 1.9788124278094557e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6348272,
      "step": 22675
    },
    {
      "epoch": 252.0,
      "grad_norm": 0.00027100101578980684,
      "learning_rate": 1.9778522915816594e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6349664,
      "step": 22680
    },
    {
      "epoch": 252.05555555555554,
      "grad_norm": 0.00016859378956723958,
      "learning_rate": 1.9768922358756014e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6351056,
      "step": 22685
    },
    {
      "epoch": 252.11111111111111,
      "grad_norm": 0.0007609858876094222,
      "learning_rate": 1.9759322608393353e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6352464,
      "step": 22690
    },
    {
      "epoch": 252.16666666666666,
      "grad_norm": 0.00025114690652117133,
      "learning_rate": 1.9749723666208992e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6353840,
      "step": 22695
    },
    {
      "epoch": 252.22222222222223,
      "grad_norm": 0.03131844848394394,
      "learning_rate": 1.9740125533683235e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6355232,
      "step": 22700
    },
    {
      "epoch": 252.27777777777777,
      "grad_norm": 0.0008284890209324658,
      "learning_rate": 1.9730528212296208e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6356672,
      "step": 22705
    },
    {
      "epoch": 252.33333333333334,
      "grad_norm": 0.019008278846740723,
      "learning_rate": 1.9720931703527945e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6358096,
      "step": 22710
    },
    {
      "epoch": 252.38888888888889,
      "grad_norm": 0.00016047549434006214,
      "learning_rate": 1.9711336008858373e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6359520,
      "step": 22715
    },
    {
      "epoch": 252.44444444444446,
      "grad_norm": 0.00031253305496647954,
      "learning_rate": 1.9701741129767233e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6360928,
      "step": 22720
    },
    {
      "epoch": 252.5,
      "grad_norm": 0.00026511907344684005,
      "learning_rate": 1.9692147067734202e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6362352,
      "step": 22725
    },
    {
      "epoch": 252.55555555555554,
      "grad_norm": 0.00021887623006477952,
      "learning_rate": 1.96825538242388e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6363712,
      "step": 22730
    },
    {
      "epoch": 252.61111111111111,
      "grad_norm": 0.0010957588674500585,
      "learning_rate": 1.967296140076041e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6365136,
      "step": 22735
    },
    {
      "epoch": 252.66666666666666,
      "grad_norm": 0.00011645564518403262,
      "learning_rate": 1.966336979877833e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6366528,
      "step": 22740
    },
    {
      "epoch": 252.72222222222223,
      "grad_norm": 0.0012835506349802017,
      "learning_rate": 1.9653779019771678e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6367968,
      "step": 22745
    },
    {
      "epoch": 252.77777777777777,
      "grad_norm": 6.871991354273632e-05,
      "learning_rate": 1.9644189065219488e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6369360,
      "step": 22750
    },
    {
      "epoch": 252.83333333333334,
      "grad_norm": 0.002382248407229781,
      "learning_rate": 1.9634599936600655e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6370752,
      "step": 22755
    },
    {
      "epoch": 252.88888888888889,
      "grad_norm": 0.00031002642936073244,
      "learning_rate": 1.9625011635393935e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6372128,
      "step": 22760
    },
    {
      "epoch": 252.94444444444446,
      "grad_norm": 0.00018940468726214021,
      "learning_rate": 1.9615424163077963e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6373488,
      "step": 22765
    },
    {
      "epoch": 253.0,
      "grad_norm": 0.0010869039688259363,
      "learning_rate": 1.9605837521131263e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6374912,
      "step": 22770
    },
    {
      "epoch": 253.05555555555554,
      "grad_norm": 7.317907147808e-05,
      "learning_rate": 1.9596251711032192e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6376352,
      "step": 22775
    },
    {
      "epoch": 253.11111111111111,
      "grad_norm": 0.0001645721058594063,
      "learning_rate": 1.958666673425903e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6377776,
      "step": 22780
    },
    {
      "epoch": 253.16666666666666,
      "grad_norm": 8.265260839834809e-05,
      "learning_rate": 1.957708259228987e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6379152,
      "step": 22785
    },
    {
      "epoch": 253.22222222222223,
      "grad_norm": 0.00018141977488994598,
      "learning_rate": 1.956749928660273e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6380528,
      "step": 22790
    },
    {
      "epoch": 253.27777777777777,
      "grad_norm": 0.0014374322490766644,
      "learning_rate": 1.955791681867547e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6381840,
      "step": 22795
    },
    {
      "epoch": 253.33333333333334,
      "grad_norm": 0.0025158815551549196,
      "learning_rate": 1.9548335189985824e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6383248,
      "step": 22800
    },
    {
      "epoch": 253.33333333333334,
      "eval_loss": 0.8212028741836548,
      "eval_runtime": 1.2472,
      "eval_samples_per_second": 32.073,
      "eval_steps_per_second": 16.036,
      "num_input_tokens_seen": 6383248,
      "step": 22800
    },
    {
      "epoch": 253.38888888888889,
      "grad_norm": 8.279320172732696e-05,
      "learning_rate": 1.9538754402011396e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6384608,
      "step": 22805
    },
    {
      "epoch": 253.44444444444446,
      "grad_norm": 0.00017828017007559538,
      "learning_rate": 1.952917445622968e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6385968,
      "step": 22810
    },
    {
      "epoch": 253.5,
      "grad_norm": 0.00012617858010344207,
      "learning_rate": 1.9519595354118005e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6387376,
      "step": 22815
    },
    {
      "epoch": 253.55555555555554,
      "grad_norm": 0.0006729629239998758,
      "learning_rate": 1.951001709715361e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6388800,
      "step": 22820
    },
    {
      "epoch": 253.61111111111111,
      "grad_norm": 0.00026014159084297717,
      "learning_rate": 1.9500439686813556e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6390176,
      "step": 22825
    },
    {
      "epoch": 253.66666666666666,
      "grad_norm": 0.00013863174535799772,
      "learning_rate": 1.949086312457482e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6391568,
      "step": 22830
    },
    {
      "epoch": 253.72222222222223,
      "grad_norm": 0.0011083963327109814,
      "learning_rate": 1.9481287411914223e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6392976,
      "step": 22835
    },
    {
      "epoch": 253.77777777777777,
      "grad_norm": 0.00013341351586859673,
      "learning_rate": 1.9471712550308457e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6394400,
      "step": 22840
    },
    {
      "epoch": 253.83333333333334,
      "grad_norm": 0.00013912691792938858,
      "learning_rate": 1.946213854123409e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6395840,
      "step": 22845
    },
    {
      "epoch": 253.88888888888889,
      "grad_norm": 0.00024783017579466105,
      "learning_rate": 1.9452565386167554e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6397232,
      "step": 22850
    },
    {
      "epoch": 253.94444444444446,
      "grad_norm": 0.000639656966086477,
      "learning_rate": 1.9442993086585142e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6398624,
      "step": 22855
    },
    {
      "epoch": 254.0,
      "grad_norm": 0.00017186014156322926,
      "learning_rate": 1.9433421643963043e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6400048,
      "step": 22860
    },
    {
      "epoch": 254.05555555555554,
      "grad_norm": 0.014807071536779404,
      "learning_rate": 1.942385105977727e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6401504,
      "step": 22865
    },
    {
      "epoch": 254.11111111111111,
      "grad_norm": 8.751917630434036e-05,
      "learning_rate": 1.9414281335503743e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6402896,
      "step": 22870
    },
    {
      "epoch": 254.16666666666666,
      "grad_norm": 0.0007945431279949844,
      "learning_rate": 1.9404712472618232e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6404288,
      "step": 22875
    },
    {
      "epoch": 254.22222222222223,
      "grad_norm": 0.0002510209451429546,
      "learning_rate": 1.939514447259636e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6405664,
      "step": 22880
    },
    {
      "epoch": 254.27777777777777,
      "grad_norm": 0.00012487186177168041,
      "learning_rate": 1.938557733691365e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6407072,
      "step": 22885
    },
    {
      "epoch": 254.33333333333334,
      "grad_norm": 8.026791329029948e-05,
      "learning_rate": 1.9376011067045476e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6408496,
      "step": 22890
    },
    {
      "epoch": 254.38888888888889,
      "grad_norm": 0.0005887945299036801,
      "learning_rate": 1.9366445664467065e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6409888,
      "step": 22895
    },
    {
      "epoch": 254.44444444444446,
      "grad_norm": 0.002377064200118184,
      "learning_rate": 1.9356881130653533e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6411280,
      "step": 22900
    },
    {
      "epoch": 254.5,
      "grad_norm": 0.00015738452202640474,
      "learning_rate": 1.9347317467079846e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6412672,
      "step": 22905
    },
    {
      "epoch": 254.55555555555554,
      "grad_norm": 0.0002455187786836177,
      "learning_rate": 1.9337754675220836e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6414048,
      "step": 22910
    },
    {
      "epoch": 254.61111111111111,
      "grad_norm": 0.0007405683281831443,
      "learning_rate": 1.9328192756551218e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6415456,
      "step": 22915
    },
    {
      "epoch": 254.66666666666666,
      "grad_norm": 0.00013971497537568212,
      "learning_rate": 1.931863171254555e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6416832,
      "step": 22920
    },
    {
      "epoch": 254.72222222222223,
      "grad_norm": 0.00015472581435460597,
      "learning_rate": 1.930907154467826e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6418224,
      "step": 22925
    },
    {
      "epoch": 254.77777777777777,
      "grad_norm": 0.0005010470631532371,
      "learning_rate": 1.9299512254423673e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6419632,
      "step": 22930
    },
    {
      "epoch": 254.83333333333334,
      "grad_norm": 0.00015741981042083353,
      "learning_rate": 1.9289953843255914e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6421088,
      "step": 22935
    },
    {
      "epoch": 254.88888888888889,
      "grad_norm": 0.0006683429819531739,
      "learning_rate": 1.9280396312649048e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6422496,
      "step": 22940
    },
    {
      "epoch": 254.94444444444446,
      "grad_norm": 0.0008141775033436716,
      "learning_rate": 1.9270839664076936e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6423824,
      "step": 22945
    },
    {
      "epoch": 255.0,
      "grad_norm": 0.00027920151478610933,
      "learning_rate": 1.9261283899013345e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6425200,
      "step": 22950
    },
    {
      "epoch": 255.05555555555554,
      "grad_norm": 0.001432459568604827,
      "learning_rate": 1.92517290189319e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6426640,
      "step": 22955
    },
    {
      "epoch": 255.11111111111111,
      "grad_norm": 0.0006246387492865324,
      "learning_rate": 1.924217502530607e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6428032,
      "step": 22960
    },
    {
      "epoch": 255.16666666666666,
      "grad_norm": 0.006431194022297859,
      "learning_rate": 1.9232621919609207e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6429440,
      "step": 22965
    },
    {
      "epoch": 255.22222222222223,
      "grad_norm": 0.00047905242536216974,
      "learning_rate": 1.9223069703314534e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6430816,
      "step": 22970
    },
    {
      "epoch": 255.27777777777777,
      "grad_norm": 0.0001524380495538935,
      "learning_rate": 1.92135183778951e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6432144,
      "step": 22975
    },
    {
      "epoch": 255.33333333333334,
      "grad_norm": 0.00016183672414626926,
      "learning_rate": 1.9203967944823857e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6433520,
      "step": 22980
    },
    {
      "epoch": 255.38888888888889,
      "grad_norm": 0.0009447407792322338,
      "learning_rate": 1.9194418405573588e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6434976,
      "step": 22985
    },
    {
      "epoch": 255.44444444444446,
      "grad_norm": 0.000805627612862736,
      "learning_rate": 1.9184869761616954e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6436384,
      "step": 22990
    },
    {
      "epoch": 255.5,
      "grad_norm": 0.0005732370191253722,
      "learning_rate": 1.9175322014426495e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6437792,
      "step": 22995
    },
    {
      "epoch": 255.55555555555554,
      "grad_norm": 0.0022068677935749292,
      "learning_rate": 1.9165775165474565e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6439168,
      "step": 23000
    },
    {
      "epoch": 255.55555555555554,
      "eval_loss": 0.7977734804153442,
      "eval_runtime": 1.2366,
      "eval_samples_per_second": 32.346,
      "eval_steps_per_second": 16.173,
      "num_input_tokens_seen": 6439168,
      "step": 23000
    },
    {
      "epoch": 255.61111111111111,
      "grad_norm": 0.0036849556490778923,
      "learning_rate": 1.9156229216233434e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6440576,
      "step": 23005
    },
    {
      "epoch": 255.66666666666666,
      "grad_norm": 0.001149207353591919,
      "learning_rate": 1.9146684168175184e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6441968,
      "step": 23010
    },
    {
      "epoch": 255.72222222222223,
      "grad_norm": 0.00038872502045705914,
      "learning_rate": 1.9137140022771796e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6443408,
      "step": 23015
    },
    {
      "epoch": 255.77777777777777,
      "grad_norm": 0.00041680640424601734,
      "learning_rate": 1.9127596781495103e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6444768,
      "step": 23020
    },
    {
      "epoch": 255.83333333333334,
      "grad_norm": 0.010144868865609169,
      "learning_rate": 1.9118054445816767e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6446160,
      "step": 23025
    },
    {
      "epoch": 255.88888888888889,
      "grad_norm": 0.00010748981003416702,
      "learning_rate": 1.9108513017208356e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6447600,
      "step": 23030
    },
    {
      "epoch": 255.94444444444446,
      "grad_norm": 8.457517105853185e-05,
      "learning_rate": 1.9098972497141287e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6449024,
      "step": 23035
    },
    {
      "epoch": 256.0,
      "grad_norm": 0.005374704021960497,
      "learning_rate": 1.9089432887086806e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6450432,
      "step": 23040
    },
    {
      "epoch": 256.05555555555554,
      "grad_norm": 0.0001011609856504947,
      "learning_rate": 1.9079894188516056e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6451856,
      "step": 23045
    },
    {
      "epoch": 256.1111111111111,
      "grad_norm": 0.002416664268821478,
      "learning_rate": 1.907035640290002e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6453280,
      "step": 23050
    },
    {
      "epoch": 256.1666666666667,
      "grad_norm": 0.0002515008090995252,
      "learning_rate": 1.9060819531709534e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6454688,
      "step": 23055
    },
    {
      "epoch": 256.22222222222223,
      "grad_norm": 0.006301973480731249,
      "learning_rate": 1.9051283576415325e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6456080,
      "step": 23060
    },
    {
      "epoch": 256.27777777777777,
      "grad_norm": 0.000784632982686162,
      "learning_rate": 1.904174853848793e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6457536,
      "step": 23065
    },
    {
      "epoch": 256.3333333333333,
      "grad_norm": 0.00017985529848374426,
      "learning_rate": 1.903221441939779e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6458928,
      "step": 23070
    },
    {
      "epoch": 256.3888888888889,
      "grad_norm": 0.0008243639604188502,
      "learning_rate": 1.9022681220615194e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6460336,
      "step": 23075
    },
    {
      "epoch": 256.44444444444446,
      "grad_norm": 0.0014792155707255006,
      "learning_rate": 1.9013148943610255e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6461744,
      "step": 23080
    },
    {
      "epoch": 256.5,
      "grad_norm": 0.0046288142912089825,
      "learning_rate": 1.9003617589852998e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6463184,
      "step": 23085
    },
    {
      "epoch": 256.55555555555554,
      "grad_norm": 0.005851684138178825,
      "learning_rate": 1.899408716081326e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6464544,
      "step": 23090
    },
    {
      "epoch": 256.6111111111111,
      "grad_norm": 0.00015005521709099412,
      "learning_rate": 1.898455765796075e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6465936,
      "step": 23095
    },
    {
      "epoch": 256.6666666666667,
      "grad_norm": 0.005519680213183165,
      "learning_rate": 1.8975029082765053e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6467328,
      "step": 23100
    },
    {
      "epoch": 256.72222222222223,
      "grad_norm": 0.0001415079168509692,
      "learning_rate": 1.8965501436695577e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6468704,
      "step": 23105
    },
    {
      "epoch": 256.77777777777777,
      "grad_norm": 0.0004432682762853801,
      "learning_rate": 1.895597472122161e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6470096,
      "step": 23110
    },
    {
      "epoch": 256.8333333333333,
      "grad_norm": 0.0003752092889044434,
      "learning_rate": 1.894644893781231e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6471488,
      "step": 23115
    },
    {
      "epoch": 256.8888888888889,
      "grad_norm": 0.0010331699158996344,
      "learning_rate": 1.893692408793665e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6472864,
      "step": 23120
    },
    {
      "epoch": 256.94444444444446,
      "grad_norm": 0.00030724884709343314,
      "learning_rate": 1.8927400173063493e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6474272,
      "step": 23125
    },
    {
      "epoch": 257.0,
      "grad_norm": 0.0005439304513856769,
      "learning_rate": 1.891787719466154e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6475680,
      "step": 23130
    },
    {
      "epoch": 257.05555555555554,
      "grad_norm": 0.0012195574818179011,
      "learning_rate": 1.8908355154199346e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6477088,
      "step": 23135
    },
    {
      "epoch": 257.1111111111111,
      "grad_norm": 6.680477235931903e-05,
      "learning_rate": 1.8898834053145357e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6478496,
      "step": 23140
    },
    {
      "epoch": 257.1666666666667,
      "grad_norm": 0.00020834327733609825,
      "learning_rate": 1.8889313892967813e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6479904,
      "step": 23145
    },
    {
      "epoch": 257.22222222222223,
      "grad_norm": 0.00010873577411985025,
      "learning_rate": 1.8879794675134863e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6481280,
      "step": 23150
    },
    {
      "epoch": 257.27777777777777,
      "grad_norm": 0.0005442913970910013,
      "learning_rate": 1.8870276401114494e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6482688,
      "step": 23155
    },
    {
      "epoch": 257.3333333333333,
      "grad_norm": 0.0004181239928584546,
      "learning_rate": 1.886075907237453e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6484096,
      "step": 23160
    },
    {
      "epoch": 257.3888888888889,
      "grad_norm": 0.00014379814092535526,
      "learning_rate": 1.8851242690382672e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6485504,
      "step": 23165
    },
    {
      "epoch": 257.44444444444446,
      "grad_norm": 0.00024854837101884186,
      "learning_rate": 1.884172725660645e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6486928,
      "step": 23170
    },
    {
      "epoch": 257.5,
      "grad_norm": 0.00012210225395392627,
      "learning_rate": 1.8832212772513277e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6488288,
      "step": 23175
    },
    {
      "epoch": 257.55555555555554,
      "grad_norm": 0.0005536149838007987,
      "learning_rate": 1.8822699239570414e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6489664,
      "step": 23180
    },
    {
      "epoch": 257.6111111111111,
      "grad_norm": 0.00013691376079805195,
      "learning_rate": 1.8813186659244943e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6491088,
      "step": 23185
    },
    {
      "epoch": 257.6666666666667,
      "grad_norm": 0.0007103057578206062,
      "learning_rate": 1.880367503300385e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6492464,
      "step": 23190
    },
    {
      "epoch": 257.72222222222223,
      "grad_norm": 0.00013514523743651807,
      "learning_rate": 1.8794164362313927e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6493840,
      "step": 23195
    },
    {
      "epoch": 257.77777777777777,
      "grad_norm": 0.0001768245710991323,
      "learning_rate": 1.878465464864185e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6495280,
      "step": 23200
    },
    {
      "epoch": 257.77777777777777,
      "eval_loss": 0.772354245185852,
      "eval_runtime": 1.2569,
      "eval_samples_per_second": 31.825,
      "eval_steps_per_second": 15.913,
      "num_input_tokens_seen": 6495280,
      "step": 23200
    },
    {
      "epoch": 257.8333333333333,
      "grad_norm": 0.00029422674560919404,
      "learning_rate": 1.877514589345414e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6496720,
      "step": 23205
    },
    {
      "epoch": 257.8888888888889,
      "grad_norm": 0.0010385316563770175,
      "learning_rate": 1.876563809821715e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6498144,
      "step": 23210
    },
    {
      "epoch": 257.94444444444446,
      "grad_norm": 0.00200998829677701,
      "learning_rate": 1.8756131264397106e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6499504,
      "step": 23215
    },
    {
      "epoch": 258.0,
      "grad_norm": 0.002955016680061817,
      "learning_rate": 1.87466253934601e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6500912,
      "step": 23220
    },
    {
      "epoch": 258.05555555555554,
      "grad_norm": 0.00011333912698319182,
      "learning_rate": 1.8737120486872033e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6502272,
      "step": 23225
    },
    {
      "epoch": 258.1111111111111,
      "grad_norm": 0.0003297474468126893,
      "learning_rate": 1.8727616546098696e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6503632,
      "step": 23230
    },
    {
      "epoch": 258.1666666666667,
      "grad_norm": 0.00013465501251630485,
      "learning_rate": 1.8718113572605716e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6505040,
      "step": 23235
    },
    {
      "epoch": 258.22222222222223,
      "grad_norm": 0.0012469239300116897,
      "learning_rate": 1.8708611567858554e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6506448,
      "step": 23240
    },
    {
      "epoch": 258.27777777777777,
      "grad_norm": 0.0009254709002561867,
      "learning_rate": 1.8699110533322565e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6507888,
      "step": 23245
    },
    {
      "epoch": 258.3333333333333,
      "grad_norm": 0.00016200989193748683,
      "learning_rate": 1.8689610470462897e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6509328,
      "step": 23250
    },
    {
      "epoch": 258.3888888888889,
      "grad_norm": 0.00015521651948802173,
      "learning_rate": 1.8680111380744604e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6510704,
      "step": 23255
    },
    {
      "epoch": 258.44444444444446,
      "grad_norm": 0.0037102713249623775,
      "learning_rate": 1.8670613265632564e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6512096,
      "step": 23260
    },
    {
      "epoch": 258.5,
      "grad_norm": 0.001646155142225325,
      "learning_rate": 1.866111612659149e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6513456,
      "step": 23265
    },
    {
      "epoch": 258.55555555555554,
      "grad_norm": 7.21548858564347e-05,
      "learning_rate": 1.8651619965085967e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6514896,
      "step": 23270
    },
    {
      "epoch": 258.6111111111111,
      "grad_norm": 0.00024740019580349326,
      "learning_rate": 1.8642124782580433e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6516336,
      "step": 23275
    },
    {
      "epoch": 258.6666666666667,
      "grad_norm": 0.0005715980660170317,
      "learning_rate": 1.8632630580539144e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6517744,
      "step": 23280
    },
    {
      "epoch": 258.72222222222223,
      "grad_norm": 0.03863895311951637,
      "learning_rate": 1.862313736042625e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6519120,
      "step": 23285
    },
    {
      "epoch": 258.77777777777777,
      "grad_norm": 0.0005108375335112214,
      "learning_rate": 1.8613645123705703e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6520512,
      "step": 23290
    },
    {
      "epoch": 258.8333333333333,
      "grad_norm": 9.850930655375123e-05,
      "learning_rate": 1.8604153871841328e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6521904,
      "step": 23295
    },
    {
      "epoch": 258.8888888888889,
      "grad_norm": 0.0009171649580821395,
      "learning_rate": 1.859466360629682e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6523296,
      "step": 23300
    },
    {
      "epoch": 258.94444444444446,
      "grad_norm": 0.0003854781680274755,
      "learning_rate": 1.8585174328535666e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6524672,
      "step": 23305
    },
    {
      "epoch": 259.0,
      "grad_norm": 0.004305514972656965,
      "learning_rate": 1.857568604002124e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6526112,
      "step": 23310
    },
    {
      "epoch": 259.05555555555554,
      "grad_norm": 0.00014940126857254654,
      "learning_rate": 1.8566198742216774e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6527456,
      "step": 23315
    },
    {
      "epoch": 259.1111111111111,
      "grad_norm": 0.008264485746622086,
      "learning_rate": 1.85567124365853e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6528848,
      "step": 23320
    },
    {
      "epoch": 259.1666666666667,
      "grad_norm": 0.00020373688312247396,
      "learning_rate": 1.854722712458975e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6530288,
      "step": 23325
    },
    {
      "epoch": 259.22222222222223,
      "grad_norm": 0.0028232343029230833,
      "learning_rate": 1.853774280769286e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6531712,
      "step": 23330
    },
    {
      "epoch": 259.27777777777777,
      "grad_norm": 0.001041584531776607,
      "learning_rate": 1.852825948735724e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6533104,
      "step": 23335
    },
    {
      "epoch": 259.3333333333333,
      "grad_norm": 0.0010187787702307105,
      "learning_rate": 1.851877716504534e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6534480,
      "step": 23340
    },
    {
      "epoch": 259.3888888888889,
      "grad_norm": 0.00049620121717453,
      "learning_rate": 1.8509295842219448e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6535872,
      "step": 23345
    },
    {
      "epoch": 259.44444444444446,
      "grad_norm": 0.014199519529938698,
      "learning_rate": 1.8499815520341697e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6537264,
      "step": 23350
    },
    {
      "epoch": 259.5,
      "grad_norm": 9.634549496695399e-05,
      "learning_rate": 1.8490336200874094e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6538688,
      "step": 23355
    },
    {
      "epoch": 259.55555555555554,
      "grad_norm": 0.0001696105900919065,
      "learning_rate": 1.848085788527844e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6540096,
      "step": 23360
    },
    {
      "epoch": 259.6111111111111,
      "grad_norm": 0.000357688928488642,
      "learning_rate": 1.847138057501644e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6541504,
      "step": 23365
    },
    {
      "epoch": 259.6666666666667,
      "grad_norm": 0.00034823687747120857,
      "learning_rate": 1.8461904271549582e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6542912,
      "step": 23370
    },
    {
      "epoch": 259.72222222222223,
      "grad_norm": 9.04979751794599e-05,
      "learning_rate": 1.845242897633926e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6544336,
      "step": 23375
    },
    {
      "epoch": 259.77777777777777,
      "grad_norm": 0.0017098826356232166,
      "learning_rate": 1.844295469084667e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6545696,
      "step": 23380
    },
    {
      "epoch": 259.8333333333333,
      "grad_norm": 0.0002704447542782873,
      "learning_rate": 1.843348141653286e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6547024,
      "step": 23385
    },
    {
      "epoch": 259.8888888888889,
      "grad_norm": 0.00011164740135427564,
      "learning_rate": 1.842400915485874e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6548400,
      "step": 23390
    },
    {
      "epoch": 259.94444444444446,
      "grad_norm": 0.0009143622010014951,
      "learning_rate": 1.8414537907285053e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6549824,
      "step": 23395
    },
    {
      "epoch": 260.0,
      "grad_norm": 0.00030002323910593987,
      "learning_rate": 1.840506767527237e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6551264,
      "step": 23400
    },
    {
      "epoch": 260.0,
      "eval_loss": 0.8360864520072937,
      "eval_runtime": 1.24,
      "eval_samples_per_second": 32.259,
      "eval_steps_per_second": 16.129,
      "num_input_tokens_seen": 6551264,
      "step": 23400
    },
    {
      "epoch": 260.05555555555554,
      "grad_norm": 0.00035894240136258304,
      "learning_rate": 1.8395598460281137e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6552720,
      "step": 23405
    },
    {
      "epoch": 260.1111111111111,
      "grad_norm": 0.0007885087979957461,
      "learning_rate": 1.838613026377161e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6554144,
      "step": 23410
    },
    {
      "epoch": 260.1666666666667,
      "grad_norm": 0.00014111690688878298,
      "learning_rate": 1.8376663087203917e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6555504,
      "step": 23415
    },
    {
      "epoch": 260.22222222222223,
      "grad_norm": 0.00139983044937253,
      "learning_rate": 1.8367196932038014e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6556960,
      "step": 23420
    },
    {
      "epoch": 260.27777777777777,
      "grad_norm": 0.0006328194867819548,
      "learning_rate": 1.8357731799733686e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6558336,
      "step": 23425
    },
    {
      "epoch": 260.3333333333333,
      "grad_norm": 0.00011314106086501852,
      "learning_rate": 1.8348267691750586e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6559744,
      "step": 23430
    },
    {
      "epoch": 260.3888888888889,
      "grad_norm": 0.00011091140186181292,
      "learning_rate": 1.833880460954821e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6561152,
      "step": 23435
    },
    {
      "epoch": 260.44444444444446,
      "grad_norm": 0.005669133737683296,
      "learning_rate": 1.8329342554585866e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6562544,
      "step": 23440
    },
    {
      "epoch": 260.5,
      "grad_norm": 0.0007554431795142591,
      "learning_rate": 1.8319881528322735e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6563888,
      "step": 23445
    },
    {
      "epoch": 260.55555555555554,
      "grad_norm": 0.00014101099804975092,
      "learning_rate": 1.8310421532217815e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6565264,
      "step": 23450
    },
    {
      "epoch": 260.6111111111111,
      "grad_norm": 0.0002700122131500393,
      "learning_rate": 1.8300962567729958e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6566656,
      "step": 23455
    },
    {
      "epoch": 260.6666666666667,
      "grad_norm": 8.023475675145164e-05,
      "learning_rate": 1.8291504636317866e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6568048,
      "step": 23460
    },
    {
      "epoch": 260.72222222222223,
      "grad_norm": 0.00035602739080786705,
      "learning_rate": 1.8282047739440055e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6569472,
      "step": 23465
    },
    {
      "epoch": 260.77777777777777,
      "grad_norm": 0.0004946375265717506,
      "learning_rate": 1.8272591878554903e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6570880,
      "step": 23470
    },
    {
      "epoch": 260.8333333333333,
      "grad_norm": 0.009792454540729523,
      "learning_rate": 1.8263137055120638e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6572304,
      "step": 23475
    },
    {
      "epoch": 260.8888888888889,
      "grad_norm": 8.575551328249276e-05,
      "learning_rate": 1.8253683270595295e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6573728,
      "step": 23480
    },
    {
      "epoch": 260.94444444444446,
      "grad_norm": 9.439748828299344e-05,
      "learning_rate": 1.824423052643677e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6575120,
      "step": 23485
    },
    {
      "epoch": 261.0,
      "grad_norm": 0.0001518354838481173,
      "learning_rate": 1.82347788241028e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6576512,
      "step": 23490
    },
    {
      "epoch": 261.05555555555554,
      "grad_norm": 0.00043854693649336696,
      "learning_rate": 1.8225328165050942e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6577920,
      "step": 23495
    },
    {
      "epoch": 261.1111111111111,
      "grad_norm": 0.0007033903384581208,
      "learning_rate": 1.821587855073863e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6579312,
      "step": 23500
    },
    {
      "epoch": 261.1666666666667,
      "grad_norm": 0.0007961745141074061,
      "learning_rate": 1.8206429982623086e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6580688,
      "step": 23505
    },
    {
      "epoch": 261.22222222222223,
      "grad_norm": 0.00028341118013486266,
      "learning_rate": 1.8196982462161416e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6582080,
      "step": 23510
    },
    {
      "epoch": 261.27777777777777,
      "grad_norm": 0.012841613031923771,
      "learning_rate": 1.818753599081055e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6583504,
      "step": 23515
    },
    {
      "epoch": 261.3333333333333,
      "grad_norm": 6.340147228911519e-05,
      "learning_rate": 1.817809057002724e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6584928,
      "step": 23520
    },
    {
      "epoch": 261.3888888888889,
      "grad_norm": 0.0002779863716568798,
      "learning_rate": 1.8168646201268096e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6586336,
      "step": 23525
    },
    {
      "epoch": 261.44444444444446,
      "grad_norm": 0.00030055452953092754,
      "learning_rate": 1.8159202885989557e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6587760,
      "step": 23530
    },
    {
      "epoch": 261.5,
      "grad_norm": 0.003390561556443572,
      "learning_rate": 1.814976062564789e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6589168,
      "step": 23535
    },
    {
      "epoch": 261.55555555555554,
      "grad_norm": 0.0012394976802170277,
      "learning_rate": 1.8140319421699234e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6590544,
      "step": 23540
    },
    {
      "epoch": 261.6111111111111,
      "grad_norm": 0.0009266218403354287,
      "learning_rate": 1.8130879275599515e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6591968,
      "step": 23545
    },
    {
      "epoch": 261.6666666666667,
      "grad_norm": 0.0008243187330663204,
      "learning_rate": 1.8121440188804544e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6593440,
      "step": 23550
    },
    {
      "epoch": 261.72222222222223,
      "grad_norm": 7.886788807809353e-05,
      "learning_rate": 1.811200216276993e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6594816,
      "step": 23555
    },
    {
      "epoch": 261.77777777777777,
      "grad_norm": 0.001697426661849022,
      "learning_rate": 1.810256519895115e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6596192,
      "step": 23560
    },
    {
      "epoch": 261.8333333333333,
      "grad_norm": 0.00027863855939358473,
      "learning_rate": 1.8093129298803494e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6597600,
      "step": 23565
    },
    {
      "epoch": 261.8888888888889,
      "grad_norm": 0.0012716833734884858,
      "learning_rate": 1.808369446378209e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6598976,
      "step": 23570
    },
    {
      "epoch": 261.94444444444446,
      "grad_norm": 0.004705403000116348,
      "learning_rate": 1.8074260695341914e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6600368,
      "step": 23575
    },
    {
      "epoch": 262.0,
      "grad_norm": 0.0003944597265217453,
      "learning_rate": 1.8064827994937782e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6601760,
      "step": 23580
    },
    {
      "epoch": 262.05555555555554,
      "grad_norm": 0.0013463094364851713,
      "learning_rate": 1.8055396364024317e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6603152,
      "step": 23585
    },
    {
      "epoch": 262.1111111111111,
      "grad_norm": 0.0002094280207529664,
      "learning_rate": 1.804596580405601e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6604592,
      "step": 23590
    },
    {
      "epoch": 262.1666666666667,
      "grad_norm": 0.00017367494001518935,
      "learning_rate": 1.8036536316487174e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6606032,
      "step": 23595
    },
    {
      "epoch": 262.22222222222223,
      "grad_norm": 0.0001991972530959174,
      "learning_rate": 1.802710790277193e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6607424,
      "step": 23600
    },
    {
      "epoch": 262.22222222222223,
      "eval_loss": 0.7746791243553162,
      "eval_runtime": 1.2328,
      "eval_samples_per_second": 32.446,
      "eval_steps_per_second": 16.223,
      "num_input_tokens_seen": 6607424,
      "step": 23600
    },
    {
      "epoch": 262.27777777777777,
      "grad_norm": 0.0003255538467783481,
      "learning_rate": 1.801768056436429e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6608768,
      "step": 23605
    },
    {
      "epoch": 262.3333333333333,
      "grad_norm": 0.007711958605796099,
      "learning_rate": 1.8008254302718035e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6610176,
      "step": 23610
    },
    {
      "epoch": 262.3888888888889,
      "grad_norm": 0.0005064439028501511,
      "learning_rate": 1.7998829119286837e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6611552,
      "step": 23615
    },
    {
      "epoch": 262.44444444444446,
      "grad_norm": 0.00043951478437520564,
      "learning_rate": 1.798940501552418e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6612944,
      "step": 23620
    },
    {
      "epoch": 262.5,
      "grad_norm": 0.0003709330048877746,
      "learning_rate": 1.797998199288336e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6614272,
      "step": 23625
    },
    {
      "epoch": 262.55555555555554,
      "grad_norm": 3.816053140326403e-05,
      "learning_rate": 1.7970560052817543e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6615680,
      "step": 23630
    },
    {
      "epoch": 262.6111111111111,
      "grad_norm": 0.00014464790001511574,
      "learning_rate": 1.7961139196779702e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6617072,
      "step": 23635
    },
    {
      "epoch": 262.6666666666667,
      "grad_norm": 0.002627609297633171,
      "learning_rate": 1.7951719426222647e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6618464,
      "step": 23640
    },
    {
      "epoch": 262.72222222222223,
      "grad_norm": 0.00019935831369366497,
      "learning_rate": 1.794230074259904e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6619888,
      "step": 23645
    },
    {
      "epoch": 262.77777777777777,
      "grad_norm": 0.00016357497952412814,
      "learning_rate": 1.7932883147361336e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6621296,
      "step": 23650
    },
    {
      "epoch": 262.8333333333333,
      "grad_norm": 5.193253673496656e-05,
      "learning_rate": 1.7923466641961865e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6622704,
      "step": 23655
    },
    {
      "epoch": 262.8888888888889,
      "grad_norm": 0.0009364917641505599,
      "learning_rate": 1.791405122785278e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6624112,
      "step": 23660
    },
    {
      "epoch": 262.94444444444446,
      "grad_norm": 0.00028952580760233104,
      "learning_rate": 1.7904636906486037e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6625504,
      "step": 23665
    },
    {
      "epoch": 263.0,
      "grad_norm": 5.034007699578069e-05,
      "learning_rate": 1.7895223679313448e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6626880,
      "step": 23670
    },
    {
      "epoch": 263.05555555555554,
      "grad_norm": 0.0003008747298736125,
      "learning_rate": 1.7885811547786653e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6628288,
      "step": 23675
    },
    {
      "epoch": 263.1111111111111,
      "grad_norm": 0.0011007323628291488,
      "learning_rate": 1.7876400513357115e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6629648,
      "step": 23680
    },
    {
      "epoch": 263.1666666666667,
      "grad_norm": 0.0002368421555729583,
      "learning_rate": 1.7866990577476146e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6631040,
      "step": 23685
    },
    {
      "epoch": 263.22222222222223,
      "grad_norm": 0.0004594626370817423,
      "learning_rate": 1.7857581741594863e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6632448,
      "step": 23690
    },
    {
      "epoch": 263.27777777777777,
      "grad_norm": 0.0005674426211044192,
      "learning_rate": 1.7848174007164237e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6633824,
      "step": 23695
    },
    {
      "epoch": 263.3333333333333,
      "grad_norm": 0.00047473417362198234,
      "learning_rate": 1.7838767375635052e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6635280,
      "step": 23700
    },
    {
      "epoch": 263.3888888888889,
      "grad_norm": 0.0004116552881896496,
      "learning_rate": 1.782936184845793e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6636656,
      "step": 23705
    },
    {
      "epoch": 263.44444444444446,
      "grad_norm": 0.0005371543229557574,
      "learning_rate": 1.7819957427083334e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6638048,
      "step": 23710
    },
    {
      "epoch": 263.5,
      "grad_norm": 0.0003938727895729244,
      "learning_rate": 1.7810554112961516e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6639488,
      "step": 23715
    },
    {
      "epoch": 263.55555555555554,
      "grad_norm": 0.0003115166909992695,
      "learning_rate": 1.7801151907542607e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6640880,
      "step": 23720
    },
    {
      "epoch": 263.6111111111111,
      "grad_norm": 0.00012925696501042694,
      "learning_rate": 1.7791750812276547e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6642288,
      "step": 23725
    },
    {
      "epoch": 263.6666666666667,
      "grad_norm": 0.0013488266849890351,
      "learning_rate": 1.778235082861309e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6643664,
      "step": 23730
    },
    {
      "epoch": 263.72222222222223,
      "grad_norm": 0.007924212142825127,
      "learning_rate": 1.777295195800184e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6645056,
      "step": 23735
    },
    {
      "epoch": 263.77777777777777,
      "grad_norm": 0.0009889433858916163,
      "learning_rate": 1.7763554201892215e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6646448,
      "step": 23740
    },
    {
      "epoch": 263.8333333333333,
      "grad_norm": 0.00010395058779977262,
      "learning_rate": 1.7754157561733476e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6647840,
      "step": 23745
    },
    {
      "epoch": 263.8888888888889,
      "grad_norm": 0.00012330517347436398,
      "learning_rate": 1.7744762038974702e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6649248,
      "step": 23750
    },
    {
      "epoch": 263.94444444444446,
      "grad_norm": 0.0007002812926657498,
      "learning_rate": 1.7735367635064788e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6650688,
      "step": 23755
    },
    {
      "epoch": 264.0,
      "grad_norm": 0.00012468457862269133,
      "learning_rate": 1.7725974351452474e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6652080,
      "step": 23760
    },
    {
      "epoch": 264.05555555555554,
      "grad_norm": 4.725301550934091e-05,
      "learning_rate": 1.771658218958634e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6653472,
      "step": 23765
    },
    {
      "epoch": 264.1111111111111,
      "grad_norm": 0.00020065084390807897,
      "learning_rate": 1.770719115091475e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6654880,
      "step": 23770
    },
    {
      "epoch": 264.1666666666667,
      "grad_norm": 0.0007739725988358259,
      "learning_rate": 1.7697801236885935e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6656256,
      "step": 23775
    },
    {
      "epoch": 264.22222222222223,
      "grad_norm": 0.0003149996045976877,
      "learning_rate": 1.7688412448947944e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6657664,
      "step": 23780
    },
    {
      "epoch": 264.27777777777777,
      "grad_norm": 0.00035549860331229866,
      "learning_rate": 1.767902478854862e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6659040,
      "step": 23785
    },
    {
      "epoch": 264.3333333333333,
      "grad_norm": 0.0013803527690470219,
      "learning_rate": 1.766963825713569e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6660384,
      "step": 23790
    },
    {
      "epoch": 264.3888888888889,
      "grad_norm": 0.0002879180246964097,
      "learning_rate": 1.766025285615665e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6661792,
      "step": 23795
    },
    {
      "epoch": 264.44444444444446,
      "grad_norm": 0.00016627901641186327,
      "learning_rate": 1.7650868587058854e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6663168,
      "step": 23800
    },
    {
      "epoch": 264.44444444444446,
      "eval_loss": 0.8496090173721313,
      "eval_runtime": 1.2621,
      "eval_samples_per_second": 31.693,
      "eval_steps_per_second": 15.847,
      "num_input_tokens_seen": 6663168,
      "step": 23800
    },
    {
      "epoch": 264.5,
      "grad_norm": 0.00019500225607771426,
      "learning_rate": 1.7641485451289484e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6664608,
      "step": 23805
    },
    {
      "epoch": 264.55555555555554,
      "grad_norm": 0.0003120337787549943,
      "learning_rate": 1.7632103450295534e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6665952,
      "step": 23810
    },
    {
      "epoch": 264.6111111111111,
      "grad_norm": 0.001276901108212769,
      "learning_rate": 1.762272258552381e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6667376,
      "step": 23815
    },
    {
      "epoch": 264.6666666666667,
      "grad_norm": 8.286058437079191e-05,
      "learning_rate": 1.7613342858420988e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6668784,
      "step": 23820
    },
    {
      "epoch": 264.72222222222223,
      "grad_norm": 0.003219309262931347,
      "learning_rate": 1.760396427043351e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6670176,
      "step": 23825
    },
    {
      "epoch": 264.77777777777777,
      "grad_norm": 0.0005347960395738482,
      "learning_rate": 1.7594586823007696e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6671584,
      "step": 23830
    },
    {
      "epoch": 264.8333333333333,
      "grad_norm": 0.00038511151797138155,
      "learning_rate": 1.7585210517589646e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6673024,
      "step": 23835
    },
    {
      "epoch": 264.8888888888889,
      "grad_norm": 0.0014054232742637396,
      "learning_rate": 1.7575835355625314e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6674448,
      "step": 23840
    },
    {
      "epoch": 264.94444444444446,
      "grad_norm": 0.00020960732945241034,
      "learning_rate": 1.756646133856048e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6675840,
      "step": 23845
    },
    {
      "epoch": 265.0,
      "grad_norm": 0.0002415698254480958,
      "learning_rate": 1.7557088467840714e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6677248,
      "step": 23850
    },
    {
      "epoch": 265.05555555555554,
      "grad_norm": 0.0002592316013760865,
      "learning_rate": 1.7547716744911438e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6678608,
      "step": 23855
    },
    {
      "epoch": 265.1111111111111,
      "grad_norm": 0.00041637246613390744,
      "learning_rate": 1.7538346171217902e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6680000,
      "step": 23860
    },
    {
      "epoch": 265.1666666666667,
      "grad_norm": 3.9579161239089444e-05,
      "learning_rate": 1.7528976748205146e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6681376,
      "step": 23865
    },
    {
      "epoch": 265.22222222222223,
      "grad_norm": 0.0007225691806524992,
      "learning_rate": 1.751960847731807e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6682768,
      "step": 23870
    },
    {
      "epoch": 265.27777777777777,
      "grad_norm": 0.0007676203385926783,
      "learning_rate": 1.7510241360001362e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6684208,
      "step": 23875
    },
    {
      "epoch": 265.3333333333333,
      "grad_norm": 8.167833584593609e-05,
      "learning_rate": 1.7500875397699562e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6685632,
      "step": 23880
    },
    {
      "epoch": 265.3888888888889,
      "grad_norm": 0.0005478225648403168,
      "learning_rate": 1.7491510591857015e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6687040,
      "step": 23885
    },
    {
      "epoch": 265.44444444444446,
      "grad_norm": 5.1032933697570115e-05,
      "learning_rate": 1.7482146943917896e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6688480,
      "step": 23890
    },
    {
      "epoch": 265.5,
      "grad_norm": 0.0004301243752706796,
      "learning_rate": 1.7472784455326185e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6689872,
      "step": 23895
    },
    {
      "epoch": 265.55555555555554,
      "grad_norm": 0.0019341037841513753,
      "learning_rate": 1.746342312752572e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6691280,
      "step": 23900
    },
    {
      "epoch": 265.6111111111111,
      "grad_norm": 0.0006346441805362701,
      "learning_rate": 1.74540629619601e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6692640,
      "step": 23905
    },
    {
      "epoch": 265.6666666666667,
      "grad_norm": 0.001266303239390254,
      "learning_rate": 1.7444703960072815e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6694048,
      "step": 23910
    },
    {
      "epoch": 265.72222222222223,
      "grad_norm": 0.0001387308439007029,
      "learning_rate": 1.7435346123307118e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6695440,
      "step": 23915
    },
    {
      "epoch": 265.77777777777777,
      "grad_norm": 0.000507791992276907,
      "learning_rate": 1.742598945310611e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6696832,
      "step": 23920
    },
    {
      "epoch": 265.8333333333333,
      "grad_norm": 0.000279098137980327,
      "learning_rate": 1.741663395091272e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6698288,
      "step": 23925
    },
    {
      "epoch": 265.8888888888889,
      "grad_norm": 0.0015902499435469508,
      "learning_rate": 1.7407279618169657e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6699712,
      "step": 23930
    },
    {
      "epoch": 265.94444444444446,
      "grad_norm": 0.00014242579345591366,
      "learning_rate": 1.73979264563195e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6701120,
      "step": 23935
    },
    {
      "epoch": 266.0,
      "grad_norm": 0.00016139417130034417,
      "learning_rate": 1.7388574466804625e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6702544,
      "step": 23940
    },
    {
      "epoch": 266.05555555555554,
      "grad_norm": 0.00019056405290029943,
      "learning_rate": 1.7379223651067207e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6703952,
      "step": 23945
    },
    {
      "epoch": 266.1111111111111,
      "grad_norm": 0.0002879663079511374,
      "learning_rate": 1.736987401054928e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6705360,
      "step": 23950
    },
    {
      "epoch": 266.1666666666667,
      "grad_norm": 0.0001311632222495973,
      "learning_rate": 1.736052554669266e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6706752,
      "step": 23955
    },
    {
      "epoch": 266.22222222222223,
      "grad_norm": 0.0002779131755232811,
      "learning_rate": 1.7351178260939007e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6708176,
      "step": 23960
    },
    {
      "epoch": 266.27777777777777,
      "grad_norm": 0.00026216040714643896,
      "learning_rate": 1.7341832154729794e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6709584,
      "step": 23965
    },
    {
      "epoch": 266.3333333333333,
      "grad_norm": 0.00030443360446952283,
      "learning_rate": 1.7332487229506286e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6710912,
      "step": 23970
    },
    {
      "epoch": 266.3888888888889,
      "grad_norm": 0.00011334290320519358,
      "learning_rate": 1.732314348670961e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6712272,
      "step": 23975
    },
    {
      "epoch": 266.44444444444446,
      "grad_norm": 4.7627505409764126e-05,
      "learning_rate": 1.7313800927780686e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6713648,
      "step": 23980
    },
    {
      "epoch": 266.5,
      "grad_norm": 9.102743933908641e-05,
      "learning_rate": 1.7304459554160245e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6715040,
      "step": 23985
    },
    {
      "epoch": 266.55555555555554,
      "grad_norm": 0.004731886088848114,
      "learning_rate": 1.7295119367288853e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6716432,
      "step": 23990
    },
    {
      "epoch": 266.6111111111111,
      "grad_norm": 0.0015257291961461306,
      "learning_rate": 1.728578036860688e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6717808,
      "step": 23995
    },
    {
      "epoch": 266.6666666666667,
      "grad_norm": 4.168119630776346e-05,
      "learning_rate": 1.7276442559554513e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6719216,
      "step": 24000
    },
    {
      "epoch": 266.6666666666667,
      "eval_loss": 0.8076408505439758,
      "eval_runtime": 1.2393,
      "eval_samples_per_second": 32.276,
      "eval_steps_per_second": 16.138,
      "num_input_tokens_seen": 6719216,
      "step": 24000
    },
    {
      "epoch": 266.72222222222223,
      "grad_norm": 0.001547609455883503,
      "learning_rate": 1.726710594157177e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6720592,
      "step": 24005
    },
    {
      "epoch": 266.77777777777777,
      "grad_norm": 0.0001229832851095125,
      "learning_rate": 1.725777051609846e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6722032,
      "step": 24010
    },
    {
      "epoch": 266.8333333333333,
      "grad_norm": 7.625164289493114e-05,
      "learning_rate": 1.7248436284574228e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6723456,
      "step": 24015
    },
    {
      "epoch": 266.8888888888889,
      "grad_norm": 0.0010745293693616986,
      "learning_rate": 1.723910324843855e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6724880,
      "step": 24020
    },
    {
      "epoch": 266.94444444444446,
      "grad_norm": 0.00019408250227570534,
      "learning_rate": 1.722977140913067e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6726304,
      "step": 24025
    },
    {
      "epoch": 267.0,
      "grad_norm": 0.0002871383912861347,
      "learning_rate": 1.7220440768089688e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6727728,
      "step": 24030
    },
    {
      "epoch": 267.05555555555554,
      "grad_norm": 8.791597792878747e-05,
      "learning_rate": 1.7211111326754505e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6729120,
      "step": 24035
    },
    {
      "epoch": 267.1111111111111,
      "grad_norm": 0.0002679396129678935,
      "learning_rate": 1.720178308656383e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6730560,
      "step": 24040
    },
    {
      "epoch": 267.1666666666667,
      "grad_norm": 0.00019049046386498958,
      "learning_rate": 1.719245604895621e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6731952,
      "step": 24045
    },
    {
      "epoch": 267.22222222222223,
      "grad_norm": 6.147799285827205e-05,
      "learning_rate": 1.7183130215369972e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6733344,
      "step": 24050
    },
    {
      "epoch": 267.27777777777777,
      "grad_norm": 7.747585914330557e-05,
      "learning_rate": 1.7173805587243292e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6734752,
      "step": 24055
    },
    {
      "epoch": 267.3333333333333,
      "grad_norm": 0.00028479649336077273,
      "learning_rate": 1.7164482166014147e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6736176,
      "step": 24060
    },
    {
      "epoch": 267.3888888888889,
      "grad_norm": 7.294727402040735e-05,
      "learning_rate": 1.7155159953120313e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6737552,
      "step": 24065
    },
    {
      "epoch": 267.44444444444446,
      "grad_norm": 0.00222992617636919,
      "learning_rate": 1.714583894999941e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6738944,
      "step": 24070
    },
    {
      "epoch": 267.5,
      "grad_norm": 0.0002239741152152419,
      "learning_rate": 1.7136519158088826e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6740352,
      "step": 24075
    },
    {
      "epoch": 267.55555555555554,
      "grad_norm": 0.0002900495601352304,
      "learning_rate": 1.712720057882581e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6741760,
      "step": 24080
    },
    {
      "epoch": 267.6111111111111,
      "grad_norm": 0.000234832419664599,
      "learning_rate": 1.7117883213647413e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6743152,
      "step": 24085
    },
    {
      "epoch": 267.6666666666667,
      "grad_norm": 0.00187577900942415,
      "learning_rate": 1.710856706399046e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6744608,
      "step": 24090
    },
    {
      "epoch": 267.72222222222223,
      "grad_norm": 0.0002694493450690061,
      "learning_rate": 1.7099252131291648e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6746016,
      "step": 24095
    },
    {
      "epoch": 267.77777777777777,
      "grad_norm": 9.425476309843361e-05,
      "learning_rate": 1.708993841698744e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6747344,
      "step": 24100
    },
    {
      "epoch": 267.8333333333333,
      "grad_norm": 0.00011310425179544836,
      "learning_rate": 1.7080625922514132e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6748768,
      "step": 24105
    },
    {
      "epoch": 267.8888888888889,
      "grad_norm": 0.00010691332863643765,
      "learning_rate": 1.7071314649307836e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6750176,
      "step": 24110
    },
    {
      "epoch": 267.94444444444446,
      "grad_norm": 0.0001322476309724152,
      "learning_rate": 1.7062004598804448e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6751600,
      "step": 24115
    },
    {
      "epoch": 268.0,
      "grad_norm": 0.00027471338398754597,
      "learning_rate": 1.7052695772439702e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6753008,
      "step": 24120
    },
    {
      "epoch": 268.05555555555554,
      "grad_norm": 0.0002748063416220248,
      "learning_rate": 1.7043388171649154e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6754416,
      "step": 24125
    },
    {
      "epoch": 268.1111111111111,
      "grad_norm": 0.0008554973173886538,
      "learning_rate": 1.7034081797868127e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6755792,
      "step": 24130
    },
    {
      "epoch": 268.1666666666667,
      "grad_norm": 0.0003512033144943416,
      "learning_rate": 1.70247766525318e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6757216,
      "step": 24135
    },
    {
      "epoch": 268.22222222222223,
      "grad_norm": 5.728884570999071e-05,
      "learning_rate": 1.701547273707514e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6758624,
      "step": 24140
    },
    {
      "epoch": 268.27777777777777,
      "grad_norm": 0.0038225504104048014,
      "learning_rate": 1.7006170052932916e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6760032,
      "step": 24145
    },
    {
      "epoch": 268.3333333333333,
      "grad_norm": 0.0011275000870227814,
      "learning_rate": 1.6996868601539735e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6761424,
      "step": 24150
    },
    {
      "epoch": 268.3888888888889,
      "grad_norm": 0.00013510511780623347,
      "learning_rate": 1.6987568384329977e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6762800,
      "step": 24155
    },
    {
      "epoch": 268.44444444444446,
      "grad_norm": 0.00010062935325549915,
      "learning_rate": 1.6978269402737866e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6764192,
      "step": 24160
    },
    {
      "epoch": 268.5,
      "grad_norm": 0.0009673179592937231,
      "learning_rate": 1.696897165819743e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6765616,
      "step": 24165
    },
    {
      "epoch": 268.55555555555554,
      "grad_norm": 0.00014845987607259303,
      "learning_rate": 1.6959675152142487e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6766992,
      "step": 24170
    },
    {
      "epoch": 268.6111111111111,
      "grad_norm": 0.00015240472566802055,
      "learning_rate": 1.6950379886006667e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6768352,
      "step": 24175
    },
    {
      "epoch": 268.6666666666667,
      "grad_norm": 7.352999091381207e-05,
      "learning_rate": 1.6941085861223438e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6769744,
      "step": 24180
    },
    {
      "epoch": 268.72222222222223,
      "grad_norm": 0.0005605408223345876,
      "learning_rate": 1.6931793079226034e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6771168,
      "step": 24185
    },
    {
      "epoch": 268.77777777777777,
      "grad_norm": 0.0004760559822898358,
      "learning_rate": 1.692250154144754e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6772512,
      "step": 24190
    },
    {
      "epoch": 268.8333333333333,
      "grad_norm": 0.0001438349427189678,
      "learning_rate": 1.6913211249320807e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6773920,
      "step": 24195
    },
    {
      "epoch": 268.8888888888889,
      "grad_norm": 0.000206856188015081,
      "learning_rate": 1.6903922204278522e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6775344,
      "step": 24200
    },
    {
      "epoch": 268.8888888888889,
      "eval_loss": 0.8437361717224121,
      "eval_runtime": 1.2408,
      "eval_samples_per_second": 32.238,
      "eval_steps_per_second": 16.119,
      "num_input_tokens_seen": 6775344,
      "step": 24200
    },
    {
      "epoch": 268.94444444444446,
      "grad_norm": 0.0003932443796657026,
      "learning_rate": 1.6894634407753186e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6776736,
      "step": 24205
    },
    {
      "epoch": 269.0,
      "grad_norm": 0.00021411836496554315,
      "learning_rate": 1.6885347861177077e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6778144,
      "step": 24210
    },
    {
      "epoch": 269.05555555555554,
      "grad_norm": 0.0006199597846716642,
      "learning_rate": 1.6876062565982298e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6779552,
      "step": 24215
    },
    {
      "epoch": 269.1111111111111,
      "grad_norm": 0.0029545847792178392,
      "learning_rate": 1.6866778523600774e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6780960,
      "step": 24220
    },
    {
      "epoch": 269.1666666666667,
      "grad_norm": 7.801531319273636e-05,
      "learning_rate": 1.6857495735464195e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6782352,
      "step": 24225
    },
    {
      "epoch": 269.22222222222223,
      "grad_norm": 0.000226469652261585,
      "learning_rate": 1.6848214203004115e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6783696,
      "step": 24230
    },
    {
      "epoch": 269.27777777777777,
      "grad_norm": 0.0002969536290038377,
      "learning_rate": 1.6838933927651835e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6785104,
      "step": 24235
    },
    {
      "epoch": 269.3333333333333,
      "grad_norm": 5.205552588449791e-05,
      "learning_rate": 1.6829654910838506e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6786496,
      "step": 24240
    },
    {
      "epoch": 269.3888888888889,
      "grad_norm": 0.0019971139263361692,
      "learning_rate": 1.6820377153995065e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6787888,
      "step": 24245
    },
    {
      "epoch": 269.44444444444446,
      "grad_norm": 8.012314356165007e-05,
      "learning_rate": 1.681110065855226e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6789280,
      "step": 24250
    },
    {
      "epoch": 269.5,
      "grad_norm": 0.00015775881183799356,
      "learning_rate": 1.6801825425940642e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6790656,
      "step": 24255
    },
    {
      "epoch": 269.55555555555554,
      "grad_norm": 0.0003262018726672977,
      "learning_rate": 1.679255145759056e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6792064,
      "step": 24260
    },
    {
      "epoch": 269.6111111111111,
      "grad_norm": 7.317669223994017e-05,
      "learning_rate": 1.6783278754932187e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6793472,
      "step": 24265
    },
    {
      "epoch": 269.6666666666667,
      "grad_norm": 6.782935815863311e-05,
      "learning_rate": 1.6774007319395496e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6794880,
      "step": 24270
    },
    {
      "epoch": 269.72222222222223,
      "grad_norm": 8.040846296353266e-05,
      "learning_rate": 1.6764737152410243e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6796304,
      "step": 24275
    },
    {
      "epoch": 269.77777777777777,
      "grad_norm": 0.00016724324086681008,
      "learning_rate": 1.6755468255406016e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6797696,
      "step": 24280
    },
    {
      "epoch": 269.8333333333333,
      "grad_norm": 0.00046460743760690093,
      "learning_rate": 1.674620062981219e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6799088,
      "step": 24285
    },
    {
      "epoch": 269.8888888888889,
      "grad_norm": 6.588666292373091e-05,
      "learning_rate": 1.6736934277057947e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6800480,
      "step": 24290
    },
    {
      "epoch": 269.94444444444446,
      "grad_norm": 0.00028373513487167656,
      "learning_rate": 1.6727669198572286e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6801888,
      "step": 24295
    },
    {
      "epoch": 270.0,
      "grad_norm": 0.00017529810429550707,
      "learning_rate": 1.6718405395783984e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6803344,
      "step": 24300
    },
    {
      "epoch": 270.05555555555554,
      "grad_norm": 0.00024186975497286767,
      "learning_rate": 1.6709142870121643e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6804736,
      "step": 24305
    },
    {
      "epoch": 270.1111111111111,
      "grad_norm": 0.00014045344141777605,
      "learning_rate": 1.669988162301367e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6806112,
      "step": 24310
    },
    {
      "epoch": 270.1666666666667,
      "grad_norm": 0.001223648083396256,
      "learning_rate": 1.6690621655888243e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6807488,
      "step": 24315
    },
    {
      "epoch": 270.22222222222223,
      "grad_norm": 0.0030730084981769323,
      "learning_rate": 1.6681362970173386e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6808896,
      "step": 24320
    },
    {
      "epoch": 270.27777777777777,
      "grad_norm": 0.00015250359138008207,
      "learning_rate": 1.6672105567296904e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6810320,
      "step": 24325
    },
    {
      "epoch": 270.3333333333333,
      "grad_norm": 0.00042481510899960995,
      "learning_rate": 1.666284944868639e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6811712,
      "step": 24330
    },
    {
      "epoch": 270.3888888888889,
      "grad_norm": 0.000896067766007036,
      "learning_rate": 1.665359461576927e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6813120,
      "step": 24335
    },
    {
      "epoch": 270.44444444444446,
      "grad_norm": 7.268170884344727e-05,
      "learning_rate": 1.6644341069972736e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6814528,
      "step": 24340
    },
    {
      "epoch": 270.5,
      "grad_norm": 0.00047881650971248746,
      "learning_rate": 1.6635088812723813e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6815904,
      "step": 24345
    },
    {
      "epoch": 270.55555555555554,
      "grad_norm": 0.00024363945703953505,
      "learning_rate": 1.6625837845449328e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6817344,
      "step": 24350
    },
    {
      "epoch": 270.6111111111111,
      "grad_norm": 7.984997500898317e-05,
      "learning_rate": 1.6616588169575874e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6818736,
      "step": 24355
    },
    {
      "epoch": 270.6666666666667,
      "grad_norm": 0.002714406466111541,
      "learning_rate": 1.6607339786529878e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6820144,
      "step": 24360
    },
    {
      "epoch": 270.72222222222223,
      "grad_norm": 0.00043729765457101166,
      "learning_rate": 1.659809269773756e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6821504,
      "step": 24365
    },
    {
      "epoch": 270.77777777777777,
      "grad_norm": 0.0015869883354753256,
      "learning_rate": 1.658884690462493e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6822880,
      "step": 24370
    },
    {
      "epoch": 270.8333333333333,
      "grad_norm": 0.00016947442782111466,
      "learning_rate": 1.6579602408617813e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6824272,
      "step": 24375
    },
    {
      "epoch": 270.8888888888889,
      "grad_norm": 0.0005417908541858196,
      "learning_rate": 1.657035921114181e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6825680,
      "step": 24380
    },
    {
      "epoch": 270.94444444444446,
      "grad_norm": 9.654730092734098e-05,
      "learning_rate": 1.656111731362236e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6827088,
      "step": 24385
    },
    {
      "epoch": 271.0,
      "grad_norm": 0.0009246039553545415,
      "learning_rate": 1.6551876717484666e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6828512,
      "step": 24390
    },
    {
      "epoch": 271.05555555555554,
      "grad_norm": 0.0005186398630030453,
      "learning_rate": 1.6542637424153752e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6829904,
      "step": 24395
    },
    {
      "epoch": 271.1111111111111,
      "grad_norm": 0.0005026845610700548,
      "learning_rate": 1.6533399435054418e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6831344,
      "step": 24400
    },
    {
      "epoch": 271.1111111111111,
      "eval_loss": 0.8418861627578735,
      "eval_runtime": 1.2361,
      "eval_samples_per_second": 32.36,
      "eval_steps_per_second": 16.18,
      "num_input_tokens_seen": 6831344,
      "step": 24400
    },
    {
      "epoch": 271.1666666666667,
      "grad_norm": 0.0002532988437451422,
      "learning_rate": 1.6524162751611304e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6832704,
      "step": 24405
    },
    {
      "epoch": 271.22222222222223,
      "grad_norm": 7.592852489324287e-05,
      "learning_rate": 1.6514927375248796e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6834096,
      "step": 24410
    },
    {
      "epoch": 271.27777777777777,
      "grad_norm": 0.0007101345108821988,
      "learning_rate": 1.6505693307391127e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6835472,
      "step": 24415
    },
    {
      "epoch": 271.3333333333333,
      "grad_norm": 0.0036856040824204683,
      "learning_rate": 1.6496460549462288e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6836848,
      "step": 24420
    },
    {
      "epoch": 271.3888888888889,
      "grad_norm": 0.0015932482201606035,
      "learning_rate": 1.6487229102886097e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6838240,
      "step": 24425
    },
    {
      "epoch": 271.44444444444446,
      "grad_norm": 0.0004227947210893035,
      "learning_rate": 1.6477998969086155e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6839584,
      "step": 24430
    },
    {
      "epoch": 271.5,
      "grad_norm": 9.889754437608644e-05,
      "learning_rate": 1.646877014948587e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6841008,
      "step": 24435
    },
    {
      "epoch": 271.55555555555554,
      "grad_norm": 0.0019123787060379982,
      "learning_rate": 1.6459542645508433e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6842464,
      "step": 24440
    },
    {
      "epoch": 271.6111111111111,
      "grad_norm": 0.00046045560156926513,
      "learning_rate": 1.6450316458576852e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6843824,
      "step": 24445
    },
    {
      "epoch": 271.6666666666667,
      "grad_norm": 0.0031976138707250357,
      "learning_rate": 1.6441091590113912e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6845248,
      "step": 24450
    },
    {
      "epoch": 271.72222222222223,
      "grad_norm": 4.014712612843141e-05,
      "learning_rate": 1.6431868041542213e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6846656,
      "step": 24455
    },
    {
      "epoch": 271.77777777777777,
      "grad_norm": 0.00012664021051023155,
      "learning_rate": 1.6422645814284123e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6848048,
      "step": 24460
    },
    {
      "epoch": 271.8333333333333,
      "grad_norm": 0.00043339954572729766,
      "learning_rate": 1.6413424909761846e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6849456,
      "step": 24465
    },
    {
      "epoch": 271.8888888888889,
      "grad_norm": 5.4465683206217363e-05,
      "learning_rate": 1.640420532939736e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6850928,
      "step": 24470
    },
    {
      "epoch": 271.94444444444446,
      "grad_norm": 0.001840950921177864,
      "learning_rate": 1.639498707461242e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6852304,
      "step": 24475
    },
    {
      "epoch": 272.0,
      "grad_norm": 0.026836786419153214,
      "learning_rate": 1.6385770146828614e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6853712,
      "step": 24480
    },
    {
      "epoch": 272.05555555555554,
      "grad_norm": 0.001488148933276534,
      "learning_rate": 1.637655454746731e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6855120,
      "step": 24485
    },
    {
      "epoch": 272.1111111111111,
      "grad_norm": 0.0011487763840705156,
      "learning_rate": 1.6367340277949658e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6856496,
      "step": 24490
    },
    {
      "epoch": 272.1666666666667,
      "grad_norm": 0.006656864657998085,
      "learning_rate": 1.635812733969663e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6857872,
      "step": 24495
    },
    {
      "epoch": 272.22222222222223,
      "grad_norm": 0.0010776671115309,
      "learning_rate": 1.634891573412896e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6859296,
      "step": 24500
    },
    {
      "epoch": 272.27777777777777,
      "grad_norm": 5.6523738749092445e-05,
      "learning_rate": 1.6339705462667196e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6860688,
      "step": 24505
    },
    {
      "epoch": 272.3333333333333,
      "grad_norm": 0.00021274981554597616,
      "learning_rate": 1.633049652673169e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6862048,
      "step": 24510
    },
    {
      "epoch": 272.3888888888889,
      "grad_norm": 7.406563963741064e-05,
      "learning_rate": 1.632128892774256e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6863520,
      "step": 24515
    },
    {
      "epoch": 272.44444444444446,
      "grad_norm": 7.659081893507391e-05,
      "learning_rate": 1.6312082667119737e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6864928,
      "step": 24520
    },
    {
      "epoch": 272.5,
      "grad_norm": 0.000555959704797715,
      "learning_rate": 1.630287774628296e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6866352,
      "step": 24525
    },
    {
      "epoch": 272.55555555555554,
      "grad_norm": 0.00030934083042666316,
      "learning_rate": 1.6293674166651718e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6867744,
      "step": 24530
    },
    {
      "epoch": 272.6111111111111,
      "grad_norm": 0.00010354380356147885,
      "learning_rate": 1.6284471929645338e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6869152,
      "step": 24535
    },
    {
      "epoch": 272.6666666666667,
      "grad_norm": 7.932097651064396e-05,
      "learning_rate": 1.627527103668291e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6870560,
      "step": 24540
    },
    {
      "epoch": 272.72222222222223,
      "grad_norm": 0.00026676352717913687,
      "learning_rate": 1.6266071489183327e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6872000,
      "step": 24545
    },
    {
      "epoch": 272.77777777777777,
      "grad_norm": 0.000265771261183545,
      "learning_rate": 1.6256873288565283e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6873360,
      "step": 24550
    },
    {
      "epoch": 272.8333333333333,
      "grad_norm": 9.72200941760093e-05,
      "learning_rate": 1.6247676436247245e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6874736,
      "step": 24555
    },
    {
      "epoch": 272.8888888888889,
      "grad_norm": 6.112870323704556e-05,
      "learning_rate": 1.6238480933647486e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6876128,
      "step": 24560
    },
    {
      "epoch": 272.94444444444446,
      "grad_norm": 0.00044164867722429335,
      "learning_rate": 1.6229286782184083e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6877504,
      "step": 24565
    },
    {
      "epoch": 273.0,
      "grad_norm": 0.00019192670879419893,
      "learning_rate": 1.622009398327487e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6878928,
      "step": 24570
    },
    {
      "epoch": 273.05555555555554,
      "grad_norm": 0.0008168210624717176,
      "learning_rate": 1.6210902538337502e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6880336,
      "step": 24575
    },
    {
      "epoch": 273.1111111111111,
      "grad_norm": 0.00028422550531104207,
      "learning_rate": 1.6201712448789413e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6881760,
      "step": 24580
    },
    {
      "epoch": 273.1666666666667,
      "grad_norm": 6.685696280328557e-05,
      "learning_rate": 1.6192523716047827e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6883184,
      "step": 24585
    },
    {
      "epoch": 273.22222222222223,
      "grad_norm": 4.310777148930356e-05,
      "learning_rate": 1.6183336341529776e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6884560,
      "step": 24590
    },
    {
      "epoch": 273.27777777777777,
      "grad_norm": 0.0012170441914349794,
      "learning_rate": 1.6174150326652047e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6885904,
      "step": 24595
    },
    {
      "epoch": 273.3333333333333,
      "grad_norm": 0.00013410394603852183,
      "learning_rate": 1.6164965672831256e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6887344,
      "step": 24600
    },
    {
      "epoch": 273.3333333333333,
      "eval_loss": 0.8429500460624695,
      "eval_runtime": 1.2388,
      "eval_samples_per_second": 32.289,
      "eval_steps_per_second": 16.145,
      "num_input_tokens_seen": 6887344,
      "step": 24600
    },
    {
      "epoch": 273.3888888888889,
      "grad_norm": 0.0007012240239419043,
      "learning_rate": 1.6155782381483784e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6888752,
      "step": 24605
    },
    {
      "epoch": 273.44444444444446,
      "grad_norm": 8.518266986357048e-05,
      "learning_rate": 1.6146600454025813e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6890160,
      "step": 24610
    },
    {
      "epoch": 273.5,
      "grad_norm": 0.0001547517313156277,
      "learning_rate": 1.6137419891873317e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6891568,
      "step": 24615
    },
    {
      "epoch": 273.55555555555554,
      "grad_norm": 0.00012158121535321698,
      "learning_rate": 1.6128240696442038e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6893008,
      "step": 24620
    },
    {
      "epoch": 273.6111111111111,
      "grad_norm": 0.0009377019014209509,
      "learning_rate": 1.611906286914753e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6894384,
      "step": 24625
    },
    {
      "epoch": 273.6666666666667,
      "grad_norm": 0.001963432179763913,
      "learning_rate": 1.6109886411405144e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6895792,
      "step": 24630
    },
    {
      "epoch": 273.72222222222223,
      "grad_norm": 0.0005163002642802894,
      "learning_rate": 1.6100711324629985e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6897200,
      "step": 24635
    },
    {
      "epoch": 273.77777777777777,
      "grad_norm": 0.00041289758519269526,
      "learning_rate": 1.609153761023698e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6898624,
      "step": 24640
    },
    {
      "epoch": 273.8333333333333,
      "grad_norm": 0.00421943562105298,
      "learning_rate": 1.608236526964083e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6900032,
      "step": 24645
    },
    {
      "epoch": 273.8888888888889,
      "grad_norm": 7.583315164083615e-05,
      "learning_rate": 1.607319430425601e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6901440,
      "step": 24650
    },
    {
      "epoch": 273.94444444444446,
      "grad_norm": 5.110514030093327e-05,
      "learning_rate": 1.606402471549682e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6902848,
      "step": 24655
    },
    {
      "epoch": 274.0,
      "grad_norm": 0.00029362773057073355,
      "learning_rate": 1.6054856504777312e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6904192,
      "step": 24660
    },
    {
      "epoch": 274.05555555555554,
      "grad_norm": 6.426241452572867e-05,
      "learning_rate": 1.6045689673511334e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6905600,
      "step": 24665
    },
    {
      "epoch": 274.1111111111111,
      "grad_norm": 0.002485057804733515,
      "learning_rate": 1.6036524223112548e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6907024,
      "step": 24670
    },
    {
      "epoch": 274.1666666666667,
      "grad_norm": 0.007052526343613863,
      "learning_rate": 1.602736015499436e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6908448,
      "step": 24675
    },
    {
      "epoch": 274.22222222222223,
      "grad_norm": 8.519378752680495e-05,
      "learning_rate": 1.601819747057e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6909856,
      "step": 24680
    },
    {
      "epoch": 274.27777777777777,
      "grad_norm": 9.462550951866433e-05,
      "learning_rate": 1.6009036171252465e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6911248,
      "step": 24685
    },
    {
      "epoch": 274.3333333333333,
      "grad_norm": 0.0006193735753186047,
      "learning_rate": 1.599987625845453e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6912656,
      "step": 24690
    },
    {
      "epoch": 274.3888888888889,
      "grad_norm": 3.868449493893422e-05,
      "learning_rate": 1.599071773358879e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6914096,
      "step": 24695
    },
    {
      "epoch": 274.44444444444446,
      "grad_norm": 0.0001323220058111474,
      "learning_rate": 1.598156059806758e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6915456,
      "step": 24700
    },
    {
      "epoch": 274.5,
      "grad_norm": 0.0003500396851450205,
      "learning_rate": 1.5972404853303062e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6916848,
      "step": 24705
    },
    {
      "epoch": 274.55555555555554,
      "grad_norm": 0.0005245236097835004,
      "learning_rate": 1.5963250500707172e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6918272,
      "step": 24710
    },
    {
      "epoch": 274.6111111111111,
      "grad_norm": 6.311519973678514e-05,
      "learning_rate": 1.5954097541691612e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6919728,
      "step": 24715
    },
    {
      "epoch": 274.6666666666667,
      "grad_norm": 6.489538645837456e-05,
      "learning_rate": 1.5944945977667884e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6921120,
      "step": 24720
    },
    {
      "epoch": 274.72222222222223,
      "grad_norm": 0.0002571332734078169,
      "learning_rate": 1.593579581004729e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6922448,
      "step": 24725
    },
    {
      "epoch": 274.77777777777777,
      "grad_norm": 0.0003106450894847512,
      "learning_rate": 1.592664704024088e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6923840,
      "step": 24730
    },
    {
      "epoch": 274.8333333333333,
      "grad_norm": 0.0003974416758865118,
      "learning_rate": 1.591749966965953e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6925264,
      "step": 24735
    },
    {
      "epoch": 274.8888888888889,
      "grad_norm": 0.0003277530486229807,
      "learning_rate": 1.5908353699713856e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6926640,
      "step": 24740
    },
    {
      "epoch": 274.94444444444446,
      "grad_norm": 6.408792251022533e-05,
      "learning_rate": 1.5899209131814298e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6928032,
      "step": 24745
    },
    {
      "epoch": 275.0,
      "grad_norm": 4.951010851073079e-05,
      "learning_rate": 1.5890065967371067e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6929456,
      "step": 24750
    },
    {
      "epoch": 275.05555555555554,
      "grad_norm": 0.00018757049110718071,
      "learning_rate": 1.5880924207794144e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6930880,
      "step": 24755
    },
    {
      "epoch": 275.1111111111111,
      "grad_norm": 0.0008108359761536121,
      "learning_rate": 1.5871783854493298e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6932336,
      "step": 24760
    },
    {
      "epoch": 275.1666666666667,
      "grad_norm": 0.00027408264577388763,
      "learning_rate": 1.5862644908878106e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6933728,
      "step": 24765
    },
    {
      "epoch": 275.22222222222223,
      "grad_norm": 0.00031478155869990587,
      "learning_rate": 1.5853507372357885e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6935200,
      "step": 24770
    },
    {
      "epoch": 275.27777777777777,
      "grad_norm": 0.0008142181322909892,
      "learning_rate": 1.5844371246341776e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6936608,
      "step": 24775
    },
    {
      "epoch": 275.3333333333333,
      "grad_norm": 0.0001201740451506339,
      "learning_rate": 1.5835236532238674e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6938048,
      "step": 24780
    },
    {
      "epoch": 275.3888888888889,
      "grad_norm": 6.93701149430126e-05,
      "learning_rate": 1.582610323145727e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6939440,
      "step": 24785
    },
    {
      "epoch": 275.44444444444446,
      "grad_norm": 0.0004988076980225742,
      "learning_rate": 1.5816971345406035e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6940784,
      "step": 24790
    },
    {
      "epoch": 275.5,
      "grad_norm": 6.0752583522116765e-05,
      "learning_rate": 1.5807840875493225e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6942208,
      "step": 24795
    },
    {
      "epoch": 275.55555555555554,
      "grad_norm": 0.002120048040524125,
      "learning_rate": 1.5798711823126854e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6943632,
      "step": 24800
    },
    {
      "epoch": 275.55555555555554,
      "eval_loss": 0.8463692665100098,
      "eval_runtime": 1.2371,
      "eval_samples_per_second": 32.333,
      "eval_steps_per_second": 16.166,
      "num_input_tokens_seen": 6943632,
      "step": 24800
    },
    {
      "epoch": 275.6111111111111,
      "grad_norm": 0.0002684856590349227,
      "learning_rate": 1.578958418971477e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6945024,
      "step": 24805
    },
    {
      "epoch": 275.6666666666667,
      "grad_norm": 0.00016792012320365757,
      "learning_rate": 1.578045797666453e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6946400,
      "step": 24810
    },
    {
      "epoch": 275.72222222222223,
      "grad_norm": 0.002866263035684824,
      "learning_rate": 1.5771333185383548e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6947760,
      "step": 24815
    },
    {
      "epoch": 275.77777777777777,
      "grad_norm": 0.0015837053069844842,
      "learning_rate": 1.576220981727895e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6949152,
      "step": 24820
    },
    {
      "epoch": 275.8333333333333,
      "grad_norm": 0.0007131223101168871,
      "learning_rate": 1.575308787375769e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6950560,
      "step": 24825
    },
    {
      "epoch": 275.8888888888889,
      "grad_norm": 0.001420552609488368,
      "learning_rate": 1.5743967356226492e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 6951952,
      "step": 24830
    },
    {
      "epoch": 275.94444444444446,
      "grad_norm": 0.00011350333807058632,
      "learning_rate": 1.5734848266091835e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6953344,
      "step": 24835
    },
    {
      "epoch": 276.0,
      "grad_norm": 0.0006273019243963063,
      "learning_rate": 1.572573060476001e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6954752,
      "step": 24840
    },
    {
      "epoch": 276.05555555555554,
      "grad_norm": 0.005983490962535143,
      "learning_rate": 1.5716614373637085e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6956160,
      "step": 24845
    },
    {
      "epoch": 276.1111111111111,
      "grad_norm": 0.00029321631882339716,
      "learning_rate": 1.570749957412887e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6957536,
      "step": 24850
    },
    {
      "epoch": 276.1666666666667,
      "grad_norm": 9.639679046813399e-05,
      "learning_rate": 1.5698386207641013e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6958960,
      "step": 24855
    },
    {
      "epoch": 276.22222222222223,
      "grad_norm": 0.00013702042633667588,
      "learning_rate": 1.5689274275578884e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6960320,
      "step": 24860
    },
    {
      "epoch": 276.27777777777777,
      "grad_norm": 5.19555906066671e-05,
      "learning_rate": 1.5680163779347667e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 6961696,
      "step": 24865
    },
    {
      "epoch": 276.3333333333333,
      "grad_norm": 0.0006680803489871323,
      "learning_rate": 1.5671054720352327e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6963136,
      "step": 24870
    },
    {
      "epoch": 276.3888888888889,
      "grad_norm": 2.41420578956604,
      "learning_rate": 1.566194709999757e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 6964560,
      "step": 24875
    },
    {
      "epoch": 276.44444444444446,
      "grad_norm": 0.0006038753781467676,
      "learning_rate": 1.5652840919687933e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 6965952,
      "step": 24880
    },
    {
      "epoch": 276.5,
      "grad_norm": 0.0005312502617016435,
      "learning_rate": 1.5643736180827676e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6967344,
      "step": 24885
    },
    {
      "epoch": 276.55555555555554,
      "grad_norm": 0.0006216559559106827,
      "learning_rate": 1.5634632884820878e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 6968768,
      "step": 24890
    },
    {
      "epoch": 276.6111111111111,
      "grad_norm": 7.142610411392525e-05,
      "learning_rate": 1.5625531033071395e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6970128,
      "step": 24895
    },
    {
      "epoch": 276.6666666666667,
      "grad_norm": 2.6765930670080706e-05,
      "learning_rate": 1.5616430626982828e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6971568,
      "step": 24900
    },
    {
      "epoch": 276.72222222222223,
      "grad_norm": 0.00035078387008979917,
      "learning_rate": 1.5607331667958575e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6972992,
      "step": 24905
    },
    {
      "epoch": 276.77777777777777,
      "grad_norm": 0.00020892049360554665,
      "learning_rate": 1.5598234157401824e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6974368,
      "step": 24910
    },
    {
      "epoch": 276.8333333333333,
      "grad_norm": 0.0011623125756159425,
      "learning_rate": 1.5589138096715503e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6975792,
      "step": 24915
    },
    {
      "epoch": 276.8888888888889,
      "grad_norm": 0.002121103461831808,
      "learning_rate": 1.5580043487302365e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6977184,
      "step": 24920
    },
    {
      "epoch": 276.94444444444446,
      "grad_norm": 0.011019556783139706,
      "learning_rate": 1.5570950330564888e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6978592,
      "step": 24925
    },
    {
      "epoch": 277.0,
      "grad_norm": 0.0002348109701415524,
      "learning_rate": 1.5561858627905367e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6980000,
      "step": 24930
    },
    {
      "epoch": 277.05555555555554,
      "grad_norm": 0.0014233659021556377,
      "learning_rate": 1.5552768380725857e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6981392,
      "step": 24935
    },
    {
      "epoch": 277.1111111111111,
      "grad_norm": 0.00013322566519491374,
      "learning_rate": 1.5543679590428183e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6982768,
      "step": 24940
    },
    {
      "epoch": 277.1666666666667,
      "grad_norm": 0.0002912166528403759,
      "learning_rate": 1.5534592258413943e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6984160,
      "step": 24945
    },
    {
      "epoch": 277.22222222222223,
      "grad_norm": 0.0006310981698334217,
      "learning_rate": 1.5525506386084538e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6985568,
      "step": 24950
    },
    {
      "epoch": 277.27777777777777,
      "grad_norm": 0.00024148607917595655,
      "learning_rate": 1.55164219748411e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6986976,
      "step": 24955
    },
    {
      "epoch": 277.3333333333333,
      "grad_norm": 0.0003859718271996826,
      "learning_rate": 1.550733902608459e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6988400,
      "step": 24960
    },
    {
      "epoch": 277.3888888888889,
      "grad_norm": 0.00012032013910356909,
      "learning_rate": 1.549825754121568e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6989792,
      "step": 24965
    },
    {
      "epoch": 277.44444444444446,
      "grad_norm": 0.008198127150535583,
      "learning_rate": 1.5489177521634864e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6991200,
      "step": 24970
    },
    {
      "epoch": 277.5,
      "grad_norm": 0.002721476135775447,
      "learning_rate": 1.5480098968742402e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6992592,
      "step": 24975
    },
    {
      "epoch": 277.55555555555554,
      "grad_norm": 6.117167504271492e-05,
      "learning_rate": 1.5471021883938304e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6994016,
      "step": 24980
    },
    {
      "epoch": 277.6111111111111,
      "grad_norm": 0.00015958957374095917,
      "learning_rate": 1.546194626862238e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6995408,
      "step": 24985
    },
    {
      "epoch": 277.6666666666667,
      "grad_norm": 0.011015018448233604,
      "learning_rate": 1.5452872124194216e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6996864,
      "step": 24990
    },
    {
      "epoch": 277.72222222222223,
      "grad_norm": 0.0009400811395607889,
      "learning_rate": 1.5443799452053136e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6998256,
      "step": 24995
    },
    {
      "epoch": 277.77777777777777,
      "grad_norm": 0.06172480806708336,
      "learning_rate": 1.543472825359828e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 6999632,
      "step": 25000
    },
    {
      "epoch": 277.77777777777777,
      "eval_loss": 0.8016790151596069,
      "eval_runtime": 1.2382,
      "eval_samples_per_second": 32.306,
      "eval_steps_per_second": 16.153,
      "num_input_tokens_seen": 6999632,
      "step": 25000
    },
    {
      "epoch": 277.8333333333333,
      "grad_norm": 0.0001265479513676837,
      "learning_rate": 1.5425658530228522e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7001024,
      "step": 25005
    },
    {
      "epoch": 277.8888888888889,
      "grad_norm": 0.002094038762152195,
      "learning_rate": 1.5416590283342546e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7002400,
      "step": 25010
    },
    {
      "epoch": 277.94444444444446,
      "grad_norm": 0.0006401396822184324,
      "learning_rate": 1.5407523514338783e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7003824,
      "step": 25015
    },
    {
      "epoch": 278.0,
      "grad_norm": 0.0022947986144572496,
      "learning_rate": 1.539845822461543e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7005184,
      "step": 25020
    },
    {
      "epoch": 278.05555555555554,
      "grad_norm": 0.0005689089302904904,
      "learning_rate": 1.538939441557048e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7006608,
      "step": 25025
    },
    {
      "epoch": 278.1111111111111,
      "grad_norm": 8.177760901162401e-05,
      "learning_rate": 1.5380332088601696e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7008000,
      "step": 25030
    },
    {
      "epoch": 278.1666666666667,
      "grad_norm": 0.00014700807514600456,
      "learning_rate": 1.537127124510658e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7009424,
      "step": 25035
    },
    {
      "epoch": 278.22222222222223,
      "grad_norm": 0.00035544534330256283,
      "learning_rate": 1.5362211886482457e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7010816,
      "step": 25040
    },
    {
      "epoch": 278.27777777777777,
      "grad_norm": 0.0006522185867652297,
      "learning_rate": 1.5353154014126363e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7012192,
      "step": 25045
    },
    {
      "epoch": 278.3333333333333,
      "grad_norm": 0.0011163922026753426,
      "learning_rate": 1.534409762943515e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7013568,
      "step": 25050
    },
    {
      "epoch": 278.3888888888889,
      "grad_norm": 0.0063320850022137165,
      "learning_rate": 1.5335042733805438e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7014976,
      "step": 25055
    },
    {
      "epoch": 278.44444444444446,
      "grad_norm": 0.00014337374886963516,
      "learning_rate": 1.532598932863358e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7016384,
      "step": 25060
    },
    {
      "epoch": 278.5,
      "grad_norm": 9.523312473902479e-05,
      "learning_rate": 1.531693741531574e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7017824,
      "step": 25065
    },
    {
      "epoch": 278.55555555555554,
      "grad_norm": 0.0007058596238493919,
      "learning_rate": 1.5307886995247844e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7019184,
      "step": 25070
    },
    {
      "epoch": 278.6111111111111,
      "grad_norm": 0.00011556546814972535,
      "learning_rate": 1.529883806982557e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7020608,
      "step": 25075
    },
    {
      "epoch": 278.6666666666667,
      "grad_norm": 0.0002628339279908687,
      "learning_rate": 1.5289790640444376e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7022016,
      "step": 25080
    },
    {
      "epoch": 278.72222222222223,
      "grad_norm": 0.00022728020849172026,
      "learning_rate": 1.5280744708499494e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7023408,
      "step": 25085
    },
    {
      "epoch": 278.77777777777777,
      "grad_norm": 0.00425796490162611,
      "learning_rate": 1.527170027538591e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7024800,
      "step": 25090
    },
    {
      "epoch": 278.8333333333333,
      "grad_norm": 0.01816270500421524,
      "learning_rate": 1.5262657342498407e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7026208,
      "step": 25095
    },
    {
      "epoch": 278.8888888888889,
      "grad_norm": 0.00013517959450837225,
      "learning_rate": 1.52536159112315e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7027568,
      "step": 25100
    },
    {
      "epoch": 278.94444444444446,
      "grad_norm": 0.001697822823189199,
      "learning_rate": 1.5244575982979497e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7028992,
      "step": 25105
    },
    {
      "epoch": 279.0,
      "grad_norm": 0.0008284483337774873,
      "learning_rate": 1.5235537559136487e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7030400,
      "step": 25110
    },
    {
      "epoch": 279.05555555555554,
      "grad_norm": 0.00019029685063287616,
      "learning_rate": 1.5226500641096286e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7031792,
      "step": 25115
    },
    {
      "epoch": 279.1111111111111,
      "grad_norm": 0.0007507503032684326,
      "learning_rate": 1.5217465230252509e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7033136,
      "step": 25120
    },
    {
      "epoch": 279.1666666666667,
      "grad_norm": 0.0016239408869296312,
      "learning_rate": 1.5208431327998523e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7034560,
      "step": 25125
    },
    {
      "epoch": 279.22222222222223,
      "grad_norm": 0.00042624215711839497,
      "learning_rate": 1.5199398935727477e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7036016,
      "step": 25130
    },
    {
      "epoch": 279.27777777777777,
      "grad_norm": 0.0001561204408062622,
      "learning_rate": 1.5190368054832282e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7037392,
      "step": 25135
    },
    {
      "epoch": 279.3333333333333,
      "grad_norm": 0.0002637266297824681,
      "learning_rate": 1.5181338686705601e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7038816,
      "step": 25140
    },
    {
      "epoch": 279.3888888888889,
      "grad_norm": 0.0008836215129122138,
      "learning_rate": 1.5172310832739889e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7040192,
      "step": 25145
    },
    {
      "epoch": 279.44444444444446,
      "grad_norm": 0.00019047597015742213,
      "learning_rate": 1.5163284494327346e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7041552,
      "step": 25150
    },
    {
      "epoch": 279.5,
      "grad_norm": 0.0005375973996706307,
      "learning_rate": 1.5154259672859952e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7042960,
      "step": 25155
    },
    {
      "epoch": 279.55555555555554,
      "grad_norm": 0.00027925980975851417,
      "learning_rate": 1.5145236369729452e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7044400,
      "step": 25160
    },
    {
      "epoch": 279.6111111111111,
      "grad_norm": 0.0004839178582187742,
      "learning_rate": 1.5136214586327335e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7045824,
      "step": 25165
    },
    {
      "epoch": 279.6666666666667,
      "grad_norm": 0.006695502437651157,
      "learning_rate": 1.5127194324044885e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7047200,
      "step": 25170
    },
    {
      "epoch": 279.72222222222223,
      "grad_norm": 0.00018939882284030318,
      "learning_rate": 1.5118175584273148e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7048592,
      "step": 25175
    },
    {
      "epoch": 279.77777777777777,
      "grad_norm": 0.0006255248445086181,
      "learning_rate": 1.5109158368402909e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7050032,
      "step": 25180
    },
    {
      "epoch": 279.8333333333333,
      "grad_norm": 0.0009528881637379527,
      "learning_rate": 1.5100142677824753e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7051424,
      "step": 25185
    },
    {
      "epoch": 279.8888888888889,
      "grad_norm": 0.00010140361700905487,
      "learning_rate": 1.509112851392901e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7052864,
      "step": 25190
    },
    {
      "epoch": 279.94444444444446,
      "grad_norm": 0.000847971357870847,
      "learning_rate": 1.5082115878105763e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7054272,
      "step": 25195
    },
    {
      "epoch": 280.0,
      "grad_norm": 0.0009921794990077615,
      "learning_rate": 1.5073104771744892e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7055664,
      "step": 25200
    },
    {
      "epoch": 280.0,
      "eval_loss": 0.8742315173149109,
      "eval_runtime": 1.2306,
      "eval_samples_per_second": 32.504,
      "eval_steps_per_second": 16.252,
      "num_input_tokens_seen": 7055664,
      "step": 25200
    },
    {
      "epoch": 280.05555555555554,
      "grad_norm": 0.0008826030534692109,
      "learning_rate": 1.5064095196236006e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7057072,
      "step": 25205
    },
    {
      "epoch": 280.1111111111111,
      "grad_norm": 0.0004739362630061805,
      "learning_rate": 1.50550871529685e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7058496,
      "step": 25210
    },
    {
      "epoch": 280.1666666666667,
      "grad_norm": 0.00011695719149429351,
      "learning_rate": 1.5046080643331546e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7059920,
      "step": 25215
    },
    {
      "epoch": 280.22222222222223,
      "grad_norm": 8.954438817454502e-05,
      "learning_rate": 1.5037075668714028e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7061360,
      "step": 25220
    },
    {
      "epoch": 280.27777777777777,
      "grad_norm": 0.00016587066056672484,
      "learning_rate": 1.5028072230504656e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7062720,
      "step": 25225
    },
    {
      "epoch": 280.3333333333333,
      "grad_norm": 0.0005773008451797068,
      "learning_rate": 1.5019070330091861e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7064160,
      "step": 25230
    },
    {
      "epoch": 280.3888888888889,
      "grad_norm": 0.00028884358471259475,
      "learning_rate": 1.5010069968863843e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7065536,
      "step": 25235
    },
    {
      "epoch": 280.44444444444446,
      "grad_norm": 0.002733692992478609,
      "learning_rate": 1.5001071148208584e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7066912,
      "step": 25240
    },
    {
      "epoch": 280.5,
      "grad_norm": 2.5724164515850134e-05,
      "learning_rate": 1.49920738695138e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7068304,
      "step": 25245
    },
    {
      "epoch": 280.55555555555554,
      "grad_norm": 0.0007047737599350512,
      "learning_rate": 1.4983078134166995e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7069680,
      "step": 25250
    },
    {
      "epoch": 280.6111111111111,
      "grad_norm": 9.560709440847859e-05,
      "learning_rate": 1.4974083943555428e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7071104,
      "step": 25255
    },
    {
      "epoch": 280.6666666666667,
      "grad_norm": 0.00022760583669878542,
      "learning_rate": 1.496509129906611e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7072544,
      "step": 25260
    },
    {
      "epoch": 280.72222222222223,
      "grad_norm": 0.002685222076252103,
      "learning_rate": 1.4956100202085809e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7073920,
      "step": 25265
    },
    {
      "epoch": 280.77777777777777,
      "grad_norm": 9.237686754204333e-05,
      "learning_rate": 1.4947110654001093e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7075296,
      "step": 25270
    },
    {
      "epoch": 280.8333333333333,
      "grad_norm": 7.322017336264253e-05,
      "learning_rate": 1.4938122656198234e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7076656,
      "step": 25275
    },
    {
      "epoch": 280.8888888888889,
      "grad_norm": 0.003075400134548545,
      "learning_rate": 1.4929136210063316e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7078032,
      "step": 25280
    },
    {
      "epoch": 280.94444444444446,
      "grad_norm": 5.452250115922652e-05,
      "learning_rate": 1.4920151316982146e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7079440,
      "step": 25285
    },
    {
      "epoch": 281.0,
      "grad_norm": 0.00015393385547213256,
      "learning_rate": 1.4911167978340312e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7080832,
      "step": 25290
    },
    {
      "epoch": 281.05555555555554,
      "grad_norm": 0.00021057692356407642,
      "learning_rate": 1.4902186195523166e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7082224,
      "step": 25295
    },
    {
      "epoch": 281.1111111111111,
      "grad_norm": 0.00022335881658364087,
      "learning_rate": 1.4893205969915805e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7083664,
      "step": 25300
    },
    {
      "epoch": 281.1666666666667,
      "grad_norm": 0.0003214880998712033,
      "learning_rate": 1.4884227302903086e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7085120,
      "step": 25305
    },
    {
      "epoch": 281.22222222222223,
      "grad_norm": 0.00038593471981585026,
      "learning_rate": 1.4875250195869653e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7086528,
      "step": 25310
    },
    {
      "epoch": 281.27777777777777,
      "grad_norm": 0.00024415136431343853,
      "learning_rate": 1.4866274650199862e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7087920,
      "step": 25315
    },
    {
      "epoch": 281.3333333333333,
      "grad_norm": 0.00025597005151212215,
      "learning_rate": 1.485730066727788e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7089312,
      "step": 25320
    },
    {
      "epoch": 281.3888888888889,
      "grad_norm": 5.766805770690553e-05,
      "learning_rate": 1.4848328248487586e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7090672,
      "step": 25325
    },
    {
      "epoch": 281.44444444444446,
      "grad_norm": 0.0001437136234017089,
      "learning_rate": 1.4839357395212656e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7092096,
      "step": 25330
    },
    {
      "epoch": 281.5,
      "grad_norm": 6.759310781490058e-05,
      "learning_rate": 1.4830388108836502e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7093472,
      "step": 25335
    },
    {
      "epoch": 281.55555555555554,
      "grad_norm": 0.010366247966885567,
      "learning_rate": 1.4821420390742299e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7094848,
      "step": 25340
    },
    {
      "epoch": 281.6111111111111,
      "grad_norm": 0.0008312824065797031,
      "learning_rate": 1.4812454242312979e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7096272,
      "step": 25345
    },
    {
      "epoch": 281.6666666666667,
      "grad_norm": 0.00015992640692275017,
      "learning_rate": 1.4803489664931253e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7097648,
      "step": 25350
    },
    {
      "epoch": 281.72222222222223,
      "grad_norm": 0.0007492418517358601,
      "learning_rate": 1.4794526659979544e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7099088,
      "step": 25355
    },
    {
      "epoch": 281.77777777777777,
      "grad_norm": 0.0004521100199781358,
      "learning_rate": 1.4785565228840086e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7100496,
      "step": 25360
    },
    {
      "epoch": 281.8333333333333,
      "grad_norm": 0.00021559967717621475,
      "learning_rate": 1.4776605372894819e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7101904,
      "step": 25365
    },
    {
      "epoch": 281.8888888888889,
      "grad_norm": 0.0002771465224213898,
      "learning_rate": 1.4767647093525488e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7103296,
      "step": 25370
    },
    {
      "epoch": 281.94444444444446,
      "grad_norm": 0.0006994395516812801,
      "learning_rate": 1.4758690392113566e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7104656,
      "step": 25375
    },
    {
      "epoch": 282.0,
      "grad_norm": 0.000749420200008899,
      "learning_rate": 1.4749735270040276e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7106048,
      "step": 25380
    },
    {
      "epoch": 282.05555555555554,
      "grad_norm": 7.277084660017863e-05,
      "learning_rate": 1.4740781728686623e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7107472,
      "step": 25385
    },
    {
      "epoch": 282.1111111111111,
      "grad_norm": 0.0034723218996077776,
      "learning_rate": 1.4731829769433358e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7108832,
      "step": 25390
    },
    {
      "epoch": 282.1666666666667,
      "grad_norm": 0.00020053752814419568,
      "learning_rate": 1.4722879393660976e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7110256,
      "step": 25395
    },
    {
      "epoch": 282.22222222222223,
      "grad_norm": 0.0004754025721922517,
      "learning_rate": 1.4713930602749748e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7111664,
      "step": 25400
    },
    {
      "epoch": 282.22222222222223,
      "eval_loss": 0.8130613565444946,
      "eval_runtime": 1.3253,
      "eval_samples_per_second": 30.183,
      "eval_steps_per_second": 15.091,
      "num_input_tokens_seen": 7111664,
      "step": 25400
    },
    {
      "epoch": 282.27777777777777,
      "grad_norm": 0.006063325330615044,
      "learning_rate": 1.470498339807968e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7113072,
      "step": 25405
    },
    {
      "epoch": 282.3333333333333,
      "grad_norm": 0.0008038334781304002,
      "learning_rate": 1.4696037781030542e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7114480,
      "step": 25410
    },
    {
      "epoch": 282.3888888888889,
      "grad_norm": 0.0008995175594463944,
      "learning_rate": 1.4687093752981876e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7115888,
      "step": 25415
    },
    {
      "epoch": 282.44444444444446,
      "grad_norm": 0.00018787223962135613,
      "learning_rate": 1.4678151315312943e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7117280,
      "step": 25420
    },
    {
      "epoch": 282.5,
      "grad_norm": 4.736927439807914e-05,
      "learning_rate": 1.4669210469402789e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7118656,
      "step": 25425
    },
    {
      "epoch": 282.55555555555554,
      "grad_norm": 0.0019289241172373295,
      "learning_rate": 1.4660271216630218e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7120048,
      "step": 25430
    },
    {
      "epoch": 282.6111111111111,
      "grad_norm": 0.00020261536701582372,
      "learning_rate": 1.4651333558373748e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7121440,
      "step": 25435
    },
    {
      "epoch": 282.6666666666667,
      "grad_norm": 8.040454849833623e-05,
      "learning_rate": 1.4642397496011707e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7122880,
      "step": 25440
    },
    {
      "epoch": 282.72222222222223,
      "grad_norm": 0.026656927540898323,
      "learning_rate": 1.4633463030922129e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7124304,
      "step": 25445
    },
    {
      "epoch": 282.77777777777777,
      "grad_norm": 0.0004584265116136521,
      "learning_rate": 1.462453016448282e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7125696,
      "step": 25450
    },
    {
      "epoch": 282.8333333333333,
      "grad_norm": 0.000469999824417755,
      "learning_rate": 1.4615598898071354e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7127120,
      "step": 25455
    },
    {
      "epoch": 282.8888888888889,
      "grad_norm": 0.00010169154847972095,
      "learning_rate": 1.4606669233065026e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7128496,
      "step": 25460
    },
    {
      "epoch": 282.94444444444446,
      "grad_norm": 2.4970560843939893e-05,
      "learning_rate": 1.4597741170840914e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7129936,
      "step": 25465
    },
    {
      "epoch": 283.0,
      "grad_norm": 0.0002079926198348403,
      "learning_rate": 1.4588814712775853e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7131328,
      "step": 25470
    },
    {
      "epoch": 283.05555555555554,
      "grad_norm": 8.049424650380388e-05,
      "learning_rate": 1.4579889860246382e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7132720,
      "step": 25475
    },
    {
      "epoch": 283.1111111111111,
      "grad_norm": 0.00035982366534881294,
      "learning_rate": 1.457096661462885e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7134096,
      "step": 25480
    },
    {
      "epoch": 283.1666666666667,
      "grad_norm": 0.0005054686916992068,
      "learning_rate": 1.4562044977299322e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7135520,
      "step": 25485
    },
    {
      "epoch": 283.22222222222223,
      "grad_norm": 0.0001741862070048228,
      "learning_rate": 1.4553124949633623e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7136944,
      "step": 25490
    },
    {
      "epoch": 283.27777777777777,
      "grad_norm": 0.00013860533363185823,
      "learning_rate": 1.4544206533007354e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7138352,
      "step": 25495
    },
    {
      "epoch": 283.3333333333333,
      "grad_norm": 0.0003049607330467552,
      "learning_rate": 1.4535289728795821e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7139744,
      "step": 25500
    },
    {
      "epoch": 283.3888888888889,
      "grad_norm": 0.0005112240905873477,
      "learning_rate": 1.4526374538374132e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7141136,
      "step": 25505
    },
    {
      "epoch": 283.44444444444446,
      "grad_norm": 0.0003843115409836173,
      "learning_rate": 1.4517460963117097e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7142544,
      "step": 25510
    },
    {
      "epoch": 283.5,
      "grad_norm": 0.00036345722037367523,
      "learning_rate": 1.4508549004399314e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7143968,
      "step": 25515
    },
    {
      "epoch": 283.55555555555554,
      "grad_norm": 0.00011194149556104094,
      "learning_rate": 1.449963866359513e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7145328,
      "step": 25520
    },
    {
      "epoch": 283.6111111111111,
      "grad_norm": 0.0003745217400137335,
      "learning_rate": 1.4490729942078607e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7146704,
      "step": 25525
    },
    {
      "epoch": 283.6666666666667,
      "grad_norm": 0.0001772013056324795,
      "learning_rate": 1.4481822841223608e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7148112,
      "step": 25530
    },
    {
      "epoch": 283.72222222222223,
      "grad_norm": 6.233544263523072e-05,
      "learning_rate": 1.4472917362403704e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7149520,
      "step": 25535
    },
    {
      "epoch": 283.77777777777777,
      "grad_norm": 0.0007866815431043506,
      "learning_rate": 1.4464013506992224e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7150896,
      "step": 25540
    },
    {
      "epoch": 283.8333333333333,
      "grad_norm": 4.327472561271861e-05,
      "learning_rate": 1.4455111276362277e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7152336,
      "step": 25545
    },
    {
      "epoch": 283.8888888888889,
      "grad_norm": 0.001094874576665461,
      "learning_rate": 1.4446210671886676e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7153728,
      "step": 25550
    },
    {
      "epoch": 283.94444444444446,
      "grad_norm": 3.13758609991055e-05,
      "learning_rate": 1.4437311694938015e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7155136,
      "step": 25555
    },
    {
      "epoch": 284.0,
      "grad_norm": 0.0006801199051551521,
      "learning_rate": 1.442841434688864e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7156512,
      "step": 25560
    },
    {
      "epoch": 284.05555555555554,
      "grad_norm": 0.0007317289710044861,
      "learning_rate": 1.4419518629110615e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7157936,
      "step": 25565
    },
    {
      "epoch": 284.1111111111111,
      "grad_norm": 0.0002459230017848313,
      "learning_rate": 1.4410624542975778e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7159360,
      "step": 25570
    },
    {
      "epoch": 284.1666666666667,
      "grad_norm": 7.895327871665359e-05,
      "learning_rate": 1.4401732089855724e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7160704,
      "step": 25575
    },
    {
      "epoch": 284.22222222222223,
      "grad_norm": 7.952541636768728e-05,
      "learning_rate": 1.4392841271121754e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7162144,
      "step": 25580
    },
    {
      "epoch": 284.27777777777777,
      "grad_norm": 0.0001504182000644505,
      "learning_rate": 1.438395208814497e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7163536,
      "step": 25585
    },
    {
      "epoch": 284.3333333333333,
      "grad_norm": 0.00010609529272187501,
      "learning_rate": 1.4375064542296174e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7164960,
      "step": 25590
    },
    {
      "epoch": 284.3888888888889,
      "grad_norm": 0.00029290365637280047,
      "learning_rate": 1.4366178634945946e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7166320,
      "step": 25595
    },
    {
      "epoch": 284.44444444444446,
      "grad_norm": 0.0003895979607477784,
      "learning_rate": 1.4357294367464616e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7167744,
      "step": 25600
    },
    {
      "epoch": 284.44444444444446,
      "eval_loss": 0.8320045471191406,
      "eval_runtime": 1.2459,
      "eval_samples_per_second": 32.106,
      "eval_steps_per_second": 16.053,
      "num_input_tokens_seen": 7167744,
      "step": 25600
    },
    {
      "epoch": 284.5,
      "grad_norm": 8.186914055841044e-05,
      "learning_rate": 1.434841174122224e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7169152,
      "step": 25605
    },
    {
      "epoch": 284.55555555555554,
      "grad_norm": 0.0002018507948378101,
      "learning_rate": 1.4339530757588615e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7170560,
      "step": 25610
    },
    {
      "epoch": 284.6111111111111,
      "grad_norm": 0.00022077331959735602,
      "learning_rate": 1.433065141793333e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7171984,
      "step": 25615
    },
    {
      "epoch": 284.6666666666667,
      "grad_norm": 0.0005140257999300957,
      "learning_rate": 1.4321773723625665e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7173376,
      "step": 25620
    },
    {
      "epoch": 284.72222222222223,
      "grad_norm": 3.1790725188329816e-05,
      "learning_rate": 1.4312897676034693e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7174784,
      "step": 25625
    },
    {
      "epoch": 284.77777777777777,
      "grad_norm": 0.0003575462906155735,
      "learning_rate": 1.4304023276529188e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7176176,
      "step": 25630
    },
    {
      "epoch": 284.8333333333333,
      "grad_norm": 0.0051523419097065926,
      "learning_rate": 1.4295150526477712e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7177584,
      "step": 25635
    },
    {
      "epoch": 284.8888888888889,
      "grad_norm": 0.0012712174793705344,
      "learning_rate": 1.4286279427248562e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7178944,
      "step": 25640
    },
    {
      "epoch": 284.94444444444446,
      "grad_norm": 5.377150955609977e-05,
      "learning_rate": 1.4277409980209747e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7180368,
      "step": 25645
    },
    {
      "epoch": 285.0,
      "grad_norm": 0.0002822915557771921,
      "learning_rate": 1.4268542186729061e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7181728,
      "step": 25650
    },
    {
      "epoch": 285.05555555555554,
      "grad_norm": 0.0014115856029093266,
      "learning_rate": 1.4259676048174043e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7183120,
      "step": 25655
    },
    {
      "epoch": 285.1111111111111,
      "grad_norm": 0.0006623719818890095,
      "learning_rate": 1.4250811565911937e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7184528,
      "step": 25660
    },
    {
      "epoch": 285.1666666666667,
      "grad_norm": 0.0001178332677227445,
      "learning_rate": 1.4241948741309782e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7185904,
      "step": 25665
    },
    {
      "epoch": 285.22222222222223,
      "grad_norm": 3.4918051824206486e-05,
      "learning_rate": 1.4233087575734317e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7187344,
      "step": 25670
    },
    {
      "epoch": 285.27777777777777,
      "grad_norm": 0.0007178197847679257,
      "learning_rate": 1.422422807055206e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7188720,
      "step": 25675
    },
    {
      "epoch": 285.3333333333333,
      "grad_norm": 0.0005124403978697956,
      "learning_rate": 1.4215370227129243e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7190144,
      "step": 25680
    },
    {
      "epoch": 285.3888888888889,
      "grad_norm": 0.00019002362387254834,
      "learning_rate": 1.4206514046831876e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7191536,
      "step": 25685
    },
    {
      "epoch": 285.44444444444446,
      "grad_norm": 0.00016515990137122571,
      "learning_rate": 1.419765953102567e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7192912,
      "step": 25690
    },
    {
      "epoch": 285.5,
      "grad_norm": 0.0002472124178893864,
      "learning_rate": 1.4188806681076125e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7194304,
      "step": 25695
    },
    {
      "epoch": 285.55555555555554,
      "grad_norm": 6.141992344055325e-05,
      "learning_rate": 1.4179955498348443e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7195744,
      "step": 25700
    },
    {
      "epoch": 285.6111111111111,
      "grad_norm": 0.001474739401601255,
      "learning_rate": 1.4171105984207605e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7197120,
      "step": 25705
    },
    {
      "epoch": 285.6666666666667,
      "grad_norm": 0.001534299342893064,
      "learning_rate": 1.4162258140018304e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7198560,
      "step": 25710
    },
    {
      "epoch": 285.72222222222223,
      "grad_norm": 0.020610760897397995,
      "learning_rate": 1.4153411967144986e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7199968,
      "step": 25715
    },
    {
      "epoch": 285.77777777777777,
      "grad_norm": 0.0010256451787427068,
      "learning_rate": 1.4144567466951864e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7201408,
      "step": 25720
    },
    {
      "epoch": 285.8333333333333,
      "grad_norm": 0.00011182211164850742,
      "learning_rate": 1.4135724640802844e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7202768,
      "step": 25725
    },
    {
      "epoch": 285.8888888888889,
      "grad_norm": 0.00031385826878249645,
      "learning_rate": 1.4126883490061615e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7204128,
      "step": 25730
    },
    {
      "epoch": 285.94444444444446,
      "grad_norm": 0.0012504961341619492,
      "learning_rate": 1.4118044016091603e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7205536,
      "step": 25735
    },
    {
      "epoch": 286.0,
      "grad_norm": 0.0007798405713401735,
      "learning_rate": 1.410920622025594e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7206960,
      "step": 25740
    },
    {
      "epoch": 286.05555555555554,
      "grad_norm": 0.0003081990871578455,
      "learning_rate": 1.4100370103917554e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7208352,
      "step": 25745
    },
    {
      "epoch": 286.1111111111111,
      "grad_norm": 0.00011426405399106443,
      "learning_rate": 1.409153566843907e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7209696,
      "step": 25750
    },
    {
      "epoch": 286.1666666666667,
      "grad_norm": 0.00012650150165427476,
      "learning_rate": 1.408270291518286e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7211120,
      "step": 25755
    },
    {
      "epoch": 286.22222222222223,
      "grad_norm": 0.0005682236514985561,
      "learning_rate": 1.407387184551107e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7212528,
      "step": 25760
    },
    {
      "epoch": 286.27777777777777,
      "grad_norm": 0.0003421927394811064,
      "learning_rate": 1.4065042460785532e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7213920,
      "step": 25765
    },
    {
      "epoch": 286.3333333333333,
      "grad_norm": 8.035104110604152e-05,
      "learning_rate": 1.405621476236787e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7215344,
      "step": 25770
    },
    {
      "epoch": 286.3888888888889,
      "grad_norm": 0.00018662743968889117,
      "learning_rate": 1.4047388751619423e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7216752,
      "step": 25775
    },
    {
      "epoch": 286.44444444444446,
      "grad_norm": 0.00048068902106024325,
      "learning_rate": 1.4038564429901264e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7218096,
      "step": 25780
    },
    {
      "epoch": 286.5,
      "grad_norm": 0.0001091586091206409,
      "learning_rate": 1.4029741798574227e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7219520,
      "step": 25785
    },
    {
      "epoch": 286.55555555555554,
      "grad_norm": 0.0005683627096004784,
      "learning_rate": 1.402092085899886e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7220928,
      "step": 25790
    },
    {
      "epoch": 286.6111111111111,
      "grad_norm": 0.001835145172663033,
      "learning_rate": 1.4012101612535464e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7222336,
      "step": 25795
    },
    {
      "epoch": 286.6666666666667,
      "grad_norm": 0.0003616136673372239,
      "learning_rate": 1.4003284060544092e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7223696,
      "step": 25800
    },
    {
      "epoch": 286.6666666666667,
      "eval_loss": 0.8105586767196655,
      "eval_runtime": 1.2442,
      "eval_samples_per_second": 32.15,
      "eval_steps_per_second": 16.075,
      "num_input_tokens_seen": 7223696,
      "step": 25800
    },
    {
      "epoch": 286.72222222222223,
      "grad_norm": 0.001843258156441152,
      "learning_rate": 1.3994468204384504e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7225088,
      "step": 25805
    },
    {
      "epoch": 286.77777777777777,
      "grad_norm": 0.00010641574044711888,
      "learning_rate": 1.398565404541622e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7226448,
      "step": 25810
    },
    {
      "epoch": 286.8333333333333,
      "grad_norm": 7.287586049642414e-05,
      "learning_rate": 1.3976841584998513e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7227872,
      "step": 25815
    },
    {
      "epoch": 286.8888888888889,
      "grad_norm": 0.0001512315939180553,
      "learning_rate": 1.3968030824490352e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7229296,
      "step": 25820
    },
    {
      "epoch": 286.94444444444446,
      "grad_norm": 0.0010240748524665833,
      "learning_rate": 1.3959221765250469e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7230704,
      "step": 25825
    },
    {
      "epoch": 287.0,
      "grad_norm": 0.00021942067542113364,
      "learning_rate": 1.3950414408637343e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 7232128,
      "step": 25830
    },
    {
      "epoch": 287.05555555555554,
      "grad_norm": 0.00014662333705928177,
      "learning_rate": 1.3941608756009166e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7233552,
      "step": 25835
    },
    {
      "epoch": 287.1111111111111,
      "grad_norm": 0.0004293948004487902,
      "learning_rate": 1.3932804808723898e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7234976,
      "step": 25840
    },
    {
      "epoch": 287.1666666666667,
      "grad_norm": 0.002400208031758666,
      "learning_rate": 1.3924002568139194e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7236352,
      "step": 25845
    },
    {
      "epoch": 287.22222222222223,
      "grad_norm": 0.00011347616964485496,
      "learning_rate": 1.3915202035612485e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7237744,
      "step": 25850
    },
    {
      "epoch": 287.27777777777777,
      "grad_norm": 5.101295027998276e-05,
      "learning_rate": 1.3906403212500935e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7239152,
      "step": 25855
    },
    {
      "epoch": 287.3333333333333,
      "grad_norm": 0.0004824016068596393,
      "learning_rate": 1.3897606100161409e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7240560,
      "step": 25860
    },
    {
      "epoch": 287.3888888888889,
      "grad_norm": 0.001104316907003522,
      "learning_rate": 1.388881069995055e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7241984,
      "step": 25865
    },
    {
      "epoch": 287.44444444444446,
      "grad_norm": 0.0008393674506805837,
      "learning_rate": 1.3880017013224708e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7243392,
      "step": 25870
    },
    {
      "epoch": 287.5,
      "grad_norm": 0.00021444815502036363,
      "learning_rate": 1.3871225041339984e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7244784,
      "step": 25875
    },
    {
      "epoch": 287.55555555555554,
      "grad_norm": 0.0017253791447728872,
      "learning_rate": 1.386243478565222e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7246176,
      "step": 25880
    },
    {
      "epoch": 287.6111111111111,
      "grad_norm": 0.0074039651080966,
      "learning_rate": 1.3853646247516966e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7247552,
      "step": 25885
    },
    {
      "epoch": 287.6666666666667,
      "grad_norm": 0.00044560665264725685,
      "learning_rate": 1.3844859428289545e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7248960,
      "step": 25890
    },
    {
      "epoch": 287.72222222222223,
      "grad_norm": 0.000321825296850875,
      "learning_rate": 1.3836074329324984e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7250368,
      "step": 25895
    },
    {
      "epoch": 287.77777777777777,
      "grad_norm": 0.0002751658612396568,
      "learning_rate": 1.3827290951978044e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7251760,
      "step": 25900
    },
    {
      "epoch": 287.8333333333333,
      "grad_norm": 0.00016773615789134055,
      "learning_rate": 1.381850929760326e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7253120,
      "step": 25905
    },
    {
      "epoch": 287.8888888888889,
      "grad_norm": 0.00019838423759210855,
      "learning_rate": 1.3809729367554842e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7254496,
      "step": 25910
    },
    {
      "epoch": 287.94444444444446,
      "grad_norm": 0.002676280913874507,
      "learning_rate": 1.3800951163186784e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7255904,
      "step": 25915
    },
    {
      "epoch": 288.0,
      "grad_norm": 7.631279731867835e-05,
      "learning_rate": 1.3792174685852801e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7257296,
      "step": 25920
    },
    {
      "epoch": 288.05555555555554,
      "grad_norm": 0.0019748429767787457,
      "learning_rate": 1.378339993690632e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7258688,
      "step": 25925
    },
    {
      "epoch": 288.1111111111111,
      "grad_norm": 0.00014257317525334656,
      "learning_rate": 1.3774626917700523e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7260064,
      "step": 25930
    },
    {
      "epoch": 288.1666666666667,
      "grad_norm": 0.0028045978397130966,
      "learning_rate": 1.3765855629588334e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7261472,
      "step": 25935
    },
    {
      "epoch": 288.22222222222223,
      "grad_norm": 3.8704030885128304e-05,
      "learning_rate": 1.3757086073922374e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7262896,
      "step": 25940
    },
    {
      "epoch": 288.27777777777777,
      "grad_norm": 0.00010717968689277768,
      "learning_rate": 1.3748318252055038e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7264288,
      "step": 25945
    },
    {
      "epoch": 288.3333333333333,
      "grad_norm": 3.702402318594977e-05,
      "learning_rate": 1.3739552165338416e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7265696,
      "step": 25950
    },
    {
      "epoch": 288.3888888888889,
      "grad_norm": 0.00044134550262242556,
      "learning_rate": 1.3730787815124354e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7267088,
      "step": 25955
    },
    {
      "epoch": 288.44444444444446,
      "grad_norm": 0.00015895356773398817,
      "learning_rate": 1.3722025202764443e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7268496,
      "step": 25960
    },
    {
      "epoch": 288.5,
      "grad_norm": 4.443739089765586e-05,
      "learning_rate": 1.371326432960997e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7269888,
      "step": 25965
    },
    {
      "epoch": 288.55555555555554,
      "grad_norm": 0.0006817211979068816,
      "learning_rate": 1.3704505197011969e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7271296,
      "step": 25970
    },
    {
      "epoch": 288.6111111111111,
      "grad_norm": 0.0017305028159171343,
      "learning_rate": 1.3695747806321224e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7272720,
      "step": 25975
    },
    {
      "epoch": 288.6666666666667,
      "grad_norm": 0.000271631870418787,
      "learning_rate": 1.3686992158888212e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7274160,
      "step": 25980
    },
    {
      "epoch": 288.72222222222223,
      "grad_norm": 0.0010771527886390686,
      "learning_rate": 1.367823825606319e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7275552,
      "step": 25985
    },
    {
      "epoch": 288.77777777777777,
      "grad_norm": 0.0005111330538056791,
      "learning_rate": 1.36694860991961e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7276960,
      "step": 25990
    },
    {
      "epoch": 288.8333333333333,
      "grad_norm": 0.009939797222614288,
      "learning_rate": 1.3660735689636636e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7278352,
      "step": 25995
    },
    {
      "epoch": 288.8888888888889,
      "grad_norm": 6.380332342814654e-05,
      "learning_rate": 1.365198702873424e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7279760,
      "step": 26000
    },
    {
      "epoch": 288.8888888888889,
      "eval_loss": 0.8208203315734863,
      "eval_runtime": 1.2385,
      "eval_samples_per_second": 32.298,
      "eval_steps_per_second": 16.149,
      "num_input_tokens_seen": 7279760,
      "step": 26000
    },
    {
      "epoch": 288.94444444444446,
      "grad_norm": 0.00048055502702482045,
      "learning_rate": 1.364324011783804e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7281200,
      "step": 26005
    },
    {
      "epoch": 289.0,
      "grad_norm": 0.0005804190295748413,
      "learning_rate": 1.3634494958296934e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7282608,
      "step": 26010
    },
    {
      "epoch": 289.05555555555554,
      "grad_norm": 0.00027756165945902467,
      "learning_rate": 1.3625751551459542e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7283984,
      "step": 26015
    },
    {
      "epoch": 289.1111111111111,
      "grad_norm": 0.0005261824699118733,
      "learning_rate": 1.3617009898674188e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7285408,
      "step": 26020
    },
    {
      "epoch": 289.1666666666667,
      "grad_norm": 0.0001952702587004751,
      "learning_rate": 1.3608270001288967e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7286816,
      "step": 26025
    },
    {
      "epoch": 289.22222222222223,
      "grad_norm": 0.001326034776866436,
      "learning_rate": 1.359953186065166e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7288224,
      "step": 26030
    },
    {
      "epoch": 289.27777777777777,
      "grad_norm": 0.000496271182782948,
      "learning_rate": 1.3590795478109814e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7289600,
      "step": 26035
    },
    {
      "epoch": 289.3333333333333,
      "grad_norm": 0.0005158482235856354,
      "learning_rate": 1.3582060855010675e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7290976,
      "step": 26040
    },
    {
      "epoch": 289.3888888888889,
      "grad_norm": 0.00011649323278106749,
      "learning_rate": 1.3573327992701245e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7292368,
      "step": 26045
    },
    {
      "epoch": 289.44444444444446,
      "grad_norm": 4.588222145684995e-05,
      "learning_rate": 1.356459689252823e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7293808,
      "step": 26050
    },
    {
      "epoch": 289.5,
      "grad_norm": 4.490537685342133e-05,
      "learning_rate": 1.3555867555838087e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7295184,
      "step": 26055
    },
    {
      "epoch": 289.55555555555554,
      "grad_norm": 0.0001239300618181005,
      "learning_rate": 1.3547139983976975e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7296576,
      "step": 26060
    },
    {
      "epoch": 289.6111111111111,
      "grad_norm": 0.0003084968193434179,
      "learning_rate": 1.3538414178290815e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7297968,
      "step": 26065
    },
    {
      "epoch": 289.6666666666667,
      "grad_norm": 8.963309664977714e-05,
      "learning_rate": 1.3529690140125209e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7299344,
      "step": 26070
    },
    {
      "epoch": 289.72222222222223,
      "grad_norm": 0.00021490197104867548,
      "learning_rate": 1.352096787082553e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7300736,
      "step": 26075
    },
    {
      "epoch": 289.77777777777777,
      "grad_norm": 0.0002107614855049178,
      "learning_rate": 1.3512247371736871e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 7302128,
      "step": 26080
    },
    {
      "epoch": 289.8333333333333,
      "grad_norm": 7.284287858055905e-05,
      "learning_rate": 1.3503528644204022e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7303536,
      "step": 26085
    },
    {
      "epoch": 289.8888888888889,
      "grad_norm": 0.0021696840412914753,
      "learning_rate": 1.349481168957153e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7304960,
      "step": 26090
    },
    {
      "epoch": 289.94444444444446,
      "grad_norm": 0.0016695569502189755,
      "learning_rate": 1.3486096509183665e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 7306368,
      "step": 26095
    },
    {
      "epoch": 290.0,
      "grad_norm": 4.081864833831787,
      "learning_rate": 1.3477383104384406e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 7307760,
      "step": 26100
    },
    {
      "epoch": 290.05555555555554,
      "grad_norm": 0.00013716850662603974,
      "learning_rate": 1.3468671476517481e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7309168,
      "step": 26105
    },
    {
      "epoch": 290.1111111111111,
      "grad_norm": 0.00018694759637583047,
      "learning_rate": 1.3459961626926326e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7310624,
      "step": 26110
    },
    {
      "epoch": 290.1666666666667,
      "grad_norm": 0.0004975213669240475,
      "learning_rate": 1.3451253556954101e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7312048,
      "step": 26115
    },
    {
      "epoch": 290.22222222222223,
      "grad_norm": 0.0002218622394138947,
      "learning_rate": 1.3442547267943717e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7313456,
      "step": 26120
    },
    {
      "epoch": 290.27777777777777,
      "grad_norm": 0.0007186591974459589,
      "learning_rate": 1.3433842761237774e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7314816,
      "step": 26125
    },
    {
      "epoch": 290.3333333333333,
      "grad_norm": 0.0001856472808867693,
      "learning_rate": 1.3425140038178639e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7316208,
      "step": 26130
    },
    {
      "epoch": 290.3888888888889,
      "grad_norm": 0.0001574783236719668,
      "learning_rate": 1.3416439100108358e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7317600,
      "step": 26135
    },
    {
      "epoch": 290.44444444444446,
      "grad_norm": 0.0004238700494170189,
      "learning_rate": 1.3407739948368734e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7318976,
      "step": 26140
    },
    {
      "epoch": 290.5,
      "grad_norm": 0.0001970376179087907,
      "learning_rate": 1.3399042584301298e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7320416,
      "step": 26145
    },
    {
      "epoch": 290.55555555555554,
      "grad_norm": 0.00046382893924601376,
      "learning_rate": 1.3390347009247272e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7321824,
      "step": 26150
    },
    {
      "epoch": 290.6111111111111,
      "grad_norm": 0.0001357643777737394,
      "learning_rate": 1.3381653224547635e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7323200,
      "step": 26155
    },
    {
      "epoch": 290.6666666666667,
      "grad_norm": 0.00019356470147613436,
      "learning_rate": 1.3372961231543086e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7324576,
      "step": 26160
    },
    {
      "epoch": 290.72222222222223,
      "grad_norm": 0.00011981304851360619,
      "learning_rate": 1.3364271031574016e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7326048,
      "step": 26165
    },
    {
      "epoch": 290.77777777777777,
      "grad_norm": 0.0002564874303061515,
      "learning_rate": 1.335558262598059e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7327392,
      "step": 26170
    },
    {
      "epoch": 290.8333333333333,
      "grad_norm": 0.006567174568772316,
      "learning_rate": 1.3346896016102645e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 7328800,
      "step": 26175
    },
    {
      "epoch": 290.8888888888889,
      "grad_norm": 0.00010177500371355563,
      "learning_rate": 1.3338211203279788e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7330240,
      "step": 26180
    },
    {
      "epoch": 290.94444444444446,
      "grad_norm": 0.0013370797969400883,
      "learning_rate": 1.3329528188851303e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7331632,
      "step": 26185
    },
    {
      "epoch": 291.0,
      "grad_norm": 5.8310211898060516e-05,
      "learning_rate": 1.3320846974156242e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 7333008,
      "step": 26190
    },
    {
      "epoch": 291.05555555555554,
      "grad_norm": 0.0001833763235481456,
      "learning_rate": 1.3312167560533337e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7334384,
      "step": 26195
    },
    {
      "epoch": 291.1111111111111,
      "grad_norm": 0.00011231168900849298,
      "learning_rate": 1.3303489949321082e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7335792,
      "step": 26200
    },
    {
      "epoch": 291.1111111111111,
      "eval_loss": 0.8450155258178711,
      "eval_runtime": 1.2495,
      "eval_samples_per_second": 32.013,
      "eval_steps_per_second": 16.007,
      "num_input_tokens_seen": 7335792,
      "step": 26200
    },
    {
      "epoch": 291.1666666666667,
      "grad_norm": 0.0001166355432360433,
      "learning_rate": 1.3294814141857653e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7337120,
      "step": 26205
    },
    {
      "epoch": 291.22222222222223,
      "grad_norm": 0.00018555363931227475,
      "learning_rate": 1.3286140139480992e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7338496,
      "step": 26210
    },
    {
      "epoch": 291.27777777777777,
      "grad_norm": 0.0002220892783952877,
      "learning_rate": 1.3277467943528719e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7339888,
      "step": 26215
    },
    {
      "epoch": 291.3333333333333,
      "grad_norm": 0.00038329200469888747,
      "learning_rate": 1.3268797555338203e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7341328,
      "step": 26220
    },
    {
      "epoch": 291.3888888888889,
      "grad_norm": 3.1846891943132505e-05,
      "learning_rate": 1.3260128976246533e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7342768,
      "step": 26225
    },
    {
      "epoch": 291.44444444444446,
      "grad_norm": 4.4399843318387866e-05,
      "learning_rate": 1.32514622075905e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7344160,
      "step": 26230
    },
    {
      "epoch": 291.5,
      "grad_norm": 0.0001634586078580469,
      "learning_rate": 1.3242797250706638e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7345536,
      "step": 26235
    },
    {
      "epoch": 291.55555555555554,
      "grad_norm": 0.0001704490277916193,
      "learning_rate": 1.3234134106931195e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7346960,
      "step": 26240
    },
    {
      "epoch": 291.6111111111111,
      "grad_norm": 0.0004644959117285907,
      "learning_rate": 1.322547277760013e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7348320,
      "step": 26245
    },
    {
      "epoch": 291.6666666666667,
      "grad_norm": 0.0018567637307569385,
      "learning_rate": 1.3216813264049132e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7349696,
      "step": 26250
    },
    {
      "epoch": 291.72222222222223,
      "grad_norm": 4.918082049698569e-05,
      "learning_rate": 1.32081555676136e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7351120,
      "step": 26255
    },
    {
      "epoch": 291.77777777777777,
      "grad_norm": 0.0013975809561088681,
      "learning_rate": 1.3199499689628674e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7352560,
      "step": 26260
    },
    {
      "epoch": 291.8333333333333,
      "grad_norm": 5.923798744333908e-05,
      "learning_rate": 1.3190845631429192e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7354000,
      "step": 26265
    },
    {
      "epoch": 291.8888888888889,
      "grad_norm": 0.007023580837994814,
      "learning_rate": 1.3182193394349704e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7355440,
      "step": 26270
    },
    {
      "epoch": 291.94444444444446,
      "grad_norm": 9.014931129058823e-05,
      "learning_rate": 1.3173542979724507e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7356864,
      "step": 26275
    },
    {
      "epoch": 292.0,
      "grad_norm": 0.00018327565339859575,
      "learning_rate": 1.3164894388887617e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7358272,
      "step": 26280
    },
    {
      "epoch": 292.05555555555554,
      "grad_norm": 0.00026003748644143343,
      "learning_rate": 1.3156247623172727e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7359632,
      "step": 26285
    },
    {
      "epoch": 292.1111111111111,
      "grad_norm": 0.0006843256414867938,
      "learning_rate": 1.3147602683913302e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7361008,
      "step": 26290
    },
    {
      "epoch": 292.1666666666667,
      "grad_norm": 0.0005133668310008943,
      "learning_rate": 1.3138959572442481e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7362400,
      "step": 26295
    },
    {
      "epoch": 292.22222222222223,
      "grad_norm": 7.395504508167505e-05,
      "learning_rate": 1.3130318290093146e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7363808,
      "step": 26300
    },
    {
      "epoch": 292.27777777777777,
      "grad_norm": 0.0007301965379156172,
      "learning_rate": 1.3121678838197909e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7365232,
      "step": 26305
    },
    {
      "epoch": 292.3333333333333,
      "grad_norm": 0.0001572415785631165,
      "learning_rate": 1.3113041218089056e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7366672,
      "step": 26310
    },
    {
      "epoch": 292.3888888888889,
      "grad_norm": 0.00014043354894965887,
      "learning_rate": 1.3104405431098626e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7368096,
      "step": 26315
    },
    {
      "epoch": 292.44444444444446,
      "grad_norm": 0.000661539554130286,
      "learning_rate": 1.3095771478558377e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7369472,
      "step": 26320
    },
    {
      "epoch": 292.5,
      "grad_norm": 8.003078255569562e-05,
      "learning_rate": 1.3087139361799766e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7370896,
      "step": 26325
    },
    {
      "epoch": 292.55555555555554,
      "grad_norm": 3.4173743188148364e-05,
      "learning_rate": 1.3078509082153964e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7372256,
      "step": 26330
    },
    {
      "epoch": 292.6111111111111,
      "grad_norm": 0.00016592402243986726,
      "learning_rate": 1.3069880640951885e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7373680,
      "step": 26335
    },
    {
      "epoch": 292.6666666666667,
      "grad_norm": 0.0003915464913006872,
      "learning_rate": 1.3061254039524123e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7375120,
      "step": 26340
    },
    {
      "epoch": 292.72222222222223,
      "grad_norm": 0.00042814287007786334,
      "learning_rate": 1.3052629279201028e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7376560,
      "step": 26345
    },
    {
      "epoch": 292.77777777777777,
      "grad_norm": 3.8121223042253405e-05,
      "learning_rate": 1.3044006361312633e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7377936,
      "step": 26350
    },
    {
      "epoch": 292.8333333333333,
      "grad_norm": 0.00036821424146182835,
      "learning_rate": 1.30353852871887e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7379312,
      "step": 26355
    },
    {
      "epoch": 292.8888888888889,
      "grad_norm": 3.3145064662676305e-05,
      "learning_rate": 1.302676605815873e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7380688,
      "step": 26360
    },
    {
      "epoch": 292.94444444444446,
      "grad_norm": 5.845587293151766e-05,
      "learning_rate": 1.3018148675551884e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7382096,
      "step": 26365
    },
    {
      "epoch": 293.0,
      "grad_norm": 0.00016113232413772494,
      "learning_rate": 1.3009533140697094e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7383440,
      "step": 26370
    },
    {
      "epoch": 293.05555555555554,
      "grad_norm": 0.0008397736819460988,
      "learning_rate": 1.3000919454922966e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7384784,
      "step": 26375
    },
    {
      "epoch": 293.1111111111111,
      "grad_norm": 9.959132876247168e-05,
      "learning_rate": 1.299230761955785e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7386208,
      "step": 26380
    },
    {
      "epoch": 293.1666666666667,
      "grad_norm": 4.828892269870266e-05,
      "learning_rate": 1.2983697635929807e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7387600,
      "step": 26385
    },
    {
      "epoch": 293.22222222222223,
      "grad_norm": 0.00011963184078922495,
      "learning_rate": 1.2975089505366584e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7389040,
      "step": 26390
    },
    {
      "epoch": 293.27777777777777,
      "grad_norm": 0.0004844827635679394,
      "learning_rate": 1.2966483229195683e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7390432,
      "step": 26395
    },
    {
      "epoch": 293.3333333333333,
      "grad_norm": 8.373806485906243e-05,
      "learning_rate": 1.2957878808744283e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7391808,
      "step": 26400
    },
    {
      "epoch": 293.3333333333333,
      "eval_loss": 0.8503596186637878,
      "eval_runtime": 1.2358,
      "eval_samples_per_second": 32.368,
      "eval_steps_per_second": 16.184,
      "num_input_tokens_seen": 7391808,
      "step": 26400
    },
    {
      "epoch": 293.3888888888889,
      "grad_norm": 0.00045024172868579626,
      "learning_rate": 1.294927624533931e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7393200,
      "step": 26405
    },
    {
      "epoch": 293.44444444444446,
      "grad_norm": 0.00024361873511224985,
      "learning_rate": 1.2940675540307378e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7394640,
      "step": 26410
    },
    {
      "epoch": 293.5,
      "grad_norm": 0.003889176994562149,
      "learning_rate": 1.2932076694974814e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7396016,
      "step": 26415
    },
    {
      "epoch": 293.55555555555554,
      "grad_norm": 0.00010977595229633152,
      "learning_rate": 1.2923479710667682e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7397424,
      "step": 26420
    },
    {
      "epoch": 293.6111111111111,
      "grad_norm": 6.017927444190718e-05,
      "learning_rate": 1.2914884588711751e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7398816,
      "step": 26425
    },
    {
      "epoch": 293.6666666666667,
      "grad_norm": 0.0001752168609527871,
      "learning_rate": 1.2906291330432475e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7400224,
      "step": 26430
    },
    {
      "epoch": 293.72222222222223,
      "grad_norm": 0.023027142509818077,
      "learning_rate": 1.2897699937155055e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7401632,
      "step": 26435
    },
    {
      "epoch": 293.77777777777777,
      "grad_norm": 0.00014365461538545787,
      "learning_rate": 1.2889110410204403e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7403040,
      "step": 26440
    },
    {
      "epoch": 293.8333333333333,
      "grad_norm": 0.0001811549736885354,
      "learning_rate": 1.2880522750905111e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7404448,
      "step": 26445
    },
    {
      "epoch": 293.8888888888889,
      "grad_norm": 0.0005546101019717753,
      "learning_rate": 1.2871936960581523e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7405808,
      "step": 26450
    },
    {
      "epoch": 293.94444444444446,
      "grad_norm": 6.14754535490647e-05,
      "learning_rate": 1.2863353040557658e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7407232,
      "step": 26455
    },
    {
      "epoch": 294.0,
      "grad_norm": 0.0001647568424232304,
      "learning_rate": 1.2854770992157273e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7408640,
      "step": 26460
    },
    {
      "epoch": 294.05555555555554,
      "grad_norm": 0.00015296645869966596,
      "learning_rate": 1.2846190816703835e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7410048,
      "step": 26465
    },
    {
      "epoch": 294.1111111111111,
      "grad_norm": 4.956752673024312e-05,
      "learning_rate": 1.2837612515520498e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7411456,
      "step": 26470
    },
    {
      "epoch": 294.1666666666667,
      "grad_norm": 6.23162486590445e-05,
      "learning_rate": 1.2829036089930163e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7412912,
      "step": 26475
    },
    {
      "epoch": 294.22222222222223,
      "grad_norm": 4.19275056628976e-05,
      "learning_rate": 1.2820461541255412e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7414320,
      "step": 26480
    },
    {
      "epoch": 294.27777777777777,
      "grad_norm": 9.651368600316346e-05,
      "learning_rate": 1.2811888870818543e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7415792,
      "step": 26485
    },
    {
      "epoch": 294.3333333333333,
      "grad_norm": 0.004291854333132505,
      "learning_rate": 1.2803318079941581e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7417216,
      "step": 26490
    },
    {
      "epoch": 294.3888888888889,
      "grad_norm": 0.00020028497965540737,
      "learning_rate": 1.2794749169946235e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7418608,
      "step": 26495
    },
    {
      "epoch": 294.44444444444446,
      "grad_norm": 5.966507160337642e-05,
      "learning_rate": 1.2786182142153952e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7420000,
      "step": 26500
    },
    {
      "epoch": 294.5,
      "grad_norm": 0.001758609781973064,
      "learning_rate": 1.2777616997885878e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7421376,
      "step": 26505
    },
    {
      "epoch": 294.55555555555554,
      "grad_norm": 0.0007603613194078207,
      "learning_rate": 1.2769053738462847e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7422768,
      "step": 26510
    },
    {
      "epoch": 294.6111111111111,
      "grad_norm": 0.0005701407790184021,
      "learning_rate": 1.2760492365205434e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7424128,
      "step": 26515
    },
    {
      "epoch": 294.6666666666667,
      "grad_norm": 0.00011200126027688384,
      "learning_rate": 1.2751932879433919e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7425504,
      "step": 26520
    },
    {
      "epoch": 294.72222222222223,
      "grad_norm": 0.00015817642270121723,
      "learning_rate": 1.2743375282468267e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7426848,
      "step": 26525
    },
    {
      "epoch": 294.77777777777777,
      "grad_norm": 0.00011054021160816774,
      "learning_rate": 1.2734819575628182e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7428240,
      "step": 26530
    },
    {
      "epoch": 294.8333333333333,
      "grad_norm": 0.0002332645672140643,
      "learning_rate": 1.2726265760233039e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7429632,
      "step": 26535
    },
    {
      "epoch": 294.8888888888889,
      "grad_norm": 0.03853954002261162,
      "learning_rate": 1.271771383760197e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7431008,
      "step": 26540
    },
    {
      "epoch": 294.94444444444446,
      "grad_norm": 0.003050224855542183,
      "learning_rate": 1.2709163809053764e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7432416,
      "step": 26545
    },
    {
      "epoch": 295.0,
      "grad_norm": 0.00047508993884548545,
      "learning_rate": 1.2700615675906963e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7433808,
      "step": 26550
    },
    {
      "epoch": 295.05555555555554,
      "grad_norm": 4.594837810145691e-05,
      "learning_rate": 1.269206943947978e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7435200,
      "step": 26555
    },
    {
      "epoch": 295.1111111111111,
      "grad_norm": 6.94455811753869e-05,
      "learning_rate": 1.2683525101090177e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7436608,
      "step": 26560
    },
    {
      "epoch": 295.1666666666667,
      "grad_norm": 5.3688847401645035e-05,
      "learning_rate": 1.2674982662055765e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7438016,
      "step": 26565
    },
    {
      "epoch": 295.22222222222223,
      "grad_norm": 0.0005520188133232296,
      "learning_rate": 1.2666442123693922e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7439376,
      "step": 26570
    },
    {
      "epoch": 295.27777777777777,
      "grad_norm": 3.9840499084675685e-05,
      "learning_rate": 1.265790348732169e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7440784,
      "step": 26575
    },
    {
      "epoch": 295.3333333333333,
      "grad_norm": 0.0004113873583264649,
      "learning_rate": 1.264936675425584e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7442176,
      "step": 26580
    },
    {
      "epoch": 295.3888888888889,
      "grad_norm": 9.359977411804721e-05,
      "learning_rate": 1.2640831925812852e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7443568,
      "step": 26585
    },
    {
      "epoch": 295.44444444444446,
      "grad_norm": 8.582478039897978e-05,
      "learning_rate": 1.263229900330889e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7444992,
      "step": 26590
    },
    {
      "epoch": 295.5,
      "grad_norm": 0.00014328656834550202,
      "learning_rate": 1.2623767988059843e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7446368,
      "step": 26595
    },
    {
      "epoch": 295.55555555555554,
      "grad_norm": 0.0022429670207202435,
      "learning_rate": 1.2615238881381309e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7447808,
      "step": 26600
    },
    {
      "epoch": 295.55555555555554,
      "eval_loss": 0.8601922988891602,
      "eval_runtime": 1.238,
      "eval_samples_per_second": 32.309,
      "eval_steps_per_second": 16.154,
      "num_input_tokens_seen": 7447808,
      "step": 26600
    },
    {
      "epoch": 295.6111111111111,
      "grad_norm": 0.00017768102406989783,
      "learning_rate": 1.2606711684588568e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7449200,
      "step": 26605
    },
    {
      "epoch": 295.6666666666667,
      "grad_norm": 0.0002744766534306109,
      "learning_rate": 1.2598186398996636e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7450560,
      "step": 26610
    },
    {
      "epoch": 295.72222222222223,
      "grad_norm": 8.819909271551296e-05,
      "learning_rate": 1.2589663025920207e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7451952,
      "step": 26615
    },
    {
      "epoch": 295.77777777777777,
      "grad_norm": 3.712128091137856e-05,
      "learning_rate": 1.2581141566673705e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7453360,
      "step": 26620
    },
    {
      "epoch": 295.8333333333333,
      "grad_norm": 8.789601997705176e-05,
      "learning_rate": 1.257262202257124e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7454768,
      "step": 26625
    },
    {
      "epoch": 295.8888888888889,
      "grad_norm": 0.0013103443197906017,
      "learning_rate": 1.2564104394926618e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7456176,
      "step": 26630
    },
    {
      "epoch": 295.94444444444446,
      "grad_norm": 8.960602281149477e-05,
      "learning_rate": 1.2555588685053383e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7457584,
      "step": 26635
    },
    {
      "epoch": 296.0,
      "grad_norm": 0.0004034583689644933,
      "learning_rate": 1.2547074894264762e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7458944,
      "step": 26640
    },
    {
      "epoch": 296.05555555555554,
      "grad_norm": 0.0021617813035845757,
      "learning_rate": 1.2538563023873679e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7460352,
      "step": 26645
    },
    {
      "epoch": 296.1111111111111,
      "grad_norm": 0.03300900384783745,
      "learning_rate": 1.2530053075192789e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7461776,
      "step": 26650
    },
    {
      "epoch": 296.1666666666667,
      "grad_norm": 0.00038132438203319907,
      "learning_rate": 1.252154504953441e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7463184,
      "step": 26655
    },
    {
      "epoch": 296.22222222222223,
      "grad_norm": 0.0007526854169555008,
      "learning_rate": 1.25130389482106e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7464528,
      "step": 26660
    },
    {
      "epoch": 296.27777777777777,
      "grad_norm": 0.00011122233991045505,
      "learning_rate": 1.2504534772533116e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7465904,
      "step": 26665
    },
    {
      "epoch": 296.3333333333333,
      "grad_norm": 7.361786265391856e-05,
      "learning_rate": 1.2496032523813387e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7467312,
      "step": 26670
    },
    {
      "epoch": 296.3888888888889,
      "grad_norm": 0.000350397196598351,
      "learning_rate": 1.2487532203362576e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7468736,
      "step": 26675
    },
    {
      "epoch": 296.44444444444446,
      "grad_norm": 7.841470505809411e-05,
      "learning_rate": 1.247903381249155e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7470176,
      "step": 26680
    },
    {
      "epoch": 296.5,
      "grad_norm": 0.004379710182547569,
      "learning_rate": 1.2470537352510853e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7471616,
      "step": 26685
    },
    {
      "epoch": 296.55555555555554,
      "grad_norm": 0.0009204487432725728,
      "learning_rate": 1.2462042824730758e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7472992,
      "step": 26690
    },
    {
      "epoch": 296.6111111111111,
      "grad_norm": 0.00030465685995295644,
      "learning_rate": 1.245355023046122e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7474368,
      "step": 26695
    },
    {
      "epoch": 296.6666666666667,
      "grad_norm": 6.74836992402561e-05,
      "learning_rate": 1.2445059571011896e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7475776,
      "step": 26700
    },
    {
      "epoch": 296.72222222222223,
      "grad_norm": 2.8202142857480794e-05,
      "learning_rate": 1.2436570847692173e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7477120,
      "step": 26705
    },
    {
      "epoch": 296.77777777777777,
      "grad_norm": 0.00012223060184624046,
      "learning_rate": 1.2428084061811096e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7478544,
      "step": 26710
    },
    {
      "epoch": 296.8333333333333,
      "grad_norm": 5.793133095721714e-05,
      "learning_rate": 1.2419599214677447e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7479952,
      "step": 26715
    },
    {
      "epoch": 296.8888888888889,
      "grad_norm": 0.002560560591518879,
      "learning_rate": 1.2411116307599702e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7481328,
      "step": 26720
    },
    {
      "epoch": 296.94444444444446,
      "grad_norm": 4.058533158968203e-05,
      "learning_rate": 1.2402635341886016e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7482784,
      "step": 26725
    },
    {
      "epoch": 297.0,
      "grad_norm": 0.00013994783512316644,
      "learning_rate": 1.2394156318844278e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7484208,
      "step": 26730
    },
    {
      "epoch": 297.05555555555554,
      "grad_norm": 0.00026524311397224665,
      "learning_rate": 1.2385679239782039e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7485600,
      "step": 26735
    },
    {
      "epoch": 297.1111111111111,
      "grad_norm": 0.00023418849741574377,
      "learning_rate": 1.2377204106006585e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7486960,
      "step": 26740
    },
    {
      "epoch": 297.1666666666667,
      "grad_norm": 0.00017062995175365359,
      "learning_rate": 1.2368730918824891e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7488304,
      "step": 26745
    },
    {
      "epoch": 297.22222222222223,
      "grad_norm": 9.229899296769872e-05,
      "learning_rate": 1.236025967954362e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7489760,
      "step": 26750
    },
    {
      "epoch": 297.27777777777777,
      "grad_norm": 5.055155997979455e-05,
      "learning_rate": 1.2351790389469153e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7491136,
      "step": 26755
    },
    {
      "epoch": 297.3333333333333,
      "grad_norm": 0.0006956947036087513,
      "learning_rate": 1.234332304990755e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7492544,
      "step": 26760
    },
    {
      "epoch": 297.3888888888889,
      "grad_norm": 0.0008249765378423035,
      "learning_rate": 1.2334857662164593e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7493952,
      "step": 26765
    },
    {
      "epoch": 297.44444444444446,
      "grad_norm": 5.627415157505311e-05,
      "learning_rate": 1.2326394227545743e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7495376,
      "step": 26770
    },
    {
      "epoch": 297.5,
      "grad_norm": 0.00015280711522791535,
      "learning_rate": 1.2317932747356162e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7496752,
      "step": 26775
    },
    {
      "epoch": 297.55555555555554,
      "grad_norm": 0.00019728961342480034,
      "learning_rate": 1.2309473222900726e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7498176,
      "step": 26780
    },
    {
      "epoch": 297.6111111111111,
      "grad_norm": 0.00019320020510349423,
      "learning_rate": 1.2301015655484006e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7499552,
      "step": 26785
    },
    {
      "epoch": 297.6666666666667,
      "grad_norm": 0.0017684089252725244,
      "learning_rate": 1.2292560046410245e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7500960,
      "step": 26790
    },
    {
      "epoch": 297.72222222222223,
      "grad_norm": 0.00010496834147488698,
      "learning_rate": 1.228410639698343e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7502384,
      "step": 26795
    },
    {
      "epoch": 297.77777777777777,
      "grad_norm": 0.00012240410433150828,
      "learning_rate": 1.2275654708507195e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7503824,
      "step": 26800
    },
    {
      "epoch": 297.77777777777777,
      "eval_loss": 0.8605717420578003,
      "eval_runtime": 1.2338,
      "eval_samples_per_second": 32.421,
      "eval_steps_per_second": 16.21,
      "num_input_tokens_seen": 7503824,
      "step": 26800
    },
    {
      "epoch": 297.8333333333333,
      "grad_norm": 0.00012981540930923074,
      "learning_rate": 1.2267204982284908e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7505216,
      "step": 26805
    },
    {
      "epoch": 297.8888888888889,
      "grad_norm": 0.00011704143253155053,
      "learning_rate": 1.2258757219619635e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7506608,
      "step": 26810
    },
    {
      "epoch": 297.94444444444446,
      "grad_norm": 0.00020732991106342524,
      "learning_rate": 1.2250311421814104e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7508000,
      "step": 26815
    },
    {
      "epoch": 298.0,
      "grad_norm": 0.0017872145399451256,
      "learning_rate": 1.2241867590170772e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7509456,
      "step": 26820
    },
    {
      "epoch": 298.05555555555554,
      "grad_norm": 9.342228440800682e-05,
      "learning_rate": 1.2233425725991799e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 7510784,
      "step": 26825
    },
    {
      "epoch": 298.1111111111111,
      "grad_norm": 3.9040485717123374e-05,
      "learning_rate": 1.2224985830579003e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7512176,
      "step": 26830
    },
    {
      "epoch": 298.1666666666667,
      "grad_norm": 8.24832241050899e-05,
      "learning_rate": 1.2216547905233944e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7513568,
      "step": 26835
    },
    {
      "epoch": 298.22222222222223,
      "grad_norm": 0.00039446147275157273,
      "learning_rate": 1.2208111951257842e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 7514992,
      "step": 26840
    },
    {
      "epoch": 298.27777777777777,
      "grad_norm": 0.0001811567199183628,
      "learning_rate": 1.2199677969951622e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 7516400,
      "step": 26845
    },
    {
      "epoch": 298.3333333333333,
      "grad_norm": 5.218713704380207e-05,
      "learning_rate": 1.2191245962615927e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7517856,
      "step": 26850
    },
    {
      "epoch": 298.3888888888889,
      "grad_norm": 0.000361867540050298,
      "learning_rate": 1.218281593055106e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7519232,
      "step": 26855
    },
    {
      "epoch": 298.44444444444446,
      "grad_norm": 0.0011180712608620524,
      "learning_rate": 1.217438787505705e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7520672,
      "step": 26860
    },
    {
      "epoch": 298.5,
      "grad_norm": 0.00036426453152671456,
      "learning_rate": 1.2165961797433615e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7522080,
      "step": 26865
    },
    {
      "epoch": 298.55555555555554,
      "grad_norm": 0.006417266093194485,
      "learning_rate": 1.215753769898014e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7523488,
      "step": 26870
    },
    {
      "epoch": 298.6111111111111,
      "grad_norm": 0.0014014403568580747,
      "learning_rate": 1.2149115580995755e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7524848,
      "step": 26875
    },
    {
      "epoch": 298.6666666666667,
      "grad_norm": 0.003680715337395668,
      "learning_rate": 1.2140695444779227e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7526224,
      "step": 26880
    },
    {
      "epoch": 298.72222222222223,
      "grad_norm": 0.00019874291319865733,
      "learning_rate": 1.2132277291629066e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7527600,
      "step": 26885
    },
    {
      "epoch": 298.77777777777777,
      "grad_norm": 0.002616644138470292,
      "learning_rate": 1.2123861122843458e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7528992,
      "step": 26890
    },
    {
      "epoch": 298.8333333333333,
      "grad_norm": 0.004218318033963442,
      "learning_rate": 1.2115446939720271e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7530416,
      "step": 26895
    },
    {
      "epoch": 298.8888888888889,
      "grad_norm": 0.00014723086496815085,
      "learning_rate": 1.210703474355708e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7531792,
      "step": 26900
    },
    {
      "epoch": 298.94444444444446,
      "grad_norm": 0.0017604818567633629,
      "learning_rate": 1.2098624535651164e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7533184,
      "step": 26905
    },
    {
      "epoch": 299.0,
      "grad_norm": 0.00033796660136431456,
      "learning_rate": 1.2090216317299477e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7534608,
      "step": 26910
    },
    {
      "epoch": 299.05555555555554,
      "grad_norm": 0.00015858749975450337,
      "learning_rate": 1.2081810089798668e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7536000,
      "step": 26915
    },
    {
      "epoch": 299.1111111111111,
      "grad_norm": 6.073172698961571e-05,
      "learning_rate": 1.2073405854445072e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7537376,
      "step": 26920
    },
    {
      "epoch": 299.1666666666667,
      "grad_norm": 3.3280852221651e-05,
      "learning_rate": 1.206500361253474e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7538784,
      "step": 26925
    },
    {
      "epoch": 299.22222222222223,
      "grad_norm": 0.008995480835437775,
      "learning_rate": 1.2056603365363409e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7540176,
      "step": 26930
    },
    {
      "epoch": 299.27777777777777,
      "grad_norm": 0.010270675644278526,
      "learning_rate": 1.2048205114226487e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7541568,
      "step": 26935
    },
    {
      "epoch": 299.3333333333333,
      "grad_norm": 0.00017988367471843958,
      "learning_rate": 1.2039808860419102e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7542992,
      "step": 26940
    },
    {
      "epoch": 299.3888888888889,
      "grad_norm": 4.5257322199177e-05,
      "learning_rate": 1.2031414605236066e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7544416,
      "step": 26945
    },
    {
      "epoch": 299.44444444444446,
      "grad_norm": 0.0014423339162021875,
      "learning_rate": 1.2023022349971862e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7545840,
      "step": 26950
    },
    {
      "epoch": 299.5,
      "grad_norm": 0.000877690443303436,
      "learning_rate": 1.20146320959207e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7547264,
      "step": 26955
    },
    {
      "epoch": 299.55555555555554,
      "grad_norm": 0.00013406069774646312,
      "learning_rate": 1.2006243844376445e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7548608,
      "step": 26960
    },
    {
      "epoch": 299.6111111111111,
      "grad_norm": 0.00010265012679155916,
      "learning_rate": 1.1997857596632678e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7549984,
      "step": 26965
    },
    {
      "epoch": 299.6666666666667,
      "grad_norm": 0.00020703354675788432,
      "learning_rate": 1.1989473353982672e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7551408,
      "step": 26970
    },
    {
      "epoch": 299.72222222222223,
      "grad_norm": 9.605368541087955e-05,
      "learning_rate": 1.198109111771937e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7552800,
      "step": 26975
    },
    {
      "epoch": 299.77777777777777,
      "grad_norm": 0.00029394921148195863,
      "learning_rate": 1.197271088913543e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7554208,
      "step": 26980
    },
    {
      "epoch": 299.8333333333333,
      "grad_norm": 0.00014707459195051342,
      "learning_rate": 1.1964332669523182e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7555648,
      "step": 26985
    },
    {
      "epoch": 299.8888888888889,
      "grad_norm": 0.0003243055834900588,
      "learning_rate": 1.1955956460174645e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7557040,
      "step": 26990
    },
    {
      "epoch": 299.94444444444446,
      "grad_norm": 0.00013473666331265122,
      "learning_rate": 1.1947582262381552e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7558448,
      "step": 26995
    },
    {
      "epoch": 300.0,
      "grad_norm": 0.0004761978634633124,
      "learning_rate": 1.1939210077435293e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7559856,
      "step": 27000
    },
    {
      "epoch": 300.0,
      "eval_loss": 0.8232558965682983,
      "eval_runtime": 1.2347,
      "eval_samples_per_second": 32.396,
      "eval_steps_per_second": 16.198,
      "num_input_tokens_seen": 7559856,
      "step": 27000
    },
    {
      "epoch": 300.05555555555554,
      "grad_norm": 4.9173395382240415e-05,
      "learning_rate": 1.193083990662697e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7561312,
      "step": 27005
    },
    {
      "epoch": 300.1111111111111,
      "grad_norm": 0.00019649520982056856,
      "learning_rate": 1.192247175124738e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7562688,
      "step": 27010
    },
    {
      "epoch": 300.1666666666667,
      "grad_norm": 0.00019599171355366707,
      "learning_rate": 1.191410561258698e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7564112,
      "step": 27015
    },
    {
      "epoch": 300.22222222222223,
      "grad_norm": 0.00036574137629941106,
      "learning_rate": 1.1905741491935944e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7565504,
      "step": 27020
    },
    {
      "epoch": 300.27777777777777,
      "grad_norm": 4.040405110572465e-05,
      "learning_rate": 1.1897379390584129e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7566928,
      "step": 27025
    },
    {
      "epoch": 300.3333333333333,
      "grad_norm": 0.00036967010237276554,
      "learning_rate": 1.1889019309821062e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7568352,
      "step": 27030
    },
    {
      "epoch": 300.3888888888889,
      "grad_norm": 0.00011205236660316586,
      "learning_rate": 1.188066125093599e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7569792,
      "step": 27035
    },
    {
      "epoch": 300.44444444444446,
      "grad_norm": 0.0001846320228651166,
      "learning_rate": 1.1872305215217811e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7571232,
      "step": 27040
    },
    {
      "epoch": 300.5,
      "grad_norm": 0.00021745743288192898,
      "learning_rate": 1.186395120395514e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7572624,
      "step": 27045
    },
    {
      "epoch": 300.55555555555554,
      "grad_norm": 6.747058068867773e-05,
      "learning_rate": 1.1855599218436283e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7574000,
      "step": 27050
    },
    {
      "epoch": 300.6111111111111,
      "grad_norm": 5.5287659051828086e-05,
      "learning_rate": 1.1847249259949209e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7575392,
      "step": 27055
    },
    {
      "epoch": 300.6666666666667,
      "grad_norm": 3.803117579082027e-05,
      "learning_rate": 1.1838901329781574e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7576784,
      "step": 27060
    },
    {
      "epoch": 300.72222222222223,
      "grad_norm": 4.2831390601349995e-05,
      "learning_rate": 1.1830555429220758e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7578176,
      "step": 27065
    },
    {
      "epoch": 300.77777777777777,
      "grad_norm": 5.229647649684921e-05,
      "learning_rate": 1.1822211559553784e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7579568,
      "step": 27070
    },
    {
      "epoch": 300.8333333333333,
      "grad_norm": 0.00032189724151976407,
      "learning_rate": 1.18138697220674e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7581008,
      "step": 27075
    },
    {
      "epoch": 300.8888888888889,
      "grad_norm": 0.0018371950136497617,
      "learning_rate": 1.1805529918048e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7582368,
      "step": 27080
    },
    {
      "epoch": 300.94444444444446,
      "grad_norm": 4.438486575963907e-05,
      "learning_rate": 1.1797192148781702e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7583744,
      "step": 27085
    },
    {
      "epoch": 301.0,
      "grad_norm": 0.000541142129804939,
      "learning_rate": 1.1788856415554297e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7585184,
      "step": 27090
    },
    {
      "epoch": 301.05555555555554,
      "grad_norm": 0.00019915327720809728,
      "learning_rate": 1.1780522719651249e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7586608,
      "step": 27095
    },
    {
      "epoch": 301.1111111111111,
      "grad_norm": 0.0001461817737435922,
      "learning_rate": 1.1772191062357721e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7588032,
      "step": 27100
    },
    {
      "epoch": 301.1666666666667,
      "grad_norm": 3.359903712407686e-05,
      "learning_rate": 1.1763861444958573e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7589408,
      "step": 27105
    },
    {
      "epoch": 301.22222222222223,
      "grad_norm": 0.00011036951036658138,
      "learning_rate": 1.1755533868738317e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7590816,
      "step": 27110
    },
    {
      "epoch": 301.27777777777777,
      "grad_norm": 0.00503143947571516,
      "learning_rate": 1.1747208334981185e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7592176,
      "step": 27115
    },
    {
      "epoch": 301.3333333333333,
      "grad_norm": 0.00045732862781733274,
      "learning_rate": 1.1738884844971067e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7593600,
      "step": 27120
    },
    {
      "epoch": 301.3888888888889,
      "grad_norm": 0.0003553881833795458,
      "learning_rate": 1.1730563399991563e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7595008,
      "step": 27125
    },
    {
      "epoch": 301.44444444444446,
      "grad_norm": 0.0018682655645534396,
      "learning_rate": 1.1722244001325938e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7596448,
      "step": 27130
    },
    {
      "epoch": 301.5,
      "grad_norm": 0.0004886770038865507,
      "learning_rate": 1.1713926650257137e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7597872,
      "step": 27135
    },
    {
      "epoch": 301.55555555555554,
      "grad_norm": 0.00635748403146863,
      "learning_rate": 1.170561134806781e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7599232,
      "step": 27140
    },
    {
      "epoch": 301.6111111111111,
      "grad_norm": 3.799217665800825e-05,
      "learning_rate": 1.1697298096040287e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7600576,
      "step": 27145
    },
    {
      "epoch": 301.6666666666667,
      "grad_norm": 0.0007110161241143942,
      "learning_rate": 1.1688986895456567e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7601968,
      "step": 27150
    },
    {
      "epoch": 301.72222222222223,
      "grad_norm": 5.1182541938032955e-05,
      "learning_rate": 1.1680677747598349e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7603360,
      "step": 27155
    },
    {
      "epoch": 301.77777777777777,
      "grad_norm": 0.00026219451683573425,
      "learning_rate": 1.1672370653746995e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7604768,
      "step": 27160
    },
    {
      "epoch": 301.8333333333333,
      "grad_norm": 0.0003303597623016685,
      "learning_rate": 1.166406561518357e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7606176,
      "step": 27165
    },
    {
      "epoch": 301.8888888888889,
      "grad_norm": 0.0001179177052108571,
      "learning_rate": 1.1655762633188826e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7607584,
      "step": 27170
    },
    {
      "epoch": 301.94444444444446,
      "grad_norm": 0.00047234425437636673,
      "learning_rate": 1.1647461709043172e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7608976,
      "step": 27175
    },
    {
      "epoch": 302.0,
      "grad_norm": 0.00023355057055596262,
      "learning_rate": 1.1639162844026722e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7610368,
      "step": 27180
    },
    {
      "epoch": 302.05555555555554,
      "grad_norm": 0.0009061955497600138,
      "learning_rate": 1.163086603941927e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7611776,
      "step": 27185
    },
    {
      "epoch": 302.1111111111111,
      "grad_norm": 0.0003647196863312274,
      "learning_rate": 1.1622571296500273e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7613136,
      "step": 27190
    },
    {
      "epoch": 302.1666666666667,
      "grad_norm": 5.593056630459614e-05,
      "learning_rate": 1.1614278616548904e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7614496,
      "step": 27195
    },
    {
      "epoch": 302.22222222222223,
      "grad_norm": 0.00019231977057643235,
      "learning_rate": 1.1605988000843986e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7615904,
      "step": 27200
    },
    {
      "epoch": 302.22222222222223,
      "eval_loss": 0.8202501535415649,
      "eval_runtime": 1.2398,
      "eval_samples_per_second": 32.262,
      "eval_steps_per_second": 16.131,
      "num_input_tokens_seen": 7615904,
      "step": 27200
    },
    {
      "epoch": 302.27777777777777,
      "grad_norm": 0.00012731371680274606,
      "learning_rate": 1.1597699450664028e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7617328,
      "step": 27205
    },
    {
      "epoch": 302.3333333333333,
      "grad_norm": 0.00013933572336100042,
      "learning_rate": 1.1589412967287252e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7618784,
      "step": 27210
    },
    {
      "epoch": 302.3888888888889,
      "grad_norm": 4.528572753770277e-05,
      "learning_rate": 1.1581128551991514e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7620144,
      "step": 27215
    },
    {
      "epoch": 302.44444444444446,
      "grad_norm": 0.0044064209796488285,
      "learning_rate": 1.1572846206054383e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7621520,
      "step": 27220
    },
    {
      "epoch": 302.5,
      "grad_norm": 0.0029069255106151104,
      "learning_rate": 1.1564565930753113e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7622928,
      "step": 27225
    },
    {
      "epoch": 302.55555555555554,
      "grad_norm": 5.39559950993862e-05,
      "learning_rate": 1.1556287727364606e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7624304,
      "step": 27230
    },
    {
      "epoch": 302.6111111111111,
      "grad_norm": 0.00020244003098923713,
      "learning_rate": 1.1548011597165489e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7625728,
      "step": 27235
    },
    {
      "epoch": 302.6666666666667,
      "grad_norm": 0.00017960539844352752,
      "learning_rate": 1.1539737541432019e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7627120,
      "step": 27240
    },
    {
      "epoch": 302.72222222222223,
      "grad_norm": 2.6515795980230905e-05,
      "learning_rate": 1.1531465561440174e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7628544,
      "step": 27245
    },
    {
      "epoch": 302.77777777777777,
      "grad_norm": 4.9579761252971366e-05,
      "learning_rate": 1.1523195658465605e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7629936,
      "step": 27250
    },
    {
      "epoch": 302.8333333333333,
      "grad_norm": 0.0006606702227145433,
      "learning_rate": 1.1514927833783618e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7631344,
      "step": 27255
    },
    {
      "epoch": 302.8888888888889,
      "grad_norm": 0.0017658794531598687,
      "learning_rate": 1.150666208866922e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7632704,
      "step": 27260
    },
    {
      "epoch": 302.94444444444446,
      "grad_norm": 0.0010439867619425058,
      "learning_rate": 1.1498398424397106e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7634128,
      "step": 27265
    },
    {
      "epoch": 303.0,
      "grad_norm": 0.00010933280282188207,
      "learning_rate": 1.1490136842241628e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7635536,
      "step": 27270
    },
    {
      "epoch": 303.05555555555554,
      "grad_norm": 3.596952956286259e-05,
      "learning_rate": 1.1481877343476813e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7636928,
      "step": 27275
    },
    {
      "epoch": 303.1111111111111,
      "grad_norm": 0.0001346189237665385,
      "learning_rate": 1.14736199293764e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7638304,
      "step": 27280
    },
    {
      "epoch": 303.1666666666667,
      "grad_norm": 0.0027643588837236166,
      "learning_rate": 1.1465364601213771e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7639696,
      "step": 27285
    },
    {
      "epoch": 303.22222222222223,
      "grad_norm": 0.00015798883396200836,
      "learning_rate": 1.1457111360262012e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7641072,
      "step": 27290
    },
    {
      "epoch": 303.27777777777777,
      "grad_norm": 0.00027947398484684527,
      "learning_rate": 1.1448860207793869e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7642448,
      "step": 27295
    },
    {
      "epoch": 303.3333333333333,
      "grad_norm": 0.001432853052392602,
      "learning_rate": 1.144061114508177e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7643856,
      "step": 27300
    },
    {
      "epoch": 303.3888888888889,
      "grad_norm": 0.00010862420458579436,
      "learning_rate": 1.1432364173397842e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7645264,
      "step": 27305
    },
    {
      "epoch": 303.44444444444446,
      "grad_norm": 0.0032745657954365015,
      "learning_rate": 1.1424119294013852e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7646672,
      "step": 27310
    },
    {
      "epoch": 303.5,
      "grad_norm": 0.00012360142136458308,
      "learning_rate": 1.1415876508201279e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7648080,
      "step": 27315
    },
    {
      "epoch": 303.55555555555554,
      "grad_norm": 4.079699647263624e-05,
      "learning_rate": 1.140763581723125e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7649504,
      "step": 27320
    },
    {
      "epoch": 303.6111111111111,
      "grad_norm": 2.6604020604281686e-05,
      "learning_rate": 1.1399397222374588e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7650912,
      "step": 27325
    },
    {
      "epoch": 303.6666666666667,
      "grad_norm": 7.680043927393854e-05,
      "learning_rate": 1.1391160724901804e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7652336,
      "step": 27330
    },
    {
      "epoch": 303.72222222222223,
      "grad_norm": 0.00019239471293985844,
      "learning_rate": 1.138292632608304e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7653744,
      "step": 27335
    },
    {
      "epoch": 303.77777777777777,
      "grad_norm": 0.00010498771007405594,
      "learning_rate": 1.1374694027188174e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7655184,
      "step": 27340
    },
    {
      "epoch": 303.8333333333333,
      "grad_norm": 7.78041867306456e-05,
      "learning_rate": 1.1366463829486711e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7656608,
      "step": 27345
    },
    {
      "epoch": 303.8888888888889,
      "grad_norm": 0.0007083378732204437,
      "learning_rate": 1.1358235734247849e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7657984,
      "step": 27350
    },
    {
      "epoch": 303.94444444444446,
      "grad_norm": 2.9547711164923385e-05,
      "learning_rate": 1.1350009742740478e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7659376,
      "step": 27355
    },
    {
      "epoch": 304.0,
      "grad_norm": 0.00024462465080432594,
      "learning_rate": 1.134178585623313e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7660800,
      "step": 27360
    },
    {
      "epoch": 304.05555555555554,
      "grad_norm": 0.0002009818417718634,
      "learning_rate": 1.1333564075994047e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7662224,
      "step": 27365
    },
    {
      "epoch": 304.1111111111111,
      "grad_norm": 0.0002163554891012609,
      "learning_rate": 1.1325344403291133e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7663648,
      "step": 27370
    },
    {
      "epoch": 304.1666666666667,
      "grad_norm": 0.0010049345437437296,
      "learning_rate": 1.1317126839391951e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7665024,
      "step": 27375
    },
    {
      "epoch": 304.22222222222223,
      "grad_norm": 0.00018884165911003947,
      "learning_rate": 1.1308911385563766e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7666448,
      "step": 27380
    },
    {
      "epoch": 304.27777777777777,
      "grad_norm": 0.00019352113304194063,
      "learning_rate": 1.1300698043073494e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7667840,
      "step": 27385
    },
    {
      "epoch": 304.3333333333333,
      "grad_norm": 0.0004032762080896646,
      "learning_rate": 1.1292486813187736e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7669232,
      "step": 27390
    },
    {
      "epoch": 304.3888888888889,
      "grad_norm": 0.00013985966506879777,
      "learning_rate": 1.1284277697172782e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7670608,
      "step": 27395
    },
    {
      "epoch": 304.44444444444446,
      "grad_norm": 0.000392166810343042,
      "learning_rate": 1.127607069629456e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7672000,
      "step": 27400
    },
    {
      "epoch": 304.44444444444446,
      "eval_loss": 0.8417047262191772,
      "eval_runtime": 1.24,
      "eval_samples_per_second": 32.258,
      "eval_steps_per_second": 16.129,
      "num_input_tokens_seen": 7672000,
      "step": 27400
    },
    {
      "epoch": 304.5,
      "grad_norm": 0.00012160490587120876,
      "learning_rate": 1.1267865811818701e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7673360,
      "step": 27405
    },
    {
      "epoch": 304.55555555555554,
      "grad_norm": 0.000264530855929479,
      "learning_rate": 1.1259663045010513e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7674720,
      "step": 27410
    },
    {
      "epoch": 304.6111111111111,
      "grad_norm": 3.0222225177567452e-05,
      "learning_rate": 1.1251462397134957e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7676144,
      "step": 27415
    },
    {
      "epoch": 304.6666666666667,
      "grad_norm": 0.00010821411706274375,
      "learning_rate": 1.1243263869456664e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7677568,
      "step": 27420
    },
    {
      "epoch": 304.72222222222223,
      "grad_norm": 0.00020652421517297626,
      "learning_rate": 1.1235067463239967e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7678992,
      "step": 27425
    },
    {
      "epoch": 304.77777777777777,
      "grad_norm": 3.3913169318111613e-05,
      "learning_rate": 1.122687317974884e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7680384,
      "step": 27430
    },
    {
      "epoch": 304.8333333333333,
      "grad_norm": 0.00018816182273440063,
      "learning_rate": 1.1218681020246963e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7681760,
      "step": 27435
    },
    {
      "epoch": 304.8888888888889,
      "grad_norm": 3.76842544937972e-05,
      "learning_rate": 1.1210490985997652e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7683168,
      "step": 27440
    },
    {
      "epoch": 304.94444444444446,
      "grad_norm": 0.0002064568834612146,
      "learning_rate": 1.1202303078263917e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7684576,
      "step": 27445
    },
    {
      "epoch": 305.0,
      "grad_norm": 6.053039032849483e-05,
      "learning_rate": 1.1194117298308451e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7685984,
      "step": 27450
    },
    {
      "epoch": 305.05555555555554,
      "grad_norm": 0.0011447478318586946,
      "learning_rate": 1.1185933647393585e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7687376,
      "step": 27455
    },
    {
      "epoch": 305.1111111111111,
      "grad_norm": 5.182814857107587e-05,
      "learning_rate": 1.1177752126781354e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7688816,
      "step": 27460
    },
    {
      "epoch": 305.1666666666667,
      "grad_norm": 0.00021167394879739732,
      "learning_rate": 1.1169572737733441e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7690224,
      "step": 27465
    },
    {
      "epoch": 305.22222222222223,
      "grad_norm": 0.0001328808139078319,
      "learning_rate": 1.1161395481511216e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7691616,
      "step": 27470
    },
    {
      "epoch": 305.27777777777777,
      "grad_norm": 0.0003936415887437761,
      "learning_rate": 1.1153220359375722e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7692992,
      "step": 27475
    },
    {
      "epoch": 305.3333333333333,
      "grad_norm": 0.00014547987666446716,
      "learning_rate": 1.114504737258765e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7694384,
      "step": 27480
    },
    {
      "epoch": 305.3888888888889,
      "grad_norm": 0.00033405094291083515,
      "learning_rate": 1.1136876522407393e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7695824,
      "step": 27485
    },
    {
      "epoch": 305.44444444444446,
      "grad_norm": 2.849012344086077e-05,
      "learning_rate": 1.1128707810094985e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7697216,
      "step": 27490
    },
    {
      "epoch": 305.5,
      "grad_norm": 0.00021487107733264565,
      "learning_rate": 1.1120541236910157e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7698608,
      "step": 27495
    },
    {
      "epoch": 305.55555555555554,
      "grad_norm": 6.304743146756664e-05,
      "learning_rate": 1.111237680411229e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7699968,
      "step": 27500
    },
    {
      "epoch": 305.6111111111111,
      "grad_norm": 3.0814986530458555e-05,
      "learning_rate": 1.1104214512960433e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7701360,
      "step": 27505
    },
    {
      "epoch": 305.6666666666667,
      "grad_norm": 0.0006370891351252794,
      "learning_rate": 1.1096054364713327e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7702752,
      "step": 27510
    },
    {
      "epoch": 305.72222222222223,
      "grad_norm": 0.0001348406367469579,
      "learning_rate": 1.1087896360629371e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7704128,
      "step": 27515
    },
    {
      "epoch": 305.77777777777777,
      "grad_norm": 0.00010153440234716982,
      "learning_rate": 1.107974050196662e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7705520,
      "step": 27520
    },
    {
      "epoch": 305.8333333333333,
      "grad_norm": 0.0006301813991740346,
      "learning_rate": 1.1071586789982816e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7706928,
      "step": 27525
    },
    {
      "epoch": 305.8888888888889,
      "grad_norm": 0.0003621359937824309,
      "learning_rate": 1.1063435225935373e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7708336,
      "step": 27530
    },
    {
      "epoch": 305.94444444444446,
      "grad_norm": 0.00018141661712434143,
      "learning_rate": 1.1055285811081348e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7709728,
      "step": 27535
    },
    {
      "epoch": 306.0,
      "grad_norm": 6.635152385570109e-05,
      "learning_rate": 1.1047138546677499e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7711104,
      "step": 27540
    },
    {
      "epoch": 306.05555555555554,
      "grad_norm": 0.00032804239890538156,
      "learning_rate": 1.1038993433980219e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7712512,
      "step": 27545
    },
    {
      "epoch": 306.1111111111111,
      "grad_norm": 7.609315071022138e-05,
      "learning_rate": 1.1030850474245597e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7713936,
      "step": 27550
    },
    {
      "epoch": 306.1666666666667,
      "grad_norm": 3.7796289689140394e-05,
      "learning_rate": 1.102270966872939e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7715312,
      "step": 27555
    },
    {
      "epoch": 306.22222222222223,
      "grad_norm": 0.008743722923099995,
      "learning_rate": 1.1014571018687e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7716656,
      "step": 27560
    },
    {
      "epoch": 306.27777777777777,
      "grad_norm": 0.0011517725652083755,
      "learning_rate": 1.1006434525373502e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7718064,
      "step": 27565
    },
    {
      "epoch": 306.3333333333333,
      "grad_norm": 0.020929794758558273,
      "learning_rate": 1.0998300190043664e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7719456,
      "step": 27570
    },
    {
      "epoch": 306.3888888888889,
      "grad_norm": 0.00012833133223466575,
      "learning_rate": 1.0990168013951882e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7720864,
      "step": 27575
    },
    {
      "epoch": 306.44444444444446,
      "grad_norm": 9.517595754005015e-05,
      "learning_rate": 1.0982037998352263e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7722240,
      "step": 27580
    },
    {
      "epoch": 306.5,
      "grad_norm": 0.000311037729261443,
      "learning_rate": 1.0973910144498534e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7723648,
      "step": 27585
    },
    {
      "epoch": 306.55555555555554,
      "grad_norm": 0.00036231486592441797,
      "learning_rate": 1.0965784453644123e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7725040,
      "step": 27590
    },
    {
      "epoch": 306.6111111111111,
      "grad_norm": 0.00029141988488845527,
      "learning_rate": 1.0957660927042127e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7726400,
      "step": 27595
    },
    {
      "epoch": 306.6666666666667,
      "grad_norm": 0.00016282890283036977,
      "learning_rate": 1.094953956594527e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7727808,
      "step": 27600
    },
    {
      "epoch": 306.6666666666667,
      "eval_loss": 0.8962135314941406,
      "eval_runtime": 1.2393,
      "eval_samples_per_second": 32.277,
      "eval_steps_per_second": 16.138,
      "num_input_tokens_seen": 7727808,
      "step": 27600
    },
    {
      "epoch": 306.72222222222223,
      "grad_norm": 0.001801114296540618,
      "learning_rate": 1.0941420371605981e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7729200,
      "step": 27605
    },
    {
      "epoch": 306.77777777777777,
      "grad_norm": 4.045546302222647e-05,
      "learning_rate": 1.0933303345276354e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7730640,
      "step": 27610
    },
    {
      "epoch": 306.8333333333333,
      "grad_norm": 0.0001410128315910697,
      "learning_rate": 1.0925188488208112e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7732016,
      "step": 27615
    },
    {
      "epoch": 306.8888888888889,
      "grad_norm": 0.0006790750776417553,
      "learning_rate": 1.0917075801652694e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7733440,
      "step": 27620
    },
    {
      "epoch": 306.94444444444446,
      "grad_norm": 8.040993998292834e-05,
      "learning_rate": 1.0908965286861151e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7734832,
      "step": 27625
    },
    {
      "epoch": 307.0,
      "grad_norm": 0.0007850764668546617,
      "learning_rate": 1.090085694508425e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7736224,
      "step": 27630
    },
    {
      "epoch": 307.05555555555554,
      "grad_norm": 0.00017799524357542396,
      "learning_rate": 1.089275077757238e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7737616,
      "step": 27635
    },
    {
      "epoch": 307.1111111111111,
      "grad_norm": 5.973679435555823e-05,
      "learning_rate": 1.0884646785575633e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7739040,
      "step": 27640
    },
    {
      "epoch": 307.1666666666667,
      "grad_norm": 0.0002570217184256762,
      "learning_rate": 1.0876544970343728e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7740464,
      "step": 27645
    },
    {
      "epoch": 307.22222222222223,
      "grad_norm": 9.915742703014985e-05,
      "learning_rate": 1.0868445333126082e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7741872,
      "step": 27650
    },
    {
      "epoch": 307.27777777777777,
      "grad_norm": 9.870691428659484e-05,
      "learning_rate": 1.0860347875171745e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7743264,
      "step": 27655
    },
    {
      "epoch": 307.3333333333333,
      "grad_norm": 0.00016311623039655387,
      "learning_rate": 1.0852252597729465e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7744672,
      "step": 27660
    },
    {
      "epoch": 307.3888888888889,
      "grad_norm": 0.00040020080632530153,
      "learning_rate": 1.0844159502047615e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7746048,
      "step": 27665
    },
    {
      "epoch": 307.44444444444446,
      "grad_norm": 4.260494461050257e-05,
      "learning_rate": 1.0836068589374265e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7747456,
      "step": 27670
    },
    {
      "epoch": 307.5,
      "grad_norm": 0.0005975241074338555,
      "learning_rate": 1.0827979860957144e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7748864,
      "step": 27675
    },
    {
      "epoch": 307.55555555555554,
      "grad_norm": 0.0005120302084833384,
      "learning_rate": 1.0819893318043615e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7750272,
      "step": 27680
    },
    {
      "epoch": 307.6111111111111,
      "grad_norm": 0.00034679094096645713,
      "learning_rate": 1.0811808961880734e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7751616,
      "step": 27685
    },
    {
      "epoch": 307.6666666666667,
      "grad_norm": 0.0003599826595745981,
      "learning_rate": 1.080372679371522e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7753024,
      "step": 27690
    },
    {
      "epoch": 307.72222222222223,
      "grad_norm": 3.867434134008363e-05,
      "learning_rate": 1.0795646814793428e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7754400,
      "step": 27695
    },
    {
      "epoch": 307.77777777777777,
      "grad_norm": 0.001123861875385046,
      "learning_rate": 1.078756902636141e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7755840,
      "step": 27700
    },
    {
      "epoch": 307.8333333333333,
      "grad_norm": 5.358320049708709e-05,
      "learning_rate": 1.077949342966485e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7757248,
      "step": 27705
    },
    {
      "epoch": 307.8888888888889,
      "grad_norm": 0.00014618290879298002,
      "learning_rate": 1.0771420025949103e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7758640,
      "step": 27710
    },
    {
      "epoch": 307.94444444444446,
      "grad_norm": 0.00920435506850481,
      "learning_rate": 1.0763348816459204e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7759968,
      "step": 27715
    },
    {
      "epoch": 308.0,
      "grad_norm": 0.0003584905352909118,
      "learning_rate": 1.0755279802439816e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7761408,
      "step": 27720
    },
    {
      "epoch": 308.05555555555554,
      "grad_norm": 0.0014296548906713724,
      "learning_rate": 1.0747212985135293e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7762816,
      "step": 27725
    },
    {
      "epoch": 308.1111111111111,
      "grad_norm": 6.623641820624471e-05,
      "learning_rate": 1.073914836578965e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7764208,
      "step": 27730
    },
    {
      "epoch": 308.1666666666667,
      "grad_norm": 0.00025705661391839385,
      "learning_rate": 1.0731085945646529e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7765584,
      "step": 27735
    },
    {
      "epoch": 308.22222222222223,
      "grad_norm": 4.242975774104707e-05,
      "learning_rate": 1.0723025725949285e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7766976,
      "step": 27740
    },
    {
      "epoch": 308.27777777777777,
      "grad_norm": 5.3173076594248414e-05,
      "learning_rate": 1.0714967707940875e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7768336,
      "step": 27745
    },
    {
      "epoch": 308.3333333333333,
      "grad_norm": 0.0001461070351069793,
      "learning_rate": 1.0706911892863963e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7769776,
      "step": 27750
    },
    {
      "epoch": 308.3888888888889,
      "grad_norm": 0.0004915533936582506,
      "learning_rate": 1.0698858281960866e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7771168,
      "step": 27755
    },
    {
      "epoch": 308.44444444444446,
      "grad_norm": 4.222615825710818e-05,
      "learning_rate": 1.069080687647353e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7772544,
      "step": 27760
    },
    {
      "epoch": 308.5,
      "grad_norm": 0.00011179553985130042,
      "learning_rate": 1.0682757677643596e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7773984,
      "step": 27765
    },
    {
      "epoch": 308.55555555555554,
      "grad_norm": 0.0002323411899851635,
      "learning_rate": 1.0674710686712359e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7775408,
      "step": 27770
    },
    {
      "epoch": 308.6111111111111,
      "grad_norm": 0.0005681578768417239,
      "learning_rate": 1.0666665904920756e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7776784,
      "step": 27775
    },
    {
      "epoch": 308.6666666666667,
      "grad_norm": 5.9294008678989485e-05,
      "learning_rate": 1.0658623333509385e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7778192,
      "step": 27780
    },
    {
      "epoch": 308.72222222222223,
      "grad_norm": 0.00012429521302692592,
      "learning_rate": 1.0650582973718532e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7779632,
      "step": 27785
    },
    {
      "epoch": 308.77777777777777,
      "grad_norm": 0.00020953304192516953,
      "learning_rate": 1.0642544826788098e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7781008,
      "step": 27790
    },
    {
      "epoch": 308.8333333333333,
      "grad_norm": 5.444104681373574e-05,
      "learning_rate": 1.063450889395769e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7782384,
      "step": 27795
    },
    {
      "epoch": 308.8888888888889,
      "grad_norm": 4.65339508082252e-05,
      "learning_rate": 1.062647517646653e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7783744,
      "step": 27800
    },
    {
      "epoch": 308.8888888888889,
      "eval_loss": 0.8646701574325562,
      "eval_runtime": 1.2546,
      "eval_samples_per_second": 31.882,
      "eval_steps_per_second": 15.941,
      "num_input_tokens_seen": 7783744,
      "step": 27800
    },
    {
      "epoch": 308.94444444444446,
      "grad_norm": 0.006628353614360094,
      "learning_rate": 1.0618443675553527e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7785152,
      "step": 27805
    },
    {
      "epoch": 309.0,
      "grad_norm": 0.00012300204252824187,
      "learning_rate": 1.0610414392457247e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7786560,
      "step": 27810
    },
    {
      "epoch": 309.05555555555554,
      "grad_norm": 0.00011058952077291906,
      "learning_rate": 1.0602387328415888e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7787952,
      "step": 27815
    },
    {
      "epoch": 309.1111111111111,
      "grad_norm": 0.004208895843476057,
      "learning_rate": 1.0594362484667347e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7789376,
      "step": 27820
    },
    {
      "epoch": 309.1666666666667,
      "grad_norm": 0.00015174405416473746,
      "learning_rate": 1.0586339862449132e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7790800,
      "step": 27825
    },
    {
      "epoch": 309.22222222222223,
      "grad_norm": 8.852131577441469e-05,
      "learning_rate": 1.0578319462998445e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7792192,
      "step": 27830
    },
    {
      "epoch": 309.27777777777777,
      "grad_norm": 8.073874050751328e-05,
      "learning_rate": 1.057030128755214e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7793584,
      "step": 27835
    },
    {
      "epoch": 309.3333333333333,
      "grad_norm": 0.00012578642053995281,
      "learning_rate": 1.0562285337346703e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7795024,
      "step": 27840
    },
    {
      "epoch": 309.3888888888889,
      "grad_norm": 2.173181746911723e-05,
      "learning_rate": 1.0554271613618308e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7796432,
      "step": 27845
    },
    {
      "epoch": 309.44444444444446,
      "grad_norm": 6.270238372962922e-05,
      "learning_rate": 1.054626011760276e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7797824,
      "step": 27850
    },
    {
      "epoch": 309.5,
      "grad_norm": 7.779366569593549e-05,
      "learning_rate": 1.0538250850535549e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7799200,
      "step": 27855
    },
    {
      "epoch": 309.55555555555554,
      "grad_norm": 6.834678788436577e-05,
      "learning_rate": 1.0530243813651794e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7800576,
      "step": 27860
    },
    {
      "epoch": 309.6111111111111,
      "grad_norm": 0.0005596845294348896,
      "learning_rate": 1.0522239008186271e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7801936,
      "step": 27865
    },
    {
      "epoch": 309.6666666666667,
      "grad_norm": 0.002016291720792651,
      "learning_rate": 1.0514236435373434e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7803312,
      "step": 27870
    },
    {
      "epoch": 309.72222222222223,
      "grad_norm": 0.006660241633653641,
      "learning_rate": 1.0506236096447386e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7804736,
      "step": 27875
    },
    {
      "epoch": 309.77777777777777,
      "grad_norm": 6.1003414884908125e-05,
      "learning_rate": 1.049823799264186e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7806144,
      "step": 27880
    },
    {
      "epoch": 309.8333333333333,
      "grad_norm": 0.0010895782615989447,
      "learning_rate": 1.049024212519028e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7807536,
      "step": 27885
    },
    {
      "epoch": 309.8888888888889,
      "grad_norm": 0.0010420846519991755,
      "learning_rate": 1.0482248495325713e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7808944,
      "step": 27890
    },
    {
      "epoch": 309.94444444444446,
      "grad_norm": 0.003812109585851431,
      "learning_rate": 1.047425710428086e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7810336,
      "step": 27895
    },
    {
      "epoch": 310.0,
      "grad_norm": 0.0006950965034775436,
      "learning_rate": 1.0466267953288114e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7811744,
      "step": 27900
    },
    {
      "epoch": 310.05555555555554,
      "grad_norm": 8.57334744068794e-05,
      "learning_rate": 1.0458281043579482e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7813104,
      "step": 27905
    },
    {
      "epoch": 310.1111111111111,
      "grad_norm": 0.00020503034465946257,
      "learning_rate": 1.0450296376386657e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7814496,
      "step": 27910
    },
    {
      "epoch": 310.1666666666667,
      "grad_norm": 0.0003106475924141705,
      "learning_rate": 1.044231395294098e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7815888,
      "step": 27915
    },
    {
      "epoch": 310.22222222222223,
      "grad_norm": 0.00016512180445715785,
      "learning_rate": 1.0434333774473435e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7817328,
      "step": 27920
    },
    {
      "epoch": 310.27777777777777,
      "grad_norm": 0.0016340259462594986,
      "learning_rate": 1.0426355842214657e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7818752,
      "step": 27925
    },
    {
      "epoch": 310.3333333333333,
      "grad_norm": 0.0005220469320192933,
      "learning_rate": 1.0418380157394963e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7820176,
      "step": 27930
    },
    {
      "epoch": 310.3888888888889,
      "grad_norm": 0.0003603968652896583,
      "learning_rate": 1.0410406721244281e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7821568,
      "step": 27935
    },
    {
      "epoch": 310.44444444444446,
      "grad_norm": 0.0008228348451666534,
      "learning_rate": 1.0402435534992238e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7823024,
      "step": 27940
    },
    {
      "epoch": 310.5,
      "grad_norm": 0.00012130910181440413,
      "learning_rate": 1.0394466599868071e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7824400,
      "step": 27945
    },
    {
      "epoch": 310.55555555555554,
      "grad_norm": 0.00026836086180992424,
      "learning_rate": 1.0386499917100697e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7825792,
      "step": 27950
    },
    {
      "epoch": 310.6111111111111,
      "grad_norm": 6.165041850181296e-05,
      "learning_rate": 1.0378535487918692e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7827168,
      "step": 27955
    },
    {
      "epoch": 310.6666666666667,
      "grad_norm": 0.000137569586513564,
      "learning_rate": 1.037057331355025e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7828624,
      "step": 27960
    },
    {
      "epoch": 310.72222222222223,
      "grad_norm": 0.0008650908712297678,
      "learning_rate": 1.0362613395223247e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7830000,
      "step": 27965
    },
    {
      "epoch": 310.77777777777777,
      "grad_norm": 0.0003429171920288354,
      "learning_rate": 1.0354655734165212e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7831424,
      "step": 27970
    },
    {
      "epoch": 310.8333333333333,
      "grad_norm": 0.00012089763185940683,
      "learning_rate": 1.03467003316033e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7832848,
      "step": 27975
    },
    {
      "epoch": 310.8888888888889,
      "grad_norm": 0.0006959705497138202,
      "learning_rate": 1.033874718876435e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7834224,
      "step": 27980
    },
    {
      "epoch": 310.94444444444446,
      "grad_norm": 0.001117055770009756,
      "learning_rate": 1.0330796306874818e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7835600,
      "step": 27985
    },
    {
      "epoch": 311.0,
      "grad_norm": 5.0046823162119836e-05,
      "learning_rate": 1.032284768716085e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7837008,
      "step": 27990
    },
    {
      "epoch": 311.05555555555554,
      "grad_norm": 0.00012985938519705087,
      "learning_rate": 1.0314901330848206e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7838416,
      "step": 27995
    },
    {
      "epoch": 311.1111111111111,
      "grad_norm": 0.0002546800533309579,
      "learning_rate": 1.030695723916233e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7839808,
      "step": 28000
    },
    {
      "epoch": 311.1111111111111,
      "eval_loss": 0.8644084930419922,
      "eval_runtime": 1.2359,
      "eval_samples_per_second": 32.365,
      "eval_steps_per_second": 16.183,
      "num_input_tokens_seen": 7839808,
      "step": 28000
    },
    {
      "epoch": 311.1666666666667,
      "grad_norm": 5.217585930950008e-05,
      "learning_rate": 1.0299015413328289e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7841184,
      "step": 28005
    },
    {
      "epoch": 311.22222222222223,
      "grad_norm": 0.00014681753236800432,
      "learning_rate": 1.0291075854570809e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7842576,
      "step": 28010
    },
    {
      "epoch": 311.27777777777777,
      "grad_norm": 0.000192346575204283,
      "learning_rate": 1.0283138564114275e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7844016,
      "step": 28015
    },
    {
      "epoch": 311.3333333333333,
      "grad_norm": 0.00013563421089202166,
      "learning_rate": 1.027520354318273e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7845376,
      "step": 28020
    },
    {
      "epoch": 311.3888888888889,
      "grad_norm": 7.847219967516139e-05,
      "learning_rate": 1.0267270792999828e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7846768,
      "step": 28025
    },
    {
      "epoch": 311.44444444444446,
      "grad_norm": 0.0024123629555106163,
      "learning_rate": 1.0259340314788919e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7848192,
      "step": 28030
    },
    {
      "epoch": 311.5,
      "grad_norm": 3.181681677233428e-05,
      "learning_rate": 1.0251412109772979e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7849600,
      "step": 28035
    },
    {
      "epoch": 311.55555555555554,
      "grad_norm": 0.00017077199299819767,
      "learning_rate": 1.0243486179174627e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7851024,
      "step": 28040
    },
    {
      "epoch": 311.6111111111111,
      "grad_norm": 0.00039772081072442234,
      "learning_rate": 1.0235562524216158e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7852432,
      "step": 28045
    },
    {
      "epoch": 311.6666666666667,
      "grad_norm": 0.0010253245709463954,
      "learning_rate": 1.022764114611948e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7853824,
      "step": 28050
    },
    {
      "epoch": 311.72222222222223,
      "grad_norm": 0.0033072540536522865,
      "learning_rate": 1.0219722046106178e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7855216,
      "step": 28055
    },
    {
      "epoch": 311.77777777777777,
      "grad_norm": 0.00022403226466849446,
      "learning_rate": 1.0211805225397486e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7856624,
      "step": 28060
    },
    {
      "epoch": 311.8333333333333,
      "grad_norm": 0.00022198962687980384,
      "learning_rate": 1.020389068521426e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7858016,
      "step": 28065
    },
    {
      "epoch": 311.8888888888889,
      "grad_norm": 0.00016666215378791094,
      "learning_rate": 1.0195978426777039e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7859424,
      "step": 28070
    },
    {
      "epoch": 311.94444444444446,
      "grad_norm": 0.00013592852337751538,
      "learning_rate": 1.0188068451305982e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7860800,
      "step": 28075
    },
    {
      "epoch": 312.0,
      "grad_norm": 7.548820576630533e-05,
      "learning_rate": 1.0180160760020902e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7862176,
      "step": 28080
    },
    {
      "epoch": 312.05555555555554,
      "grad_norm": 3.787907917285338e-05,
      "learning_rate": 1.0172255354141278e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7863568,
      "step": 28085
    },
    {
      "epoch": 312.1111111111111,
      "grad_norm": 0.00011137204273836687,
      "learning_rate": 1.0164352234886205e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7864944,
      "step": 28090
    },
    {
      "epoch": 312.1666666666667,
      "grad_norm": 2.5613942852942273e-05,
      "learning_rate": 1.0156451403474454e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7866336,
      "step": 28095
    },
    {
      "epoch": 312.22222222222223,
      "grad_norm": 0.00022911651467438787,
      "learning_rate": 1.0148552861124443e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7867744,
      "step": 28100
    },
    {
      "epoch": 312.27777777777777,
      "grad_norm": 1.946094198501669e-05,
      "learning_rate": 1.0140656609054205e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7869152,
      "step": 28105
    },
    {
      "epoch": 312.3333333333333,
      "grad_norm": 0.00023648592468816787,
      "learning_rate": 1.0132762648481455e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7870560,
      "step": 28110
    },
    {
      "epoch": 312.3888888888889,
      "grad_norm": 0.0018463247688487172,
      "learning_rate": 1.0124870980623543e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7871968,
      "step": 28115
    },
    {
      "epoch": 312.44444444444446,
      "grad_norm": 0.00026916502974927425,
      "learning_rate": 1.0116981606697453e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7873376,
      "step": 28120
    },
    {
      "epoch": 312.5,
      "grad_norm": 7.303078018594533e-05,
      "learning_rate": 1.0109094527919838e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7874736,
      "step": 28125
    },
    {
      "epoch": 312.55555555555554,
      "grad_norm": 4.909292692900635e-05,
      "learning_rate": 1.010120974550697e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7876096,
      "step": 28130
    },
    {
      "epoch": 312.6111111111111,
      "grad_norm": 2.8164002287667245e-05,
      "learning_rate": 1.0093327260674795e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7877536,
      "step": 28135
    },
    {
      "epoch": 312.6666666666667,
      "grad_norm": 0.0002030634059337899,
      "learning_rate": 1.0085447074638878e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7878928,
      "step": 28140
    },
    {
      "epoch": 312.72222222222223,
      "grad_norm": 0.000606378132943064,
      "learning_rate": 1.0077569188614461e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7880320,
      "step": 28145
    },
    {
      "epoch": 312.77777777777777,
      "grad_norm": 0.00018659849592950195,
      "learning_rate": 1.0069693603816393e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7881760,
      "step": 28150
    },
    {
      "epoch": 312.8333333333333,
      "grad_norm": 4.517859997577034e-05,
      "learning_rate": 1.0061820321459204e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7883168,
      "step": 28155
    },
    {
      "epoch": 312.8888888888889,
      "grad_norm": 0.00010050314449472353,
      "learning_rate": 1.0053949342757038e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7884592,
      "step": 28160
    },
    {
      "epoch": 312.94444444444446,
      "grad_norm": 6.241348455660045e-05,
      "learning_rate": 1.0046080668923717e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7886032,
      "step": 28165
    },
    {
      "epoch": 313.0,
      "grad_norm": 0.0001362825569231063,
      "learning_rate": 1.003821430117267e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7887424,
      "step": 28170
    },
    {
      "epoch": 313.05555555555554,
      "grad_norm": 0.00019854317361023277,
      "learning_rate": 1.0030350240716999e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7888816,
      "step": 28175
    },
    {
      "epoch": 313.1111111111111,
      "grad_norm": 0.00020217255223542452,
      "learning_rate": 1.0022488488769449e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7890192,
      "step": 28180
    },
    {
      "epoch": 313.1666666666667,
      "grad_norm": 0.000463957927422598,
      "learning_rate": 1.0014629046542387e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7891632,
      "step": 28185
    },
    {
      "epoch": 313.22222222222223,
      "grad_norm": 0.00027654494624584913,
      "learning_rate": 1.0006771915247842e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 7893056,
      "step": 28190
    },
    {
      "epoch": 313.27777777777777,
      "grad_norm": 6.600286724278703e-05,
      "learning_rate": 9.998917096097495e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7894448,
      "step": 28195
    },
    {
      "epoch": 313.3333333333333,
      "grad_norm": 0.00030666461680084467,
      "learning_rate": 9.991064590302638e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7895872,
      "step": 28200
    },
    {
      "epoch": 313.3333333333333,
      "eval_loss": 0.8795791864395142,
      "eval_runtime": 1.2339,
      "eval_samples_per_second": 32.416,
      "eval_steps_per_second": 16.208,
      "num_input_tokens_seen": 7895872,
      "step": 28200
    },
    {
      "epoch": 313.3888888888889,
      "grad_norm": 0.00024228861730080098,
      "learning_rate": 9.983214399074241e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7897264,
      "step": 28205
    },
    {
      "epoch": 313.44444444444446,
      "grad_norm": 0.0003104136267211288,
      "learning_rate": 9.975366523622893e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7898656,
      "step": 28210
    },
    {
      "epoch": 313.5,
      "grad_norm": 4.401220940053463e-05,
      "learning_rate": 9.967520965158841e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7900048,
      "step": 28215
    },
    {
      "epoch": 313.55555555555554,
      "grad_norm": 6.703258986817673e-05,
      "learning_rate": 9.95967772489197e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7901472,
      "step": 28220
    },
    {
      "epoch": 313.6111111111111,
      "grad_norm": 0.00016926646640058607,
      "learning_rate": 9.951836804031794e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7902880,
      "step": 28225
    },
    {
      "epoch": 313.6666666666667,
      "grad_norm": 9.71919289440848e-05,
      "learning_rate": 9.943998203787489e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7904288,
      "step": 28230
    },
    {
      "epoch": 313.72222222222223,
      "grad_norm": 2.8728147299261764e-05,
      "learning_rate": 9.936161925367874e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7905696,
      "step": 28235
    },
    {
      "epoch": 313.77777777777777,
      "grad_norm": 0.0006911589880473912,
      "learning_rate": 9.928327969981386e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7907072,
      "step": 28240
    },
    {
      "epoch": 313.8333333333333,
      "grad_norm": 0.006218441762030125,
      "learning_rate": 9.920496338836135e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7908480,
      "step": 28245
    },
    {
      "epoch": 313.8888888888889,
      "grad_norm": 6.73230315442197e-05,
      "learning_rate": 9.912667033139844e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7909824,
      "step": 28250
    },
    {
      "epoch": 313.94444444444446,
      "grad_norm": 0.0002714817819651216,
      "learning_rate": 9.904840054099893e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7911184,
      "step": 28255
    },
    {
      "epoch": 314.0,
      "grad_norm": 0.00024071747611742467,
      "learning_rate": 9.897015402923312e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7912528,
      "step": 28260
    },
    {
      "epoch": 314.05555555555554,
      "grad_norm": 0.002855984028428793,
      "learning_rate": 9.889193080816744e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7913936,
      "step": 28265
    },
    {
      "epoch": 314.1111111111111,
      "grad_norm": 4.471023567020893e-05,
      "learning_rate": 9.881373088986498e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7915328,
      "step": 28270
    },
    {
      "epoch": 314.1666666666667,
      "grad_norm": 3.885240585077554e-05,
      "learning_rate": 9.873555428638523e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7916688,
      "step": 28275
    },
    {
      "epoch": 314.22222222222223,
      "grad_norm": 1.9520679416018538e-05,
      "learning_rate": 9.865740100978383e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7918128,
      "step": 28280
    },
    {
      "epoch": 314.27777777777777,
      "grad_norm": 8.899514068616554e-05,
      "learning_rate": 9.857927107211315e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7919536,
      "step": 28285
    },
    {
      "epoch": 314.3333333333333,
      "grad_norm": 0.000411531247664243,
      "learning_rate": 9.850116448542177e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7920896,
      "step": 28290
    },
    {
      "epoch": 314.3888888888889,
      "grad_norm": 0.00046503744670189917,
      "learning_rate": 9.842308126175457e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7922304,
      "step": 28295
    },
    {
      "epoch": 314.44444444444446,
      "grad_norm": 0.00011299683683319017,
      "learning_rate": 9.834502141315315e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7923744,
      "step": 28300
    },
    {
      "epoch": 314.5,
      "grad_norm": 0.0007465693051926792,
      "learning_rate": 9.82669849516552e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7925152,
      "step": 28305
    },
    {
      "epoch": 314.55555555555554,
      "grad_norm": 0.0002383781102253124,
      "learning_rate": 9.818897188929493e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7926544,
      "step": 28310
    },
    {
      "epoch": 314.6111111111111,
      "grad_norm": 0.0002545380557421595,
      "learning_rate": 9.811098223810309e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7927936,
      "step": 28315
    },
    {
      "epoch": 314.6666666666667,
      "grad_norm": 0.001689306111074984,
      "learning_rate": 9.803301601010641e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7929312,
      "step": 28320
    },
    {
      "epoch": 314.72222222222223,
      "grad_norm": 0.00032172229839488864,
      "learning_rate": 9.795507321732853e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7930688,
      "step": 28325
    },
    {
      "epoch": 314.77777777777777,
      "grad_norm": 3.711332828970626e-05,
      "learning_rate": 9.787715387178898e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7932144,
      "step": 28330
    },
    {
      "epoch": 314.8333333333333,
      "grad_norm": 6.20208156760782e-05,
      "learning_rate": 9.779925798550399e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7933536,
      "step": 28335
    },
    {
      "epoch": 314.8888888888889,
      "grad_norm": 0.0002751038409769535,
      "learning_rate": 9.772138557048619e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7934896,
      "step": 28340
    },
    {
      "epoch": 314.94444444444446,
      "grad_norm": 0.000808602781035006,
      "learning_rate": 9.764353663874426e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7936336,
      "step": 28345
    },
    {
      "epoch": 315.0,
      "grad_norm": 0.00019529862038325518,
      "learning_rate": 9.756571120228375e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7937744,
      "step": 28350
    },
    {
      "epoch": 315.05555555555554,
      "grad_norm": 0.00010285446478519589,
      "learning_rate": 9.748790927310605e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7939104,
      "step": 28355
    },
    {
      "epoch": 315.1111111111111,
      "grad_norm": 0.0007378001464530826,
      "learning_rate": 9.741013086320946e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7940496,
      "step": 28360
    },
    {
      "epoch": 315.1666666666667,
      "grad_norm": 0.002077667275443673,
      "learning_rate": 9.733237598458821e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7941888,
      "step": 28365
    },
    {
      "epoch": 315.22222222222223,
      "grad_norm": 9.07307694433257e-05,
      "learning_rate": 9.725464464923308e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7943280,
      "step": 28370
    },
    {
      "epoch": 315.27777777777777,
      "grad_norm": 4.936890400131233e-05,
      "learning_rate": 9.717693686913123e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7944688,
      "step": 28375
    },
    {
      "epoch": 315.3333333333333,
      "grad_norm": 4.034728044643998e-05,
      "learning_rate": 9.709925265626632e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7946064,
      "step": 28380
    },
    {
      "epoch": 315.3888888888889,
      "grad_norm": 0.0002146427723346278,
      "learning_rate": 9.702159202261801e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7947472,
      "step": 28385
    },
    {
      "epoch": 315.44444444444446,
      "grad_norm": 0.00011369588901288807,
      "learning_rate": 9.694395498016268e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7948880,
      "step": 28390
    },
    {
      "epoch": 315.5,
      "grad_norm": 3.745682261069305e-05,
      "learning_rate": 9.686634154087298e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7950288,
      "step": 28395
    },
    {
      "epoch": 315.55555555555554,
      "grad_norm": 0.00013789486547466367,
      "learning_rate": 9.678875171671776e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7951664,
      "step": 28400
    },
    {
      "epoch": 315.55555555555554,
      "eval_loss": 0.8324952125549316,
      "eval_runtime": 1.2497,
      "eval_samples_per_second": 32.008,
      "eval_steps_per_second": 16.004,
      "num_input_tokens_seen": 7951664,
      "step": 28400
    },
    {
      "epoch": 315.6111111111111,
      "grad_norm": 0.0003385216696187854,
      "learning_rate": 9.671118551966246e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7953120,
      "step": 28405
    },
    {
      "epoch": 315.6666666666667,
      "grad_norm": 0.0002556012768764049,
      "learning_rate": 9.66336429616686e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7954544,
      "step": 28410
    },
    {
      "epoch": 315.72222222222223,
      "grad_norm": 0.0001596042566234246,
      "learning_rate": 9.655612405469436e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7955936,
      "step": 28415
    },
    {
      "epoch": 315.77777777777777,
      "grad_norm": 0.0004977740463800728,
      "learning_rate": 9.647862881069413e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7957328,
      "step": 28420
    },
    {
      "epoch": 315.8333333333333,
      "grad_norm": 0.0001571773609612137,
      "learning_rate": 9.640115724161855e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7958736,
      "step": 28425
    },
    {
      "epoch": 315.8888888888889,
      "grad_norm": 0.00043135491432622075,
      "learning_rate": 9.632370935941483e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7960128,
      "step": 28430
    },
    {
      "epoch": 315.94444444444446,
      "grad_norm": 0.0005192215321585536,
      "learning_rate": 9.624628517602634e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7961536,
      "step": 28435
    },
    {
      "epoch": 316.0,
      "grad_norm": 0.00024108339857775718,
      "learning_rate": 9.61688847033928e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7962928,
      "step": 28440
    },
    {
      "epoch": 316.05555555555554,
      "grad_norm": 0.00035793037386611104,
      "learning_rate": 9.609150795345051e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7964336,
      "step": 28445
    },
    {
      "epoch": 316.1111111111111,
      "grad_norm": 5.692704507964663e-05,
      "learning_rate": 9.601415493813171e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7965728,
      "step": 28450
    },
    {
      "epoch": 316.1666666666667,
      "grad_norm": 0.00031262755510397255,
      "learning_rate": 9.593682566936533e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7967152,
      "step": 28455
    },
    {
      "epoch": 316.22222222222223,
      "grad_norm": 0.00017609259521123022,
      "learning_rate": 9.58595201590766e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7968512,
      "step": 28460
    },
    {
      "epoch": 316.27777777777777,
      "grad_norm": 0.0009288343135267496,
      "learning_rate": 9.578223841918681e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7969952,
      "step": 28465
    },
    {
      "epoch": 316.3333333333333,
      "grad_norm": 0.00033579207956790924,
      "learning_rate": 9.570498046161389e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7971392,
      "step": 28470
    },
    {
      "epoch": 316.3888888888889,
      "grad_norm": 0.0002904892899096012,
      "learning_rate": 9.562774629827206e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7972800,
      "step": 28475
    },
    {
      "epoch": 316.44444444444446,
      "grad_norm": 4.8248442908516154e-05,
      "learning_rate": 9.555053594107163e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7974208,
      "step": 28480
    },
    {
      "epoch": 316.5,
      "grad_norm": 6.487472273875028e-05,
      "learning_rate": 9.547334940191957e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7975584,
      "step": 28485
    },
    {
      "epoch": 316.55555555555554,
      "grad_norm": 5.4629250371363014e-05,
      "learning_rate": 9.539618669271886e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7977008,
      "step": 28490
    },
    {
      "epoch": 316.6111111111111,
      "grad_norm": 6.891554221510887e-05,
      "learning_rate": 9.531904782536904e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7978352,
      "step": 28495
    },
    {
      "epoch": 316.6666666666667,
      "grad_norm": 0.0003348141035530716,
      "learning_rate": 9.524193281176597e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7979744,
      "step": 28500
    },
    {
      "epoch": 316.72222222222223,
      "grad_norm": 3.875842594425194e-05,
      "learning_rate": 9.516484166380165e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7981120,
      "step": 28505
    },
    {
      "epoch": 316.77777777777777,
      "grad_norm": 0.00019480421906337142,
      "learning_rate": 9.508777439336447e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7982480,
      "step": 28510
    },
    {
      "epoch": 316.8333333333333,
      "grad_norm": 0.0002698124444577843,
      "learning_rate": 9.50107310123393e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7983872,
      "step": 28515
    },
    {
      "epoch": 316.8888888888889,
      "grad_norm": 0.00017410975124221295,
      "learning_rate": 9.493371153260702e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7985328,
      "step": 28520
    },
    {
      "epoch": 316.94444444444446,
      "grad_norm": 0.0001709544740151614,
      "learning_rate": 9.485671596604523e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7986704,
      "step": 28525
    },
    {
      "epoch": 317.0,
      "grad_norm": 0.00010332401870982721,
      "learning_rate": 9.477974432452738e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7988096,
      "step": 28530
    },
    {
      "epoch": 317.05555555555554,
      "grad_norm": 0.00013012647104915231,
      "learning_rate": 9.470279661992356e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7989472,
      "step": 28535
    },
    {
      "epoch": 317.1111111111111,
      "grad_norm": 3.8516769564012066e-05,
      "learning_rate": 9.462587286410021e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7990864,
      "step": 28540
    },
    {
      "epoch": 317.1666666666667,
      "grad_norm": 0.00021600838226731867,
      "learning_rate": 9.454897306891972e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7992224,
      "step": 28545
    },
    {
      "epoch": 317.22222222222223,
      "grad_norm": 0.002555015031248331,
      "learning_rate": 9.44720972462411e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7993632,
      "step": 28550
    },
    {
      "epoch": 317.27777777777777,
      "grad_norm": 0.00017200771253556013,
      "learning_rate": 9.439524540791964e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7995008,
      "step": 28555
    },
    {
      "epoch": 317.3333333333333,
      "grad_norm": 0.0009840555721893907,
      "learning_rate": 9.431841756580673e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7996432,
      "step": 28560
    },
    {
      "epoch": 317.3888888888889,
      "grad_norm": 0.00024121963360812515,
      "learning_rate": 9.42416137317503e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7997840,
      "step": 28565
    },
    {
      "epoch": 317.44444444444446,
      "grad_norm": 0.0003005914331879467,
      "learning_rate": 9.416483391759437e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 7999200,
      "step": 28570
    },
    {
      "epoch": 317.5,
      "grad_norm": 0.0004911914584226906,
      "learning_rate": 9.408807813517945e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8000560,
      "step": 28575
    },
    {
      "epoch": 317.55555555555554,
      "grad_norm": 0.0005538039840757847,
      "learning_rate": 9.401134639634221e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8002032,
      "step": 28580
    },
    {
      "epoch": 317.6111111111111,
      "grad_norm": 0.000872497505042702,
      "learning_rate": 9.393463871291555e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8003424,
      "step": 28585
    },
    {
      "epoch": 317.6666666666667,
      "grad_norm": 0.0004760580195579678,
      "learning_rate": 9.385795509672881e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8004848,
      "step": 28590
    },
    {
      "epoch": 317.72222222222223,
      "grad_norm": 0.00010110755101777613,
      "learning_rate": 9.378129555960771e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8006320,
      "step": 28595
    },
    {
      "epoch": 317.77777777777777,
      "grad_norm": 2.914661672548391e-05,
      "learning_rate": 9.370466011337392e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8007744,
      "step": 28600
    },
    {
      "epoch": 317.77777777777777,
      "eval_loss": 0.8311160802841187,
      "eval_runtime": 1.2353,
      "eval_samples_per_second": 32.38,
      "eval_steps_per_second": 16.19,
      "num_input_tokens_seen": 8007744,
      "step": 28600
    },
    {
      "epoch": 317.8333333333333,
      "grad_norm": 0.00022568869462702423,
      "learning_rate": 9.362804876984573e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8009168,
      "step": 28605
    },
    {
      "epoch": 317.8888888888889,
      "grad_norm": 8.270692342193797e-05,
      "learning_rate": 9.355146154083747e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8010544,
      "step": 28610
    },
    {
      "epoch": 317.94444444444446,
      "grad_norm": 0.00021625393128488213,
      "learning_rate": 9.347489843815987e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8011936,
      "step": 28615
    },
    {
      "epoch": 318.0,
      "grad_norm": 0.0017876782221719623,
      "learning_rate": 9.339835947362002e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8013328,
      "step": 28620
    },
    {
      "epoch": 318.05555555555554,
      "grad_norm": 0.0003453014651313424,
      "learning_rate": 9.332184465902105e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8014736,
      "step": 28625
    },
    {
      "epoch": 318.1111111111111,
      "grad_norm": 0.0025812690146267414,
      "learning_rate": 9.324535400616266e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8016160,
      "step": 28630
    },
    {
      "epoch": 318.1666666666667,
      "grad_norm": 5.321120988810435e-05,
      "learning_rate": 9.31688875268405e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8017536,
      "step": 28635
    },
    {
      "epoch": 318.22222222222223,
      "grad_norm": 0.00012331116886343807,
      "learning_rate": 9.309244523284674e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8018928,
      "step": 28640
    },
    {
      "epoch": 318.27777777777777,
      "grad_norm": 0.000209713340154849,
      "learning_rate": 9.301602713596982e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8020368,
      "step": 28645
    },
    {
      "epoch": 318.3333333333333,
      "grad_norm": 0.0002000334206968546,
      "learning_rate": 9.293963324799432e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8021712,
      "step": 28650
    },
    {
      "epoch": 318.3888888888889,
      "grad_norm": 0.0001955529151018709,
      "learning_rate": 9.286326358070104e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8023104,
      "step": 28655
    },
    {
      "epoch": 318.44444444444446,
      "grad_norm": 0.0001339978480245918,
      "learning_rate": 9.278691814586729e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8024512,
      "step": 28660
    },
    {
      "epoch": 318.5,
      "grad_norm": 0.000746772566344589,
      "learning_rate": 9.271059695526635e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8025952,
      "step": 28665
    },
    {
      "epoch": 318.55555555555554,
      "grad_norm": 0.00024072059022728354,
      "learning_rate": 9.263430002066805e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8027312,
      "step": 28670
    },
    {
      "epoch": 318.6111111111111,
      "grad_norm": 0.00019050731498282403,
      "learning_rate": 9.25580273538382e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8028720,
      "step": 28675
    },
    {
      "epoch": 318.6666666666667,
      "grad_norm": 0.0007754908874630928,
      "learning_rate": 9.248177896653907e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8030160,
      "step": 28680
    },
    {
      "epoch": 318.72222222222223,
      "grad_norm": 0.00010202859266428277,
      "learning_rate": 9.240555487052918e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8031552,
      "step": 28685
    },
    {
      "epoch": 318.77777777777777,
      "grad_norm": 0.0004147550498601049,
      "learning_rate": 9.232935507756313e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8032944,
      "step": 28690
    },
    {
      "epoch": 318.8333333333333,
      "grad_norm": 0.0002458960807416588,
      "learning_rate": 9.225317959939193e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8034352,
      "step": 28695
    },
    {
      "epoch": 318.8888888888889,
      "grad_norm": 0.0003567827807273716,
      "learning_rate": 9.217702844776287e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8035808,
      "step": 28700
    },
    {
      "epoch": 318.94444444444446,
      "grad_norm": 0.00039386810385622084,
      "learning_rate": 9.210090163441929e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8037152,
      "step": 28705
    },
    {
      "epoch": 319.0,
      "grad_norm": 0.00026996323140338063,
      "learning_rate": 9.202479917110105e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8038512,
      "step": 28710
    },
    {
      "epoch": 319.05555555555554,
      "grad_norm": 6.234634929569438e-05,
      "learning_rate": 9.194872106954392e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8039888,
      "step": 28715
    },
    {
      "epoch": 319.1111111111111,
      "grad_norm": 4.6668454160680994e-05,
      "learning_rate": 9.187266734148029e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8041280,
      "step": 28720
    },
    {
      "epoch": 319.1666666666667,
      "grad_norm": 5.635102206724696e-05,
      "learning_rate": 9.179663799863849e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8042656,
      "step": 28725
    },
    {
      "epoch": 319.22222222222223,
      "grad_norm": 0.00014219518925528973,
      "learning_rate": 9.172063305274317e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8044048,
      "step": 28730
    },
    {
      "epoch": 319.27777777777777,
      "grad_norm": 0.005697907879948616,
      "learning_rate": 9.164465251551527e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8045440,
      "step": 28735
    },
    {
      "epoch": 319.3333333333333,
      "grad_norm": 5.036991569795646e-05,
      "learning_rate": 9.156869639867205e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8046848,
      "step": 28740
    },
    {
      "epoch": 319.3888888888889,
      "grad_norm": 0.00038481614319607615,
      "learning_rate": 9.149276471392677e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8048256,
      "step": 28745
    },
    {
      "epoch": 319.44444444444446,
      "grad_norm": 9.635250899009407e-05,
      "learning_rate": 9.141685747298914e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8049664,
      "step": 28750
    },
    {
      "epoch": 319.5,
      "grad_norm": 0.00011595179239520803,
      "learning_rate": 9.13409746875649e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8051056,
      "step": 28755
    },
    {
      "epoch": 319.55555555555554,
      "grad_norm": 5.68736249988433e-05,
      "learning_rate": 9.12651163693562e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8052480,
      "step": 28760
    },
    {
      "epoch": 319.6111111111111,
      "grad_norm": 6.03499538556207e-05,
      "learning_rate": 9.11892825300614e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8053856,
      "step": 28765
    },
    {
      "epoch": 319.6666666666667,
      "grad_norm": 0.00016155760386027396,
      "learning_rate": 9.111347318137491e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8055232,
      "step": 28770
    },
    {
      "epoch": 319.72222222222223,
      "grad_norm": 0.0003956368309445679,
      "learning_rate": 9.103768833498755e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8056624,
      "step": 28775
    },
    {
      "epoch": 319.77777777777777,
      "grad_norm": 0.00010050902346847579,
      "learning_rate": 9.096192800258639e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8057984,
      "step": 28780
    },
    {
      "epoch": 319.8333333333333,
      "grad_norm": 0.00027395455981604755,
      "learning_rate": 9.088619219585443e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8059392,
      "step": 28785
    },
    {
      "epoch": 319.8888888888889,
      "grad_norm": 8.006420830497518e-05,
      "learning_rate": 9.081048092647127e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8060800,
      "step": 28790
    },
    {
      "epoch": 319.94444444444446,
      "grad_norm": 0.0014082890702411532,
      "learning_rate": 9.073479420611245e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8062224,
      "step": 28795
    },
    {
      "epoch": 320.0,
      "grad_norm": 0.001571118482388556,
      "learning_rate": 9.065913204644974e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8063616,
      "step": 28800
    },
    {
      "epoch": 320.0,
      "eval_loss": 0.8026739358901978,
      "eval_runtime": 1.2505,
      "eval_samples_per_second": 31.986,
      "eval_steps_per_second": 15.993,
      "num_input_tokens_seen": 8063616,
      "step": 28800
    },
    {
      "epoch": 320.05555555555554,
      "grad_norm": 5.4415606427937746e-05,
      "learning_rate": 9.058349445915135e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8065024,
      "step": 28805
    },
    {
      "epoch": 320.1111111111111,
      "grad_norm": 0.0003530389803927392,
      "learning_rate": 9.050788145588138e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8066416,
      "step": 28810
    },
    {
      "epoch": 320.1666666666667,
      "grad_norm": 0.00036667168023996055,
      "learning_rate": 9.043229304830039e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8067856,
      "step": 28815
    },
    {
      "epoch": 320.22222222222223,
      "grad_norm": 0.0024374164640903473,
      "learning_rate": 9.035672924806515e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8069232,
      "step": 28820
    },
    {
      "epoch": 320.27777777777777,
      "grad_norm": 3.90001769119408e-05,
      "learning_rate": 9.028119006682839e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8070624,
      "step": 28825
    },
    {
      "epoch": 320.3333333333333,
      "grad_norm": 0.00023776869056746364,
      "learning_rate": 9.020567551623935e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8071984,
      "step": 28830
    },
    {
      "epoch": 320.3888888888889,
      "grad_norm": 0.012525434605777264,
      "learning_rate": 9.013018560794318e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8073360,
      "step": 28835
    },
    {
      "epoch": 320.44444444444446,
      "grad_norm": 0.00015597425226587802,
      "learning_rate": 9.005472035358139e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8074784,
      "step": 28840
    },
    {
      "epoch": 320.5,
      "grad_norm": 0.0001270816137548536,
      "learning_rate": 8.997927976479185e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8076160,
      "step": 28845
    },
    {
      "epoch": 320.55555555555554,
      "grad_norm": 0.00038639723788946867,
      "learning_rate": 8.99038638532082e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8077600,
      "step": 28850
    },
    {
      "epoch": 320.6111111111111,
      "grad_norm": 0.0019267667084932327,
      "learning_rate": 8.982847263046065e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8079040,
      "step": 28855
    },
    {
      "epoch": 320.6666666666667,
      "grad_norm": 0.003946646116673946,
      "learning_rate": 8.975310610817555e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8080400,
      "step": 28860
    },
    {
      "epoch": 320.72222222222223,
      "grad_norm": 6.192471482791007e-05,
      "learning_rate": 8.967776429797528e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8081824,
      "step": 28865
    },
    {
      "epoch": 320.77777777777777,
      "grad_norm": 4.519483263720758e-05,
      "learning_rate": 8.960244721147842e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8083216,
      "step": 28870
    },
    {
      "epoch": 320.8333333333333,
      "grad_norm": 6.796141678933054e-05,
      "learning_rate": 8.952715486029995e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8084640,
      "step": 28875
    },
    {
      "epoch": 320.8888888888889,
      "grad_norm": 0.0001369930396322161,
      "learning_rate": 8.945188725605075e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8086016,
      "step": 28880
    },
    {
      "epoch": 320.94444444444446,
      "grad_norm": 0.000787211291026324,
      "learning_rate": 8.937664441033817e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8087392,
      "step": 28885
    },
    {
      "epoch": 321.0,
      "grad_norm": 0.0028218121733516455,
      "learning_rate": 8.930142633476549e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8088784,
      "step": 28890
    },
    {
      "epoch": 321.05555555555554,
      "grad_norm": 0.0009849766502156854,
      "learning_rate": 8.92262330409323e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8090144,
      "step": 28895
    },
    {
      "epoch": 321.1111111111111,
      "grad_norm": 0.00016660777328070253,
      "learning_rate": 8.915106454043448e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8091536,
      "step": 28900
    },
    {
      "epoch": 321.1666666666667,
      "grad_norm": 0.02342378720641136,
      "learning_rate": 8.90759208448638e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8092928,
      "step": 28905
    },
    {
      "epoch": 321.22222222222223,
      "grad_norm": 0.00047037130570970476,
      "learning_rate": 8.900080196580848e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8094336,
      "step": 28910
    },
    {
      "epoch": 321.27777777777777,
      "grad_norm": 0.00022996596817392856,
      "learning_rate": 8.892570791485267e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8095792,
      "step": 28915
    },
    {
      "epoch": 321.3333333333333,
      "grad_norm": 0.00011783569789258763,
      "learning_rate": 8.885063870357688e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8097216,
      "step": 28920
    },
    {
      "epoch": 321.3888888888889,
      "grad_norm": 0.0003772134077735245,
      "learning_rate": 8.87755943435578e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8098608,
      "step": 28925
    },
    {
      "epoch": 321.44444444444446,
      "grad_norm": 0.00011545267625479028,
      "learning_rate": 8.87005748463681e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8100000,
      "step": 28930
    },
    {
      "epoch": 321.5,
      "grad_norm": 0.00010313709208276123,
      "learning_rate": 8.862558022357681e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8101360,
      "step": 28935
    },
    {
      "epoch": 321.55555555555554,
      "grad_norm": 0.0003063284675590694,
      "learning_rate": 8.855061048674903e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8102736,
      "step": 28940
    },
    {
      "epoch": 321.6111111111111,
      "grad_norm": 3.314423520350829e-05,
      "learning_rate": 8.847566564744595e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8104176,
      "step": 28945
    },
    {
      "epoch": 321.6666666666667,
      "grad_norm": 2.976858741021715e-05,
      "learning_rate": 8.840074571722512e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8105616,
      "step": 28950
    },
    {
      "epoch": 321.72222222222223,
      "grad_norm": 0.00014454510528594255,
      "learning_rate": 8.832585070764002e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8106992,
      "step": 28955
    },
    {
      "epoch": 321.77777777777777,
      "grad_norm": 0.0016876135487109423,
      "learning_rate": 8.825098063024045e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8108416,
      "step": 28960
    },
    {
      "epoch": 321.8333333333333,
      "grad_norm": 0.0002863590489141643,
      "learning_rate": 8.817613549657244e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8109856,
      "step": 28965
    },
    {
      "epoch": 321.8888888888889,
      "grad_norm": 0.0002317656617378816,
      "learning_rate": 8.810131531817783e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8111248,
      "step": 28970
    },
    {
      "epoch": 321.94444444444446,
      "grad_norm": 0.0008176696137525141,
      "learning_rate": 8.802652010659496e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8112640,
      "step": 28975
    },
    {
      "epoch": 322.0,
      "grad_norm": 5.074317959952168e-05,
      "learning_rate": 8.795174987335827e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 8114000,
      "step": 28980
    },
    {
      "epoch": 322.05555555555554,
      "grad_norm": 0.00020894999033771455,
      "learning_rate": 8.787700462999807e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8115424,
      "step": 28985
    },
    {
      "epoch": 322.1111111111111,
      "grad_norm": 0.00038712730747647583,
      "learning_rate": 8.780228438804122e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8116800,
      "step": 28990
    },
    {
      "epoch": 322.1666666666667,
      "grad_norm": 0.00013650310575030744,
      "learning_rate": 8.772758915901032e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8118192,
      "step": 28995
    },
    {
      "epoch": 322.22222222222223,
      "grad_norm": 0.0002583055174909532,
      "learning_rate": 8.765291895442443e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8119520,
      "step": 29000
    },
    {
      "epoch": 322.22222222222223,
      "eval_loss": 0.8451299667358398,
      "eval_runtime": 1.2361,
      "eval_samples_per_second": 32.361,
      "eval_steps_per_second": 16.18,
      "num_input_tokens_seen": 8119520,
      "step": 29000
    },
    {
      "epoch": 322.27777777777777,
      "grad_norm": 0.0010508340783417225,
      "learning_rate": 8.75782737857987e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8120944,
      "step": 29005
    },
    {
      "epoch": 322.3333333333333,
      "grad_norm": 0.0004201972915325314,
      "learning_rate": 8.750365366464425e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8122384,
      "step": 29010
    },
    {
      "epoch": 322.3888888888889,
      "grad_norm": 1.2475093171815388e-05,
      "learning_rate": 8.742905860246838e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8123808,
      "step": 29015
    },
    {
      "epoch": 322.44444444444446,
      "grad_norm": 3.6616514989873394e-05,
      "learning_rate": 8.735448861077478e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8125200,
      "step": 29020
    },
    {
      "epoch": 322.5,
      "grad_norm": 0.0002139915304724127,
      "learning_rate": 8.727994370106288e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8126560,
      "step": 29025
    },
    {
      "epoch": 322.55555555555554,
      "grad_norm": 0.00010666379239410162,
      "learning_rate": 8.720542388482861e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8127968,
      "step": 29030
    },
    {
      "epoch": 322.6111111111111,
      "grad_norm": 3.638435009634122e-05,
      "learning_rate": 8.71309291735637e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8129328,
      "step": 29035
    },
    {
      "epoch": 322.6666666666667,
      "grad_norm": 1.9539280401659198e-05,
      "learning_rate": 8.705645957875621e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8130688,
      "step": 29040
    },
    {
      "epoch": 322.72222222222223,
      "grad_norm": 0.00023035275808069855,
      "learning_rate": 8.698201511189048e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8132096,
      "step": 29045
    },
    {
      "epoch": 322.77777777777777,
      "grad_norm": 0.0001458184269722551,
      "learning_rate": 8.690759578444649e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8133536,
      "step": 29050
    },
    {
      "epoch": 322.8333333333333,
      "grad_norm": 0.00020122040587011725,
      "learning_rate": 8.68332016079008e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8134960,
      "step": 29055
    },
    {
      "epoch": 322.8888888888889,
      "grad_norm": 0.0003239136131014675,
      "learning_rate": 8.6758832593726e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8136384,
      "step": 29060
    },
    {
      "epoch": 322.94444444444446,
      "grad_norm": 3.4502929338486865e-05,
      "learning_rate": 8.668448875339053e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8137776,
      "step": 29065
    },
    {
      "epoch": 323.0,
      "grad_norm": 0.00012340344255790114,
      "learning_rate": 8.661017009835933e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8139152,
      "step": 29070
    },
    {
      "epoch": 323.05555555555554,
      "grad_norm": 0.0011159752029925585,
      "learning_rate": 8.653587664009311e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8140544,
      "step": 29075
    },
    {
      "epoch": 323.1111111111111,
      "grad_norm": 7.154931518016383e-05,
      "learning_rate": 8.646160839004902e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8141952,
      "step": 29080
    },
    {
      "epoch": 323.1666666666667,
      "grad_norm": 2.5041959816007875e-05,
      "learning_rate": 8.638736535967998e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8143376,
      "step": 29085
    },
    {
      "epoch": 323.22222222222223,
      "grad_norm": 0.0008820475195534527,
      "learning_rate": 8.631314756043535e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8144736,
      "step": 29090
    },
    {
      "epoch": 323.27777777777777,
      "grad_norm": 0.00012414506636559963,
      "learning_rate": 8.62389550037603e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8146144,
      "step": 29095
    },
    {
      "epoch": 323.3333333333333,
      "grad_norm": 5.717501699109562e-05,
      "learning_rate": 8.616478770109646e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8147536,
      "step": 29100
    },
    {
      "epoch": 323.3888888888889,
      "grad_norm": 0.0001911280123749748,
      "learning_rate": 8.609064566388111e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8148928,
      "step": 29105
    },
    {
      "epoch": 323.44444444444446,
      "grad_norm": 0.00023672454699408263,
      "learning_rate": 8.601652890354815e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8150304,
      "step": 29110
    },
    {
      "epoch": 323.5,
      "grad_norm": 0.0031096467282623053,
      "learning_rate": 8.594243743152705e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8151712,
      "step": 29115
    },
    {
      "epoch": 323.55555555555554,
      "grad_norm": 4.70716149720829e-05,
      "learning_rate": 8.58683712592438e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8153104,
      "step": 29120
    },
    {
      "epoch": 323.6111111111111,
      "grad_norm": 0.001589919556863606,
      "learning_rate": 8.579433039812037e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8154496,
      "step": 29125
    },
    {
      "epoch": 323.6666666666667,
      "grad_norm": 0.0008913125493563712,
      "learning_rate": 8.572031485957466e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8155936,
      "step": 29130
    },
    {
      "epoch": 323.72222222222223,
      "grad_norm": 0.0003941661852877587,
      "learning_rate": 8.564632465502084e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8157312,
      "step": 29135
    },
    {
      "epoch": 323.77777777777777,
      "grad_norm": 0.00018623123469296843,
      "learning_rate": 8.557235979586928e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8158720,
      "step": 29140
    },
    {
      "epoch": 323.8333333333333,
      "grad_norm": 0.00017609872156754136,
      "learning_rate": 8.549842029352606e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8160144,
      "step": 29145
    },
    {
      "epoch": 323.8888888888889,
      "grad_norm": 0.00013640672841574997,
      "learning_rate": 8.542450615939376e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8161536,
      "step": 29150
    },
    {
      "epoch": 323.94444444444446,
      "grad_norm": 0.0001539865625090897,
      "learning_rate": 8.535061740487082e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8162976,
      "step": 29155
    },
    {
      "epoch": 324.0,
      "grad_norm": 0.0008419143850915134,
      "learning_rate": 8.527675404135168e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8164384,
      "step": 29160
    },
    {
      "epoch": 324.05555555555554,
      "grad_norm": 0.0006010643555782735,
      "learning_rate": 8.520291608022724e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8165760,
      "step": 29165
    },
    {
      "epoch": 324.1111111111111,
      "grad_norm": 0.00019943864026572555,
      "learning_rate": 8.512910353288398e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8167168,
      "step": 29170
    },
    {
      "epoch": 324.1666666666667,
      "grad_norm": 0.00020308338571339846,
      "learning_rate": 8.505531641070486e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8168544,
      "step": 29175
    },
    {
      "epoch": 324.22222222222223,
      "grad_norm": 0.0003147760289721191,
      "learning_rate": 8.498155472506885e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8169952,
      "step": 29180
    },
    {
      "epoch": 324.27777777777777,
      "grad_norm": 0.000132315035443753,
      "learning_rate": 8.49078184873508e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8171344,
      "step": 29185
    },
    {
      "epoch": 324.3333333333333,
      "grad_norm": 0.00010633876081556082,
      "learning_rate": 8.483410770892188e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8172832,
      "step": 29190
    },
    {
      "epoch": 324.3888888888889,
      "grad_norm": 0.00015201576752588153,
      "learning_rate": 8.476042240114909e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8174208,
      "step": 29195
    },
    {
      "epoch": 324.44444444444446,
      "grad_norm": 7.602348341606557e-05,
      "learning_rate": 8.468676257539568e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8175584,
      "step": 29200
    },
    {
      "epoch": 324.44444444444446,
      "eval_loss": 0.9005203247070312,
      "eval_runtime": 1.236,
      "eval_samples_per_second": 32.362,
      "eval_steps_per_second": 16.181,
      "num_input_tokens_seen": 8175584,
      "step": 29200
    },
    {
      "epoch": 324.5,
      "grad_norm": 0.0005280082696117461,
      "learning_rate": 8.4613128243021e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8176976,
      "step": 29205
    },
    {
      "epoch": 324.55555555555554,
      "grad_norm": 0.00022968067787587643,
      "learning_rate": 8.453951941538028e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8178368,
      "step": 29210
    },
    {
      "epoch": 324.6111111111111,
      "grad_norm": 0.00011436753993621096,
      "learning_rate": 8.446593610382495e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8179760,
      "step": 29215
    },
    {
      "epoch": 324.6666666666667,
      "grad_norm": 0.00018004632147494704,
      "learning_rate": 8.439237831970259e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8181152,
      "step": 29220
    },
    {
      "epoch": 324.72222222222223,
      "grad_norm": 0.0002781399234663695,
      "learning_rate": 8.431884607435667e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8182576,
      "step": 29225
    },
    {
      "epoch": 324.77777777777777,
      "grad_norm": 4.3221607484156266e-05,
      "learning_rate": 8.424533937912665e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8183968,
      "step": 29230
    },
    {
      "epoch": 324.8333333333333,
      "grad_norm": 0.00011445445124991238,
      "learning_rate": 8.41718582453484e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8185344,
      "step": 29235
    },
    {
      "epoch": 324.8888888888889,
      "grad_norm": 0.00012869230704382062,
      "learning_rate": 8.409840268435346e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8186752,
      "step": 29240
    },
    {
      "epoch": 324.94444444444446,
      "grad_norm": 0.0001351253013126552,
      "learning_rate": 8.402497270746976e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8188208,
      "step": 29245
    },
    {
      "epoch": 325.0,
      "grad_norm": 0.01255149208009243,
      "learning_rate": 8.395156832602095e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8189616,
      "step": 29250
    },
    {
      "epoch": 325.05555555555554,
      "grad_norm": 0.0006821950664743781,
      "learning_rate": 8.387818955132707e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8191056,
      "step": 29255
    },
    {
      "epoch": 325.1111111111111,
      "grad_norm": 0.0001913592277560383,
      "learning_rate": 8.38048363947039e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8192416,
      "step": 29260
    },
    {
      "epoch": 325.1666666666667,
      "grad_norm": 0.0002708460669964552,
      "learning_rate": 8.373150886746351e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8193824,
      "step": 29265
    },
    {
      "epoch": 325.22222222222223,
      "grad_norm": 0.0001398111489834264,
      "learning_rate": 8.365820698091397e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8195216,
      "step": 29270
    },
    {
      "epoch": 325.27777777777777,
      "grad_norm": 2.4193093850044534e-05,
      "learning_rate": 8.358493074635922e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8196624,
      "step": 29275
    },
    {
      "epoch": 325.3333333333333,
      "grad_norm": 0.00017792877042666078,
      "learning_rate": 8.351168017509948e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8198016,
      "step": 29280
    },
    {
      "epoch": 325.3888888888889,
      "grad_norm": 0.00015454983804374933,
      "learning_rate": 8.343845527843094e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8199456,
      "step": 29285
    },
    {
      "epoch": 325.44444444444446,
      "grad_norm": 6.23441010247916e-05,
      "learning_rate": 8.336525606764566e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8200864,
      "step": 29290
    },
    {
      "epoch": 325.5,
      "grad_norm": 4.5257700548972934e-05,
      "learning_rate": 8.329208255403204e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8202304,
      "step": 29295
    },
    {
      "epoch": 325.55555555555554,
      "grad_norm": 0.00019908469403162599,
      "learning_rate": 8.321893474887426e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8203712,
      "step": 29300
    },
    {
      "epoch": 325.6111111111111,
      "grad_norm": 0.0004264444578438997,
      "learning_rate": 8.31458126634526e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8205104,
      "step": 29305
    },
    {
      "epoch": 325.6666666666667,
      "grad_norm": 2.535227031330578e-05,
      "learning_rate": 8.30727163090435e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8206496,
      "step": 29310
    },
    {
      "epoch": 325.72222222222223,
      "grad_norm": 0.00045321928337216377,
      "learning_rate": 8.29996456969192e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8207888,
      "step": 29315
    },
    {
      "epoch": 325.77777777777777,
      "grad_norm": 9.890468936646357e-05,
      "learning_rate": 8.292660083834818e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8209296,
      "step": 29320
    },
    {
      "epoch": 325.8333333333333,
      "grad_norm": 7.679282134631649e-05,
      "learning_rate": 8.2853581744595e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8210704,
      "step": 29325
    },
    {
      "epoch": 325.8888888888889,
      "grad_norm": 0.00012071795208612457,
      "learning_rate": 8.278058842691991e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8212112,
      "step": 29330
    },
    {
      "epoch": 325.94444444444446,
      "grad_norm": 0.00041484888060949743,
      "learning_rate": 8.27076208965796e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8213504,
      "step": 29335
    },
    {
      "epoch": 326.0,
      "grad_norm": 0.00043354739318601787,
      "learning_rate": 8.263467916482637e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8214896,
      "step": 29340
    },
    {
      "epoch": 326.05555555555554,
      "grad_norm": 0.00017828680574893951,
      "learning_rate": 8.256176324290885e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8216272,
      "step": 29345
    },
    {
      "epoch": 326.1111111111111,
      "grad_norm": 4.611624899553135e-05,
      "learning_rate": 8.248887314207168e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8217648,
      "step": 29350
    },
    {
      "epoch": 326.1666666666667,
      "grad_norm": 7.913455192465335e-05,
      "learning_rate": 8.24160088735553e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8219088,
      "step": 29355
    },
    {
      "epoch": 326.22222222222223,
      "grad_norm": 0.0034707263112068176,
      "learning_rate": 8.234317044859629e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8220480,
      "step": 29360
    },
    {
      "epoch": 326.27777777777777,
      "grad_norm": 3.697927240864374e-05,
      "learning_rate": 8.227035787842744e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8221888,
      "step": 29365
    },
    {
      "epoch": 326.3333333333333,
      "grad_norm": 8.785456884652376e-05,
      "learning_rate": 8.219757117427721e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8223328,
      "step": 29370
    },
    {
      "epoch": 326.3888888888889,
      "grad_norm": 8.084440923994407e-05,
      "learning_rate": 8.212481034737014e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8224688,
      "step": 29375
    },
    {
      "epoch": 326.44444444444446,
      "grad_norm": 6.26368127996102e-05,
      "learning_rate": 8.205207540892707e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8226128,
      "step": 29380
    },
    {
      "epoch": 326.5,
      "grad_norm": 0.0003194510645698756,
      "learning_rate": 8.197936637016442e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8227568,
      "step": 29385
    },
    {
      "epoch": 326.55555555555554,
      "grad_norm": 0.00010721338912844658,
      "learning_rate": 8.190668324229508e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8228944,
      "step": 29390
    },
    {
      "epoch": 326.6111111111111,
      "grad_norm": 0.00011060499673476443,
      "learning_rate": 8.183402603652749e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8230352,
      "step": 29395
    },
    {
      "epoch": 326.6666666666667,
      "grad_norm": 2.38287502725143e-05,
      "learning_rate": 8.176139476406635e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8231760,
      "step": 29400
    },
    {
      "epoch": 326.6666666666667,
      "eval_loss": 0.8589199781417847,
      "eval_runtime": 1.2407,
      "eval_samples_per_second": 32.239,
      "eval_steps_per_second": 16.119,
      "num_input_tokens_seen": 8231760,
      "step": 29400
    },
    {
      "epoch": 326.72222222222223,
      "grad_norm": 6.518299778690562e-05,
      "learning_rate": 8.16887894361125e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8233120,
      "step": 29405
    },
    {
      "epoch": 326.77777777777777,
      "grad_norm": 0.0005291083944030106,
      "learning_rate": 8.161621006386233e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8234480,
      "step": 29410
    },
    {
      "epoch": 326.8333333333333,
      "grad_norm": 0.0006752801709808409,
      "learning_rate": 8.154365665850869e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8235872,
      "step": 29415
    },
    {
      "epoch": 326.8888888888889,
      "grad_norm": 0.00761764170601964,
      "learning_rate": 8.147112923124005e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8237280,
      "step": 29420
    },
    {
      "epoch": 326.94444444444446,
      "grad_norm": 2.4753762772888876e-05,
      "learning_rate": 8.13986277932412e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8238672,
      "step": 29425
    },
    {
      "epoch": 327.0,
      "grad_norm": 0.002367420820519328,
      "learning_rate": 8.132615235569277e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8240048,
      "step": 29430
    },
    {
      "epoch": 327.05555555555554,
      "grad_norm": 0.00015927293861750513,
      "learning_rate": 8.125370292977124e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8241424,
      "step": 29435
    },
    {
      "epoch": 327.1111111111111,
      "grad_norm": 0.00023090028844308108,
      "learning_rate": 8.118127952664944e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8242864,
      "step": 29440
    },
    {
      "epoch": 327.1666666666667,
      "grad_norm": 0.0006414719973690808,
      "learning_rate": 8.110888215749574e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8244256,
      "step": 29445
    },
    {
      "epoch": 327.22222222222223,
      "grad_norm": 0.002378873759880662,
      "learning_rate": 8.10365108334749e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8245696,
      "step": 29450
    },
    {
      "epoch": 327.27777777777777,
      "grad_norm": 0.00011302941129542887,
      "learning_rate": 8.096416556574743e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8247072,
      "step": 29455
    },
    {
      "epoch": 327.3333333333333,
      "grad_norm": 0.00014983727305661887,
      "learning_rate": 8.08918463654698e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8248480,
      "step": 29460
    },
    {
      "epoch": 327.3888888888889,
      "grad_norm": 0.0002525081217754632,
      "learning_rate": 8.081955324379458e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8249872,
      "step": 29465
    },
    {
      "epoch": 327.44444444444446,
      "grad_norm": 0.001798324752599001,
      "learning_rate": 8.074728621187039e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8251248,
      "step": 29470
    },
    {
      "epoch": 327.5,
      "grad_norm": 0.003138162661343813,
      "learning_rate": 8.067504528084158e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8252672,
      "step": 29475
    },
    {
      "epoch": 327.55555555555554,
      "grad_norm": 0.0002328294503968209,
      "learning_rate": 8.060283046184861e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8254080,
      "step": 29480
    },
    {
      "epoch": 327.6111111111111,
      "grad_norm": 9.539401798974723e-05,
      "learning_rate": 8.053064176602806e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8255504,
      "step": 29485
    },
    {
      "epoch": 327.6666666666667,
      "grad_norm": 6.0000031226081774e-05,
      "learning_rate": 8.045847920451216e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8256912,
      "step": 29490
    },
    {
      "epoch": 327.72222222222223,
      "grad_norm": 8.31540091894567e-05,
      "learning_rate": 8.038634278842944e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8258288,
      "step": 29495
    },
    {
      "epoch": 327.77777777777777,
      "grad_norm": 7.932051812531427e-05,
      "learning_rate": 8.031423252890408e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8259680,
      "step": 29500
    },
    {
      "epoch": 327.8333333333333,
      "grad_norm": 0.0002511402126401663,
      "learning_rate": 8.024214843705646e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8261072,
      "step": 29505
    },
    {
      "epoch": 327.8888888888889,
      "grad_norm": 1.863358193077147e-05,
      "learning_rate": 8.017009052400295e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8262448,
      "step": 29510
    },
    {
      "epoch": 327.94444444444446,
      "grad_norm": 4.896187238045968e-05,
      "learning_rate": 8.00980588008557e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8263888,
      "step": 29515
    },
    {
      "epoch": 328.0,
      "grad_norm": 7.088498387020081e-05,
      "learning_rate": 8.002605327872282e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8265328,
      "step": 29520
    },
    {
      "epoch": 328.05555555555554,
      "grad_norm": 6.3953862991184e-05,
      "learning_rate": 7.995407396870862e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8266720,
      "step": 29525
    },
    {
      "epoch": 328.1111111111111,
      "grad_norm": 0.00019638289813883603,
      "learning_rate": 7.988212088191307e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8268128,
      "step": 29530
    },
    {
      "epoch": 328.1666666666667,
      "grad_norm": 9.4542745500803e-05,
      "learning_rate": 7.98101940294324e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8269552,
      "step": 29535
    },
    {
      "epoch": 328.22222222222223,
      "grad_norm": 4.510365397436544e-05,
      "learning_rate": 7.973829342235847e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8270960,
      "step": 29540
    },
    {
      "epoch": 328.27777777777777,
      "grad_norm": 1.3453282008413225e-05,
      "learning_rate": 7.966641907177936e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8272384,
      "step": 29545
    },
    {
      "epoch": 328.3333333333333,
      "grad_norm": 0.00011133852967759594,
      "learning_rate": 7.959457098877901e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8273792,
      "step": 29550
    },
    {
      "epoch": 328.3888888888889,
      "grad_norm": 0.00010486075188964605,
      "learning_rate": 7.952274918443719e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8275184,
      "step": 29555
    },
    {
      "epoch": 328.44444444444446,
      "grad_norm": 0.00010087141708936542,
      "learning_rate": 7.945095366982983e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8276576,
      "step": 29560
    },
    {
      "epoch": 328.5,
      "grad_norm": 0.0011197593994438648,
      "learning_rate": 7.937918445602871e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8278000,
      "step": 29565
    },
    {
      "epoch": 328.55555555555554,
      "grad_norm": 2.3743505153106526e-05,
      "learning_rate": 7.930744155410145e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8279392,
      "step": 29570
    },
    {
      "epoch": 328.6111111111111,
      "grad_norm": 8.595943654654548e-05,
      "learning_rate": 7.923572497511181e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8280768,
      "step": 29575
    },
    {
      "epoch": 328.6666666666667,
      "grad_norm": 0.00011289581743767485,
      "learning_rate": 7.916403473011927e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8282176,
      "step": 29580
    },
    {
      "epoch": 328.72222222222223,
      "grad_norm": 0.006907827220857143,
      "learning_rate": 7.909237083017953e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8283584,
      "step": 29585
    },
    {
      "epoch": 328.77777777777777,
      "grad_norm": 0.0029976326040923595,
      "learning_rate": 7.902073328634389e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8284944,
      "step": 29590
    },
    {
      "epoch": 328.8333333333333,
      "grad_norm": 0.0012820129049941897,
      "learning_rate": 7.894912210965987e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8286352,
      "step": 29595
    },
    {
      "epoch": 328.8888888888889,
      "grad_norm": 0.0019162525422871113,
      "learning_rate": 7.887753731117075e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8287696,
      "step": 29600
    },
    {
      "epoch": 328.8888888888889,
      "eval_loss": 0.8993797302246094,
      "eval_runtime": 1.2394,
      "eval_samples_per_second": 32.273,
      "eval_steps_per_second": 16.136,
      "num_input_tokens_seen": 8287696,
      "step": 29600
    },
    {
      "epoch": 328.94444444444446,
      "grad_norm": 2.167954335163813e-05,
      "learning_rate": 7.880597890191587e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8289120,
      "step": 29605
    },
    {
      "epoch": 329.0,
      "grad_norm": 0.0002665576175786555,
      "learning_rate": 7.873444689293036e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8290560,
      "step": 29610
    },
    {
      "epoch": 329.05555555555554,
      "grad_norm": 6.345566362142563e-05,
      "learning_rate": 7.866294129524548e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8291968,
      "step": 29615
    },
    {
      "epoch": 329.1111111111111,
      "grad_norm": 0.0011597895063459873,
      "learning_rate": 7.859146211988811e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8293376,
      "step": 29620
    },
    {
      "epoch": 329.1666666666667,
      "grad_norm": 0.00018520731828175485,
      "learning_rate": 7.852000937788134e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8294720,
      "step": 29625
    },
    {
      "epoch": 329.22222222222223,
      "grad_norm": 4.372908006189391e-05,
      "learning_rate": 7.844858308024416e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8296160,
      "step": 29630
    },
    {
      "epoch": 329.27777777777777,
      "grad_norm": 0.00013618294906336814,
      "learning_rate": 7.837718323799122e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8297520,
      "step": 29635
    },
    {
      "epoch": 329.3333333333333,
      "grad_norm": 2.9407161491690204e-05,
      "learning_rate": 7.83058098621334e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8298896,
      "step": 29640
    },
    {
      "epoch": 329.3888888888889,
      "grad_norm": 0.00012035465624649078,
      "learning_rate": 7.823446296367739e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8300320,
      "step": 29645
    },
    {
      "epoch": 329.44444444444446,
      "grad_norm": 0.0037603683304041624,
      "learning_rate": 7.81631425536257e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8301680,
      "step": 29650
    },
    {
      "epoch": 329.5,
      "grad_norm": 0.00016824343765620142,
      "learning_rate": 7.809184864297689e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8303104,
      "step": 29655
    },
    {
      "epoch": 329.55555555555554,
      "grad_norm": 4.698933844338171e-05,
      "learning_rate": 7.802058124272532e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8304528,
      "step": 29660
    },
    {
      "epoch": 329.6111111111111,
      "grad_norm": 0.000619869795627892,
      "learning_rate": 7.79493403638614e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8305888,
      "step": 29665
    },
    {
      "epoch": 329.6666666666667,
      "grad_norm": 0.00039115993422456086,
      "learning_rate": 7.787812601737132e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8307328,
      "step": 29670
    },
    {
      "epoch": 329.72222222222223,
      "grad_norm": 2.4174129066523165e-05,
      "learning_rate": 7.780693821423715e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8308752,
      "step": 29675
    },
    {
      "epoch": 329.77777777777777,
      "grad_norm": 0.00030708801932632923,
      "learning_rate": 7.773577696543705e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8310160,
      "step": 29680
    },
    {
      "epoch": 329.8333333333333,
      "grad_norm": 4.612498014466837e-05,
      "learning_rate": 7.7664642281945e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8311520,
      "step": 29685
    },
    {
      "epoch": 329.8888888888889,
      "grad_norm": 0.00021706272673327476,
      "learning_rate": 7.759353417473072e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8312944,
      "step": 29690
    },
    {
      "epoch": 329.94444444444446,
      "grad_norm": 0.0014427837450057268,
      "learning_rate": 7.752245265476016e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8314400,
      "step": 29695
    },
    {
      "epoch": 330.0,
      "grad_norm": 5.227290603215806e-05,
      "learning_rate": 7.745139773299481e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8315840,
      "step": 29700
    },
    {
      "epoch": 330.05555555555554,
      "grad_norm": 4.003175490652211e-05,
      "learning_rate": 7.738036942039232e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8317248,
      "step": 29705
    },
    {
      "epoch": 330.1111111111111,
      "grad_norm": 3.0734776373719797e-05,
      "learning_rate": 7.73093677279062e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8318592,
      "step": 29710
    },
    {
      "epoch": 330.1666666666667,
      "grad_norm": 0.00017238393775187433,
      "learning_rate": 7.72383926664857e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8320000,
      "step": 29715
    },
    {
      "epoch": 330.22222222222223,
      "grad_norm": 5.742490611737594e-05,
      "learning_rate": 7.716744424707606e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8321392,
      "step": 29720
    },
    {
      "epoch": 330.27777777777777,
      "grad_norm": 0.00037719597457908094,
      "learning_rate": 7.709652248061858e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8322768,
      "step": 29725
    },
    {
      "epoch": 330.3333333333333,
      "grad_norm": 1.8163020286010578e-05,
      "learning_rate": 7.702562737805017e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8324144,
      "step": 29730
    },
    {
      "epoch": 330.3888888888889,
      "grad_norm": 3.843099329969846e-05,
      "learning_rate": 7.695475895030365e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8325584,
      "step": 29735
    },
    {
      "epoch": 330.44444444444446,
      "grad_norm": 9.144903742708266e-05,
      "learning_rate": 7.6883917208308e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8327072,
      "step": 29740
    },
    {
      "epoch": 330.5,
      "grad_norm": 0.0027091163210570812,
      "learning_rate": 7.681310216298778e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8328448,
      "step": 29745
    },
    {
      "epoch": 330.55555555555554,
      "grad_norm": 6.440275319619104e-05,
      "learning_rate": 7.674231382526367e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8329872,
      "step": 29750
    },
    {
      "epoch": 330.6111111111111,
      "grad_norm": 0.0005727424286305904,
      "learning_rate": 7.667155220605198e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8331280,
      "step": 29755
    },
    {
      "epoch": 330.6666666666667,
      "grad_norm": 7.240592094603926e-05,
      "learning_rate": 7.660081731626515e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8332720,
      "step": 29760
    },
    {
      "epoch": 330.72222222222223,
      "grad_norm": 0.0008549306076020002,
      "learning_rate": 7.653010916681141e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8334112,
      "step": 29765
    },
    {
      "epoch": 330.77777777777777,
      "grad_norm": 0.000402107514673844,
      "learning_rate": 7.645942776859472e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8335472,
      "step": 29770
    },
    {
      "epoch": 330.8333333333333,
      "grad_norm": 2.3345890440396033e-05,
      "learning_rate": 7.63887731325152e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8336832,
      "step": 29775
    },
    {
      "epoch": 330.8888888888889,
      "grad_norm": 4.514936517807655e-05,
      "learning_rate": 7.63181452694685e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8338192,
      "step": 29780
    },
    {
      "epoch": 330.94444444444446,
      "grad_norm": 0.0009247826528735459,
      "learning_rate": 7.624754419034644e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8339584,
      "step": 29785
    },
    {
      "epoch": 331.0,
      "grad_norm": 2.1654172087437473e-05,
      "learning_rate": 7.6176969906036645e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8340976,
      "step": 29790
    },
    {
      "epoch": 331.05555555555554,
      "grad_norm": 2.235017927887384e-05,
      "learning_rate": 7.610642242742242e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8342352,
      "step": 29795
    },
    {
      "epoch": 331.1111111111111,
      "grad_norm": 0.0006798666436225176,
      "learning_rate": 7.603590176538322e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8343760,
      "step": 29800
    },
    {
      "epoch": 331.1111111111111,
      "eval_loss": 0.8627737164497375,
      "eval_runtime": 1.2407,
      "eval_samples_per_second": 32.239,
      "eval_steps_per_second": 16.119,
      "num_input_tokens_seen": 8343760,
      "step": 29800
    },
    {
      "epoch": 331.1666666666667,
      "grad_norm": 0.0010370357194915414,
      "learning_rate": 7.596540793079404e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8345168,
      "step": 29805
    },
    {
      "epoch": 331.22222222222223,
      "grad_norm": 0.00031194774783216417,
      "learning_rate": 7.5894940934526125e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8346576,
      "step": 29810
    },
    {
      "epoch": 331.27777777777777,
      "grad_norm": 9.422533912584186e-05,
      "learning_rate": 7.582450078744621e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8348000,
      "step": 29815
    },
    {
      "epoch": 331.3333333333333,
      "grad_norm": 0.00010635744547471404,
      "learning_rate": 7.575408750041707e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8349392,
      "step": 29820
    },
    {
      "epoch": 331.3888888888889,
      "grad_norm": 2.737028626143001e-05,
      "learning_rate": 7.568370108429732e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8350816,
      "step": 29825
    },
    {
      "epoch": 331.44444444444446,
      "grad_norm": 0.00630094762891531,
      "learning_rate": 7.561334154994154e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8352208,
      "step": 29830
    },
    {
      "epoch": 331.5,
      "grad_norm": 0.00015722440730314702,
      "learning_rate": 7.55430089081999e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8353584,
      "step": 29835
    },
    {
      "epoch": 331.55555555555554,
      "grad_norm": 0.0011397578055039048,
      "learning_rate": 7.547270316991864e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8354944,
      "step": 29840
    },
    {
      "epoch": 331.6111111111111,
      "grad_norm": 4.7853565774858e-05,
      "learning_rate": 7.5402424345939884e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8356320,
      "step": 29845
    },
    {
      "epoch": 331.6666666666667,
      "grad_norm": 0.0007562093669548631,
      "learning_rate": 7.533217244710133e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8357728,
      "step": 29850
    },
    {
      "epoch": 331.72222222222223,
      "grad_norm": 0.0017694445559754968,
      "learning_rate": 7.52619474842369e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8359152,
      "step": 29855
    },
    {
      "epoch": 331.77777777777777,
      "grad_norm": 0.00026817465550266206,
      "learning_rate": 7.519174946817597e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8360544,
      "step": 29860
    },
    {
      "epoch": 331.8333333333333,
      "grad_norm": 0.00015183730283752084,
      "learning_rate": 7.512157840974407e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8361968,
      "step": 29865
    },
    {
      "epoch": 331.8888888888889,
      "grad_norm": 0.00021635401935782284,
      "learning_rate": 7.5051434319762496e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8363376,
      "step": 29870
    },
    {
      "epoch": 331.94444444444446,
      "grad_norm": 0.0002613012620713562,
      "learning_rate": 7.498131720904822e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8364768,
      "step": 29875
    },
    {
      "epoch": 332.0,
      "grad_norm": 0.00012201054050819948,
      "learning_rate": 7.491122708841433e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8366160,
      "step": 29880
    },
    {
      "epoch": 332.05555555555554,
      "grad_norm": 0.00030327029526233673,
      "learning_rate": 7.4841163968669524e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8367584,
      "step": 29885
    },
    {
      "epoch": 332.1111111111111,
      "grad_norm": 6.223244417924434e-05,
      "learning_rate": 7.4771127860618355e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8368992,
      "step": 29890
    },
    {
      "epoch": 332.1666666666667,
      "grad_norm": 0.00020186057372484356,
      "learning_rate": 7.470111877506139e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8370384,
      "step": 29895
    },
    {
      "epoch": 332.22222222222223,
      "grad_norm": 0.00010643665154930204,
      "learning_rate": 7.463113672279479e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8371760,
      "step": 29900
    },
    {
      "epoch": 332.27777777777777,
      "grad_norm": 0.00013205308641772717,
      "learning_rate": 7.456118171461071e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8373152,
      "step": 29905
    },
    {
      "epoch": 332.3333333333333,
      "grad_norm": 3.759573519346304e-05,
      "learning_rate": 7.449125376129721e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8374576,
      "step": 29910
    },
    {
      "epoch": 332.3888888888889,
      "grad_norm": 0.00028030533576384187,
      "learning_rate": 7.442135287363788e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8375968,
      "step": 29915
    },
    {
      "epoch": 332.44444444444446,
      "grad_norm": 0.00012657091428991407,
      "learning_rate": 7.435147906241247e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8377344,
      "step": 29920
    },
    {
      "epoch": 332.5,
      "grad_norm": 0.00045980134746059775,
      "learning_rate": 7.428163233839624e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8378736,
      "step": 29925
    },
    {
      "epoch": 332.55555555555554,
      "grad_norm": 0.0001228766341228038,
      "learning_rate": 7.4211812712360525e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8380144,
      "step": 29930
    },
    {
      "epoch": 332.6111111111111,
      "grad_norm": 7.002270285738632e-05,
      "learning_rate": 7.4142020195072464e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8381536,
      "step": 29935
    },
    {
      "epoch": 332.6666666666667,
      "grad_norm": 9.550524555379525e-05,
      "learning_rate": 7.407225479729479e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8382960,
      "step": 29940
    },
    {
      "epoch": 332.72222222222223,
      "grad_norm": 3.1314633815782145e-05,
      "learning_rate": 7.400251652978632e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8384368,
      "step": 29945
    },
    {
      "epoch": 332.77777777777777,
      "grad_norm": 0.00019656585936900228,
      "learning_rate": 7.393280540330147e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8385760,
      "step": 29950
    },
    {
      "epoch": 332.8333333333333,
      "grad_norm": 0.0015386524610221386,
      "learning_rate": 7.386312142859069e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8387168,
      "step": 29955
    },
    {
      "epoch": 332.8888888888889,
      "grad_norm": 0.0004538565408438444,
      "learning_rate": 7.379346461640008e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8388560,
      "step": 29960
    },
    {
      "epoch": 332.94444444444446,
      "grad_norm": 0.0002815747575368732,
      "learning_rate": 7.372383497747149e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8389920,
      "step": 29965
    },
    {
      "epoch": 333.0,
      "grad_norm": 0.00020795133605133742,
      "learning_rate": 7.3654232522542775e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8391328,
      "step": 29970
    },
    {
      "epoch": 333.05555555555554,
      "grad_norm": 0.003836961230263114,
      "learning_rate": 7.358465726234756e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8392720,
      "step": 29975
    },
    {
      "epoch": 333.1111111111111,
      "grad_norm": 0.0001971958263311535,
      "learning_rate": 7.351510920761512e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8394112,
      "step": 29980
    },
    {
      "epoch": 333.1666666666667,
      "grad_norm": 0.0009560128673911095,
      "learning_rate": 7.344558836907067e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8395504,
      "step": 29985
    },
    {
      "epoch": 333.22222222222223,
      "grad_norm": 0.00012538475857581943,
      "learning_rate": 7.3376094757435285e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8396928,
      "step": 29990
    },
    {
      "epoch": 333.27777777777777,
      "grad_norm": 2.7171565307071432e-05,
      "learning_rate": 7.330662838342561e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8398320,
      "step": 29995
    },
    {
      "epoch": 333.3333333333333,
      "grad_norm": 0.00025128322886303067,
      "learning_rate": 7.323718925775438e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8399696,
      "step": 30000
    },
    {
      "epoch": 333.3333333333333,
      "eval_loss": 0.8769500851631165,
      "eval_runtime": 1.2347,
      "eval_samples_per_second": 32.398,
      "eval_steps_per_second": 16.199,
      "num_input_tokens_seen": 8399696,
      "step": 30000
    },
    {
      "epoch": 333.3888888888889,
      "grad_norm": 6.94663031026721e-05,
      "learning_rate": 7.316777739112985e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8401104,
      "step": 30005
    },
    {
      "epoch": 333.44444444444446,
      "grad_norm": 0.00027799277449958026,
      "learning_rate": 7.309839279425626e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8402512,
      "step": 30010
    },
    {
      "epoch": 333.5,
      "grad_norm": 0.0005379028152674437,
      "learning_rate": 7.302903547783366e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8403888,
      "step": 30015
    },
    {
      "epoch": 333.55555555555554,
      "grad_norm": 0.0001832939888117835,
      "learning_rate": 7.2959705452557644e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8405312,
      "step": 30020
    },
    {
      "epoch": 333.6111111111111,
      "grad_norm": 0.00045300525380298495,
      "learning_rate": 7.289040272911996e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8406752,
      "step": 30025
    },
    {
      "epoch": 333.6666666666667,
      "grad_norm": 8.887342119123787e-05,
      "learning_rate": 7.282112731820789e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8408192,
      "step": 30030
    },
    {
      "epoch": 333.72222222222223,
      "grad_norm": 0.00012403304572217166,
      "learning_rate": 7.275187923050447e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8409568,
      "step": 30035
    },
    {
      "epoch": 333.77777777777777,
      "grad_norm": 0.0004500896902754903,
      "learning_rate": 7.268265847668879e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8411008,
      "step": 30040
    },
    {
      "epoch": 333.8333333333333,
      "grad_norm": 0.003140217624604702,
      "learning_rate": 7.261346506743538e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8412368,
      "step": 30045
    },
    {
      "epoch": 333.8888888888889,
      "grad_norm": 6.25909015070647e-05,
      "learning_rate": 7.254429901341486e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8413744,
      "step": 30050
    },
    {
      "epoch": 333.94444444444446,
      "grad_norm": 0.0002158079732907936,
      "learning_rate": 7.247516032529356e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8415152,
      "step": 30055
    },
    {
      "epoch": 334.0,
      "grad_norm": 0.00029164020088501275,
      "learning_rate": 7.240604901373338e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8416592,
      "step": 30060
    },
    {
      "epoch": 334.05555555555554,
      "grad_norm": 0.00025333999656140804,
      "learning_rate": 7.233696508939223e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8418016,
      "step": 30065
    },
    {
      "epoch": 334.1111111111111,
      "grad_norm": 0.002731401240453124,
      "learning_rate": 7.226790856292376e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8419408,
      "step": 30070
    },
    {
      "epoch": 334.1666666666667,
      "grad_norm": 9.100643364945427e-05,
      "learning_rate": 7.219887944497727e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8420784,
      "step": 30075
    },
    {
      "epoch": 334.22222222222223,
      "grad_norm": 9.990712715080008e-05,
      "learning_rate": 7.2129877746198e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8422192,
      "step": 30080
    },
    {
      "epoch": 334.27777777777777,
      "grad_norm": 3.930641105398536e-05,
      "learning_rate": 7.20609034772268e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8423648,
      "step": 30085
    },
    {
      "epoch": 334.3333333333333,
      "grad_norm": 0.0003056906571146101,
      "learning_rate": 7.19919566487004e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8425056,
      "step": 30090
    },
    {
      "epoch": 334.3888888888889,
      "grad_norm": 6.958517042221501e-05,
      "learning_rate": 7.192303727125132e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8426464,
      "step": 30095
    },
    {
      "epoch": 334.44444444444446,
      "grad_norm": 7.777615246595815e-05,
      "learning_rate": 7.185414535550777e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8427872,
      "step": 30100
    },
    {
      "epoch": 334.5,
      "grad_norm": 0.0012227207189425826,
      "learning_rate": 7.178528091209363e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8429280,
      "step": 30105
    },
    {
      "epoch": 334.55555555555554,
      "grad_norm": 0.0018576240399852395,
      "learning_rate": 7.171644395162888e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8430656,
      "step": 30110
    },
    {
      "epoch": 334.6111111111111,
      "grad_norm": 2.6813682779902592e-05,
      "learning_rate": 7.164763448472881e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8432048,
      "step": 30115
    },
    {
      "epoch": 334.6666666666667,
      "grad_norm": 0.0001567200233694166,
      "learning_rate": 7.157885252200491e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8433488,
      "step": 30120
    },
    {
      "epoch": 334.72222222222223,
      "grad_norm": 0.00020281679462641478,
      "learning_rate": 7.151009807406403e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8434800,
      "step": 30125
    },
    {
      "epoch": 334.77777777777777,
      "grad_norm": 7.617022492922843e-05,
      "learning_rate": 7.144137115150909e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8436208,
      "step": 30130
    },
    {
      "epoch": 334.8333333333333,
      "grad_norm": 9.23649495234713e-05,
      "learning_rate": 7.1372671764938725e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8437632,
      "step": 30135
    },
    {
      "epoch": 334.8888888888889,
      "grad_norm": 1.6493677321705036e-05,
      "learning_rate": 7.130399992494705e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8439056,
      "step": 30140
    },
    {
      "epoch": 334.94444444444446,
      "grad_norm": 8.8964392489288e-05,
      "learning_rate": 7.123535564212419e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8440480,
      "step": 30145
    },
    {
      "epoch": 335.0,
      "grad_norm": 0.0015749286394566298,
      "learning_rate": 7.116673892705611e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8441872,
      "step": 30150
    },
    {
      "epoch": 335.05555555555554,
      "grad_norm": 8.807233098195866e-05,
      "learning_rate": 7.109814979032415e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8443296,
      "step": 30155
    },
    {
      "epoch": 335.1111111111111,
      "grad_norm": 7.144756818888709e-05,
      "learning_rate": 7.102958824250577e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8444640,
      "step": 30160
    },
    {
      "epoch": 335.1666666666667,
      "grad_norm": 5.8040244766743854e-05,
      "learning_rate": 7.096105429417393e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8446048,
      "step": 30165
    },
    {
      "epoch": 335.22222222222223,
      "grad_norm": 0.00023978512035682797,
      "learning_rate": 7.0892547955897506e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8447488,
      "step": 30170
    },
    {
      "epoch": 335.27777777777777,
      "grad_norm": 0.000160745665198192,
      "learning_rate": 7.0824069238241e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8448912,
      "step": 30175
    },
    {
      "epoch": 335.3333333333333,
      "grad_norm": 0.00010369186202296987,
      "learning_rate": 7.075561815176462e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8450240,
      "step": 30180
    },
    {
      "epoch": 335.3888888888889,
      "grad_norm": 9.554625285090879e-05,
      "learning_rate": 7.068719470702445e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8451632,
      "step": 30185
    },
    {
      "epoch": 335.44444444444446,
      "grad_norm": 0.0002996854018419981,
      "learning_rate": 7.061879891457229e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8452992,
      "step": 30190
    },
    {
      "epoch": 335.5,
      "grad_norm": 7.222851127153262e-05,
      "learning_rate": 7.0550430784955515e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8454400,
      "step": 30195
    },
    {
      "epoch": 335.55555555555554,
      "grad_norm": 0.00013975177716929466,
      "learning_rate": 7.048209032871752e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8455776,
      "step": 30200
    },
    {
      "epoch": 335.55555555555554,
      "eval_loss": 0.8599721789360046,
      "eval_runtime": 1.2513,
      "eval_samples_per_second": 31.968,
      "eval_steps_per_second": 15.984,
      "num_input_tokens_seen": 8455776,
      "step": 30200
    },
    {
      "epoch": 335.6111111111111,
      "grad_norm": 0.01630767621099949,
      "learning_rate": 7.0413777556397055e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8457232,
      "step": 30205
    },
    {
      "epoch": 335.6666666666667,
      "grad_norm": 0.0002666291838977486,
      "learning_rate": 7.0345492478528925e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8458608,
      "step": 30210
    },
    {
      "epoch": 335.72222222222223,
      "grad_norm": 0.00014250331150833517,
      "learning_rate": 7.02772351056436e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8460016,
      "step": 30215
    },
    {
      "epoch": 335.77777777777777,
      "grad_norm": 0.00028123214724473655,
      "learning_rate": 7.020900544826709e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8461360,
      "step": 30220
    },
    {
      "epoch": 335.8333333333333,
      "grad_norm": 1.839912511059083e-05,
      "learning_rate": 7.014080351692134e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8462800,
      "step": 30225
    },
    {
      "epoch": 335.8888888888889,
      "grad_norm": 0.0014411750016734004,
      "learning_rate": 7.0072629322124024e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8464192,
      "step": 30230
    },
    {
      "epoch": 335.94444444444446,
      "grad_norm": 4.874527803622186e-05,
      "learning_rate": 7.000448287438827e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8465616,
      "step": 30235
    },
    {
      "epoch": 336.0,
      "grad_norm": 0.0010960783110931516,
      "learning_rate": 6.993636418422331e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8466992,
      "step": 30240
    },
    {
      "epoch": 336.05555555555554,
      "grad_norm": 0.004604854620993137,
      "learning_rate": 6.986827326213383e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8468416,
      "step": 30245
    },
    {
      "epoch": 336.1111111111111,
      "grad_norm": 0.00016493590374011546,
      "learning_rate": 6.9800210118620205e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8469808,
      "step": 30250
    },
    {
      "epoch": 336.1666666666667,
      "grad_norm": 3.896892667398788e-05,
      "learning_rate": 6.973217476417876e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8471216,
      "step": 30255
    },
    {
      "epoch": 336.22222222222223,
      "grad_norm": 0.00013294444943312556,
      "learning_rate": 6.96641672093013e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8472608,
      "step": 30260
    },
    {
      "epoch": 336.27777777777777,
      "grad_norm": 5.774403325631283e-05,
      "learning_rate": 6.95961874644755e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8473984,
      "step": 30265
    },
    {
      "epoch": 336.3333333333333,
      "grad_norm": 0.0001095091356546618,
      "learning_rate": 6.952823554018476e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8475424,
      "step": 30270
    },
    {
      "epoch": 336.3888888888889,
      "grad_norm": 5.051286279922351e-05,
      "learning_rate": 6.946031144690798e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8476848,
      "step": 30275
    },
    {
      "epoch": 336.44444444444446,
      "grad_norm": 0.0001526860287413001,
      "learning_rate": 6.939241519512005e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8478256,
      "step": 30280
    },
    {
      "epoch": 336.5,
      "grad_norm": 6.790834595449269e-05,
      "learning_rate": 6.932454679529129e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8479696,
      "step": 30285
    },
    {
      "epoch": 336.55555555555554,
      "grad_norm": 6.864530587336048e-05,
      "learning_rate": 6.925670625788791e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8481104,
      "step": 30290
    },
    {
      "epoch": 336.6111111111111,
      "grad_norm": 8.187473576981574e-05,
      "learning_rate": 6.918889359337186e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8482496,
      "step": 30295
    },
    {
      "epoch": 336.6666666666667,
      "grad_norm": 0.0007354724803008139,
      "learning_rate": 6.912110881220058e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8483856,
      "step": 30300
    },
    {
      "epoch": 336.72222222222223,
      "grad_norm": 1.535751471237745e-05,
      "learning_rate": 6.905335192482735e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8485248,
      "step": 30305
    },
    {
      "epoch": 336.77777777777777,
      "grad_norm": 0.00013488925469573587,
      "learning_rate": 6.8985622941701275e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8486624,
      "step": 30310
    },
    {
      "epoch": 336.8333333333333,
      "grad_norm": 8.422742394031957e-05,
      "learning_rate": 6.89179218732669e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8487984,
      "step": 30315
    },
    {
      "epoch": 336.8888888888889,
      "grad_norm": 0.00013774778926745057,
      "learning_rate": 6.8850248729964595e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8489392,
      "step": 30320
    },
    {
      "epoch": 336.94444444444446,
      "grad_norm": 2.1983934857416898e-05,
      "learning_rate": 6.8782603522230314e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8490768,
      "step": 30325
    },
    {
      "epoch": 337.0,
      "grad_norm": 0.00013554828183259815,
      "learning_rate": 6.871498626049591e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8492160,
      "step": 30330
    },
    {
      "epoch": 337.05555555555554,
      "grad_norm": 0.00012642325600609183,
      "learning_rate": 6.8647396955188875e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8493584,
      "step": 30335
    },
    {
      "epoch": 337.1111111111111,
      "grad_norm": 0.00012586031516548246,
      "learning_rate": 6.857983561673218e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8494992,
      "step": 30340
    },
    {
      "epoch": 337.1666666666667,
      "grad_norm": 4.712257214123383e-05,
      "learning_rate": 6.851230225554467e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8496384,
      "step": 30345
    },
    {
      "epoch": 337.22222222222223,
      "grad_norm": 5.573932139668614e-05,
      "learning_rate": 6.8444796882040946e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8497792,
      "step": 30350
    },
    {
      "epoch": 337.27777777777777,
      "grad_norm": 7.244494190672413e-05,
      "learning_rate": 6.837731950663106e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8499168,
      "step": 30355
    },
    {
      "epoch": 337.3333333333333,
      "grad_norm": 0.00016062379290815443,
      "learning_rate": 6.830987013972098e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8500544,
      "step": 30360
    },
    {
      "epoch": 337.3888888888889,
      "grad_norm": 0.00029074735357426107,
      "learning_rate": 6.82424487917121e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8501968,
      "step": 30365
    },
    {
      "epoch": 337.44444444444446,
      "grad_norm": 0.00013799894077237695,
      "learning_rate": 6.8175055473001735e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8503360,
      "step": 30370
    },
    {
      "epoch": 337.5,
      "grad_norm": 2.3077613150235265e-05,
      "learning_rate": 6.8107690193982855e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8504752,
      "step": 30375
    },
    {
      "epoch": 337.55555555555554,
      "grad_norm": 5.1369624998187646e-05,
      "learning_rate": 6.804035296504385e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8506160,
      "step": 30380
    },
    {
      "epoch": 337.6111111111111,
      "grad_norm": 3.359504626132548e-05,
      "learning_rate": 6.797304379656916e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8507552,
      "step": 30385
    },
    {
      "epoch": 337.6666666666667,
      "grad_norm": 3.73289076378569e-05,
      "learning_rate": 6.790576269893861e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8508960,
      "step": 30390
    },
    {
      "epoch": 337.72222222222223,
      "grad_norm": 0.0008028990705497563,
      "learning_rate": 6.783850968252772e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8510352,
      "step": 30395
    },
    {
      "epoch": 337.77777777777777,
      "grad_norm": 0.00019037924357689917,
      "learning_rate": 6.777128475770789e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8511760,
      "step": 30400
    },
    {
      "epoch": 337.77777777777777,
      "eval_loss": 0.8755990266799927,
      "eval_runtime": 1.2363,
      "eval_samples_per_second": 32.354,
      "eval_steps_per_second": 16.177,
      "num_input_tokens_seen": 8511760,
      "step": 30400
    },
    {
      "epoch": 337.8333333333333,
      "grad_norm": 0.00026865286054089665,
      "learning_rate": 6.77040879348459e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8513184,
      "step": 30405
    },
    {
      "epoch": 337.8888888888889,
      "grad_norm": 0.00020653718092944473,
      "learning_rate": 6.763691922430443e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8514560,
      "step": 30410
    },
    {
      "epoch": 337.94444444444446,
      "grad_norm": 4.0872466342989355e-05,
      "learning_rate": 6.756977863644178e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8515952,
      "step": 30415
    },
    {
      "epoch": 338.0,
      "grad_norm": 7.273762457771227e-05,
      "learning_rate": 6.7502666181611804e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8517360,
      "step": 30420
    },
    {
      "epoch": 338.05555555555554,
      "grad_norm": 4.79725276818499e-05,
      "learning_rate": 6.743558187016405e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8518800,
      "step": 30425
    },
    {
      "epoch": 338.1111111111111,
      "grad_norm": 0.0003687023709062487,
      "learning_rate": 6.7368525712443925e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8520192,
      "step": 30430
    },
    {
      "epoch": 338.1666666666667,
      "grad_norm": 0.00022242273553274572,
      "learning_rate": 6.7301497718792155e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8521600,
      "step": 30435
    },
    {
      "epoch": 338.22222222222223,
      "grad_norm": 9.783949644770473e-05,
      "learning_rate": 6.723449789954544e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8522960,
      "step": 30440
    },
    {
      "epoch": 338.27777777777777,
      "grad_norm": 0.0027617854066193104,
      "learning_rate": 6.716752626503586e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8524384,
      "step": 30445
    },
    {
      "epoch": 338.3333333333333,
      "grad_norm": 0.0002343245141673833,
      "learning_rate": 6.710058282559131e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8525824,
      "step": 30450
    },
    {
      "epoch": 338.3888888888889,
      "grad_norm": 6.458385905716568e-05,
      "learning_rate": 6.703366759153545e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8527216,
      "step": 30455
    },
    {
      "epoch": 338.44444444444446,
      "grad_norm": 0.00037366882315836847,
      "learning_rate": 6.6966780573187335e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8528624,
      "step": 30460
    },
    {
      "epoch": 338.5,
      "grad_norm": 6.700481026200578e-05,
      "learning_rate": 6.689992178086174e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8530032,
      "step": 30465
    },
    {
      "epoch": 338.55555555555554,
      "grad_norm": 8.02933718659915e-05,
      "learning_rate": 6.683309122486925e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8531376,
      "step": 30470
    },
    {
      "epoch": 338.6111111111111,
      "grad_norm": 1.7848262359621003e-05,
      "learning_rate": 6.676628891551584e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8532736,
      "step": 30475
    },
    {
      "epoch": 338.6666666666667,
      "grad_norm": 0.00017927800945471972,
      "learning_rate": 6.6699514863103385e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8534160,
      "step": 30480
    },
    {
      "epoch": 338.72222222222223,
      "grad_norm": 4.760780939250253e-05,
      "learning_rate": 6.663276907792921e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8535584,
      "step": 30485
    },
    {
      "epoch": 338.77777777777777,
      "grad_norm": 4.684444866143167e-05,
      "learning_rate": 6.656605157028634e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8537024,
      "step": 30490
    },
    {
      "epoch": 338.8333333333333,
      "grad_norm": 7.964477117639035e-05,
      "learning_rate": 6.649936235046358e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8538352,
      "step": 30495
    },
    {
      "epoch": 338.8888888888889,
      "grad_norm": 0.0017864002147689462,
      "learning_rate": 6.643270142874508e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8539792,
      "step": 30500
    },
    {
      "epoch": 338.94444444444446,
      "grad_norm": 0.011930335313081741,
      "learning_rate": 6.636606881541094e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8541184,
      "step": 30505
    },
    {
      "epoch": 339.0,
      "grad_norm": 0.00027728028362616897,
      "learning_rate": 6.629946452073662e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8542576,
      "step": 30510
    },
    {
      "epoch": 339.05555555555554,
      "grad_norm": 6.707572902087122e-05,
      "learning_rate": 6.6232888554993375e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8543968,
      "step": 30515
    },
    {
      "epoch": 339.1111111111111,
      "grad_norm": 0.0002384337130934,
      "learning_rate": 6.616634092844817e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8545392,
      "step": 30520
    },
    {
      "epoch": 339.1666666666667,
      "grad_norm": 0.00014959405234549195,
      "learning_rate": 6.609982165136331e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8546784,
      "step": 30525
    },
    {
      "epoch": 339.22222222222223,
      "grad_norm": 0.0001433424185961485,
      "learning_rate": 6.603333073399706e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8548192,
      "step": 30530
    },
    {
      "epoch": 339.27777777777777,
      "grad_norm": 0.00017840362852439284,
      "learning_rate": 6.596686818660308e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8549568,
      "step": 30535
    },
    {
      "epoch": 339.3333333333333,
      "grad_norm": 2.3984732251847163e-05,
      "learning_rate": 6.590043401943066e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8550976,
      "step": 30540
    },
    {
      "epoch": 339.3888888888889,
      "grad_norm": 0.008368026465177536,
      "learning_rate": 6.583402824272494e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8552368,
      "step": 30545
    },
    {
      "epoch": 339.44444444444446,
      "grad_norm": 5.802044688607566e-05,
      "learning_rate": 6.576765086672634e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8553792,
      "step": 30550
    },
    {
      "epoch": 339.5,
      "grad_norm": 4.052025906275958e-05,
      "learning_rate": 6.57013019016712e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8555216,
      "step": 30555
    },
    {
      "epoch": 339.55555555555554,
      "grad_norm": 0.0007217798847705126,
      "learning_rate": 6.563498135779142e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8556656,
      "step": 30560
    },
    {
      "epoch": 339.6111111111111,
      "grad_norm": 4.104730760445818e-05,
      "learning_rate": 6.556868924531431e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8558048,
      "step": 30565
    },
    {
      "epoch": 339.6666666666667,
      "grad_norm": 4.470913700060919e-05,
      "learning_rate": 6.550242557446304e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8559408,
      "step": 30570
    },
    {
      "epoch": 339.72222222222223,
      "grad_norm": 0.0003423923044465482,
      "learning_rate": 6.543619035545634e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8560768,
      "step": 30575
    },
    {
      "epoch": 339.77777777777777,
      "grad_norm": 0.00045226330985315144,
      "learning_rate": 6.53699835985084e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8562208,
      "step": 30580
    },
    {
      "epoch": 339.8333333333333,
      "grad_norm": 0.000293070130283013,
      "learning_rate": 6.530380531382927e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8563584,
      "step": 30585
    },
    {
      "epoch": 339.8888888888889,
      "grad_norm": 0.0001365298085147515,
      "learning_rate": 6.523765551162433e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8564976,
      "step": 30590
    },
    {
      "epoch": 339.94444444444446,
      "grad_norm": 0.00036277988692745566,
      "learning_rate": 6.517153420209476e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8566400,
      "step": 30595
    },
    {
      "epoch": 340.0,
      "grad_norm": 0.0005279176402837038,
      "learning_rate": 6.510544139543739e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8567792,
      "step": 30600
    },
    {
      "epoch": 340.0,
      "eval_loss": 0.8545724153518677,
      "eval_runtime": 1.234,
      "eval_samples_per_second": 32.414,
      "eval_steps_per_second": 16.207,
      "num_input_tokens_seen": 8567792,
      "step": 30600
    },
    {
      "epoch": 340.05555555555554,
      "grad_norm": 0.0001235105301020667,
      "learning_rate": 6.503937710184452e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8569168,
      "step": 30605
    },
    {
      "epoch": 340.1111111111111,
      "grad_norm": 0.0001163750421255827,
      "learning_rate": 6.4973341331503954e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8570576,
      "step": 30610
    },
    {
      "epoch": 340.1666666666667,
      "grad_norm": 1.722585511743091e-05,
      "learning_rate": 6.490733409459942e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8572000,
      "step": 30615
    },
    {
      "epoch": 340.22222222222223,
      "grad_norm": 6.079008016968146e-05,
      "learning_rate": 6.484135540130995e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8573408,
      "step": 30620
    },
    {
      "epoch": 340.27777777777777,
      "grad_norm": 0.0017992990324273705,
      "learning_rate": 6.4775405261810364e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8574816,
      "step": 30625
    },
    {
      "epoch": 340.3333333333333,
      "grad_norm": 0.00048819257062859833,
      "learning_rate": 6.470948368627092e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8576208,
      "step": 30630
    },
    {
      "epoch": 340.3888888888889,
      "grad_norm": 1.4576487956219353e-05,
      "learning_rate": 6.464359068485756e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8577632,
      "step": 30635
    },
    {
      "epoch": 340.44444444444446,
      "grad_norm": 0.00014258229930419475,
      "learning_rate": 6.457772626773195e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8579056,
      "step": 30640
    },
    {
      "epoch": 340.5,
      "grad_norm": 0.00031763443257659674,
      "learning_rate": 6.451189044505104e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8580416,
      "step": 30645
    },
    {
      "epoch": 340.55555555555554,
      "grad_norm": 2.833023427228909e-05,
      "learning_rate": 6.44460832269676e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8581824,
      "step": 30650
    },
    {
      "epoch": 340.6111111111111,
      "grad_norm": 0.00028682261472567916,
      "learning_rate": 6.438030462363001e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8583200,
      "step": 30655
    },
    {
      "epoch": 340.6666666666667,
      "grad_norm": 0.00013895960000809282,
      "learning_rate": 6.431455464518205e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8584560,
      "step": 30660
    },
    {
      "epoch": 340.72222222222223,
      "grad_norm": 0.0013436718145385385,
      "learning_rate": 6.424883330176326e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8585936,
      "step": 30665
    },
    {
      "epoch": 340.77777777777777,
      "grad_norm": 0.013908260501921177,
      "learning_rate": 6.418314060350864e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8587328,
      "step": 30670
    },
    {
      "epoch": 340.8333333333333,
      "grad_norm": 3.727681178133935e-05,
      "learning_rate": 6.4117476560548895e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8588688,
      "step": 30675
    },
    {
      "epoch": 340.8888888888889,
      "grad_norm": 0.00038336351281031966,
      "learning_rate": 6.405184118301016e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8590128,
      "step": 30680
    },
    {
      "epoch": 340.94444444444446,
      "grad_norm": 0.002202624222263694,
      "learning_rate": 6.398623448101434e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8591504,
      "step": 30685
    },
    {
      "epoch": 341.0,
      "grad_norm": 0.0008926786249503493,
      "learning_rate": 6.392065646467871e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8592944,
      "step": 30690
    },
    {
      "epoch": 341.05555555555554,
      "grad_norm": 0.0006267457501962781,
      "learning_rate": 6.385510714411632e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8594352,
      "step": 30695
    },
    {
      "epoch": 341.1111111111111,
      "grad_norm": 6.595670129172504e-05,
      "learning_rate": 6.378958652943559e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8595776,
      "step": 30700
    },
    {
      "epoch": 341.1666666666667,
      "grad_norm": 0.00043826535693369806,
      "learning_rate": 6.3724094630740776e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8597152,
      "step": 30705
    },
    {
      "epoch": 341.22222222222223,
      "grad_norm": 0.00044160205288790166,
      "learning_rate": 6.365863145813136e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8598608,
      "step": 30710
    },
    {
      "epoch": 341.27777777777777,
      "grad_norm": 4.033915320178494e-05,
      "learning_rate": 6.359319702170269e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8600016,
      "step": 30715
    },
    {
      "epoch": 341.3333333333333,
      "grad_norm": 0.00015112967230379581,
      "learning_rate": 6.352779133154566e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8601424,
      "step": 30720
    },
    {
      "epoch": 341.3888888888889,
      "grad_norm": 5.748527473770082e-05,
      "learning_rate": 6.346241439774648e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8602848,
      "step": 30725
    },
    {
      "epoch": 341.44444444444446,
      "grad_norm": 0.000819922424852848,
      "learning_rate": 6.339706623038716e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8604256,
      "step": 30730
    },
    {
      "epoch": 341.5,
      "grad_norm": 0.004310184624046087,
      "learning_rate": 6.333174683954532e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8605616,
      "step": 30735
    },
    {
      "epoch": 341.55555555555554,
      "grad_norm": 0.00018864414596464485,
      "learning_rate": 6.326645623529387e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8606976,
      "step": 30740
    },
    {
      "epoch": 341.6111111111111,
      "grad_norm": 8.792054723016918e-05,
      "learning_rate": 6.320119442770156e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8608400,
      "step": 30745
    },
    {
      "epoch": 341.6666666666667,
      "grad_norm": 0.0003241691447328776,
      "learning_rate": 6.313596142683254e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8609792,
      "step": 30750
    },
    {
      "epoch": 341.72222222222223,
      "grad_norm": 5.280647746985778e-05,
      "learning_rate": 6.307075724274647e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8611200,
      "step": 30755
    },
    {
      "epoch": 341.77777777777777,
      "grad_norm": 5.07158983964473e-05,
      "learning_rate": 6.300558188549882e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8612592,
      "step": 30760
    },
    {
      "epoch": 341.8333333333333,
      "grad_norm": 2.6848720153793693e-05,
      "learning_rate": 6.29404353651403e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8613984,
      "step": 30765
    },
    {
      "epoch": 341.8888888888889,
      "grad_norm": 7.687027391511947e-05,
      "learning_rate": 6.287531769171737e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8615376,
      "step": 30770
    },
    {
      "epoch": 341.94444444444446,
      "grad_norm": 8.091750351013616e-05,
      "learning_rate": 6.2810228875272045e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8616752,
      "step": 30775
    },
    {
      "epoch": 342.0,
      "grad_norm": 0.00021155459398869425,
      "learning_rate": 6.274516892584179e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8618160,
      "step": 30780
    },
    {
      "epoch": 342.05555555555554,
      "grad_norm": 0.003174671670421958,
      "learning_rate": 6.268013785345969e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8619568,
      "step": 30785
    },
    {
      "epoch": 342.1111111111111,
      "grad_norm": 0.00014749581168871373,
      "learning_rate": 6.26151356681543e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8620928,
      "step": 30790
    },
    {
      "epoch": 342.1666666666667,
      "grad_norm": 0.00019548760610632598,
      "learning_rate": 6.255016237994981e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8622320,
      "step": 30795
    },
    {
      "epoch": 342.22222222222223,
      "grad_norm": 0.00047238069237209857,
      "learning_rate": 6.248521799886603e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8623728,
      "step": 30800
    },
    {
      "epoch": 342.22222222222223,
      "eval_loss": 0.8786500692367554,
      "eval_runtime": 1.2475,
      "eval_samples_per_second": 32.064,
      "eval_steps_per_second": 16.032,
      "num_input_tokens_seen": 8623728,
      "step": 30800
    },
    {
      "epoch": 342.27777777777777,
      "grad_norm": 2.0634337488445453e-05,
      "learning_rate": 6.242030253491798e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8625088,
      "step": 30805
    },
    {
      "epoch": 342.3333333333333,
      "grad_norm": 0.0003850400389637798,
      "learning_rate": 6.235541599811656e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8626528,
      "step": 30810
    },
    {
      "epoch": 342.3888888888889,
      "grad_norm": 0.0003361934213899076,
      "learning_rate": 6.229055839846814e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8627920,
      "step": 30815
    },
    {
      "epoch": 342.44444444444446,
      "grad_norm": 0.00045064312871545553,
      "learning_rate": 6.222572974597455e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8629312,
      "step": 30820
    },
    {
      "epoch": 342.5,
      "grad_norm": 5.5904056353028864e-05,
      "learning_rate": 6.216093005063306e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8630736,
      "step": 30825
    },
    {
      "epoch": 342.55555555555554,
      "grad_norm": 0.0002179174480261281,
      "learning_rate": 6.209615932243678e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8632112,
      "step": 30830
    },
    {
      "epoch": 342.6111111111111,
      "grad_norm": 0.0002525066665839404,
      "learning_rate": 6.203141757137399e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8633536,
      "step": 30835
    },
    {
      "epoch": 342.6666666666667,
      "grad_norm": 4.7367520892294124e-05,
      "learning_rate": 6.196670480742886e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8634976,
      "step": 30840
    },
    {
      "epoch": 342.72222222222223,
      "grad_norm": 0.00013627229782287031,
      "learning_rate": 6.190202104058074e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8636400,
      "step": 30845
    },
    {
      "epoch": 342.77777777777777,
      "grad_norm": 0.00039090332575142384,
      "learning_rate": 6.183736628080475e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8637792,
      "step": 30850
    },
    {
      "epoch": 342.8333333333333,
      "grad_norm": 0.0001371364196529612,
      "learning_rate": 6.177274053807155e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8639168,
      "step": 30855
    },
    {
      "epoch": 342.8888888888889,
      "grad_norm": 0.001232688780874014,
      "learning_rate": 6.170814382234713e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8640608,
      "step": 30860
    },
    {
      "epoch": 342.94444444444446,
      "grad_norm": 4.3582305806921795e-05,
      "learning_rate": 6.16435761435932e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8642016,
      "step": 30865
    },
    {
      "epoch": 343.0,
      "grad_norm": 0.0002299346524523571,
      "learning_rate": 6.157903751176681e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8643456,
      "step": 30870
    },
    {
      "epoch": 343.05555555555554,
      "grad_norm": 5.207226786296815e-05,
      "learning_rate": 6.151452793682066e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8644880,
      "step": 30875
    },
    {
      "epoch": 343.1111111111111,
      "grad_norm": 2.2047692255000584e-05,
      "learning_rate": 6.145004742870305e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8646304,
      "step": 30880
    },
    {
      "epoch": 343.1666666666667,
      "grad_norm": 6.009902790538035e-05,
      "learning_rate": 6.138559599735752e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8647712,
      "step": 30885
    },
    {
      "epoch": 343.22222222222223,
      "grad_norm": 9.289196896133944e-05,
      "learning_rate": 6.132117365272344e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8649104,
      "step": 30890
    },
    {
      "epoch": 343.27777777777777,
      "grad_norm": 3.461465530563146e-05,
      "learning_rate": 6.125678040473545e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8650512,
      "step": 30895
    },
    {
      "epoch": 343.3333333333333,
      "grad_norm": 0.0006724708364345133,
      "learning_rate": 6.1192416263323755e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8651888,
      "step": 30900
    },
    {
      "epoch": 343.3888888888889,
      "grad_norm": 1.3004488209844567e-05,
      "learning_rate": 6.112808123841424e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8653296,
      "step": 30905
    },
    {
      "epoch": 343.44444444444446,
      "grad_norm": 4.215330773149617e-05,
      "learning_rate": 6.106377533992805e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8654704,
      "step": 30910
    },
    {
      "epoch": 343.5,
      "grad_norm": 0.00021528812067117542,
      "learning_rate": 6.099949857778204e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8656064,
      "step": 30915
    },
    {
      "epoch": 343.55555555555554,
      "grad_norm": 0.000211497928830795,
      "learning_rate": 6.093525096188852e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8657424,
      "step": 30920
    },
    {
      "epoch": 343.6111111111111,
      "grad_norm": 0.0002890844189096242,
      "learning_rate": 6.087103250215518e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8658864,
      "step": 30925
    },
    {
      "epoch": 343.6666666666667,
      "grad_norm": 0.0011222101747989655,
      "learning_rate": 6.080684320848537e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8660272,
      "step": 30930
    },
    {
      "epoch": 343.72222222222223,
      "grad_norm": 0.0003548035747371614,
      "learning_rate": 6.074268309077794e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8661664,
      "step": 30935
    },
    {
      "epoch": 343.77777777777777,
      "grad_norm": 0.00017679441953077912,
      "learning_rate": 6.067855215892709e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8663120,
      "step": 30940
    },
    {
      "epoch": 343.8333333333333,
      "grad_norm": 0.00028583398670889437,
      "learning_rate": 6.061445042282271e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8664528,
      "step": 30945
    },
    {
      "epoch": 343.8888888888889,
      "grad_norm": 0.00012921479356009513,
      "learning_rate": 6.055037789234999e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8665968,
      "step": 30950
    },
    {
      "epoch": 343.94444444444446,
      "grad_norm": 3.560267941793427e-05,
      "learning_rate": 6.048633457738975e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8667360,
      "step": 30955
    },
    {
      "epoch": 344.0,
      "grad_norm": 0.0001647283206693828,
      "learning_rate": 6.042232048781837e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8668736,
      "step": 30960
    },
    {
      "epoch": 344.05555555555554,
      "grad_norm": 0.0002586038608569652,
      "learning_rate": 6.035833563350757e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8670112,
      "step": 30965
    },
    {
      "epoch": 344.1111111111111,
      "grad_norm": 0.00017555159865878522,
      "learning_rate": 6.0294380024324525e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8671520,
      "step": 30970
    },
    {
      "epoch": 344.1666666666667,
      "grad_norm": 0.0003423740272410214,
      "learning_rate": 6.023045367013213e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8672896,
      "step": 30975
    },
    {
      "epoch": 344.22222222222223,
      "grad_norm": 0.002603646833449602,
      "learning_rate": 6.016655658078851e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8674288,
      "step": 30980
    },
    {
      "epoch": 344.27777777777777,
      "grad_norm": 9.061284072231501e-05,
      "learning_rate": 6.010268876614753e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8675696,
      "step": 30985
    },
    {
      "epoch": 344.3333333333333,
      "grad_norm": 3.807245229836553e-05,
      "learning_rate": 6.0038850236058266e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8677120,
      "step": 30990
    },
    {
      "epoch": 344.3888888888889,
      "grad_norm": 5.121084177517332e-05,
      "learning_rate": 5.997504100036549e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8678512,
      "step": 30995
    },
    {
      "epoch": 344.44444444444446,
      "grad_norm": 0.00025277971872128546,
      "learning_rate": 5.991126106890949e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8679920,
      "step": 31000
    },
    {
      "epoch": 344.44444444444446,
      "eval_loss": 0.90406334400177,
      "eval_runtime": 1.2385,
      "eval_samples_per_second": 32.297,
      "eval_steps_per_second": 16.148,
      "num_input_tokens_seen": 8679920,
      "step": 31000
    },
    {
      "epoch": 344.5,
      "grad_norm": 2.67962968791835e-05,
      "learning_rate": 5.984751045152576e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8681360,
      "step": 31005
    },
    {
      "epoch": 344.55555555555554,
      "grad_norm": 5.5524931667605415e-05,
      "learning_rate": 5.978378915804553e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8682736,
      "step": 31010
    },
    {
      "epoch": 344.6111111111111,
      "grad_norm": 8.84058972587809e-05,
      "learning_rate": 5.972009719829547e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8684128,
      "step": 31015
    },
    {
      "epoch": 344.6666666666667,
      "grad_norm": 0.00020159591804258525,
      "learning_rate": 5.965643458209755e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8685536,
      "step": 31020
    },
    {
      "epoch": 344.72222222222223,
      "grad_norm": 4.758673094329424e-05,
      "learning_rate": 5.95928013192695e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8686912,
      "step": 31025
    },
    {
      "epoch": 344.77777777777777,
      "grad_norm": 0.0001358016743324697,
      "learning_rate": 5.952919741962423e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8688336,
      "step": 31030
    },
    {
      "epoch": 344.8333333333333,
      "grad_norm": 5.748603871325031e-05,
      "learning_rate": 5.946562289297042e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8689760,
      "step": 31035
    },
    {
      "epoch": 344.8888888888889,
      "grad_norm": 7.243036816362292e-05,
      "learning_rate": 5.9402077749111855e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8691152,
      "step": 31040
    },
    {
      "epoch": 344.94444444444446,
      "grad_norm": 0.00019228130986448377,
      "learning_rate": 5.933856199784821e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8692560,
      "step": 31045
    },
    {
      "epoch": 345.0,
      "grad_norm": 0.00020058127120137215,
      "learning_rate": 5.927507564897419e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8693984,
      "step": 31050
    },
    {
      "epoch": 345.05555555555554,
      "grad_norm": 0.0006332481862045825,
      "learning_rate": 5.9211618712280395e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8695376,
      "step": 31055
    },
    {
      "epoch": 345.1111111111111,
      "grad_norm": 2.148674138879869e-05,
      "learning_rate": 5.914819119755255e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8696784,
      "step": 31060
    },
    {
      "epoch": 345.1666666666667,
      "grad_norm": 0.0001697619736660272,
      "learning_rate": 5.908479311457205e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8698176,
      "step": 31065
    },
    {
      "epoch": 345.22222222222223,
      "grad_norm": 0.0008064741850830615,
      "learning_rate": 5.902142447311559e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8699616,
      "step": 31070
    },
    {
      "epoch": 345.27777777777777,
      "grad_norm": 9.334341302746907e-05,
      "learning_rate": 5.895808528295546e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8701040,
      "step": 31075
    },
    {
      "epoch": 345.3333333333333,
      "grad_norm": 0.0004976377822458744,
      "learning_rate": 5.889477555385941e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8702448,
      "step": 31080
    },
    {
      "epoch": 345.3888888888889,
      "grad_norm": 0.0012417472898960114,
      "learning_rate": 5.883149529559051e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8703840,
      "step": 31085
    },
    {
      "epoch": 345.44444444444446,
      "grad_norm": 8.996045653475448e-05,
      "learning_rate": 5.876824451790738e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8705200,
      "step": 31090
    },
    {
      "epoch": 345.5,
      "grad_norm": 9.899412543745711e-05,
      "learning_rate": 5.87050232305642e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8706592,
      "step": 31095
    },
    {
      "epoch": 345.55555555555554,
      "grad_norm": 0.0002752681029960513,
      "learning_rate": 5.864183144331034e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8707984,
      "step": 31100
    },
    {
      "epoch": 345.6111111111111,
      "grad_norm": 0.0007525334367528558,
      "learning_rate": 5.857866916589089e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8709360,
      "step": 31105
    },
    {
      "epoch": 345.6666666666667,
      "grad_norm": 6.656025652773678e-05,
      "learning_rate": 5.8515536408046216e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8710816,
      "step": 31110
    },
    {
      "epoch": 345.72222222222223,
      "grad_norm": 1.634328145883046e-05,
      "learning_rate": 5.845243317951208e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8712224,
      "step": 31115
    },
    {
      "epoch": 345.77777777777777,
      "grad_norm": 0.0003004850586876273,
      "learning_rate": 5.838935949001997e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8713632,
      "step": 31120
    },
    {
      "epoch": 345.8333333333333,
      "grad_norm": 7.715208630543202e-05,
      "learning_rate": 5.8326315349296476e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8714992,
      "step": 31125
    },
    {
      "epoch": 345.8888888888889,
      "grad_norm": 0.00022755445388611406,
      "learning_rate": 5.826330076706396e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8716400,
      "step": 31130
    },
    {
      "epoch": 345.94444444444446,
      "grad_norm": 1.21543262139312e-05,
      "learning_rate": 5.820031575303988e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8717792,
      "step": 31135
    },
    {
      "epoch": 346.0,
      "grad_norm": 6.553650746354833e-05,
      "learning_rate": 5.813736031693745e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8719184,
      "step": 31140
    },
    {
      "epoch": 346.05555555555554,
      "grad_norm": 0.00027902296278625727,
      "learning_rate": 5.807443446846522e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8720608,
      "step": 31145
    },
    {
      "epoch": 346.1111111111111,
      "grad_norm": 0.00011214410187676549,
      "learning_rate": 5.801153821732699e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8722032,
      "step": 31150
    },
    {
      "epoch": 346.1666666666667,
      "grad_norm": 1.2080739907105453e-05,
      "learning_rate": 5.794867157322229e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8723472,
      "step": 31155
    },
    {
      "epoch": 346.22222222222223,
      "grad_norm": 0.00010356965503888205,
      "learning_rate": 5.788583454584593e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8724880,
      "step": 31160
    },
    {
      "epoch": 346.27777777777777,
      "grad_norm": 2.799814501486253e-05,
      "learning_rate": 5.7823027144888075e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8726272,
      "step": 31165
    },
    {
      "epoch": 346.3333333333333,
      "grad_norm": 0.00021992296387907118,
      "learning_rate": 5.776024938003455e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8727712,
      "step": 31170
    },
    {
      "epoch": 346.3888888888889,
      "grad_norm": 5.40833716513589e-05,
      "learning_rate": 5.7697501260966345e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8729088,
      "step": 31175
    },
    {
      "epoch": 346.44444444444446,
      "grad_norm": 7.262943108798936e-05,
      "learning_rate": 5.7634782797360145e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8730480,
      "step": 31180
    },
    {
      "epoch": 346.5,
      "grad_norm": 0.00020789069822058082,
      "learning_rate": 5.757209399888777e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8731888,
      "step": 31185
    },
    {
      "epoch": 346.55555555555554,
      "grad_norm": 0.0031430611852556467,
      "learning_rate": 5.750943487521679e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8733264,
      "step": 31190
    },
    {
      "epoch": 346.6111111111111,
      "grad_norm": 0.000548778276424855,
      "learning_rate": 5.744680543600986e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8734624,
      "step": 31195
    },
    {
      "epoch": 346.6666666666667,
      "grad_norm": 9.380064147990197e-05,
      "learning_rate": 5.738420569092537e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8736032,
      "step": 31200
    },
    {
      "epoch": 346.6666666666667,
      "eval_loss": 0.8959587812423706,
      "eval_runtime": 1.2431,
      "eval_samples_per_second": 32.178,
      "eval_steps_per_second": 16.089,
      "num_input_tokens_seen": 8736032,
      "step": 31200
    },
    {
      "epoch": 346.72222222222223,
      "grad_norm": 0.0011324534425511956,
      "learning_rate": 5.732163564961684e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8737424,
      "step": 31205
    },
    {
      "epoch": 346.77777777777777,
      "grad_norm": 0.0002885934372898191,
      "learning_rate": 5.725909532173354e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8738784,
      "step": 31210
    },
    {
      "epoch": 346.8333333333333,
      "grad_norm": 0.00023773314023856074,
      "learning_rate": 5.719658471691977e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8740208,
      "step": 31215
    },
    {
      "epoch": 346.8888888888889,
      "grad_norm": 4.532674574875273e-05,
      "learning_rate": 5.71341038448156e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8741584,
      "step": 31220
    },
    {
      "epoch": 346.94444444444446,
      "grad_norm": 7.227906462503597e-05,
      "learning_rate": 5.707165271505635e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8742976,
      "step": 31225
    },
    {
      "epoch": 347.0,
      "grad_norm": 0.000475995650049299,
      "learning_rate": 5.700923133727271e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8744368,
      "step": 31230
    },
    {
      "epoch": 347.05555555555554,
      "grad_norm": 0.00019828724907711148,
      "learning_rate": 5.694683972109083e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 8745760,
      "step": 31235
    },
    {
      "epoch": 347.1111111111111,
      "grad_norm": 0.0014419470680877566,
      "learning_rate": 5.688447787613241e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8747136,
      "step": 31240
    },
    {
      "epoch": 347.1666666666667,
      "grad_norm": 1.7313081116299145e-05,
      "learning_rate": 5.6822145812014285e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8748528,
      "step": 31245
    },
    {
      "epoch": 347.22222222222223,
      "grad_norm": 0.001675440464168787,
      "learning_rate": 5.675984353834896e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8749952,
      "step": 31250
    },
    {
      "epoch": 347.27777777777777,
      "grad_norm": 0.00010644295252859592,
      "learning_rate": 5.66975710647441e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8751392,
      "step": 31255
    },
    {
      "epoch": 347.3333333333333,
      "grad_norm": 16.139232635498047,
      "learning_rate": 5.663532840080304e-06,
      "loss": 0.0049,
      "num_input_tokens_seen": 8752816,
      "step": 31260
    },
    {
      "epoch": 347.3888888888889,
      "grad_norm": 0.00030884516309015453,
      "learning_rate": 5.6573115556124325e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8754240,
      "step": 31265
    },
    {
      "epoch": 347.44444444444446,
      "grad_norm": 0.00016767527267802507,
      "learning_rate": 5.651093254030185e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8755616,
      "step": 31270
    },
    {
      "epoch": 347.5,
      "grad_norm": 0.00011677108705043793,
      "learning_rate": 5.644877936292514e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8756944,
      "step": 31275
    },
    {
      "epoch": 347.55555555555554,
      "grad_norm": 0.0002891678304877132,
      "learning_rate": 5.638665603357901e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8758352,
      "step": 31280
    },
    {
      "epoch": 347.6111111111111,
      "grad_norm": 0.005899432115256786,
      "learning_rate": 5.632456256184357e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8759808,
      "step": 31285
    },
    {
      "epoch": 347.6666666666667,
      "grad_norm": 0.0013826063368469477,
      "learning_rate": 5.626249895729452e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8761216,
      "step": 31290
    },
    {
      "epoch": 347.72222222222223,
      "grad_norm": 0.00010937373735941947,
      "learning_rate": 5.620046522950273e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8762624,
      "step": 31295
    },
    {
      "epoch": 347.77777777777777,
      "grad_norm": 0.00013619019591715187,
      "learning_rate": 5.613846138803464e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8764032,
      "step": 31300
    },
    {
      "epoch": 347.8333333333333,
      "grad_norm": 5.777820115326904e-05,
      "learning_rate": 5.607648744245206e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8765440,
      "step": 31305
    },
    {
      "epoch": 347.8888888888889,
      "grad_norm": 0.0008432770264334977,
      "learning_rate": 5.601454340231207e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8766816,
      "step": 31310
    },
    {
      "epoch": 347.94444444444446,
      "grad_norm": 0.00016268773470073938,
      "learning_rate": 5.595262927716724e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8768176,
      "step": 31315
    },
    {
      "epoch": 348.0,
      "grad_norm": 0.024742577224969864,
      "learning_rate": 5.589074507656561e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8769584,
      "step": 31320
    },
    {
      "epoch": 348.05555555555554,
      "grad_norm": 2.802374910970684e-05,
      "learning_rate": 5.582889081005044e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8770928,
      "step": 31325
    },
    {
      "epoch": 348.1111111111111,
      "grad_norm": 0.00028206119895912707,
      "learning_rate": 5.5767066487160316e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8772336,
      "step": 31330
    },
    {
      "epoch": 348.1666666666667,
      "grad_norm": 0.00028236390789970756,
      "learning_rate": 5.570527211742949e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8773728,
      "step": 31335
    },
    {
      "epoch": 348.22222222222223,
      "grad_norm": 8.606924529885873e-05,
      "learning_rate": 5.564350771038731e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8775168,
      "step": 31340
    },
    {
      "epoch": 348.27777777777777,
      "grad_norm": 3.4815067920135334e-05,
      "learning_rate": 5.558177327555875e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8776592,
      "step": 31345
    },
    {
      "epoch": 348.3333333333333,
      "grad_norm": 0.0004964579129591584,
      "learning_rate": 5.552006882246388e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8777952,
      "step": 31350
    },
    {
      "epoch": 348.3888888888889,
      "grad_norm": 1.560120472277049e-05,
      "learning_rate": 5.545839436061839e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8779360,
      "step": 31355
    },
    {
      "epoch": 348.44444444444446,
      "grad_norm": 0.0063384417444467545,
      "learning_rate": 5.539674989953331e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8780768,
      "step": 31360
    },
    {
      "epoch": 348.5,
      "grad_norm": 2.0196697732899338e-05,
      "learning_rate": 5.533513544871488e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8782160,
      "step": 31365
    },
    {
      "epoch": 348.55555555555554,
      "grad_norm": 8.342770888702944e-05,
      "learning_rate": 5.527355101766493e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8783584,
      "step": 31370
    },
    {
      "epoch": 348.6111111111111,
      "grad_norm": 5.440021413960494e-05,
      "learning_rate": 5.521199661588044e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8784960,
      "step": 31375
    },
    {
      "epoch": 348.6666666666667,
      "grad_norm": 3.4984324884135276e-05,
      "learning_rate": 5.5150472252853944e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8786384,
      "step": 31380
    },
    {
      "epoch": 348.72222222222223,
      "grad_norm": 0.0019710129126906395,
      "learning_rate": 5.50889779380733e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8787776,
      "step": 31385
    },
    {
      "epoch": 348.77777777777777,
      "grad_norm": 8.275095751741901e-05,
      "learning_rate": 5.5027513681021605e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8789120,
      "step": 31390
    },
    {
      "epoch": 348.8333333333333,
      "grad_norm": 8.898880332708359e-05,
      "learning_rate": 5.4966079491177545e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8790512,
      "step": 31395
    },
    {
      "epoch": 348.8888888888889,
      "grad_norm": 1.9091756257694215e-05,
      "learning_rate": 5.490467537801491e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8791888,
      "step": 31400
    },
    {
      "epoch": 348.8888888888889,
      "eval_loss": 0.8573068380355835,
      "eval_runtime": 1.245,
      "eval_samples_per_second": 32.129,
      "eval_steps_per_second": 16.064,
      "num_input_tokens_seen": 8791888,
      "step": 31400
    },
    {
      "epoch": 348.94444444444446,
      "grad_norm": 3.8239031709963456e-05,
      "learning_rate": 5.484330135100313e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8793312,
      "step": 31405
    },
    {
      "epoch": 349.0,
      "grad_norm": 2.626134482852649e-05,
      "learning_rate": 5.4781957419606785e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8794656,
      "step": 31410
    },
    {
      "epoch": 349.05555555555554,
      "grad_norm": 8.452557085547596e-05,
      "learning_rate": 5.472064359328577e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8796016,
      "step": 31415
    },
    {
      "epoch": 349.1111111111111,
      "grad_norm": 9.394881635671481e-05,
      "learning_rate": 5.4659359881495565e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8797424,
      "step": 31420
    },
    {
      "epoch": 349.1666666666667,
      "grad_norm": 2.1161671611480415e-05,
      "learning_rate": 5.4598106293686916e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8798848,
      "step": 31425
    },
    {
      "epoch": 349.22222222222223,
      "grad_norm": 0.0001249487540917471,
      "learning_rate": 5.45368828393058e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8800256,
      "step": 31430
    },
    {
      "epoch": 349.27777777777777,
      "grad_norm": 0.0012011852813884616,
      "learning_rate": 5.44756895277937e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8801632,
      "step": 31435
    },
    {
      "epoch": 349.3333333333333,
      "grad_norm": 3.709740849444643e-05,
      "learning_rate": 5.441452636858746e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8803040,
      "step": 31440
    },
    {
      "epoch": 349.3888888888889,
      "grad_norm": 8.305769006256014e-05,
      "learning_rate": 5.435339337111905e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8804432,
      "step": 31445
    },
    {
      "epoch": 349.44444444444446,
      "grad_norm": 2.522821341699455e-05,
      "learning_rate": 5.42922905448161e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8805824,
      "step": 31450
    },
    {
      "epoch": 349.5,
      "grad_norm": 8.572189108235762e-05,
      "learning_rate": 5.423121789910129e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8807216,
      "step": 31455
    },
    {
      "epoch": 349.55555555555554,
      "grad_norm": 0.0001263367012143135,
      "learning_rate": 5.417017544339287e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8808640,
      "step": 31460
    },
    {
      "epoch": 349.6111111111111,
      "grad_norm": 0.00029619369888678193,
      "learning_rate": 5.410916318710443e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8810048,
      "step": 31465
    },
    {
      "epoch": 349.6666666666667,
      "grad_norm": 0.00031627254793420434,
      "learning_rate": 5.404818113964466e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8811472,
      "step": 31470
    },
    {
      "epoch": 349.72222222222223,
      "grad_norm": 2.9133572752471082e-05,
      "learning_rate": 5.398722931041792e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8812864,
      "step": 31475
    },
    {
      "epoch": 349.77777777777777,
      "grad_norm": 4.8461733967997134e-05,
      "learning_rate": 5.392630770882367e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8814272,
      "step": 31480
    },
    {
      "epoch": 349.8333333333333,
      "grad_norm": 9.191606295644306e-06,
      "learning_rate": 5.3865416344256705e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8815680,
      "step": 31485
    },
    {
      "epoch": 349.8888888888889,
      "grad_norm": 0.002573217498138547,
      "learning_rate": 5.380455522610742e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8817104,
      "step": 31490
    },
    {
      "epoch": 349.94444444444446,
      "grad_norm": 6.51514419587329e-05,
      "learning_rate": 5.374372436376116e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8818496,
      "step": 31495
    },
    {
      "epoch": 350.0,
      "grad_norm": 7.295003160834312e-05,
      "learning_rate": 5.368292376659895e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8819888,
      "step": 31500
    },
    {
      "epoch": 350.05555555555554,
      "grad_norm": 9.09188311197795e-05,
      "learning_rate": 5.362215344399701e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8821264,
      "step": 31505
    },
    {
      "epoch": 350.1111111111111,
      "grad_norm": 0.0032250559888780117,
      "learning_rate": 5.356141340532678e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8822640,
      "step": 31510
    },
    {
      "epoch": 350.1666666666667,
      "grad_norm": 9.211852739099413e-05,
      "learning_rate": 5.350070365995522e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8824048,
      "step": 31515
    },
    {
      "epoch": 350.22222222222223,
      "grad_norm": 1.7008866052492522e-05,
      "learning_rate": 5.344002421724459e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8825456,
      "step": 31520
    },
    {
      "epoch": 350.27777777777777,
      "grad_norm": 0.00018667585391085595,
      "learning_rate": 5.337937508655228e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8826864,
      "step": 31525
    },
    {
      "epoch": 350.3333333333333,
      "grad_norm": 0.00019647074805106968,
      "learning_rate": 5.331875627723126e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8828256,
      "step": 31530
    },
    {
      "epoch": 350.3888888888889,
      "grad_norm": 1.0326992196496576e-05,
      "learning_rate": 5.325816779862963e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8829648,
      "step": 31535
    },
    {
      "epoch": 350.44444444444446,
      "grad_norm": 2.6355019144830294e-05,
      "learning_rate": 5.319760966009102e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8831024,
      "step": 31540
    },
    {
      "epoch": 350.5,
      "grad_norm": 0.00017626280896365643,
      "learning_rate": 5.3137081870954096e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8832448,
      "step": 31545
    },
    {
      "epoch": 350.55555555555554,
      "grad_norm": 3.557659874786623e-05,
      "learning_rate": 5.307658444055313e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8833856,
      "step": 31550
    },
    {
      "epoch": 350.6111111111111,
      "grad_norm": 1.6369363947887905e-05,
      "learning_rate": 5.301611737821749e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8835216,
      "step": 31555
    },
    {
      "epoch": 350.6666666666667,
      "grad_norm": 0.0008766730315983295,
      "learning_rate": 5.295568069327206e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8836592,
      "step": 31560
    },
    {
      "epoch": 350.72222222222223,
      "grad_norm": 6.964643398532644e-05,
      "learning_rate": 5.289527439503683e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8837984,
      "step": 31565
    },
    {
      "epoch": 350.77777777777777,
      "grad_norm": 0.0033467309549450874,
      "learning_rate": 5.28348984928273e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8839360,
      "step": 31570
    },
    {
      "epoch": 350.8333333333333,
      "grad_norm": 0.0003474231925792992,
      "learning_rate": 5.27745529959541e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8840752,
      "step": 31575
    },
    {
      "epoch": 350.8888888888889,
      "grad_norm": 0.00017319236940238625,
      "learning_rate": 5.271423791372335e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8842160,
      "step": 31580
    },
    {
      "epoch": 350.94444444444446,
      "grad_norm": 0.00010665522131603211,
      "learning_rate": 5.26539532554364e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8843568,
      "step": 31585
    },
    {
      "epoch": 351.0,
      "grad_norm": 0.0001585757709108293,
      "learning_rate": 5.25936990303898e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8844960,
      "step": 31590
    },
    {
      "epoch": 351.05555555555554,
      "grad_norm": 2.515514279366471e-05,
      "learning_rate": 5.253347524787555e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8846320,
      "step": 31595
    },
    {
      "epoch": 351.1111111111111,
      "grad_norm": 0.00010833062697201967,
      "learning_rate": 5.2473281917181035e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8847728,
      "step": 31600
    },
    {
      "epoch": 351.1111111111111,
      "eval_loss": 0.8748646974563599,
      "eval_runtime": 1.2401,
      "eval_samples_per_second": 32.254,
      "eval_steps_per_second": 16.127,
      "num_input_tokens_seen": 8847728,
      "step": 31600
    },
    {
      "epoch": 351.1666666666667,
      "grad_norm": 3.362995994393714e-05,
      "learning_rate": 5.241311904758864e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8849184,
      "step": 31605
    },
    {
      "epoch": 351.22222222222223,
      "grad_norm": 0.00013895463780499995,
      "learning_rate": 5.23529866483764e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8850608,
      "step": 31610
    },
    {
      "epoch": 351.27777777777777,
      "grad_norm": 0.00040172203443944454,
      "learning_rate": 5.229288472881732e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8852000,
      "step": 31615
    },
    {
      "epoch": 351.3333333333333,
      "grad_norm": 0.00018550289678387344,
      "learning_rate": 5.2232813298180025e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8853408,
      "step": 31620
    },
    {
      "epoch": 351.3888888888889,
      "grad_norm": 8.013423212105408e-05,
      "learning_rate": 5.217277236572824e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8854816,
      "step": 31625
    },
    {
      "epoch": 351.44444444444446,
      "grad_norm": 0.00019299803534522653,
      "learning_rate": 5.211276194072093e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8856208,
      "step": 31630
    },
    {
      "epoch": 351.5,
      "grad_norm": 0.0026862446684390306,
      "learning_rate": 5.205278203241254e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8857632,
      "step": 31635
    },
    {
      "epoch": 351.55555555555554,
      "grad_norm": 5.848267755936831e-05,
      "learning_rate": 5.199283265005278e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8859008,
      "step": 31640
    },
    {
      "epoch": 351.6111111111111,
      "grad_norm": 4.8110290663316846e-05,
      "learning_rate": 5.193291380288648e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8860384,
      "step": 31645
    },
    {
      "epoch": 351.6666666666667,
      "grad_norm": 0.038742486387491226,
      "learning_rate": 5.1873025500153995e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8861792,
      "step": 31650
    },
    {
      "epoch": 351.72222222222223,
      "grad_norm": 6.783664866816252e-05,
      "learning_rate": 5.181316775109071e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8863184,
      "step": 31655
    },
    {
      "epoch": 351.77777777777777,
      "grad_norm": 0.00014139882114250213,
      "learning_rate": 5.1753340564927564e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8864608,
      "step": 31660
    },
    {
      "epoch": 351.8333333333333,
      "grad_norm": 0.00029823064687661827,
      "learning_rate": 5.169354395089068e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8865984,
      "step": 31665
    },
    {
      "epoch": 351.8888888888889,
      "grad_norm": 0.0010626501170918345,
      "learning_rate": 5.1633777918201346e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8867424,
      "step": 31670
    },
    {
      "epoch": 351.94444444444446,
      "grad_norm": 0.001160282175987959,
      "learning_rate": 5.157404247607625e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8868832,
      "step": 31675
    },
    {
      "epoch": 352.0,
      "grad_norm": 6.221716466825455e-05,
      "learning_rate": 5.1514337633727454e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8870240,
      "step": 31680
    },
    {
      "epoch": 352.05555555555554,
      "grad_norm": 0.00011322381760692224,
      "learning_rate": 5.145466340036206e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8871680,
      "step": 31685
    },
    {
      "epoch": 352.1111111111111,
      "grad_norm": 0.00011786715913331136,
      "learning_rate": 5.139501978518274e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8873056,
      "step": 31690
    },
    {
      "epoch": 352.1666666666667,
      "grad_norm": 0.0015122374752536416,
      "learning_rate": 5.133540679738716e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8874496,
      "step": 31695
    },
    {
      "epoch": 352.22222222222223,
      "grad_norm": 0.00017131092317868024,
      "learning_rate": 5.127582444616838e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8875904,
      "step": 31700
    },
    {
      "epoch": 352.27777777777777,
      "grad_norm": 0.0009509765077382326,
      "learning_rate": 5.121627274071486e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8877296,
      "step": 31705
    },
    {
      "epoch": 352.3333333333333,
      "grad_norm": 0.0017238968284800649,
      "learning_rate": 5.115675169021009e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8878720,
      "step": 31710
    },
    {
      "epoch": 352.3888888888889,
      "grad_norm": 3.5355442378204316e-05,
      "learning_rate": 5.1097261303832994e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8880080,
      "step": 31715
    },
    {
      "epoch": 352.44444444444446,
      "grad_norm": 2.2233101844904013e-05,
      "learning_rate": 5.103780159075788e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8881488,
      "step": 31720
    },
    {
      "epoch": 352.5,
      "grad_norm": 0.0002984330931212753,
      "learning_rate": 5.0978372560154e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8882912,
      "step": 31725
    },
    {
      "epoch": 352.55555555555554,
      "grad_norm": 4.551077290670946e-05,
      "learning_rate": 5.091897422118619e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8884320,
      "step": 31730
    },
    {
      "epoch": 352.6111111111111,
      "grad_norm": 0.0001717193954391405,
      "learning_rate": 5.0859606583014305e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8885744,
      "step": 31735
    },
    {
      "epoch": 352.6666666666667,
      "grad_norm": 2.5156048650387675e-05,
      "learning_rate": 5.080026965479365e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8887104,
      "step": 31740
    },
    {
      "epoch": 352.72222222222223,
      "grad_norm": 0.00022632154286839068,
      "learning_rate": 5.074096344567475e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8888512,
      "step": 31745
    },
    {
      "epoch": 352.77777777777777,
      "grad_norm": 5.916264490224421e-05,
      "learning_rate": 5.0681687964803294e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8889904,
      "step": 31750
    },
    {
      "epoch": 352.8333333333333,
      "grad_norm": 6.716233474435285e-05,
      "learning_rate": 5.06224432213204e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8891328,
      "step": 31755
    },
    {
      "epoch": 352.8888888888889,
      "grad_norm": 0.0001586589205544442,
      "learning_rate": 5.056322922436224e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8892720,
      "step": 31760
    },
    {
      "epoch": 352.94444444444446,
      "grad_norm": 0.00010403097985545173,
      "learning_rate": 5.0504045983060465e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8894112,
      "step": 31765
    },
    {
      "epoch": 353.0,
      "grad_norm": 5.6729350035311654e-05,
      "learning_rate": 5.044489350654183e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8895488,
      "step": 31770
    },
    {
      "epoch": 353.05555555555554,
      "grad_norm": 5.7230474340030923e-05,
      "learning_rate": 5.038577180392831e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8896928,
      "step": 31775
    },
    {
      "epoch": 353.1111111111111,
      "grad_norm": 0.00022544997045770288,
      "learning_rate": 5.032668088433729e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8898320,
      "step": 31780
    },
    {
      "epoch": 353.1666666666667,
      "grad_norm": 2.9415452445391566e-05,
      "learning_rate": 5.02676207568814e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8899744,
      "step": 31785
    },
    {
      "epoch": 353.22222222222223,
      "grad_norm": 0.00012850339408032596,
      "learning_rate": 5.02085914306683e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8901168,
      "step": 31790
    },
    {
      "epoch": 353.27777777777777,
      "grad_norm": 0.00014702552289236337,
      "learning_rate": 5.014959291480123e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8902544,
      "step": 31795
    },
    {
      "epoch": 353.3333333333333,
      "grad_norm": 0.0005565116298384964,
      "learning_rate": 5.009062521837835e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8903952,
      "step": 31800
    },
    {
      "epoch": 353.3333333333333,
      "eval_loss": 0.905285656452179,
      "eval_runtime": 1.2501,
      "eval_samples_per_second": 31.997,
      "eval_steps_per_second": 15.998,
      "num_input_tokens_seen": 8903952,
      "step": 31800
    },
    {
      "epoch": 353.3888888888889,
      "grad_norm": 3.638598354882561e-05,
      "learning_rate": 5.003168835049324e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8905344,
      "step": 31805
    },
    {
      "epoch": 353.44444444444446,
      "grad_norm": 0.00019300768326502293,
      "learning_rate": 4.997278232023483e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8906768,
      "step": 31810
    },
    {
      "epoch": 353.5,
      "grad_norm": 0.00044065475231036544,
      "learning_rate": 4.9913907136687036e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8908192,
      "step": 31815
    },
    {
      "epoch": 353.55555555555554,
      "grad_norm": 4.70824925287161e-05,
      "learning_rate": 4.985506280892918e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8909568,
      "step": 31820
    },
    {
      "epoch": 353.6111111111111,
      "grad_norm": 0.00013765158655587584,
      "learning_rate": 4.979624934603589e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8910960,
      "step": 31825
    },
    {
      "epoch": 353.6666666666667,
      "grad_norm": 6.877411942696199e-05,
      "learning_rate": 4.97374667570768e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8912352,
      "step": 31830
    },
    {
      "epoch": 353.72222222222223,
      "grad_norm": 2.6674008040572517e-05,
      "learning_rate": 4.967871505111704e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8913728,
      "step": 31835
    },
    {
      "epoch": 353.77777777777777,
      "grad_norm": 8.804111712379381e-05,
      "learning_rate": 4.961999423721686e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8915136,
      "step": 31840
    },
    {
      "epoch": 353.8333333333333,
      "grad_norm": 0.00016808847431093454,
      "learning_rate": 4.956130432443159e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8916544,
      "step": 31845
    },
    {
      "epoch": 353.8888888888889,
      "grad_norm": 7.15073401806876e-05,
      "learning_rate": 4.950264532181215e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8917872,
      "step": 31850
    },
    {
      "epoch": 353.94444444444446,
      "grad_norm": 2.7974294425803237e-05,
      "learning_rate": 4.944401723840433e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8919296,
      "step": 31855
    },
    {
      "epoch": 354.0,
      "grad_norm": 3.894153269357048e-05,
      "learning_rate": 4.938542008324942e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8920688,
      "step": 31860
    },
    {
      "epoch": 354.05555555555554,
      "grad_norm": 0.00011348079715389758,
      "learning_rate": 4.9326853865383855e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8922080,
      "step": 31865
    },
    {
      "epoch": 354.1111111111111,
      "grad_norm": 6.514081178465858e-05,
      "learning_rate": 4.926831859383918e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8923472,
      "step": 31870
    },
    {
      "epoch": 354.1666666666667,
      "grad_norm": 0.0006969982059672475,
      "learning_rate": 4.92098142776424e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8924880,
      "step": 31875
    },
    {
      "epoch": 354.22222222222223,
      "grad_norm": 3.7409787182696164e-05,
      "learning_rate": 4.91513409258155e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8926272,
      "step": 31880
    },
    {
      "epoch": 354.27777777777777,
      "grad_norm": 0.0001693958038231358,
      "learning_rate": 4.909289854737581e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8927664,
      "step": 31885
    },
    {
      "epoch": 354.3333333333333,
      "grad_norm": 0.00016598479123786092,
      "learning_rate": 4.903448715133602e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8929056,
      "step": 31890
    },
    {
      "epoch": 354.3888888888889,
      "grad_norm": 0.0002333361335331574,
      "learning_rate": 4.897610674670372e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8930496,
      "step": 31895
    },
    {
      "epoch": 354.44444444444446,
      "grad_norm": 9.045044862432405e-05,
      "learning_rate": 4.8917757342482e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8931888,
      "step": 31900
    },
    {
      "epoch": 354.5,
      "grad_norm": 0.00013783699250780046,
      "learning_rate": 4.885943894766909e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8933264,
      "step": 31905
    },
    {
      "epoch": 354.55555555555554,
      "grad_norm": 6.209365528775379e-05,
      "learning_rate": 4.880115157125842e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8934640,
      "step": 31910
    },
    {
      "epoch": 354.6111111111111,
      "grad_norm": 0.00011946587619604543,
      "learning_rate": 4.874289522223857e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8936080,
      "step": 31915
    },
    {
      "epoch": 354.6666666666667,
      "grad_norm": 3.0918712582206354e-05,
      "learning_rate": 4.868466990959339e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8937520,
      "step": 31920
    },
    {
      "epoch": 354.72222222222223,
      "grad_norm": 0.00029277446446940303,
      "learning_rate": 4.8626475642301964e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8938896,
      "step": 31925
    },
    {
      "epoch": 354.77777777777777,
      "grad_norm": 8.210122905438766e-05,
      "learning_rate": 4.856831242933871e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8940272,
      "step": 31930
    },
    {
      "epoch": 354.8333333333333,
      "grad_norm": 3.205120447091758e-05,
      "learning_rate": 4.851018027967294e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8941680,
      "step": 31935
    },
    {
      "epoch": 354.8888888888889,
      "grad_norm": 4.367849760456011e-05,
      "learning_rate": 4.845207920226946e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8943056,
      "step": 31940
    },
    {
      "epoch": 354.94444444444446,
      "grad_norm": 0.00010663251305231825,
      "learning_rate": 4.839400920608825e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8944496,
      "step": 31945
    },
    {
      "epoch": 355.0,
      "grad_norm": 4.0915772842708975e-05,
      "learning_rate": 4.83359703000843e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8945936,
      "step": 31950
    },
    {
      "epoch": 355.05555555555554,
      "grad_norm": 9.524650522507727e-06,
      "learning_rate": 4.827796249320804e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8947344,
      "step": 31955
    },
    {
      "epoch": 355.1111111111111,
      "grad_norm": 2.971072717627976e-05,
      "learning_rate": 4.82199857944049e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8948752,
      "step": 31960
    },
    {
      "epoch": 355.1666666666667,
      "grad_norm": 0.0003188357804901898,
      "learning_rate": 4.8162040212615695e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8950160,
      "step": 31965
    },
    {
      "epoch": 355.22222222222223,
      "grad_norm": 4.0638224163558334e-05,
      "learning_rate": 4.810412575677639e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8951568,
      "step": 31970
    },
    {
      "epoch": 355.27777777777777,
      "grad_norm": 5.7265511713922024e-05,
      "learning_rate": 4.804624243581801e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8952944,
      "step": 31975
    },
    {
      "epoch": 355.3333333333333,
      "grad_norm": 6.240802758838981e-05,
      "learning_rate": 4.798839025866703e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8954368,
      "step": 31980
    },
    {
      "epoch": 355.3888888888889,
      "grad_norm": 0.0005339794442988932,
      "learning_rate": 4.793056923424491e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8955776,
      "step": 31985
    },
    {
      "epoch": 355.44444444444446,
      "grad_norm": 0.001190018723718822,
      "learning_rate": 4.78727793714683e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8957120,
      "step": 31990
    },
    {
      "epoch": 355.5,
      "grad_norm": 0.00031481904443353415,
      "learning_rate": 4.7815020679249285e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8958512,
      "step": 31995
    },
    {
      "epoch": 355.55555555555554,
      "grad_norm": 8.01494243205525e-05,
      "learning_rate": 4.775729316649483e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8959920,
      "step": 32000
    },
    {
      "epoch": 355.55555555555554,
      "eval_loss": 0.8862390518188477,
      "eval_runtime": 1.2463,
      "eval_samples_per_second": 32.095,
      "eval_steps_per_second": 16.047,
      "num_input_tokens_seen": 8959920,
      "step": 32000
    },
    {
      "epoch": 355.6111111111111,
      "grad_norm": 4.0410810470348224e-05,
      "learning_rate": 4.769959684210728e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8961296,
      "step": 32005
    },
    {
      "epoch": 355.6666666666667,
      "grad_norm": 6.609956471947953e-05,
      "learning_rate": 4.764193171498426e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8962688,
      "step": 32010
    },
    {
      "epoch": 355.72222222222223,
      "grad_norm": 0.0034187226556241512,
      "learning_rate": 4.75842977940183e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8964112,
      "step": 32015
    },
    {
      "epoch": 355.77777777777777,
      "grad_norm": 0.0003395500825718045,
      "learning_rate": 4.752669508809729e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8965552,
      "step": 32020
    },
    {
      "epoch": 355.8333333333333,
      "grad_norm": 0.00010528103302931413,
      "learning_rate": 4.746912360610445e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8966928,
      "step": 32025
    },
    {
      "epoch": 355.8888888888889,
      "grad_norm": 0.00019128168059978634,
      "learning_rate": 4.741158335691781e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8968304,
      "step": 32030
    },
    {
      "epoch": 355.94444444444446,
      "grad_norm": 1.0806358659465332e-05,
      "learning_rate": 4.7354074349410994e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8969728,
      "step": 32035
    },
    {
      "epoch": 356.0,
      "grad_norm": 2.8639149604714476e-05,
      "learning_rate": 4.729659659245245e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8971152,
      "step": 32040
    },
    {
      "epoch": 356.05555555555554,
      "grad_norm": 0.00024298517382703722,
      "learning_rate": 4.723915009490601e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8972576,
      "step": 32045
    },
    {
      "epoch": 356.1111111111111,
      "grad_norm": 2.0519044483080506e-05,
      "learning_rate": 4.718173486563077e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8974016,
      "step": 32050
    },
    {
      "epoch": 356.1666666666667,
      "grad_norm": 4.444923251867294e-05,
      "learning_rate": 4.71243509134808e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8975408,
      "step": 32055
    },
    {
      "epoch": 356.22222222222223,
      "grad_norm": 3.1440424208994955e-05,
      "learning_rate": 4.706699824730532e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8976800,
      "step": 32060
    },
    {
      "epoch": 356.27777777777777,
      "grad_norm": 9.242352098226547e-05,
      "learning_rate": 4.700967687594901e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8978192,
      "step": 32065
    },
    {
      "epoch": 356.3333333333333,
      "grad_norm": 0.0001557011710247025,
      "learning_rate": 4.69523868082514e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8979568,
      "step": 32070
    },
    {
      "epoch": 356.3888888888889,
      "grad_norm": 0.0005780462524853647,
      "learning_rate": 4.689512805304747e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8980976,
      "step": 32075
    },
    {
      "epoch": 356.44444444444446,
      "grad_norm": 0.00015135249122977257,
      "learning_rate": 4.683790061916707e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8982400,
      "step": 32080
    },
    {
      "epoch": 356.5,
      "grad_norm": 8.044917922234163e-05,
      "learning_rate": 4.678070451543551e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8983792,
      "step": 32085
    },
    {
      "epoch": 356.55555555555554,
      "grad_norm": 0.001871593645773828,
      "learning_rate": 4.6723539750673204e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8985216,
      "step": 32090
    },
    {
      "epoch": 356.6111111111111,
      "grad_norm": 0.0003828752669505775,
      "learning_rate": 4.666640633369551e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8986624,
      "step": 32095
    },
    {
      "epoch": 356.6666666666667,
      "grad_norm": 9.271038288716227e-05,
      "learning_rate": 4.660930427331323e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8988000,
      "step": 32100
    },
    {
      "epoch": 356.72222222222223,
      "grad_norm": 6.523881893372163e-05,
      "learning_rate": 4.6552233578332244e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8989344,
      "step": 32105
    },
    {
      "epoch": 356.77777777777777,
      "grad_norm": 9.419696289114654e-05,
      "learning_rate": 4.649519425755347e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8990800,
      "step": 32110
    },
    {
      "epoch": 356.8333333333333,
      "grad_norm": 6.88932414050214e-05,
      "learning_rate": 4.64381863197732e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8992176,
      "step": 32115
    },
    {
      "epoch": 356.8888888888889,
      "grad_norm": 0.00014774811279494315,
      "learning_rate": 4.638120977378269e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8993552,
      "step": 32120
    },
    {
      "epoch": 356.94444444444446,
      "grad_norm": 0.0001466180256102234,
      "learning_rate": 4.632426462836848e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8994976,
      "step": 32125
    },
    {
      "epoch": 357.0,
      "grad_norm": 6.800627306802198e-05,
      "learning_rate": 4.626735089231224e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8996384,
      "step": 32130
    },
    {
      "epoch": 357.05555555555554,
      "grad_norm": 9.136665175901726e-05,
      "learning_rate": 4.621046857439068e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8997824,
      "step": 32135
    },
    {
      "epoch": 357.1111111111111,
      "grad_norm": 0.00023899157531559467,
      "learning_rate": 4.615361768337587e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 8999216,
      "step": 32140
    },
    {
      "epoch": 357.1666666666667,
      "grad_norm": 3.830519926850684e-05,
      "learning_rate": 4.6096798228034946e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9000624,
      "step": 32145
    },
    {
      "epoch": 357.22222222222223,
      "grad_norm": 0.0001849442342063412,
      "learning_rate": 4.604001021713008e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9002080,
      "step": 32150
    },
    {
      "epoch": 357.27777777777777,
      "grad_norm": 0.0001016164751490578,
      "learning_rate": 4.598325365941883e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9003488,
      "step": 32155
    },
    {
      "epoch": 357.3333333333333,
      "grad_norm": 0.0004856271843891591,
      "learning_rate": 4.5926528563653645e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9004896,
      "step": 32160
    },
    {
      "epoch": 357.3888888888889,
      "grad_norm": 9.939793380908668e-05,
      "learning_rate": 4.5869834938582295e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9006256,
      "step": 32165
    },
    {
      "epoch": 357.44444444444446,
      "grad_norm": 0.0007880312041379511,
      "learning_rate": 4.581317279294772e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9007680,
      "step": 32170
    },
    {
      "epoch": 357.5,
      "grad_norm": 3.265481791459024e-05,
      "learning_rate": 4.57565421354878e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9009120,
      "step": 32175
    },
    {
      "epoch": 357.55555555555554,
      "grad_norm": 0.0021177579183131456,
      "learning_rate": 4.569994297493579e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9010528,
      "step": 32180
    },
    {
      "epoch": 357.6111111111111,
      "grad_norm": 0.0005892066401429474,
      "learning_rate": 4.564337532002002e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9011920,
      "step": 32185
    },
    {
      "epoch": 357.6666666666667,
      "grad_norm": 7.128343713702634e-05,
      "learning_rate": 4.55868391794638e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9013328,
      "step": 32190
    },
    {
      "epoch": 357.72222222222223,
      "grad_norm": 0.00010789400403155014,
      "learning_rate": 4.553033456198588e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9014704,
      "step": 32195
    },
    {
      "epoch": 357.77777777777777,
      "grad_norm": 4.143346450291574e-05,
      "learning_rate": 4.54738614762999e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9016096,
      "step": 32200
    },
    {
      "epoch": 357.77777777777777,
      "eval_loss": 0.8689216375350952,
      "eval_runtime": 1.248,
      "eval_samples_per_second": 32.051,
      "eval_steps_per_second": 16.025,
      "num_input_tokens_seen": 9016096,
      "step": 32200
    },
    {
      "epoch": 357.8333333333333,
      "grad_norm": 0.0002981304714921862,
      "learning_rate": 4.541741993111465e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9017456,
      "step": 32205
    },
    {
      "epoch": 357.8888888888889,
      "grad_norm": 0.00022382776660379022,
      "learning_rate": 4.536100993513423e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9018848,
      "step": 32210
    },
    {
      "epoch": 357.94444444444446,
      "grad_norm": 0.00011317271128064021,
      "learning_rate": 4.530463149705768e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9020288,
      "step": 32215
    },
    {
      "epoch": 358.0,
      "grad_norm": 1.8266882761963643e-05,
      "learning_rate": 4.524828462557934e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9021664,
      "step": 32220
    },
    {
      "epoch": 358.05555555555554,
      "grad_norm": 1.4770999769098125e-05,
      "learning_rate": 4.5191969329388625e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9023088,
      "step": 32225
    },
    {
      "epoch": 358.1111111111111,
      "grad_norm": 6.877734267618507e-05,
      "learning_rate": 4.5135685617169965e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9024528,
      "step": 32230
    },
    {
      "epoch": 358.1666666666667,
      "grad_norm": 4.156568320468068e-05,
      "learning_rate": 4.507943349760313e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9025952,
      "step": 32235
    },
    {
      "epoch": 358.22222222222223,
      "grad_norm": 3.307377846795134e-05,
      "learning_rate": 4.502321297936277e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9027360,
      "step": 32240
    },
    {
      "epoch": 358.27777777777777,
      "grad_norm": 0.00013017121818847954,
      "learning_rate": 4.496702407111888e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9028768,
      "step": 32245
    },
    {
      "epoch": 358.3333333333333,
      "grad_norm": 0.00013718358241021633,
      "learning_rate": 4.491086678153653e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9030192,
      "step": 32250
    },
    {
      "epoch": 358.3888888888889,
      "grad_norm": 0.0001539344957564026,
      "learning_rate": 4.485474111927579e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9031584,
      "step": 32255
    },
    {
      "epoch": 358.44444444444446,
      "grad_norm": 9.481378219788894e-05,
      "learning_rate": 4.479864709299197e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9032976,
      "step": 32260
    },
    {
      "epoch": 358.5,
      "grad_norm": 1.1653693945845589e-05,
      "learning_rate": 4.474258471133555e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9034400,
      "step": 32265
    },
    {
      "epoch": 358.55555555555554,
      "grad_norm": 0.00026861659716814756,
      "learning_rate": 4.4686553982952014e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9035776,
      "step": 32270
    },
    {
      "epoch": 358.6111111111111,
      "grad_norm": 0.000379179953597486,
      "learning_rate": 4.463055491648191e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9037168,
      "step": 32275
    },
    {
      "epoch": 358.6666666666667,
      "grad_norm": 4.2986615881090984e-05,
      "learning_rate": 4.457458752056112e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9038576,
      "step": 32280
    },
    {
      "epoch": 358.72222222222223,
      "grad_norm": 0.0006118403398431838,
      "learning_rate": 4.451865180382042e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9039968,
      "step": 32285
    },
    {
      "epoch": 358.77777777777777,
      "grad_norm": 2.4576303985668346e-05,
      "learning_rate": 4.4462747774885936e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9041328,
      "step": 32290
    },
    {
      "epoch": 358.8333333333333,
      "grad_norm": 0.0005673632258549333,
      "learning_rate": 4.440687544237859e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9042736,
      "step": 32295
    },
    {
      "epoch": 358.8888888888889,
      "grad_norm": 0.00029240630101412535,
      "learning_rate": 4.435103481491471e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9044176,
      "step": 32300
    },
    {
      "epoch": 358.94444444444446,
      "grad_norm": 0.0001513946772320196,
      "learning_rate": 4.429522590110569e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9045552,
      "step": 32305
    },
    {
      "epoch": 359.0,
      "grad_norm": 0.00015580537728965282,
      "learning_rate": 4.423944870955779e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9046960,
      "step": 32310
    },
    {
      "epoch": 359.05555555555554,
      "grad_norm": 7.174774509621784e-05,
      "learning_rate": 4.418370324887272e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9048384,
      "step": 32315
    },
    {
      "epoch": 359.1111111111111,
      "grad_norm": 0.00014661146269645542,
      "learning_rate": 4.412798952764699e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9049824,
      "step": 32320
    },
    {
      "epoch": 359.1666666666667,
      "grad_norm": 0.00018157176964450628,
      "learning_rate": 4.407230755447245e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9051232,
      "step": 32325
    },
    {
      "epoch": 359.22222222222223,
      "grad_norm": 0.011669319123029709,
      "learning_rate": 4.401665733793598e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9052608,
      "step": 32330
    },
    {
      "epoch": 359.27777777777777,
      "grad_norm": 0.00012560843606479466,
      "learning_rate": 4.3961038886619425e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9054064,
      "step": 32335
    },
    {
      "epoch": 359.3333333333333,
      "grad_norm": 0.00011487900337669998,
      "learning_rate": 4.39054522091e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9055456,
      "step": 32340
    },
    {
      "epoch": 359.3888888888889,
      "grad_norm": 7.912842556834221e-05,
      "learning_rate": 4.384989731394979e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9056864,
      "step": 32345
    },
    {
      "epoch": 359.44444444444446,
      "grad_norm": 0.00023288137163035572,
      "learning_rate": 4.379437420973598e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9058272,
      "step": 32350
    },
    {
      "epoch": 359.5,
      "grad_norm": 0.00018523202743381262,
      "learning_rate": 4.373888290502107e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9059680,
      "step": 32355
    },
    {
      "epoch": 359.55555555555554,
      "grad_norm": 3.3268752304138616e-05,
      "learning_rate": 4.36834234083624e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9061040,
      "step": 32360
    },
    {
      "epoch": 359.6111111111111,
      "grad_norm": 0.0004492694279178977,
      "learning_rate": 4.362799572831258e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9062432,
      "step": 32365
    },
    {
      "epoch": 359.6666666666667,
      "grad_norm": 5.2122868510195985e-05,
      "learning_rate": 4.35725998734193e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9063808,
      "step": 32370
    },
    {
      "epoch": 359.72222222222223,
      "grad_norm": 8.11831996543333e-05,
      "learning_rate": 4.3517235852225195e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9065200,
      "step": 32375
    },
    {
      "epoch": 359.77777777777777,
      "grad_norm": 2.7746002160711214e-05,
      "learning_rate": 4.346190367326822e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9066576,
      "step": 32380
    },
    {
      "epoch": 359.8333333333333,
      "grad_norm": 0.00013921508798375726,
      "learning_rate": 4.340660334508115e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9067968,
      "step": 32385
    },
    {
      "epoch": 359.8888888888889,
      "grad_norm": 0.00011511048796819523,
      "learning_rate": 4.335133487619206e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9069344,
      "step": 32390
    },
    {
      "epoch": 359.94444444444446,
      "grad_norm": 0.00040918440208770335,
      "learning_rate": 4.329609827512409e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9070784,
      "step": 32395
    },
    {
      "epoch": 360.0,
      "grad_norm": 0.0002198773727286607,
      "learning_rate": 4.324089355039531e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9072192,
      "step": 32400
    },
    {
      "epoch": 360.0,
      "eval_loss": 0.8738431930541992,
      "eval_runtime": 1.2352,
      "eval_samples_per_second": 32.383,
      "eval_steps_per_second": 16.191,
      "num_input_tokens_seen": 9072192,
      "step": 32400
    },
    {
      "epoch": 360.05555555555554,
      "grad_norm": 0.0008817660855129361,
      "learning_rate": 4.3185720710519075e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9073584,
      "step": 32405
    },
    {
      "epoch": 360.1111111111111,
      "grad_norm": 8.9602755906526e-05,
      "learning_rate": 4.3130579764003724e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9075008,
      "step": 32410
    },
    {
      "epoch": 360.1666666666667,
      "grad_norm": 0.0002891590411309153,
      "learning_rate": 4.307547071935267e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9076400,
      "step": 32415
    },
    {
      "epoch": 360.22222222222223,
      "grad_norm": 1.631138911761809e-05,
      "learning_rate": 4.302039358506435e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9077792,
      "step": 32420
    },
    {
      "epoch": 360.27777777777777,
      "grad_norm": 0.0004872725112363696,
      "learning_rate": 4.296534836963245e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9079200,
      "step": 32425
    },
    {
      "epoch": 360.3333333333333,
      "grad_norm": 0.00013250661140773445,
      "learning_rate": 4.291033508154555e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9080592,
      "step": 32430
    },
    {
      "epoch": 360.3888888888889,
      "grad_norm": 7.003609061939642e-05,
      "learning_rate": 4.285535372928748e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9082000,
      "step": 32435
    },
    {
      "epoch": 360.44444444444446,
      "grad_norm": 6.797794776502997e-05,
      "learning_rate": 4.280040432133695e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9083360,
      "step": 32440
    },
    {
      "epoch": 360.5,
      "grad_norm": 0.006912127137184143,
      "learning_rate": 4.274548686616789e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9084736,
      "step": 32445
    },
    {
      "epoch": 360.55555555555554,
      "grad_norm": 0.0002298399485880509,
      "learning_rate": 4.2690601372249364e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9086176,
      "step": 32450
    },
    {
      "epoch": 360.6111111111111,
      "grad_norm": 7.95811356510967e-05,
      "learning_rate": 4.263574784804525e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9087584,
      "step": 32455
    },
    {
      "epoch": 360.6666666666667,
      "grad_norm": 4.916285251965746e-05,
      "learning_rate": 4.258092630201479e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9088960,
      "step": 32460
    },
    {
      "epoch": 360.72222222222223,
      "grad_norm": 0.00031846831552684307,
      "learning_rate": 4.252613674261202e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9090352,
      "step": 32465
    },
    {
      "epoch": 360.77777777777777,
      "grad_norm": 7.271278445841745e-05,
      "learning_rate": 4.2471379178286224e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9091712,
      "step": 32470
    },
    {
      "epoch": 360.8333333333333,
      "grad_norm": 8.825211261864752e-05,
      "learning_rate": 4.241665361748181e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9093120,
      "step": 32475
    },
    {
      "epoch": 360.8888888888889,
      "grad_norm": 8.404746040469036e-05,
      "learning_rate": 4.2361960068637994e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9094544,
      "step": 32480
    },
    {
      "epoch": 360.94444444444446,
      "grad_norm": 0.00018843637371901423,
      "learning_rate": 4.230729854018933e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9095968,
      "step": 32485
    },
    {
      "epoch": 361.0,
      "grad_norm": 1.3443828720482998e-05,
      "learning_rate": 4.225266904056521e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9097424,
      "step": 32490
    },
    {
      "epoch": 361.05555555555554,
      "grad_norm": 1.5789721146575175e-05,
      "learning_rate": 4.21980715781903e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9098800,
      "step": 32495
    },
    {
      "epoch": 361.1111111111111,
      "grad_norm": 6.118010787758976e-05,
      "learning_rate": 4.214350616148416e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9100192,
      "step": 32500
    },
    {
      "epoch": 361.1666666666667,
      "grad_norm": 0.002554772188887,
      "learning_rate": 4.20889727988614e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9101568,
      "step": 32505
    },
    {
      "epoch": 361.22222222222223,
      "grad_norm": 0.0005271618138067424,
      "learning_rate": 4.20344714987318e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9103024,
      "step": 32510
    },
    {
      "epoch": 361.27777777777777,
      "grad_norm": 1.7206000848091207e-05,
      "learning_rate": 4.198000226950022e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9104432,
      "step": 32515
    },
    {
      "epoch": 361.3333333333333,
      "grad_norm": 0.00037010834785178304,
      "learning_rate": 4.192556511956635e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9105824,
      "step": 32520
    },
    {
      "epoch": 361.3888888888889,
      "grad_norm": 0.00013495922030415386,
      "learning_rate": 4.18711600573252e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9107232,
      "step": 32525
    },
    {
      "epoch": 361.44444444444446,
      "grad_norm": 6.852027581771836e-05,
      "learning_rate": 4.181678709116671e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9108576,
      "step": 32530
    },
    {
      "epoch": 361.5,
      "grad_norm": 4.1409839468542486e-05,
      "learning_rate": 4.1762446229475785e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9109968,
      "step": 32535
    },
    {
      "epoch": 361.55555555555554,
      "grad_norm": 9.197526378557086e-05,
      "learning_rate": 4.17081374806326e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9111360,
      "step": 32540
    },
    {
      "epoch": 361.6111111111111,
      "grad_norm": 4.271200668881647e-05,
      "learning_rate": 4.165386085301212e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9112784,
      "step": 32545
    },
    {
      "epoch": 361.6666666666667,
      "grad_norm": 0.00043604779057204723,
      "learning_rate": 4.1599616354984525e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9114208,
      "step": 32550
    },
    {
      "epoch": 361.72222222222223,
      "grad_norm": 9.578058234183118e-05,
      "learning_rate": 4.154540399491508e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9115648,
      "step": 32555
    },
    {
      "epoch": 361.77777777777777,
      "grad_norm": 0.00012053987302351743,
      "learning_rate": 4.149122378116394e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9117008,
      "step": 32560
    },
    {
      "epoch": 361.8333333333333,
      "grad_norm": 0.0001511097070761025,
      "learning_rate": 4.14370757220863e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9118400,
      "step": 32565
    },
    {
      "epoch": 361.8888888888889,
      "grad_norm": 2.8396467314451e-05,
      "learning_rate": 4.138295982603263e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9119808,
      "step": 32570
    },
    {
      "epoch": 361.94444444444446,
      "grad_norm": 3.671646845759824e-05,
      "learning_rate": 4.132887610134814e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9121200,
      "step": 32575
    },
    {
      "epoch": 362.0,
      "grad_norm": 7.705033931415528e-05,
      "learning_rate": 4.127482455637335e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9122624,
      "step": 32580
    },
    {
      "epoch": 362.05555555555554,
      "grad_norm": 4.6321241825353354e-05,
      "learning_rate": 4.1220805199443545e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9124032,
      "step": 32585
    },
    {
      "epoch": 362.1111111111111,
      "grad_norm": 6.607394607271999e-05,
      "learning_rate": 4.116681803888925e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9125424,
      "step": 32590
    },
    {
      "epoch": 362.1666666666667,
      "grad_norm": 0.00031668253359384835,
      "learning_rate": 4.111286308303605e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9126832,
      "step": 32595
    },
    {
      "epoch": 362.22222222222223,
      "grad_norm": 3.715957791428082e-05,
      "learning_rate": 4.105894034020433e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9128272,
      "step": 32600
    },
    {
      "epoch": 362.22222222222223,
      "eval_loss": 0.8963890075683594,
      "eval_runtime": 1.2383,
      "eval_samples_per_second": 32.302,
      "eval_steps_per_second": 16.151,
      "num_input_tokens_seen": 9128272,
      "step": 32600
    },
    {
      "epoch": 362.27777777777777,
      "grad_norm": 0.0005197184509597719,
      "learning_rate": 4.100504981870975e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9129664,
      "step": 32605
    },
    {
      "epoch": 362.3333333333333,
      "grad_norm": 0.00011551699572009966,
      "learning_rate": 4.0951191526862915e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9131008,
      "step": 32610
    },
    {
      "epoch": 362.3888888888889,
      "grad_norm": 0.0001720949076116085,
      "learning_rate": 4.089736547296938e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9132368,
      "step": 32615
    },
    {
      "epoch": 362.44444444444446,
      "grad_norm": 9.632544970372692e-05,
      "learning_rate": 4.08435716653299e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9133728,
      "step": 32620
    },
    {
      "epoch": 362.5,
      "grad_norm": 9.230043360730633e-05,
      "learning_rate": 4.0789810112240005e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9135152,
      "step": 32625
    },
    {
      "epoch": 362.55555555555554,
      "grad_norm": 0.0002171297383029014,
      "learning_rate": 4.073608082199057e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9136544,
      "step": 32630
    },
    {
      "epoch": 362.6111111111111,
      "grad_norm": 2.8923934223712422e-05,
      "learning_rate": 4.068238380286718e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9137952,
      "step": 32635
    },
    {
      "epoch": 362.6666666666667,
      "grad_norm": 5.208923903410323e-05,
      "learning_rate": 4.062871906315072e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9139344,
      "step": 32640
    },
    {
      "epoch": 362.72222222222223,
      "grad_norm": 0.0003897174319718033,
      "learning_rate": 4.057508661111686e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9140704,
      "step": 32645
    },
    {
      "epoch": 362.77777777777777,
      "grad_norm": 0.002538575790822506,
      "learning_rate": 4.052148645503648e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9142128,
      "step": 32650
    },
    {
      "epoch": 362.8333333333333,
      "grad_norm": 0.0002568019845057279,
      "learning_rate": 4.046791860317531e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9143520,
      "step": 32655
    },
    {
      "epoch": 362.8888888888889,
      "grad_norm": 0.00011666561476886272,
      "learning_rate": 4.041438306379431e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9144928,
      "step": 32660
    },
    {
      "epoch": 362.94444444444446,
      "grad_norm": 0.00019327984773553908,
      "learning_rate": 4.036087984514916e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9146336,
      "step": 32665
    },
    {
      "epoch": 363.0,
      "grad_norm": 8.393501047976315e-05,
      "learning_rate": 4.030740895549084e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9147776,
      "step": 32670
    },
    {
      "epoch": 363.05555555555554,
      "grad_norm": 3.576280869310722e-05,
      "learning_rate": 4.025397040306531e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9149200,
      "step": 32675
    },
    {
      "epoch": 363.1111111111111,
      "grad_norm": 0.00027524036704562604,
      "learning_rate": 4.0200564196113285e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9150576,
      "step": 32680
    },
    {
      "epoch": 363.1666666666667,
      "grad_norm": 0.000192393155884929,
      "learning_rate": 4.014719034287079e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9152000,
      "step": 32685
    },
    {
      "epoch": 363.22222222222223,
      "grad_norm": 0.00018247630214318633,
      "learning_rate": 4.0093848851568775e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9153424,
      "step": 32690
    },
    {
      "epoch": 363.27777777777777,
      "grad_norm": 3.219370773877017e-05,
      "learning_rate": 4.004053973043304e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9154800,
      "step": 32695
    },
    {
      "epoch": 363.3333333333333,
      "grad_norm": 0.00022841240570414811,
      "learning_rate": 3.998726298768465e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9156192,
      "step": 32700
    },
    {
      "epoch": 363.3888888888889,
      "grad_norm": 2.5083574655582197e-05,
      "learning_rate": 3.99340186315395e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9157600,
      "step": 32705
    },
    {
      "epoch": 363.44444444444446,
      "grad_norm": 8.587448246544227e-05,
      "learning_rate": 3.988080667020849e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9159024,
      "step": 32710
    },
    {
      "epoch": 363.5,
      "grad_norm": 0.008137897588312626,
      "learning_rate": 3.982762711189766e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9160416,
      "step": 32715
    },
    {
      "epoch": 363.55555555555554,
      "grad_norm": 0.00033224813523702323,
      "learning_rate": 3.977447996480785e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9161776,
      "step": 32720
    },
    {
      "epoch": 363.6111111111111,
      "grad_norm": 0.00040303391870111227,
      "learning_rate": 3.97213652371351e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9163168,
      "step": 32725
    },
    {
      "epoch": 363.6666666666667,
      "grad_norm": 6.11256982665509e-05,
      "learning_rate": 3.966828293707042e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9164592,
      "step": 32730
    },
    {
      "epoch": 363.72222222222223,
      "grad_norm": 0.0008620006847195327,
      "learning_rate": 3.961523307279963e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9165952,
      "step": 32735
    },
    {
      "epoch": 363.77777777777777,
      "grad_norm": 7.602987898280844e-05,
      "learning_rate": 3.956221565250382e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9167360,
      "step": 32740
    },
    {
      "epoch": 363.8333333333333,
      "grad_norm": 2.8816219128202647e-05,
      "learning_rate": 3.950923068435883e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9168768,
      "step": 32745
    },
    {
      "epoch": 363.8888888888889,
      "grad_norm": 4.943405292578973e-05,
      "learning_rate": 3.945627817653566e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9170160,
      "step": 32750
    },
    {
      "epoch": 363.94444444444446,
      "grad_norm": 5.3812371334061027e-05,
      "learning_rate": 3.9403358137200335e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9171600,
      "step": 32755
    },
    {
      "epoch": 364.0,
      "grad_norm": 0.00012073286779923365,
      "learning_rate": 3.9350470574513605e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9172992,
      "step": 32760
    },
    {
      "epoch": 364.05555555555554,
      "grad_norm": 0.000940933998208493,
      "learning_rate": 3.9297615496631525e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9174400,
      "step": 32765
    },
    {
      "epoch": 364.1111111111111,
      "grad_norm": 0.000830948119983077,
      "learning_rate": 3.924479291170505e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9175872,
      "step": 32770
    },
    {
      "epoch": 364.1666666666667,
      "grad_norm": 6.069149821996689e-05,
      "learning_rate": 3.919200282788002e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9177232,
      "step": 32775
    },
    {
      "epoch": 364.22222222222223,
      "grad_norm": 3.473543984000571e-05,
      "learning_rate": 3.913924525329726e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9178608,
      "step": 32780
    },
    {
      "epoch": 364.27777777777777,
      "grad_norm": 4.739047653856687e-05,
      "learning_rate": 3.908652019609279e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9180032,
      "step": 32785
    },
    {
      "epoch": 364.3333333333333,
      "grad_norm": 0.00023591720673721284,
      "learning_rate": 3.9033827664397364e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9181408,
      "step": 32790
    },
    {
      "epoch": 364.3888888888889,
      "grad_norm": 6.525100616272539e-05,
      "learning_rate": 3.898116766633694e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9182816,
      "step": 32795
    },
    {
      "epoch": 364.44444444444446,
      "grad_norm": 0.0001575958594912663,
      "learning_rate": 3.8928540210032225e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9184240,
      "step": 32800
    },
    {
      "epoch": 364.44444444444446,
      "eval_loss": 0.8605605363845825,
      "eval_runtime": 1.2421,
      "eval_samples_per_second": 32.202,
      "eval_steps_per_second": 16.101,
      "num_input_tokens_seen": 9184240,
      "step": 32800
    },
    {
      "epoch": 364.5,
      "grad_norm": 6.98583826306276e-05,
      "learning_rate": 3.887594530359909e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9185664,
      "step": 32805
    },
    {
      "epoch": 364.55555555555554,
      "grad_norm": 0.0019985982216894627,
      "learning_rate": 3.88233829551484e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9187056,
      "step": 32810
    },
    {
      "epoch": 364.6111111111111,
      "grad_norm": 0.0007932670414447784,
      "learning_rate": 3.877085317278581e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9188448,
      "step": 32815
    },
    {
      "epoch": 364.6666666666667,
      "grad_norm": 0.00011157626431668177,
      "learning_rate": 3.87183559646122e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9189808,
      "step": 32820
    },
    {
      "epoch": 364.72222222222223,
      "grad_norm": 0.0007150335586629808,
      "learning_rate": 3.866589133872317e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9191216,
      "step": 32825
    },
    {
      "epoch": 364.77777777777777,
      "grad_norm": 8.834508480504155e-05,
      "learning_rate": 3.861345930320948e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9192608,
      "step": 32830
    },
    {
      "epoch": 364.8333333333333,
      "grad_norm": 0.00016138704086188227,
      "learning_rate": 3.856105986615688e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9194016,
      "step": 32835
    },
    {
      "epoch": 364.8888888888889,
      "grad_norm": 0.0009172277641482651,
      "learning_rate": 3.850869303564589e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9195456,
      "step": 32840
    },
    {
      "epoch": 364.94444444444446,
      "grad_norm": 0.0002524565497878939,
      "learning_rate": 3.845635881975226e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9196800,
      "step": 32845
    },
    {
      "epoch": 365.0,
      "grad_norm": 8.216549031203613e-05,
      "learning_rate": 3.840405722654647e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9198192,
      "step": 32850
    },
    {
      "epoch": 365.05555555555554,
      "grad_norm": 0.00104390864726156,
      "learning_rate": 3.835178826409419e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9199584,
      "step": 32855
    },
    {
      "epoch": 365.1111111111111,
      "grad_norm": 0.00011603258462855592,
      "learning_rate": 3.8299551940455895e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9200960,
      "step": 32860
    },
    {
      "epoch": 365.1666666666667,
      "grad_norm": 0.00033016386441886425,
      "learning_rate": 3.824734826368703e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9202336,
      "step": 32865
    },
    {
      "epoch": 365.22222222222223,
      "grad_norm": 1.5383566278615035e-05,
      "learning_rate": 3.819517724183813e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9203760,
      "step": 32870
    },
    {
      "epoch": 365.27777777777777,
      "grad_norm": 0.0002396680210949853,
      "learning_rate": 3.8143038882954648e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9205168,
      "step": 32875
    },
    {
      "epoch": 365.3333333333333,
      "grad_norm": 0.0001390130928484723,
      "learning_rate": 3.8090933195076867e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9206640,
      "step": 32880
    },
    {
      "epoch": 365.3888888888889,
      "grad_norm": 0.0031682203989475965,
      "learning_rate": 3.8038860186240198e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9208048,
      "step": 32885
    },
    {
      "epoch": 365.44444444444446,
      "grad_norm": 5.025400605518371e-05,
      "learning_rate": 3.7986819864475026e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9209392,
      "step": 32890
    },
    {
      "epoch": 365.5,
      "grad_norm": 5.383588722907007e-05,
      "learning_rate": 3.793481223780651e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9210752,
      "step": 32895
    },
    {
      "epoch": 365.55555555555554,
      "grad_norm": 7.790033123455942e-05,
      "learning_rate": 3.788283731425496e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9212144,
      "step": 32900
    },
    {
      "epoch": 365.6111111111111,
      "grad_norm": 7.893832662375644e-05,
      "learning_rate": 3.7830895101835488e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9213568,
      "step": 32905
    },
    {
      "epoch": 365.6666666666667,
      "grad_norm": 5.659040107275359e-05,
      "learning_rate": 3.7778985608558274e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9214992,
      "step": 32910
    },
    {
      "epoch": 365.72222222222223,
      "grad_norm": 0.00044102713582105935,
      "learning_rate": 3.7727108842428443e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9216384,
      "step": 32915
    },
    {
      "epoch": 365.77777777777777,
      "grad_norm": 9.702783427201211e-05,
      "learning_rate": 3.7675264811446065e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9217792,
      "step": 32920
    },
    {
      "epoch": 365.8333333333333,
      "grad_norm": 0.0001890073181129992,
      "learning_rate": 3.7623453523605994e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9219216,
      "step": 32925
    },
    {
      "epoch": 365.8888888888889,
      "grad_norm": 0.0020806090906262398,
      "learning_rate": 3.757167498689834e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9220592,
      "step": 32930
    },
    {
      "epoch": 365.94444444444446,
      "grad_norm": 5.299601616570726e-05,
      "learning_rate": 3.7519929209307914e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9221936,
      "step": 32935
    },
    {
      "epoch": 366.0,
      "grad_norm": 3.0921775760361925e-05,
      "learning_rate": 3.746821619881463e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9223312,
      "step": 32940
    },
    {
      "epoch": 366.05555555555554,
      "grad_norm": 0.00013912181020714343,
      "learning_rate": 3.74165359633932e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9224720,
      "step": 32945
    },
    {
      "epoch": 366.1111111111111,
      "grad_norm": 0.0017749161925166845,
      "learning_rate": 3.736488851101341e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9226128,
      "step": 32950
    },
    {
      "epoch": 366.1666666666667,
      "grad_norm": 2.895328725571744e-05,
      "learning_rate": 3.7313273849640035e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9227520,
      "step": 32955
    },
    {
      "epoch": 366.22222222222223,
      "grad_norm": 3.3902746508829296e-05,
      "learning_rate": 3.7261691987232533e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9228880,
      "step": 32960
    },
    {
      "epoch": 366.27777777777777,
      "grad_norm": 0.0002325365785509348,
      "learning_rate": 3.7210142931745575e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9230240,
      "step": 32965
    },
    {
      "epoch": 366.3333333333333,
      "grad_norm": 0.0003059325972571969,
      "learning_rate": 3.7158626691128712e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9231648,
      "step": 32970
    },
    {
      "epoch": 366.3888888888889,
      "grad_norm": 1.802407314244192e-05,
      "learning_rate": 3.710714327332629e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9233008,
      "step": 32975
    },
    {
      "epoch": 366.44444444444446,
      "grad_norm": 5.643266558763571e-05,
      "learning_rate": 3.7055692686277815e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9234400,
      "step": 32980
    },
    {
      "epoch": 366.5,
      "grad_norm": 0.00015462868032045662,
      "learning_rate": 3.70042749379175e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9235776,
      "step": 32985
    },
    {
      "epoch": 366.55555555555554,
      "grad_norm": 0.00015451990475412458,
      "learning_rate": 3.6952890036174693e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9237216,
      "step": 32990
    },
    {
      "epoch": 366.6111111111111,
      "grad_norm": 4.623666245606728e-05,
      "learning_rate": 3.690153798897353e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9238656,
      "step": 32995
    },
    {
      "epoch": 366.6666666666667,
      "grad_norm": 0.0006179266492836177,
      "learning_rate": 3.6850218804233225e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9240064,
      "step": 33000
    },
    {
      "epoch": 366.6666666666667,
      "eval_loss": 0.873401939868927,
      "eval_runtime": 1.24,
      "eval_samples_per_second": 32.259,
      "eval_steps_per_second": 16.13,
      "num_input_tokens_seen": 9240064,
      "step": 33000
    },
    {
      "epoch": 366.72222222222223,
      "grad_norm": 0.002429524203762412,
      "learning_rate": 3.679893248986779e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9241456,
      "step": 33005
    },
    {
      "epoch": 366.77777777777777,
      "grad_norm": 5.206227797316387e-05,
      "learning_rate": 3.6747679053786147e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9242864,
      "step": 33010
    },
    {
      "epoch": 366.8333333333333,
      "grad_norm": 0.0001508292625658214,
      "learning_rate": 3.669645850389228e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9244272,
      "step": 33015
    },
    {
      "epoch": 366.8888888888889,
      "grad_norm": 0.0002218247827840969,
      "learning_rate": 3.664527084808514e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9245664,
      "step": 33020
    },
    {
      "epoch": 366.94444444444446,
      "grad_norm": 7.517473568441346e-05,
      "learning_rate": 3.6594116094258337e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9247056,
      "step": 33025
    },
    {
      "epoch": 367.0,
      "grad_norm": 1.7151853171526454e-05,
      "learning_rate": 3.6542994250300665e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9248496,
      "step": 33030
    },
    {
      "epoch": 367.05555555555554,
      "grad_norm": 8.133317896863446e-05,
      "learning_rate": 3.6491905324095825e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9249888,
      "step": 33035
    },
    {
      "epoch": 367.1111111111111,
      "grad_norm": 0.00011756871390389279,
      "learning_rate": 3.644084932352221e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9251248,
      "step": 33040
    },
    {
      "epoch": 367.1666666666667,
      "grad_norm": 5.612305540125817e-05,
      "learning_rate": 3.6389826256453457e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9252672,
      "step": 33045
    },
    {
      "epoch": 367.22222222222223,
      "grad_norm": 0.0003516394062899053,
      "learning_rate": 3.633883613075781e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9254080,
      "step": 33050
    },
    {
      "epoch": 367.27777777777777,
      "grad_norm": 0.00018439299310557544,
      "learning_rate": 3.6287878954298693e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9255488,
      "step": 33055
    },
    {
      "epoch": 367.3333333333333,
      "grad_norm": 8.672101830597967e-05,
      "learning_rate": 3.6236954734934354e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9256912,
      "step": 33060
    },
    {
      "epoch": 367.3888888888889,
      "grad_norm": 0.00038235506508499384,
      "learning_rate": 3.618606348051784e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9258288,
      "step": 33065
    },
    {
      "epoch": 367.44444444444446,
      "grad_norm": 0.0004952909075655043,
      "learning_rate": 3.6135205198897376e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9259664,
      "step": 33070
    },
    {
      "epoch": 367.5,
      "grad_norm": 7.236192323034629e-05,
      "learning_rate": 3.6084379897915854e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9261056,
      "step": 33075
    },
    {
      "epoch": 367.55555555555554,
      "grad_norm": 9.028545900946483e-05,
      "learning_rate": 3.6033587585411115e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9262448,
      "step": 33080
    },
    {
      "epoch": 367.6111111111111,
      "grad_norm": 0.00038971827598288655,
      "learning_rate": 3.5982828269216117e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9263840,
      "step": 33085
    },
    {
      "epoch": 367.6666666666667,
      "grad_norm": 9.149862307822332e-05,
      "learning_rate": 3.593210195715843e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9265232,
      "step": 33090
    },
    {
      "epoch": 367.72222222222223,
      "grad_norm": 1.5815821825526655e-05,
      "learning_rate": 3.5881408657060773e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9266640,
      "step": 33095
    },
    {
      "epoch": 367.77777777777777,
      "grad_norm": 0.00013177553773857653,
      "learning_rate": 3.583074837674075e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9268000,
      "step": 33100
    },
    {
      "epoch": 367.8333333333333,
      "grad_norm": 0.00011363579687895253,
      "learning_rate": 3.578012112401069e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9269424,
      "step": 33105
    },
    {
      "epoch": 367.8888888888889,
      "grad_norm": 0.00043299479875713587,
      "learning_rate": 3.5729526906677996e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9270880,
      "step": 33110
    },
    {
      "epoch": 367.94444444444446,
      "grad_norm": 6.205474346643314e-05,
      "learning_rate": 3.5678965732545007e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9272272,
      "step": 33115
    },
    {
      "epoch": 368.0,
      "grad_norm": 0.0001355481072096154,
      "learning_rate": 3.562843760940876e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9273664,
      "step": 33120
    },
    {
      "epoch": 368.05555555555554,
      "grad_norm": 0.00032646069303154945,
      "learning_rate": 3.5577942545061473e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9275088,
      "step": 33125
    },
    {
      "epoch": 368.1111111111111,
      "grad_norm": 6.183629739098251e-05,
      "learning_rate": 3.5527480547289967e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9276464,
      "step": 33130
    },
    {
      "epoch": 368.1666666666667,
      "grad_norm": 5.7969293266069144e-05,
      "learning_rate": 3.547705162387624e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9277872,
      "step": 33135
    },
    {
      "epoch": 368.22222222222223,
      "grad_norm": 1.7326947272522375e-05,
      "learning_rate": 3.542665578259699e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9279248,
      "step": 33140
    },
    {
      "epoch": 368.27777777777777,
      "grad_norm": 0.00040616426849737763,
      "learning_rate": 3.5376293031223945e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9280640,
      "step": 33145
    },
    {
      "epoch": 368.3333333333333,
      "grad_norm": 2.960425990750082e-05,
      "learning_rate": 3.5325963377523614e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9282000,
      "step": 33150
    },
    {
      "epoch": 368.3888888888889,
      "grad_norm": 6.460466829594225e-05,
      "learning_rate": 3.5275666829257536e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9283424,
      "step": 33155
    },
    {
      "epoch": 368.44444444444446,
      "grad_norm": 0.00017952462076209486,
      "learning_rate": 3.5225403394181955e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9284816,
      "step": 33160
    },
    {
      "epoch": 368.5,
      "grad_norm": 0.0004684609593823552,
      "learning_rate": 3.517517308004828e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9286240,
      "step": 33165
    },
    {
      "epoch": 368.55555555555554,
      "grad_norm": 0.0005902953562326729,
      "learning_rate": 3.512497589460251e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9287664,
      "step": 33170
    },
    {
      "epoch": 368.6111111111111,
      "grad_norm": 8.479456300847232e-05,
      "learning_rate": 3.5074811845585727e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9289040,
      "step": 33175
    },
    {
      "epoch": 368.6666666666667,
      "grad_norm": 8.609942597104236e-05,
      "learning_rate": 3.5024680940733937e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9290416,
      "step": 33180
    },
    {
      "epoch": 368.72222222222223,
      "grad_norm": 0.0002500995760783553,
      "learning_rate": 3.4974583187777852e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9291824,
      "step": 33185
    },
    {
      "epoch": 368.77777777777777,
      "grad_norm": 7.879933400545269e-05,
      "learning_rate": 3.4924518594443204e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9293200,
      "step": 33190
    },
    {
      "epoch": 368.8333333333333,
      "grad_norm": 0.00016702679567970335,
      "learning_rate": 3.4874487168450682e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9294560,
      "step": 33195
    },
    {
      "epoch": 368.8888888888889,
      "grad_norm": 0.00030812143813818693,
      "learning_rate": 3.482448891751558e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9295952,
      "step": 33200
    },
    {
      "epoch": 368.8888888888889,
      "eval_loss": 0.923679530620575,
      "eval_runtime": 1.2413,
      "eval_samples_per_second": 32.224,
      "eval_steps_per_second": 16.112,
      "num_input_tokens_seen": 9295952,
      "step": 33200
    },
    {
      "epoch": 368.94444444444446,
      "grad_norm": 0.0008176242699846625,
      "learning_rate": 3.477452384934843e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9297408,
      "step": 33205
    },
    {
      "epoch": 369.0,
      "grad_norm": 0.0022684200666844845,
      "learning_rate": 3.472459197165434e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9298832,
      "step": 33210
    },
    {
      "epoch": 369.05555555555554,
      "grad_norm": 7.770951924612746e-05,
      "learning_rate": 3.4674693292133518e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9300224,
      "step": 33215
    },
    {
      "epoch": 369.1111111111111,
      "grad_norm": 0.00016752109513618052,
      "learning_rate": 3.4624827818480977e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9301584,
      "step": 33220
    },
    {
      "epoch": 369.1666666666667,
      "grad_norm": 0.0019288038602098823,
      "learning_rate": 3.4574995558386474e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9302976,
      "step": 33225
    },
    {
      "epoch": 369.22222222222223,
      "grad_norm": 0.00010564545664237812,
      "learning_rate": 3.452519651953487e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9304384,
      "step": 33230
    },
    {
      "epoch": 369.27777777777777,
      "grad_norm": 2.4672264771652408e-05,
      "learning_rate": 3.447543070960585e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9305792,
      "step": 33235
    },
    {
      "epoch": 369.3333333333333,
      "grad_norm": 3.646976620075293e-05,
      "learning_rate": 3.4425698136273778e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9307184,
      "step": 33240
    },
    {
      "epoch": 369.3888888888889,
      "grad_norm": 0.00011341802746756002,
      "learning_rate": 3.437599880720821e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9308592,
      "step": 33245
    },
    {
      "epoch": 369.44444444444446,
      "grad_norm": 0.0002371474984101951,
      "learning_rate": 3.4326332730073267e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9309984,
      "step": 33250
    },
    {
      "epoch": 369.5,
      "grad_norm": 4.930803333991207e-05,
      "learning_rate": 3.427669991252813e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9311360,
      "step": 33255
    },
    {
      "epoch": 369.55555555555554,
      "grad_norm": 2.1235993699519895e-05,
      "learning_rate": 3.42271003622269e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9312736,
      "step": 33260
    },
    {
      "epoch": 369.6111111111111,
      "grad_norm": 0.00014425680274143815,
      "learning_rate": 3.4177534086818286e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9314112,
      "step": 33265
    },
    {
      "epoch": 369.6666666666667,
      "grad_norm": 1.3818704246659763e-05,
      "learning_rate": 3.412800109394612e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9315584,
      "step": 33270
    },
    {
      "epoch": 369.72222222222223,
      "grad_norm": 0.000360522186383605,
      "learning_rate": 3.4078501391249044e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9317008,
      "step": 33275
    },
    {
      "epoch": 369.77777777777777,
      "grad_norm": 2.6513907869230025e-05,
      "learning_rate": 3.4029034986360453e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9318416,
      "step": 33280
    },
    {
      "epoch": 369.8333333333333,
      "grad_norm": 0.0014038709923624992,
      "learning_rate": 3.397960188690877e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9319776,
      "step": 33285
    },
    {
      "epoch": 369.8888888888889,
      "grad_norm": 0.001240986050106585,
      "learning_rate": 3.393020210051717e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9321200,
      "step": 33290
    },
    {
      "epoch": 369.94444444444446,
      "grad_norm": 0.0002172648091800511,
      "learning_rate": 3.3880835634803655e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9322624,
      "step": 33295
    },
    {
      "epoch": 370.0,
      "grad_norm": 0.0001060038703144528,
      "learning_rate": 3.383150249738126e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9324064,
      "step": 33300
    },
    {
      "epoch": 370.05555555555554,
      "grad_norm": 3.259710138081573e-05,
      "learning_rate": 3.3782202695857663e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9325472,
      "step": 33305
    },
    {
      "epoch": 370.1111111111111,
      "grad_norm": 3.227982597309165e-05,
      "learning_rate": 3.373293623783558e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9326880,
      "step": 33310
    },
    {
      "epoch": 370.1666666666667,
      "grad_norm": 0.00010119222861249,
      "learning_rate": 3.368370313091257e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9328304,
      "step": 33315
    },
    {
      "epoch": 370.22222222222223,
      "grad_norm": 8.354769670404494e-05,
      "learning_rate": 3.363450338268087e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9329648,
      "step": 33320
    },
    {
      "epoch": 370.27777777777777,
      "grad_norm": 0.00010631128679960966,
      "learning_rate": 3.358533700072783e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9331040,
      "step": 33325
    },
    {
      "epoch": 370.3333333333333,
      "grad_norm": 0.0001678566331975162,
      "learning_rate": 3.3536203992635377e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9332448,
      "step": 33330
    },
    {
      "epoch": 370.3888888888889,
      "grad_norm": 7.31184845790267e-05,
      "learning_rate": 3.348710436598057e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9333872,
      "step": 33335
    },
    {
      "epoch": 370.44444444444446,
      "grad_norm": 8.093902579275891e-05,
      "learning_rate": 3.3438038128335155e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9335264,
      "step": 33340
    },
    {
      "epoch": 370.5,
      "grad_norm": 6.369650509441271e-05,
      "learning_rate": 3.338900528726571e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9336672,
      "step": 33345
    },
    {
      "epoch": 370.55555555555554,
      "grad_norm": 5.031626278650947e-05,
      "learning_rate": 3.3340005850333812e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9338000,
      "step": 33350
    },
    {
      "epoch": 370.6111111111111,
      "grad_norm": 8.316320599988103e-05,
      "learning_rate": 3.329103982509568e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9339376,
      "step": 33355
    },
    {
      "epoch": 370.6666666666667,
      "grad_norm": 0.00013301610306371003,
      "learning_rate": 3.324210721910259e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9340784,
      "step": 33360
    },
    {
      "epoch": 370.72222222222223,
      "grad_norm": 6.751055479981005e-05,
      "learning_rate": 3.319320803990053e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9342192,
      "step": 33365
    },
    {
      "epoch": 370.77777777777777,
      "grad_norm": 0.00019862569752149284,
      "learning_rate": 3.3144342295030274e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9343568,
      "step": 33370
    },
    {
      "epoch": 370.8333333333333,
      "grad_norm": 0.0004029884294141084,
      "learning_rate": 3.309550999202765e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9344928,
      "step": 33375
    },
    {
      "epoch": 370.8888888888889,
      "grad_norm": 0.0015894530806690454,
      "learning_rate": 3.3046711138423197e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9346336,
      "step": 33380
    },
    {
      "epoch": 370.94444444444446,
      "grad_norm": 0.0002334397577214986,
      "learning_rate": 3.2997945741742255e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9347776,
      "step": 33385
    },
    {
      "epoch": 371.0,
      "grad_norm": 0.00013332215894479305,
      "learning_rate": 3.2949213809505082e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9349184,
      "step": 33390
    },
    {
      "epoch": 371.05555555555554,
      "grad_norm": 0.0006720647215843201,
      "learning_rate": 3.2900515349226834e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9350592,
      "step": 33395
    },
    {
      "epoch": 371.1111111111111,
      "grad_norm": 3.873289460898377e-05,
      "learning_rate": 3.285185036841731e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9352016,
      "step": 33400
    },
    {
      "epoch": 371.1111111111111,
      "eval_loss": 0.8651801943778992,
      "eval_runtime": 1.248,
      "eval_samples_per_second": 32.052,
      "eval_steps_per_second": 16.026,
      "num_input_tokens_seen": 9352016,
      "step": 33400
    },
    {
      "epoch": 371.1666666666667,
      "grad_norm": 0.0012358477106317878,
      "learning_rate": 3.2803218874581377e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9353392,
      "step": 33405
    },
    {
      "epoch": 371.22222222222223,
      "grad_norm": 0.00010008500248659402,
      "learning_rate": 3.2754620875218494e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9354816,
      "step": 33410
    },
    {
      "epoch": 371.27777777777777,
      "grad_norm": 0.0002007932198466733,
      "learning_rate": 3.2706056377823146e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9356160,
      "step": 33415
    },
    {
      "epoch": 371.3333333333333,
      "grad_norm": 2.6169580451096408e-05,
      "learning_rate": 3.2657525389884647e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9357568,
      "step": 33420
    },
    {
      "epoch": 371.3888888888889,
      "grad_norm": 3.828995977528393e-05,
      "learning_rate": 3.260902791888698e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9358992,
      "step": 33425
    },
    {
      "epoch": 371.44444444444446,
      "grad_norm": 2.275631959491875e-05,
      "learning_rate": 3.2560563972309166e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9360432,
      "step": 33430
    },
    {
      "epoch": 371.5,
      "grad_norm": 6.378500984283164e-05,
      "learning_rate": 3.251213355762489e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9361840,
      "step": 33435
    },
    {
      "epoch": 371.55555555555554,
      "grad_norm": 4.7301720769610256e-05,
      "learning_rate": 3.2463736682302707e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9363232,
      "step": 33440
    },
    {
      "epoch": 371.6111111111111,
      "grad_norm": 4.4100554077886045e-05,
      "learning_rate": 3.2415373353806124e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9364640,
      "step": 33445
    },
    {
      "epoch": 371.6666666666667,
      "grad_norm": 0.00027134723495692015,
      "learning_rate": 3.236704357959322e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9365984,
      "step": 33450
    },
    {
      "epoch": 371.72222222222223,
      "grad_norm": 7.380745955742896e-05,
      "learning_rate": 3.2318747367117154e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9367376,
      "step": 33455
    },
    {
      "epoch": 371.77777777777777,
      "grad_norm": 0.002951089758425951,
      "learning_rate": 3.227048472382585e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9368768,
      "step": 33460
    },
    {
      "epoch": 371.8333333333333,
      "grad_norm": 0.0002295162994414568,
      "learning_rate": 3.2222255657161915e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9370192,
      "step": 33465
    },
    {
      "epoch": 371.8888888888889,
      "grad_norm": 6.871105142636225e-05,
      "learning_rate": 3.2174060174562924e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9371568,
      "step": 33470
    },
    {
      "epoch": 371.94444444444446,
      "grad_norm": 0.00029530824394896626,
      "learning_rate": 3.2125898283461298e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9372992,
      "step": 33475
    },
    {
      "epoch": 372.0,
      "grad_norm": 1.6156735000549816e-05,
      "learning_rate": 3.207776999128406e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9374416,
      "step": 33480
    },
    {
      "epoch": 372.05555555555554,
      "grad_norm": 0.0001241381251020357,
      "learning_rate": 3.202967530545331e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9375824,
      "step": 33485
    },
    {
      "epoch": 372.1111111111111,
      "grad_norm": 0.00025128989364020526,
      "learning_rate": 3.1981614233385778e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9377168,
      "step": 33490
    },
    {
      "epoch": 372.1666666666667,
      "grad_norm": 7.942088268464431e-05,
      "learning_rate": 3.1933586782493115e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9378560,
      "step": 33495
    },
    {
      "epoch": 372.22222222222223,
      "grad_norm": 0.00028346406179480255,
      "learning_rate": 3.188559296018184e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9379936,
      "step": 33500
    },
    {
      "epoch": 372.27777777777777,
      "grad_norm": 6.179814226925373e-05,
      "learning_rate": 3.1837632773853098e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9381312,
      "step": 33505
    },
    {
      "epoch": 372.3333333333333,
      "grad_norm": 6.92777248332277e-05,
      "learning_rate": 3.178970623090294e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9382688,
      "step": 33510
    },
    {
      "epoch": 372.3888888888889,
      "grad_norm": 0.0005479419487528503,
      "learning_rate": 3.174181333872234e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9384048,
      "step": 33515
    },
    {
      "epoch": 372.44444444444446,
      "grad_norm": 2.434295674902387e-05,
      "learning_rate": 3.169395410469686e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9385472,
      "step": 33520
    },
    {
      "epoch": 372.5,
      "grad_norm": 3.499534432194196e-05,
      "learning_rate": 3.164612853620713e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9386896,
      "step": 33525
    },
    {
      "epoch": 372.55555555555554,
      "grad_norm": 0.00010183866834267974,
      "learning_rate": 3.1598336640628333e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9388272,
      "step": 33530
    },
    {
      "epoch": 372.6111111111111,
      "grad_norm": 7.553931936854497e-05,
      "learning_rate": 3.155057842533063e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9389696,
      "step": 33535
    },
    {
      "epoch": 372.6666666666667,
      "grad_norm": 0.00021728983847424388,
      "learning_rate": 3.1502853897678984e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9391104,
      "step": 33540
    },
    {
      "epoch": 372.72222222222223,
      "grad_norm": 3.924971315427683e-05,
      "learning_rate": 3.1455163065033017e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9392512,
      "step": 33545
    },
    {
      "epoch": 372.77777777777777,
      "grad_norm": 6.03760163357947e-05,
      "learning_rate": 3.140750593474734e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9393872,
      "step": 33550
    },
    {
      "epoch": 372.8333333333333,
      "grad_norm": 5.971331847831607e-05,
      "learning_rate": 3.1359882514171294e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9395280,
      "step": 33555
    },
    {
      "epoch": 372.8888888888889,
      "grad_norm": 2.9445196560118347e-05,
      "learning_rate": 3.1312292810648903e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9396704,
      "step": 33560
    },
    {
      "epoch": 372.94444444444446,
      "grad_norm": 7.232294592540711e-05,
      "learning_rate": 3.1264736831519204e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9398144,
      "step": 33565
    },
    {
      "epoch": 373.0,
      "grad_norm": 0.0007076594047248363,
      "learning_rate": 3.1217214584115863e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9399584,
      "step": 33570
    },
    {
      "epoch": 373.05555555555554,
      "grad_norm": 0.00017995746748056263,
      "learning_rate": 3.116972607576746e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9400992,
      "step": 33575
    },
    {
      "epoch": 373.1111111111111,
      "grad_norm": 0.0002735976886469871,
      "learning_rate": 3.1122271313797303e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9402368,
      "step": 33580
    },
    {
      "epoch": 373.1666666666667,
      "grad_norm": 4.098413774045184e-05,
      "learning_rate": 3.107485030552343e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9403744,
      "step": 33585
    },
    {
      "epoch": 373.22222222222223,
      "grad_norm": 5.362758747651242e-05,
      "learning_rate": 3.1027463058258848e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9405136,
      "step": 33590
    },
    {
      "epoch": 373.27777777777777,
      "grad_norm": 8.969258487923071e-05,
      "learning_rate": 3.0980109579311273e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9406512,
      "step": 33595
    },
    {
      "epoch": 373.3333333333333,
      "grad_norm": 8.21998473838903e-05,
      "learning_rate": 3.093278987598314e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9407968,
      "step": 33600
    },
    {
      "epoch": 373.3333333333333,
      "eval_loss": 0.9285857081413269,
      "eval_runtime": 1.2354,
      "eval_samples_per_second": 32.377,
      "eval_steps_per_second": 16.189,
      "num_input_tokens_seen": 9407968,
      "step": 33600
    },
    {
      "epoch": 373.3888888888889,
      "grad_norm": 5.354523818823509e-05,
      "learning_rate": 3.0885503955571826e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9409376,
      "step": 33605
    },
    {
      "epoch": 373.44444444444446,
      "grad_norm": 1.3380661584960762e-05,
      "learning_rate": 3.0838251825369313e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9410752,
      "step": 33610
    },
    {
      "epoch": 373.5,
      "grad_norm": 0.004892196971923113,
      "learning_rate": 3.0791033492662517e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9412144,
      "step": 33615
    },
    {
      "epoch": 373.55555555555554,
      "grad_norm": 0.0008157311240211129,
      "learning_rate": 3.0743848964733203e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9413552,
      "step": 33620
    },
    {
      "epoch": 373.6111111111111,
      "grad_norm": 0.0008147642365656793,
      "learning_rate": 3.0696698248857625e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9414912,
      "step": 33625
    },
    {
      "epoch": 373.6666666666667,
      "grad_norm": 0.00036167335929349065,
      "learning_rate": 3.0649581352307192e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9416336,
      "step": 33630
    },
    {
      "epoch": 373.72222222222223,
      "grad_norm": 0.017160307615995407,
      "learning_rate": 3.060249828234776e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9417728,
      "step": 33635
    },
    {
      "epoch": 373.77777777777777,
      "grad_norm": 0.0008107605390250683,
      "learning_rate": 3.055544904624025e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9419120,
      "step": 33640
    },
    {
      "epoch": 373.8333333333333,
      "grad_norm": 1.65702476806473e-05,
      "learning_rate": 3.050843365124026e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9420560,
      "step": 33645
    },
    {
      "epoch": 373.8888888888889,
      "grad_norm": 1.9428507584962063e-05,
      "learning_rate": 3.0461452104598083e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9421904,
      "step": 33650
    },
    {
      "epoch": 373.94444444444446,
      "grad_norm": 3.1658779334975407e-05,
      "learning_rate": 3.0414504413558836e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9423312,
      "step": 33655
    },
    {
      "epoch": 374.0,
      "grad_norm": 0.00013473878789227456,
      "learning_rate": 3.0367590585362564e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9424752,
      "step": 33660
    },
    {
      "epoch": 374.05555555555554,
      "grad_norm": 9.863647574093193e-05,
      "learning_rate": 3.0320710627243813e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9426208,
      "step": 33665
    },
    {
      "epoch": 374.1111111111111,
      "grad_norm": 0.0003111370315309614,
      "learning_rate": 3.027386454643222e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9427552,
      "step": 33670
    },
    {
      "epoch": 374.1666666666667,
      "grad_norm": 8.919542597141117e-05,
      "learning_rate": 3.0227052350151914e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9428928,
      "step": 33675
    },
    {
      "epoch": 374.22222222222223,
      "grad_norm": 0.00023267578217200935,
      "learning_rate": 3.0180274045621957e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9430288,
      "step": 33680
    },
    {
      "epoch": 374.27777777777777,
      "grad_norm": 0.00011555825039977208,
      "learning_rate": 3.013352964005625e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9431712,
      "step": 33685
    },
    {
      "epoch": 374.3333333333333,
      "grad_norm": 0.0014882683753967285,
      "learning_rate": 3.0086819140663218e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9433104,
      "step": 33690
    },
    {
      "epoch": 374.3888888888889,
      "grad_norm": 0.0001917836780194193,
      "learning_rate": 3.0040142554646265e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9434496,
      "step": 33695
    },
    {
      "epoch": 374.44444444444446,
      "grad_norm": 0.00014154693053569645,
      "learning_rate": 2.999349988920361e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9435888,
      "step": 33700
    },
    {
      "epoch": 374.5,
      "grad_norm": 6.266387208597735e-05,
      "learning_rate": 2.994689115152796e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9437296,
      "step": 33705
    },
    {
      "epoch": 374.55555555555554,
      "grad_norm": 0.00032821117201820016,
      "learning_rate": 2.9900316348807105e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9438688,
      "step": 33710
    },
    {
      "epoch": 374.6111111111111,
      "grad_norm": 0.0014995145611464977,
      "learning_rate": 2.985377548822338e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9440096,
      "step": 33715
    },
    {
      "epoch": 374.6666666666667,
      "grad_norm": 6.873450911371037e-05,
      "learning_rate": 2.980726857695404e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9441488,
      "step": 33720
    },
    {
      "epoch": 374.72222222222223,
      "grad_norm": 0.00033862789859995246,
      "learning_rate": 2.9760795622171017e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9442880,
      "step": 33725
    },
    {
      "epoch": 374.77777777777777,
      "grad_norm": 0.0007960483781062067,
      "learning_rate": 2.971435663104094e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9444288,
      "step": 33730
    },
    {
      "epoch": 374.8333333333333,
      "grad_norm": 2.868884075724054e-05,
      "learning_rate": 2.9667951610725385e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9445664,
      "step": 33735
    },
    {
      "epoch": 374.8888888888889,
      "grad_norm": 7.736693078186363e-05,
      "learning_rate": 2.9621580568380575e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9447008,
      "step": 33740
    },
    {
      "epoch": 374.94444444444446,
      "grad_norm": 0.0010154295014217496,
      "learning_rate": 2.9575243511157453e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9448464,
      "step": 33745
    },
    {
      "epoch": 375.0,
      "grad_norm": 9.473228419665247e-05,
      "learning_rate": 2.952894044620186e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9449888,
      "step": 33750
    },
    {
      "epoch": 375.05555555555554,
      "grad_norm": 8.45498580019921e-05,
      "learning_rate": 2.948267138065419e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9451264,
      "step": 33755
    },
    {
      "epoch": 375.1111111111111,
      "grad_norm": 0.00010224144352832809,
      "learning_rate": 2.943643632164983e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9452624,
      "step": 33760
    },
    {
      "epoch": 375.1666666666667,
      "grad_norm": 4.2949435737682506e-05,
      "learning_rate": 2.939023527631879e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9454096,
      "step": 33765
    },
    {
      "epoch": 375.22222222222223,
      "grad_norm": 6.710195884807035e-05,
      "learning_rate": 2.934406825178576e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9455504,
      "step": 33770
    },
    {
      "epoch": 375.27777777777777,
      "grad_norm": 0.00013680224947165698,
      "learning_rate": 2.9297935255170357e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9456912,
      "step": 33775
    },
    {
      "epoch": 375.3333333333333,
      "grad_norm": 2.390089321124833e-05,
      "learning_rate": 2.925183629358691e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9458304,
      "step": 33780
    },
    {
      "epoch": 375.3888888888889,
      "grad_norm": 0.0001531100133433938,
      "learning_rate": 2.9205771374144346e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9459744,
      "step": 33785
    },
    {
      "epoch": 375.44444444444446,
      "grad_norm": 0.0002842645044438541,
      "learning_rate": 2.915974050394657e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9461120,
      "step": 33790
    },
    {
      "epoch": 375.5,
      "grad_norm": 0.001104877912439406,
      "learning_rate": 2.9113743690092067e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9462512,
      "step": 33795
    },
    {
      "epoch": 375.55555555555554,
      "grad_norm": 0.0009178849868476391,
      "learning_rate": 2.906778093967402e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9463920,
      "step": 33800
    },
    {
      "epoch": 375.55555555555554,
      "eval_loss": 0.9073829650878906,
      "eval_runtime": 1.2429,
      "eval_samples_per_second": 32.182,
      "eval_steps_per_second": 16.091,
      "num_input_tokens_seen": 9463920,
      "step": 33800
    },
    {
      "epoch": 375.6111111111111,
      "grad_norm": 7.602602272527292e-05,
      "learning_rate": 2.9021852259780656e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9465312,
      "step": 33805
    },
    {
      "epoch": 375.6666666666667,
      "grad_norm": 0.0001659805711824447,
      "learning_rate": 2.8975957657494583e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9466736,
      "step": 33810
    },
    {
      "epoch": 375.72222222222223,
      "grad_norm": 3.988064418081194e-05,
      "learning_rate": 2.8930097139893417e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9468112,
      "step": 33815
    },
    {
      "epoch": 375.77777777777777,
      "grad_norm": 8.552025246899575e-05,
      "learning_rate": 2.888427071404945e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9469536,
      "step": 33820
    },
    {
      "epoch": 375.8333333333333,
      "grad_norm": 3.08713351842016e-05,
      "learning_rate": 2.8838478387029606e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9470928,
      "step": 33825
    },
    {
      "epoch": 375.8888888888889,
      "grad_norm": 5.743449219153263e-05,
      "learning_rate": 2.8792720165895737e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9472304,
      "step": 33830
    },
    {
      "epoch": 375.94444444444446,
      "grad_norm": 5.98211663600523e-05,
      "learning_rate": 2.874699605770423e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9473712,
      "step": 33835
    },
    {
      "epoch": 376.0,
      "grad_norm": 0.0004546926938928664,
      "learning_rate": 2.8701306069506383e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9475136,
      "step": 33840
    },
    {
      "epoch": 376.05555555555554,
      "grad_norm": 0.0012857766123488545,
      "learning_rate": 2.8655650208348178e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9476544,
      "step": 33845
    },
    {
      "epoch": 376.1111111111111,
      "grad_norm": 4.9282574764220044e-05,
      "learning_rate": 2.8610028481270257e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9477936,
      "step": 33850
    },
    {
      "epoch": 376.1666666666667,
      "grad_norm": 0.00011471178731881082,
      "learning_rate": 2.856444089530813e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9479312,
      "step": 33855
    },
    {
      "epoch": 376.22222222222223,
      "grad_norm": 5.0426820962456986e-05,
      "learning_rate": 2.8518887457491955e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9480688,
      "step": 33860
    },
    {
      "epoch": 376.27777777777777,
      "grad_norm": 6.651330477325246e-05,
      "learning_rate": 2.8473368174846666e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9482128,
      "step": 33865
    },
    {
      "epoch": 376.3333333333333,
      "grad_norm": 7.004178041825071e-05,
      "learning_rate": 2.842788305439184e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9483568,
      "step": 33870
    },
    {
      "epoch": 376.3888888888889,
      "grad_norm": 4.8427045840071514e-05,
      "learning_rate": 2.8382432103141925e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9484928,
      "step": 33875
    },
    {
      "epoch": 376.44444444444446,
      "grad_norm": 7.493988232454285e-05,
      "learning_rate": 2.833701532810598e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9486336,
      "step": 33880
    },
    {
      "epoch": 376.5,
      "grad_norm": 0.00015345749852713197,
      "learning_rate": 2.8291632736287877e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9487744,
      "step": 33885
    },
    {
      "epoch": 376.55555555555554,
      "grad_norm": 3.71784481103532e-05,
      "learning_rate": 2.824628433468615e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9489088,
      "step": 33890
    },
    {
      "epoch": 376.6111111111111,
      "grad_norm": 0.00014995936362538487,
      "learning_rate": 2.8200970130294073e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9490480,
      "step": 33895
    },
    {
      "epoch": 376.6666666666667,
      "grad_norm": 0.00035489685251377523,
      "learning_rate": 2.8155690130099775e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9491872,
      "step": 33900
    },
    {
      "epoch": 376.72222222222223,
      "grad_norm": 0.0002417651703581214,
      "learning_rate": 2.8110444341085895e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9493280,
      "step": 33905
    },
    {
      "epoch": 376.77777777777777,
      "grad_norm": 2.2546581021742895e-05,
      "learning_rate": 2.806523277022996e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9494672,
      "step": 33910
    },
    {
      "epoch": 376.8333333333333,
      "grad_norm": 2.9401240681181662e-05,
      "learning_rate": 2.802005542450409e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9496080,
      "step": 33915
    },
    {
      "epoch": 376.8888888888889,
      "grad_norm": 5.326988684828393e-05,
      "learning_rate": 2.797491231087526e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9497504,
      "step": 33920
    },
    {
      "epoch": 376.94444444444446,
      "grad_norm": 2.3479324227082543e-05,
      "learning_rate": 2.7929803436305137e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9498928,
      "step": 33925
    },
    {
      "epoch": 377.0,
      "grad_norm": 0.0017630604561418295,
      "learning_rate": 2.788472880774998e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9500336,
      "step": 33930
    },
    {
      "epoch": 377.05555555555554,
      "grad_norm": 9.264640539186075e-05,
      "learning_rate": 2.7839688432160977e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9501760,
      "step": 33935
    },
    {
      "epoch": 377.1111111111111,
      "grad_norm": 0.0002470083418302238,
      "learning_rate": 2.779468231648383e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9503136,
      "step": 33940
    },
    {
      "epoch": 377.1666666666667,
      "grad_norm": 0.00021012141951359808,
      "learning_rate": 2.774971046765906e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9504560,
      "step": 33945
    },
    {
      "epoch": 377.22222222222223,
      "grad_norm": 0.00011853982869070023,
      "learning_rate": 2.770477289262194e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9505936,
      "step": 33950
    },
    {
      "epoch": 377.27777777777777,
      "grad_norm": 4.699034252553247e-05,
      "learning_rate": 2.765986959830233e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9507328,
      "step": 33955
    },
    {
      "epoch": 377.3333333333333,
      "grad_norm": 0.0003333112981636077,
      "learning_rate": 2.761500059162492e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9508768,
      "step": 33960
    },
    {
      "epoch": 377.3888888888889,
      "grad_norm": 0.0019085102248936892,
      "learning_rate": 2.757016587950914e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9510112,
      "step": 33965
    },
    {
      "epoch": 377.44444444444446,
      "grad_norm": 3.185322930221446e-05,
      "learning_rate": 2.752536546886897e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9511536,
      "step": 33970
    },
    {
      "epoch": 377.5,
      "grad_norm": 0.000154229870531708,
      "learning_rate": 2.7480599366613234e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9512976,
      "step": 33975
    },
    {
      "epoch": 377.55555555555554,
      "grad_norm": 2.7023306756746024e-05,
      "learning_rate": 2.7435867579645473e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9514368,
      "step": 33980
    },
    {
      "epoch": 377.6111111111111,
      "grad_norm": 0.00013275268429424614,
      "learning_rate": 2.739117011486378e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9515776,
      "step": 33985
    },
    {
      "epoch": 377.6666666666667,
      "grad_norm": 0.00022971644648350775,
      "learning_rate": 2.7346506979161216e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9517152,
      "step": 33990
    },
    {
      "epoch": 377.72222222222223,
      "grad_norm": 8.086296293186024e-05,
      "learning_rate": 2.7301878179425227e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9518544,
      "step": 33995
    },
    {
      "epoch": 377.77777777777777,
      "grad_norm": 0.000178770293132402,
      "learning_rate": 2.7257283722538244e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9519984,
      "step": 34000
    },
    {
      "epoch": 377.77777777777777,
      "eval_loss": 0.885639488697052,
      "eval_runtime": 1.237,
      "eval_samples_per_second": 32.336,
      "eval_steps_per_second": 16.168,
      "num_input_tokens_seen": 9519984,
      "step": 34000
    },
    {
      "epoch": 377.8333333333333,
      "grad_norm": 0.00029067249852232635,
      "learning_rate": 2.7212723615377326e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9521376,
      "step": 34005
    },
    {
      "epoch": 377.8888888888889,
      "grad_norm": 0.00028662747354246676,
      "learning_rate": 2.7168197864814145e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9522784,
      "step": 34010
    },
    {
      "epoch": 377.94444444444446,
      "grad_norm": 0.0015060149598866701,
      "learning_rate": 2.712370647771509e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9524192,
      "step": 34015
    },
    {
      "epoch": 378.0,
      "grad_norm": 0.0011226892238482833,
      "learning_rate": 2.707924946094137e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9525568,
      "step": 34020
    },
    {
      "epoch": 378.05555555555554,
      "grad_norm": 2.6806461391970515e-05,
      "learning_rate": 2.7034826821348723e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9526944,
      "step": 34025
    },
    {
      "epoch": 378.1111111111111,
      "grad_norm": 0.0019948261324316263,
      "learning_rate": 2.6990438565787786e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9528272,
      "step": 34030
    },
    {
      "epoch": 378.1666666666667,
      "grad_norm": 2.8271335395402275e-05,
      "learning_rate": 2.6946084701103714e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9529680,
      "step": 34035
    },
    {
      "epoch": 378.22222222222223,
      "grad_norm": 9.238358325092122e-05,
      "learning_rate": 2.6901765234136428e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9531104,
      "step": 34040
    },
    {
      "epoch": 378.27777777777777,
      "grad_norm": 6.63168029859662e-05,
      "learning_rate": 2.685748017172063e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9532512,
      "step": 34045
    },
    {
      "epoch": 378.3333333333333,
      "grad_norm": 4.2266670789103955e-05,
      "learning_rate": 2.681322952068549e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9533920,
      "step": 34050
    },
    {
      "epoch": 378.3888888888889,
      "grad_norm": 2.3963779312907718e-05,
      "learning_rate": 2.6769013287855137e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9535344,
      "step": 34055
    },
    {
      "epoch": 378.44444444444446,
      "grad_norm": 1.867367245722562e-05,
      "learning_rate": 2.6724831480048286e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9536704,
      "step": 34060
    },
    {
      "epoch": 378.5,
      "grad_norm": 1.836300543800462e-05,
      "learning_rate": 2.66806841040782e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9538128,
      "step": 34065
    },
    {
      "epoch": 378.55555555555554,
      "grad_norm": 9.033128299051896e-05,
      "learning_rate": 2.6636571166753083e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9539552,
      "step": 34070
    },
    {
      "epoch": 378.6111111111111,
      "grad_norm": 0.0031349242199212313,
      "learning_rate": 2.6592492674875598e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9540960,
      "step": 34075
    },
    {
      "epoch": 378.6666666666667,
      "grad_norm": 3.507890141918324e-05,
      "learning_rate": 2.6548448635243305e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9542336,
      "step": 34080
    },
    {
      "epoch": 378.72222222222223,
      "grad_norm": 2.5933473807526752e-05,
      "learning_rate": 2.650443905464828e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9543712,
      "step": 34085
    },
    {
      "epoch": 378.77777777777777,
      "grad_norm": 0.0006047241040505469,
      "learning_rate": 2.646046393987739e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9545104,
      "step": 34090
    },
    {
      "epoch": 378.8333333333333,
      "grad_norm": 7.652908243471757e-05,
      "learning_rate": 2.64165232977121e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9546512,
      "step": 34095
    },
    {
      "epoch": 378.8888888888889,
      "grad_norm": 2.0770552509929985e-05,
      "learning_rate": 2.6372617134928695e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9547920,
      "step": 34100
    },
    {
      "epoch": 378.94444444444446,
      "grad_norm": 0.00016597886860836297,
      "learning_rate": 2.6328745458297943e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9549328,
      "step": 34105
    },
    {
      "epoch": 379.0,
      "grad_norm": 7.344120240304619e-05,
      "learning_rate": 2.6284908274585546e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9550704,
      "step": 34110
    },
    {
      "epoch": 379.05555555555554,
      "grad_norm": 0.001344231073744595,
      "learning_rate": 2.6241105590551595e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9552128,
      "step": 34115
    },
    {
      "epoch": 379.1111111111111,
      "grad_norm": 0.00014128477778285742,
      "learning_rate": 2.6197337412951105e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9553568,
      "step": 34120
    },
    {
      "epoch": 379.1666666666667,
      "grad_norm": 0.0003604391822591424,
      "learning_rate": 2.6153603748533705e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9554976,
      "step": 34125
    },
    {
      "epoch": 379.22222222222223,
      "grad_norm": 8.830511069390923e-05,
      "learning_rate": 2.6109904604043585e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9556416,
      "step": 34130
    },
    {
      "epoch": 379.27777777777777,
      "grad_norm": 0.00022185107809491456,
      "learning_rate": 2.6066239986219765e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9557824,
      "step": 34135
    },
    {
      "epoch": 379.3333333333333,
      "grad_norm": 0.00017931664478965104,
      "learning_rate": 2.602260990179592e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9559216,
      "step": 34140
    },
    {
      "epoch": 379.3888888888889,
      "grad_norm": 0.00013884517829865217,
      "learning_rate": 2.5979014357500248e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9560624,
      "step": 34145
    },
    {
      "epoch": 379.44444444444446,
      "grad_norm": 3.130299228359945e-05,
      "learning_rate": 2.5935453360055844e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9562032,
      "step": 34150
    },
    {
      "epoch": 379.5,
      "grad_norm": 0.00016429107927251607,
      "learning_rate": 2.5891926916180283e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9563376,
      "step": 34155
    },
    {
      "epoch": 379.55555555555554,
      "grad_norm": 0.0004033127333968878,
      "learning_rate": 2.5848435032585883e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9564736,
      "step": 34160
    },
    {
      "epoch": 379.6111111111111,
      "grad_norm": 2.209595731983427e-05,
      "learning_rate": 2.58049777159797e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9566096,
      "step": 34165
    },
    {
      "epoch": 379.6666666666667,
      "grad_norm": 2.9472230380633846e-05,
      "learning_rate": 2.576155497306332e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9567504,
      "step": 34170
    },
    {
      "epoch": 379.72222222222223,
      "grad_norm": 0.00028025032952427864,
      "learning_rate": 2.57181668105331e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9568896,
      "step": 34175
    },
    {
      "epoch": 379.77777777777777,
      "grad_norm": 0.0002322059590369463,
      "learning_rate": 2.567481323508014e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9570304,
      "step": 34180
    },
    {
      "epoch": 379.8333333333333,
      "grad_norm": 3.879984069499187e-05,
      "learning_rate": 2.5631494253389954e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9571680,
      "step": 34185
    },
    {
      "epoch": 379.8888888888889,
      "grad_norm": 4.4572181650437415e-05,
      "learning_rate": 2.5588209872142997e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9573104,
      "step": 34190
    },
    {
      "epoch": 379.94444444444446,
      "grad_norm": 7.616846414748579e-05,
      "learning_rate": 2.5544960098014186e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9574544,
      "step": 34195
    },
    {
      "epoch": 380.0,
      "grad_norm": 5.479701576405205e-05,
      "learning_rate": 2.550174493767318e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9575936,
      "step": 34200
    },
    {
      "epoch": 380.0,
      "eval_loss": 0.8791100382804871,
      "eval_runtime": 1.2361,
      "eval_samples_per_second": 32.359,
      "eval_steps_per_second": 16.179,
      "num_input_tokens_seen": 9575936,
      "step": 34200
    },
    {
      "epoch": 380.05555555555554,
      "grad_norm": 2.4390996259171516e-05,
      "learning_rate": 2.545856439778438e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9577312,
      "step": 34205
    },
    {
      "epoch": 380.1111111111111,
      "grad_norm": 3.569132968550548e-05,
      "learning_rate": 2.541541848500667e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9578704,
      "step": 34210
    },
    {
      "epoch": 380.1666666666667,
      "grad_norm": 4.230644481140189e-05,
      "learning_rate": 2.5372307205993733e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9580112,
      "step": 34215
    },
    {
      "epoch": 380.22222222222223,
      "grad_norm": 3.533854760462418e-05,
      "learning_rate": 2.5329230567393917e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9581536,
      "step": 34220
    },
    {
      "epoch": 380.27777777777777,
      "grad_norm": 0.00016576975758653134,
      "learning_rate": 2.5286188575850164e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9582944,
      "step": 34225
    },
    {
      "epoch": 380.3333333333333,
      "grad_norm": 0.00014401243242900819,
      "learning_rate": 2.5243181237999984e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9584352,
      "step": 34230
    },
    {
      "epoch": 380.3888888888889,
      "grad_norm": 3.599932824727148e-05,
      "learning_rate": 2.520020856047578e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9585776,
      "step": 34235
    },
    {
      "epoch": 380.44444444444446,
      "grad_norm": 4.015624654130079e-05,
      "learning_rate": 2.515727054990438e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9587136,
      "step": 34240
    },
    {
      "epoch": 380.5,
      "grad_norm": 0.0001376406871713698,
      "learning_rate": 2.511436721290747e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9588576,
      "step": 34245
    },
    {
      "epoch": 380.55555555555554,
      "grad_norm": 0.00019244910799898207,
      "learning_rate": 2.5071498556101164e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9589968,
      "step": 34250
    },
    {
      "epoch": 380.6111111111111,
      "grad_norm": 2.629325354064349e-05,
      "learning_rate": 2.5028664586096485e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9591344,
      "step": 34255
    },
    {
      "epoch": 380.6666666666667,
      "grad_norm": 0.00012947604409419,
      "learning_rate": 2.498586530949881e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9592752,
      "step": 34260
    },
    {
      "epoch": 380.72222222222223,
      "grad_norm": 5.927858001086861e-05,
      "learning_rate": 2.4943100732908427e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9594208,
      "step": 34265
    },
    {
      "epoch": 380.77777777777777,
      "grad_norm": 0.0007041478529572487,
      "learning_rate": 2.4900370862920188e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9595568,
      "step": 34270
    },
    {
      "epoch": 380.8333333333333,
      "grad_norm": 3.786749948631041e-05,
      "learning_rate": 2.4857675706123518e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9596944,
      "step": 34275
    },
    {
      "epoch": 380.8888888888889,
      "grad_norm": 0.0002618228318169713,
      "learning_rate": 2.4815015269102543e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9598336,
      "step": 34280
    },
    {
      "epoch": 380.94444444444446,
      "grad_norm": 7.940131035866216e-05,
      "learning_rate": 2.477238955843611e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9599776,
      "step": 34285
    },
    {
      "epoch": 381.0,
      "grad_norm": 2.1145984646864235e-05,
      "learning_rate": 2.4729798580697573e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9601152,
      "step": 34290
    },
    {
      "epoch": 381.05555555555554,
      "grad_norm": 0.00017324727377854288,
      "learning_rate": 2.4687242342455034e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9602512,
      "step": 34295
    },
    {
      "epoch": 381.1111111111111,
      "grad_norm": 0.00017171361832879484,
      "learning_rate": 2.4644720850271196e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9603920,
      "step": 34300
    },
    {
      "epoch": 381.1666666666667,
      "grad_norm": 9.832879004534334e-05,
      "learning_rate": 2.4602234110703364e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9605344,
      "step": 34305
    },
    {
      "epoch": 381.22222222222223,
      "grad_norm": 4.8937821702566e-05,
      "learning_rate": 2.4559782130303576e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9606752,
      "step": 34310
    },
    {
      "epoch": 381.27777777777777,
      "grad_norm": 6.693515024380758e-05,
      "learning_rate": 2.451736491561843e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9608176,
      "step": 34315
    },
    {
      "epoch": 381.3333333333333,
      "grad_norm": 0.001584912184625864,
      "learning_rate": 2.4474982473189163e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9609552,
      "step": 34320
    },
    {
      "epoch": 381.3888888888889,
      "grad_norm": 6.089130329200998e-05,
      "learning_rate": 2.4432634809551796e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9610944,
      "step": 34325
    },
    {
      "epoch": 381.44444444444446,
      "grad_norm": 1.693418380455114e-05,
      "learning_rate": 2.439032193123675e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9612352,
      "step": 34330
    },
    {
      "epoch": 381.5,
      "grad_norm": 0.0002869878662750125,
      "learning_rate": 2.4348043844769297e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9613760,
      "step": 34335
    },
    {
      "epoch": 381.55555555555554,
      "grad_norm": 2.991107976413332e-05,
      "learning_rate": 2.4305800556669146e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9615136,
      "step": 34340
    },
    {
      "epoch": 381.6111111111111,
      "grad_norm": 2.545840834500268e-05,
      "learning_rate": 2.426359207345083e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9616544,
      "step": 34345
    },
    {
      "epoch": 381.6666666666667,
      "grad_norm": 0.0006275554769672453,
      "learning_rate": 2.4221418401623396e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9617936,
      "step": 34350
    },
    {
      "epoch": 381.72222222222223,
      "grad_norm": 0.0016074738232418895,
      "learning_rate": 2.4179279547690557e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9619344,
      "step": 34355
    },
    {
      "epoch": 381.77777777777777,
      "grad_norm": 0.000497060245834291,
      "learning_rate": 2.413717551815062e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9620720,
      "step": 34360
    },
    {
      "epoch": 381.8333333333333,
      "grad_norm": 3.939084126614034e-05,
      "learning_rate": 2.409510631949666e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9622144,
      "step": 34365
    },
    {
      "epoch": 381.8888888888889,
      "grad_norm": 0.00019263208378106356,
      "learning_rate": 2.405307195821618e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9623520,
      "step": 34370
    },
    {
      "epoch": 381.94444444444446,
      "grad_norm": 2.5515761080896482e-05,
      "learning_rate": 2.4011072440791372e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9624960,
      "step": 34375
    },
    {
      "epoch": 382.0,
      "grad_norm": 0.00013852929987479,
      "learning_rate": 2.3969107773699233e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9626368,
      "step": 34380
    },
    {
      "epoch": 382.05555555555554,
      "grad_norm": 3.192263829987496e-05,
      "learning_rate": 2.3927177963411096e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9627792,
      "step": 34385
    },
    {
      "epoch": 382.1111111111111,
      "grad_norm": 0.00021244653908070177,
      "learning_rate": 2.3885283016393144e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9629168,
      "step": 34390
    },
    {
      "epoch": 382.1666666666667,
      "grad_norm": 0.00015994183195289224,
      "learning_rate": 2.3843422939106076e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9630544,
      "step": 34395
    },
    {
      "epoch": 382.22222222222223,
      "grad_norm": 4.1852610593196005e-05,
      "learning_rate": 2.380159773800525e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9631952,
      "step": 34400
    },
    {
      "epoch": 382.22222222222223,
      "eval_loss": 0.8856326937675476,
      "eval_runtime": 1.2396,
      "eval_samples_per_second": 32.267,
      "eval_steps_per_second": 16.134,
      "num_input_tokens_seen": 9631952,
      "step": 34400
    },
    {
      "epoch": 382.27777777777777,
      "grad_norm": 0.00013976915215607733,
      "learning_rate": 2.3759807419540675e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9633392,
      "step": 34405
    },
    {
      "epoch": 382.3333333333333,
      "grad_norm": 5.220733510213904e-05,
      "learning_rate": 2.3718051990156835e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9634816,
      "step": 34410
    },
    {
      "epoch": 382.3888888888889,
      "grad_norm": 9.55941723077558e-05,
      "learning_rate": 2.367633145629311e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9636176,
      "step": 34415
    },
    {
      "epoch": 382.44444444444446,
      "grad_norm": 0.0005137936095707119,
      "learning_rate": 2.363464582438316e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9637568,
      "step": 34420
    },
    {
      "epoch": 382.5,
      "grad_norm": 9.436794061912224e-05,
      "learning_rate": 2.3592995100855526e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9638992,
      "step": 34425
    },
    {
      "epoch": 382.55555555555554,
      "grad_norm": 0.00019835054990835488,
      "learning_rate": 2.3551379292133273e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9640400,
      "step": 34430
    },
    {
      "epoch": 382.6111111111111,
      "grad_norm": 3.745269714272581e-05,
      "learning_rate": 2.3509798404634047e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9641760,
      "step": 34435
    },
    {
      "epoch": 382.6666666666667,
      "grad_norm": 0.0006999897886998951,
      "learning_rate": 2.346825244477019e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9643184,
      "step": 34440
    },
    {
      "epoch": 382.72222222222223,
      "grad_norm": 0.00011402325617382303,
      "learning_rate": 2.3426741418948545e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9644592,
      "step": 34445
    },
    {
      "epoch": 382.77777777777777,
      "grad_norm": 0.00011087709572166204,
      "learning_rate": 2.3385265333570715e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9645984,
      "step": 34450
    },
    {
      "epoch": 382.8333333333333,
      "grad_norm": 0.00041162213892675936,
      "learning_rate": 2.334382419503278e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9647376,
      "step": 34455
    },
    {
      "epoch": 382.8888888888889,
      "grad_norm": 5.600256554316729e-05,
      "learning_rate": 2.3302418009725465e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9648752,
      "step": 34460
    },
    {
      "epoch": 382.94444444444446,
      "grad_norm": 9.273958130506799e-05,
      "learning_rate": 2.326104678403415e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9650160,
      "step": 34465
    },
    {
      "epoch": 383.0,
      "grad_norm": 1.687787334958557e-05,
      "learning_rate": 2.321971052433883e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9651584,
      "step": 34470
    },
    {
      "epoch": 383.05555555555554,
      "grad_norm": 2.9880236979806796e-05,
      "learning_rate": 2.3178409237014004e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9653008,
      "step": 34475
    },
    {
      "epoch": 383.1111111111111,
      "grad_norm": 1.2634078302653506e-05,
      "learning_rate": 2.313714292842889e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9654400,
      "step": 34480
    },
    {
      "epoch": 383.1666666666667,
      "grad_norm": 0.00012058848369633779,
      "learning_rate": 2.309591160494734e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9655840,
      "step": 34485
    },
    {
      "epoch": 383.22222222222223,
      "grad_norm": 2.2429090677178465e-05,
      "learning_rate": 2.305471527292763e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9657216,
      "step": 34490
    },
    {
      "epoch": 383.27777777777777,
      "grad_norm": 5.774948658654466e-05,
      "learning_rate": 2.3013553938722817e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9658624,
      "step": 34495
    },
    {
      "epoch": 383.3333333333333,
      "grad_norm": 0.0016806928906589746,
      "learning_rate": 2.297242760868043e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9660032,
      "step": 34500
    },
    {
      "epoch": 383.3888888888889,
      "grad_norm": 1.6504538507433608e-05,
      "learning_rate": 2.2931336289142735e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9661424,
      "step": 34505
    },
    {
      "epoch": 383.44444444444446,
      "grad_norm": 9.737534128362313e-05,
      "learning_rate": 2.289027998644655e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9662816,
      "step": 34510
    },
    {
      "epoch": 383.5,
      "grad_norm": 2.9790422559017316e-05,
      "learning_rate": 2.2849258706923228e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9664192,
      "step": 34515
    },
    {
      "epoch": 383.55555555555554,
      "grad_norm": 4.979869845556095e-05,
      "learning_rate": 2.2808272456898705e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9665600,
      "step": 34520
    },
    {
      "epoch": 383.6111111111111,
      "grad_norm": 0.006312322802841663,
      "learning_rate": 2.2767321242693707e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9667008,
      "step": 34525
    },
    {
      "epoch": 383.6666666666667,
      "grad_norm": 0.0005827315617352724,
      "learning_rate": 2.272640507062329e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9668400,
      "step": 34530
    },
    {
      "epoch": 383.72222222222223,
      "grad_norm": 7.562113023595884e-05,
      "learning_rate": 2.2685523946997382e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9669840,
      "step": 34535
    },
    {
      "epoch": 383.77777777777777,
      "grad_norm": 2.348621273995377e-05,
      "learning_rate": 2.2644677878120245e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9671216,
      "step": 34540
    },
    {
      "epoch": 383.8333333333333,
      "grad_norm": 4.807260847883299e-05,
      "learning_rate": 2.2603866870290897e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9672624,
      "step": 34545
    },
    {
      "epoch": 383.8888888888889,
      "grad_norm": 9.94262081803754e-05,
      "learning_rate": 2.256309092980294e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9674032,
      "step": 34550
    },
    {
      "epoch": 383.94444444444446,
      "grad_norm": 9.400385897606611e-05,
      "learning_rate": 2.252235006294448e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9675456,
      "step": 34555
    },
    {
      "epoch": 384.0,
      "grad_norm": 0.0002977158292196691,
      "learning_rate": 2.2481644275998333e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9676864,
      "step": 34560
    },
    {
      "epoch": 384.05555555555554,
      "grad_norm": 7.704232120886445e-05,
      "learning_rate": 2.2440973575241832e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9678256,
      "step": 34565
    },
    {
      "epoch": 384.1111111111111,
      "grad_norm": 1.9070497728534974e-05,
      "learning_rate": 2.240033796694685e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9679632,
      "step": 34570
    },
    {
      "epoch": 384.1666666666667,
      "grad_norm": 0.000580270541831851,
      "learning_rate": 2.235973745737999e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9680992,
      "step": 34575
    },
    {
      "epoch": 384.22222222222223,
      "grad_norm": 8.79329745657742e-05,
      "learning_rate": 2.2319172052802263e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9682416,
      "step": 34580
    },
    {
      "epoch": 384.27777777777777,
      "grad_norm": 0.00347969401627779,
      "learning_rate": 2.2278641759469477e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9683808,
      "step": 34585
    },
    {
      "epoch": 384.3333333333333,
      "grad_norm": 0.0011130261700600386,
      "learning_rate": 2.2238146583631825e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9685232,
      "step": 34590
    },
    {
      "epoch": 384.3888888888889,
      "grad_norm": 5.218640944804065e-05,
      "learning_rate": 2.2197686531534256e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9686592,
      "step": 34595
    },
    {
      "epoch": 384.44444444444446,
      "grad_norm": 0.000192221617908217,
      "learning_rate": 2.2157261609416087e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9687936,
      "step": 34600
    },
    {
      "epoch": 384.44444444444446,
      "eval_loss": 0.903516411781311,
      "eval_runtime": 1.2397,
      "eval_samples_per_second": 32.266,
      "eval_steps_per_second": 16.133,
      "num_input_tokens_seen": 9687936,
      "step": 34600
    },
    {
      "epoch": 384.5,
      "grad_norm": 4.488029662752524e-05,
      "learning_rate": 2.211687182351149e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9689328,
      "step": 34605
    },
    {
      "epoch": 384.55555555555554,
      "grad_norm": 4.995822746423073e-05,
      "learning_rate": 2.2076517180048993e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9690768,
      "step": 34610
    },
    {
      "epoch": 384.6111111111111,
      "grad_norm": 7.55256405682303e-05,
      "learning_rate": 2.2036197685251834e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9692144,
      "step": 34615
    },
    {
      "epoch": 384.6666666666667,
      "grad_norm": 0.001551701338030398,
      "learning_rate": 2.199591334533771e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9693552,
      "step": 34620
    },
    {
      "epoch": 384.72222222222223,
      "grad_norm": 0.00014970255142543465,
      "learning_rate": 2.1955664166519036e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9694992,
      "step": 34625
    },
    {
      "epoch": 384.77777777777777,
      "grad_norm": 0.000548429845366627,
      "learning_rate": 2.1915450155002793e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9696384,
      "step": 34630
    },
    {
      "epoch": 384.8333333333333,
      "grad_norm": 0.00011471965990494937,
      "learning_rate": 2.187527131699038e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9697808,
      "step": 34635
    },
    {
      "epoch": 384.8888888888889,
      "grad_norm": 0.00018100820307154208,
      "learning_rate": 2.18351276586779e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9699264,
      "step": 34640
    },
    {
      "epoch": 384.94444444444446,
      "grad_norm": 0.00031622970709577203,
      "learning_rate": 2.1795019186256092e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9700672,
      "step": 34645
    },
    {
      "epoch": 385.0,
      "grad_norm": 8.772194269113243e-05,
      "learning_rate": 2.1754945905910094e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9702048,
      "step": 34650
    },
    {
      "epoch": 385.05555555555554,
      "grad_norm": 7.928022387204692e-05,
      "learning_rate": 2.171490782381977e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9703440,
      "step": 34655
    },
    {
      "epoch": 385.1111111111111,
      "grad_norm": 4.686410102294758e-05,
      "learning_rate": 2.1674904946159425e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9704784,
      "step": 34660
    },
    {
      "epoch": 385.1666666666667,
      "grad_norm": 0.000597420905251056,
      "learning_rate": 2.16349372790981e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9706192,
      "step": 34665
    },
    {
      "epoch": 385.22222222222223,
      "grad_norm": 0.001330843660980463,
      "learning_rate": 2.159500482879928e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9707616,
      "step": 34670
    },
    {
      "epoch": 385.27777777777777,
      "grad_norm": 3.574360380298458e-05,
      "learning_rate": 2.155510760142096e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9708960,
      "step": 34675
    },
    {
      "epoch": 385.3333333333333,
      "grad_norm": 0.0008816500194370747,
      "learning_rate": 2.151524560311588e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9710352,
      "step": 34680
    },
    {
      "epoch": 385.3888888888889,
      "grad_norm": 5.6855700677260756e-05,
      "learning_rate": 2.147541884003129e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9711744,
      "step": 34685
    },
    {
      "epoch": 385.44444444444446,
      "grad_norm": 0.00025490359985269606,
      "learning_rate": 2.1435627318308895e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9713120,
      "step": 34690
    },
    {
      "epoch": 385.5,
      "grad_norm": 0.000874298217240721,
      "learning_rate": 2.139587104408511e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9714512,
      "step": 34695
    },
    {
      "epoch": 385.55555555555554,
      "grad_norm": 0.0006416967371478677,
      "learning_rate": 2.1356150023490783e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9715968,
      "step": 34700
    },
    {
      "epoch": 385.6111111111111,
      "grad_norm": 0.0006105630891397595,
      "learning_rate": 2.1316464262651464e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9717344,
      "step": 34705
    },
    {
      "epoch": 385.6666666666667,
      "grad_norm": 0.0010290234349668026,
      "learning_rate": 2.1276813767687224e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9718752,
      "step": 34710
    },
    {
      "epoch": 385.72222222222223,
      "grad_norm": 0.00019813327526208013,
      "learning_rate": 2.123719854471254e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9720160,
      "step": 34715
    },
    {
      "epoch": 385.77777777777777,
      "grad_norm": 0.00011327327956678346,
      "learning_rate": 2.119761859983668e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9721552,
      "step": 34720
    },
    {
      "epoch": 385.8333333333333,
      "grad_norm": 2.2760226784157567e-05,
      "learning_rate": 2.1158073939163386e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9722976,
      "step": 34725
    },
    {
      "epoch": 385.8888888888889,
      "grad_norm": 2.4196167942136526e-05,
      "learning_rate": 2.111856456879088e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9724352,
      "step": 34730
    },
    {
      "epoch": 385.94444444444446,
      "grad_norm": 0.0005369142163544893,
      "learning_rate": 2.1079090494811993e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9725776,
      "step": 34735
    },
    {
      "epoch": 386.0,
      "grad_norm": 8.895708015188575e-05,
      "learning_rate": 2.103965172331418e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9727216,
      "step": 34740
    },
    {
      "epoch": 386.05555555555554,
      "grad_norm": 0.0001009109037113376,
      "learning_rate": 2.100024826037933e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9728592,
      "step": 34745
    },
    {
      "epoch": 386.1111111111111,
      "grad_norm": 0.0001797382574295625,
      "learning_rate": 2.0960880112084027e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9730016,
      "step": 34750
    },
    {
      "epoch": 386.1666666666667,
      "grad_norm": 3.180788553436287e-05,
      "learning_rate": 2.092154728449927e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9731424,
      "step": 34755
    },
    {
      "epoch": 386.22222222222223,
      "grad_norm": 0.00011635449482128024,
      "learning_rate": 2.0882249783690687e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9732800,
      "step": 34760
    },
    {
      "epoch": 386.27777777777777,
      "grad_norm": 9.417233377462253e-05,
      "learning_rate": 2.084298761571851e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9734208,
      "step": 34765
    },
    {
      "epoch": 386.3333333333333,
      "grad_norm": 0.0013457599561661482,
      "learning_rate": 2.080376078663737e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9735616,
      "step": 34770
    },
    {
      "epoch": 386.3888888888889,
      "grad_norm": 2.035372017417103e-05,
      "learning_rate": 2.0764569302496593e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9736992,
      "step": 34775
    },
    {
      "epoch": 386.44444444444446,
      "grad_norm": 9.69969987636432e-05,
      "learning_rate": 2.0725413169339957e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9738400,
      "step": 34780
    },
    {
      "epoch": 386.5,
      "grad_norm": 0.00014349078992381692,
      "learning_rate": 2.068629239320588e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9739744,
      "step": 34785
    },
    {
      "epoch": 386.55555555555554,
      "grad_norm": 3.641479634097777e-05,
      "learning_rate": 2.064720698012726e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9741168,
      "step": 34790
    },
    {
      "epoch": 386.6111111111111,
      "grad_norm": 3.9588703657500446e-05,
      "learning_rate": 2.0608156936131522e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9742544,
      "step": 34795
    },
    {
      "epoch": 386.6666666666667,
      "grad_norm": 0.0008106732275336981,
      "learning_rate": 2.056914226724074e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9743968,
      "step": 34800
    },
    {
      "epoch": 386.6666666666667,
      "eval_loss": 0.8925191164016724,
      "eval_runtime": 1.2406,
      "eval_samples_per_second": 32.242,
      "eval_steps_per_second": 16.121,
      "num_input_tokens_seen": 9743968,
      "step": 34800
    },
    {
      "epoch": 386.72222222222223,
      "grad_norm": 2.589833275123965e-05,
      "learning_rate": 2.0530162979471385e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9745376,
      "step": 34805
    },
    {
      "epoch": 386.77777777777777,
      "grad_norm": 0.0019077076576650143,
      "learning_rate": 2.0491219078834667e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9746704,
      "step": 34810
    },
    {
      "epoch": 386.8333333333333,
      "grad_norm": 2.5951301722670905e-05,
      "learning_rate": 2.045231057133612e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9748096,
      "step": 34815
    },
    {
      "epoch": 386.8888888888889,
      "grad_norm": 0.00038646807661280036,
      "learning_rate": 2.0413437462975944e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9749488,
      "step": 34820
    },
    {
      "epoch": 386.94444444444446,
      "grad_norm": 9.37111399252899e-05,
      "learning_rate": 2.0374599759748843e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9750912,
      "step": 34825
    },
    {
      "epoch": 387.0,
      "grad_norm": 2.291425880684983e-05,
      "learning_rate": 2.033579746764419e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9752320,
      "step": 34830
    },
    {
      "epoch": 387.05555555555554,
      "grad_norm": 0.00020863157988060266,
      "learning_rate": 2.029703059264565e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9753712,
      "step": 34835
    },
    {
      "epoch": 387.1111111111111,
      "grad_norm": 8.614209946244955e-05,
      "learning_rate": 2.02582991407316e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9755088,
      "step": 34840
    },
    {
      "epoch": 387.1666666666667,
      "grad_norm": 0.00021389228641055524,
      "learning_rate": 2.0219603117874992e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9756464,
      "step": 34845
    },
    {
      "epoch": 387.22222222222223,
      "grad_norm": 0.00028599263168871403,
      "learning_rate": 2.0180942530043156e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9757824,
      "step": 34850
    },
    {
      "epoch": 387.27777777777777,
      "grad_norm": 4.630421972251497e-05,
      "learning_rate": 2.0142317383198107e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9759232,
      "step": 34855
    },
    {
      "epoch": 387.3333333333333,
      "grad_norm": 8.446641732007265e-05,
      "learning_rate": 2.0103727683296243e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9760656,
      "step": 34860
    },
    {
      "epoch": 387.3888888888889,
      "grad_norm": 0.0006373659707605839,
      "learning_rate": 2.0065173436288636e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9762032,
      "step": 34865
    },
    {
      "epoch": 387.44444444444446,
      "grad_norm": 0.0003239427169319242,
      "learning_rate": 2.002665464812087e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9763424,
      "step": 34870
    },
    {
      "epoch": 387.5,
      "grad_norm": 0.0001164862114819698,
      "learning_rate": 1.998817132473291e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9764880,
      "step": 34875
    },
    {
      "epoch": 387.55555555555554,
      "grad_norm": 0.00011942334094783291,
      "learning_rate": 1.9949723472059507e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9766304,
      "step": 34880
    },
    {
      "epoch": 387.6111111111111,
      "grad_norm": 0.000138535862788558,
      "learning_rate": 1.9911311096029726e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9767728,
      "step": 34885
    },
    {
      "epoch": 387.6666666666667,
      "grad_norm": 2.7339176085661165e-05,
      "learning_rate": 1.9872934202567224e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9769120,
      "step": 34890
    },
    {
      "epoch": 387.72222222222223,
      "grad_norm": 5.113648148835637e-05,
      "learning_rate": 1.9834592797590257e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9770544,
      "step": 34895
    },
    {
      "epoch": 387.77777777777777,
      "grad_norm": 2.3622145818080753e-05,
      "learning_rate": 1.979628688701149e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9771984,
      "step": 34900
    },
    {
      "epoch": 387.8333333333333,
      "grad_norm": 0.00033312858431600034,
      "learning_rate": 1.9758016476738193e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9773424,
      "step": 34905
    },
    {
      "epoch": 387.8888888888889,
      "grad_norm": 0.0002504807198420167,
      "learning_rate": 1.971978157267221e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9774816,
      "step": 34910
    },
    {
      "epoch": 387.94444444444446,
      "grad_norm": 1.5085941413417459e-05,
      "learning_rate": 1.968158218070973e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9776192,
      "step": 34915
    },
    {
      "epoch": 388.0,
      "grad_norm": 0.00014662505418527871,
      "learning_rate": 1.9643418306741682e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9777568,
      "step": 34920
    },
    {
      "epoch": 388.05555555555554,
      "grad_norm": 0.002048749942332506,
      "learning_rate": 1.9605289956653337e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9778976,
      "step": 34925
    },
    {
      "epoch": 388.1111111111111,
      "grad_norm": 0.00011035148781957105,
      "learning_rate": 1.9567197136324626e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9780416,
      "step": 34930
    },
    {
      "epoch": 388.1666666666667,
      "grad_norm": 0.003127373056486249,
      "learning_rate": 1.9529139851629935e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9781840,
      "step": 34935
    },
    {
      "epoch": 388.22222222222223,
      "grad_norm": 0.0004697504045907408,
      "learning_rate": 1.949111810843812e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9783232,
      "step": 34940
    },
    {
      "epoch": 388.27777777777777,
      "grad_norm": 1.3478154869517311e-05,
      "learning_rate": 1.9453131912612694e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9784656,
      "step": 34945
    },
    {
      "epoch": 388.3333333333333,
      "grad_norm": 0.0004858262836933136,
      "learning_rate": 1.941518127001149e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9786080,
      "step": 34950
    },
    {
      "epoch": 388.3888888888889,
      "grad_norm": 0.00012491224333643913,
      "learning_rate": 1.9377266186487107e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9787440,
      "step": 34955
    },
    {
      "epoch": 388.44444444444446,
      "grad_norm": 3.0174092898960225e-05,
      "learning_rate": 1.9339386667886483e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9788832,
      "step": 34960
    },
    {
      "epoch": 388.5,
      "grad_norm": 1.5089024600456469e-05,
      "learning_rate": 1.9301542720051024e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9790240,
      "step": 34965
    },
    {
      "epoch": 388.55555555555554,
      "grad_norm": 9.226038673659787e-05,
      "learning_rate": 1.926373434881684e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9791600,
      "step": 34970
    },
    {
      "epoch": 388.6111111111111,
      "grad_norm": 5.248967499937862e-05,
      "learning_rate": 1.9225961560014468e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9793008,
      "step": 34975
    },
    {
      "epoch": 388.6666666666667,
      "grad_norm": 0.0007596084615215659,
      "learning_rate": 1.918822435946885e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9794416,
      "step": 34980
    },
    {
      "epoch": 388.72222222222223,
      "grad_norm": 1.857976894825697e-05,
      "learning_rate": 1.915052275299961e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9795792,
      "step": 34985
    },
    {
      "epoch": 388.77777777777777,
      "grad_norm": 0.00025922589702531695,
      "learning_rate": 1.9112856746420854e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9797232,
      "step": 34990
    },
    {
      "epoch": 388.8333333333333,
      "grad_norm": 0.000716281880158931,
      "learning_rate": 1.907522634554104e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9798624,
      "step": 34995
    },
    {
      "epoch": 388.8888888888889,
      "grad_norm": 0.00032136094523593783,
      "learning_rate": 1.9037631556163337e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9800016,
      "step": 35000
    },
    {
      "epoch": 388.8888888888889,
      "eval_loss": 0.882737934589386,
      "eval_runtime": 1.241,
      "eval_samples_per_second": 32.232,
      "eval_steps_per_second": 16.116,
      "num_input_tokens_seen": 9800016,
      "step": 35000
    },
    {
      "epoch": 388.94444444444446,
      "grad_norm": 0.00027498332201503217,
      "learning_rate": 1.9000072384085272e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9801376,
      "step": 35005
    },
    {
      "epoch": 389.0,
      "grad_norm": 2.807745295285713e-05,
      "learning_rate": 1.8962548835098987e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9802816,
      "step": 35010
    },
    {
      "epoch": 389.05555555555554,
      "grad_norm": 0.0008489297470077872,
      "learning_rate": 1.8925060914991077e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9804256,
      "step": 35015
    },
    {
      "epoch": 389.1111111111111,
      "grad_norm": 0.0002571908989921212,
      "learning_rate": 1.888760862954264e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9805648,
      "step": 35020
    },
    {
      "epoch": 389.1666666666667,
      "grad_norm": 8.730767149245366e-05,
      "learning_rate": 1.8850191984529309e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9807040,
      "step": 35025
    },
    {
      "epoch": 389.22222222222223,
      "grad_norm": 1.8593056665849872e-05,
      "learning_rate": 1.8812810985721186e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9808416,
      "step": 35030
    },
    {
      "epoch": 389.27777777777777,
      "grad_norm": 0.0006656120531260967,
      "learning_rate": 1.8775465638882856e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9809808,
      "step": 35035
    },
    {
      "epoch": 389.3333333333333,
      "grad_norm": 4.31704283982981e-05,
      "learning_rate": 1.8738155949773517e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9811232,
      "step": 35040
    },
    {
      "epoch": 389.3888888888889,
      "grad_norm": 0.00014280523464549333,
      "learning_rate": 1.8700881924146707e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9812640,
      "step": 35045
    },
    {
      "epoch": 389.44444444444446,
      "grad_norm": 4.1174484067596495e-05,
      "learning_rate": 1.8663643567750577e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9814048,
      "step": 35050
    },
    {
      "epoch": 389.5,
      "grad_norm": 0.00011804117821156979,
      "learning_rate": 1.8626440886327813e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9815424,
      "step": 35055
    },
    {
      "epoch": 389.55555555555554,
      "grad_norm": 0.004287404008209705,
      "learning_rate": 1.8589273885615432e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9816816,
      "step": 35060
    },
    {
      "epoch": 389.6111111111111,
      "grad_norm": 1.4053623999643605e-05,
      "learning_rate": 1.8552142571345133e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9818272,
      "step": 35065
    },
    {
      "epoch": 389.6666666666667,
      "grad_norm": 7.766136695863679e-05,
      "learning_rate": 1.8515046949243025e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9819712,
      "step": 35070
    },
    {
      "epoch": 389.72222222222223,
      "grad_norm": 2.94803630822571e-05,
      "learning_rate": 1.8477987025029674e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9821056,
      "step": 35075
    },
    {
      "epoch": 389.77777777777777,
      "grad_norm": 0.00013581728853750974,
      "learning_rate": 1.8440962804420232e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9822448,
      "step": 35080
    },
    {
      "epoch": 389.8333333333333,
      "grad_norm": 0.00011432919563958421,
      "learning_rate": 1.8403974293124265e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9823824,
      "step": 35085
    },
    {
      "epoch": 389.8888888888889,
      "grad_norm": 1.8636215827427804e-05,
      "learning_rate": 1.8367021496845854e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9825248,
      "step": 35090
    },
    {
      "epoch": 389.94444444444446,
      "grad_norm": 5.962107752566226e-05,
      "learning_rate": 1.8330104421283662e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9826624,
      "step": 35095
    },
    {
      "epoch": 390.0,
      "grad_norm": 2.5659757739049383e-05,
      "learning_rate": 1.8293223072130717e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9828016,
      "step": 35100
    },
    {
      "epoch": 390.05555555555554,
      "grad_norm": 0.003855578601360321,
      "learning_rate": 1.8256377455074525e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9829408,
      "step": 35105
    },
    {
      "epoch": 390.1111111111111,
      "grad_norm": 0.0005487148300744593,
      "learning_rate": 1.8219567575797263e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9830832,
      "step": 35110
    },
    {
      "epoch": 390.1666666666667,
      "grad_norm": 0.00016423231863882393,
      "learning_rate": 1.8182793439975365e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9832288,
      "step": 35115
    },
    {
      "epoch": 390.22222222222223,
      "grad_norm": 7.763583562336862e-05,
      "learning_rate": 1.8146055053279958e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9833664,
      "step": 35120
    },
    {
      "epoch": 390.27777777777777,
      "grad_norm": 8.130490459734574e-05,
      "learning_rate": 1.8109352421376486e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9835072,
      "step": 35125
    },
    {
      "epoch": 390.3333333333333,
      "grad_norm": 2.9784196158288978e-05,
      "learning_rate": 1.8072685549924972e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9836464,
      "step": 35130
    },
    {
      "epoch": 390.3888888888889,
      "grad_norm": 3.063831536564976e-05,
      "learning_rate": 1.8036054444579982e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9837840,
      "step": 35135
    },
    {
      "epoch": 390.44444444444446,
      "grad_norm": 0.00021094188559800386,
      "learning_rate": 1.7999459110990407e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9839264,
      "step": 35140
    },
    {
      "epoch": 390.5,
      "grad_norm": 1.0143561667064205e-05,
      "learning_rate": 1.7962899554799712e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9840640,
      "step": 35145
    },
    {
      "epoch": 390.55555555555554,
      "grad_norm": 0.00015282757522072643,
      "learning_rate": 1.7926375781645937e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9842016,
      "step": 35150
    },
    {
      "epoch": 390.6111111111111,
      "grad_norm": 0.00010064275556942448,
      "learning_rate": 1.7889887797161359e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9843408,
      "step": 35155
    },
    {
      "epoch": 390.6666666666667,
      "grad_norm": 0.0014688092051073909,
      "learning_rate": 1.7853435606973028e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9844800,
      "step": 35160
    },
    {
      "epoch": 390.72222222222223,
      "grad_norm": 0.00017379976634401828,
      "learning_rate": 1.781701921670223e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9846160,
      "step": 35165
    },
    {
      "epoch": 390.77777777777777,
      "grad_norm": 3.7503079511225224e-05,
      "learning_rate": 1.7780638631964886e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9847664,
      "step": 35170
    },
    {
      "epoch": 390.8333333333333,
      "grad_norm": 7.03298719599843e-05,
      "learning_rate": 1.7744293858371314e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9849072,
      "step": 35175
    },
    {
      "epoch": 390.8888888888889,
      "grad_norm": 0.0001430392439942807,
      "learning_rate": 1.770798490152631e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9850480,
      "step": 35180
    },
    {
      "epoch": 390.94444444444446,
      "grad_norm": 0.00011630238441284746,
      "learning_rate": 1.767171176702917e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9851840,
      "step": 35185
    },
    {
      "epoch": 391.0,
      "grad_norm": 0.00027791986940428615,
      "learning_rate": 1.7635474460473755e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9853232,
      "step": 35190
    },
    {
      "epoch": 391.05555555555554,
      "grad_norm": 0.00022633235494140536,
      "learning_rate": 1.7599272987448206e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9854640,
      "step": 35195
    },
    {
      "epoch": 391.1111111111111,
      "grad_norm": 0.0012380540138110518,
      "learning_rate": 1.7563107353535362e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9856016,
      "step": 35200
    },
    {
      "epoch": 391.1111111111111,
      "eval_loss": 0.9022579193115234,
      "eval_runtime": 1.2571,
      "eval_samples_per_second": 31.818,
      "eval_steps_per_second": 15.909,
      "num_input_tokens_seen": 9856016,
      "step": 35200
    },
    {
      "epoch": 391.1666666666667,
      "grad_norm": 0.005653598345816135,
      "learning_rate": 1.7526977564312263e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9857408,
      "step": 35205
    },
    {
      "epoch": 391.22222222222223,
      "grad_norm": 5.2414805395528674e-05,
      "learning_rate": 1.7490883625350701e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9858784,
      "step": 35210
    },
    {
      "epoch": 391.27777777777777,
      "grad_norm": 3.6053093936061487e-05,
      "learning_rate": 1.7454825542216807e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9860192,
      "step": 35215
    },
    {
      "epoch": 391.3333333333333,
      "grad_norm": 0.0002528171462472528,
      "learning_rate": 1.7418803320471105e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9861552,
      "step": 35220
    },
    {
      "epoch": 391.3888888888889,
      "grad_norm": 2.55177710641874e-05,
      "learning_rate": 1.7382816965668737e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9862960,
      "step": 35225
    },
    {
      "epoch": 391.44444444444446,
      "grad_norm": 0.0009139031171798706,
      "learning_rate": 1.7346866483359285e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9864384,
      "step": 35230
    },
    {
      "epoch": 391.5,
      "grad_norm": 9.96902454062365e-05,
      "learning_rate": 1.7310951879086657e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9865824,
      "step": 35235
    },
    {
      "epoch": 391.55555555555554,
      "grad_norm": 9.602982754586264e-05,
      "learning_rate": 1.7275073158389471e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9867232,
      "step": 35240
    },
    {
      "epoch": 391.6111111111111,
      "grad_norm": 1.5099119991646148e-05,
      "learning_rate": 1.723923032680061e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9868624,
      "step": 35245
    },
    {
      "epoch": 391.6666666666667,
      "grad_norm": 6.29771820968017e-05,
      "learning_rate": 1.7203423389847428e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9870016,
      "step": 35250
    },
    {
      "epoch": 391.72222222222223,
      "grad_norm": 0.0001944807154359296,
      "learning_rate": 1.7167652353051928e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9871424,
      "step": 35255
    },
    {
      "epoch": 391.77777777777777,
      "grad_norm": 0.00019966508261859417,
      "learning_rate": 1.7131917221930333e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9872864,
      "step": 35260
    },
    {
      "epoch": 391.8333333333333,
      "grad_norm": 0.00017386532272212207,
      "learning_rate": 1.7096218001993513e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9874240,
      "step": 35265
    },
    {
      "epoch": 391.8888888888889,
      "grad_norm": 7.31818363419734e-05,
      "learning_rate": 1.706055469874676e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9875680,
      "step": 35270
    },
    {
      "epoch": 391.94444444444446,
      "grad_norm": 6.596732418984175e-05,
      "learning_rate": 1.702492731768976e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9877056,
      "step": 35275
    },
    {
      "epoch": 392.0,
      "grad_norm": 0.00012911780504509807,
      "learning_rate": 1.6989335864316724e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9878496,
      "step": 35280
    },
    {
      "epoch": 392.05555555555554,
      "grad_norm": 0.00015736764180473983,
      "learning_rate": 1.6953780344116265e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9879904,
      "step": 35285
    },
    {
      "epoch": 392.1111111111111,
      "grad_norm": 4.46478406956885e-05,
      "learning_rate": 1.6918260762571497e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9881296,
      "step": 35290
    },
    {
      "epoch": 392.1666666666667,
      "grad_norm": 3.919289156328887e-05,
      "learning_rate": 1.6882777125160093e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9882624,
      "step": 35295
    },
    {
      "epoch": 392.22222222222223,
      "grad_norm": 0.0015449762577190995,
      "learning_rate": 1.6847329437353899e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9884032,
      "step": 35300
    },
    {
      "epoch": 392.27777777777777,
      "grad_norm": 0.00015981023898348212,
      "learning_rate": 1.6811917704619511e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9885424,
      "step": 35305
    },
    {
      "epoch": 392.3333333333333,
      "grad_norm": 2.5935261874110438e-05,
      "learning_rate": 1.67765419324179e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9886832,
      "step": 35310
    },
    {
      "epoch": 392.3888888888889,
      "grad_norm": 0.009235210716724396,
      "learning_rate": 1.6741202126204364e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9888208,
      "step": 35315
    },
    {
      "epoch": 392.44444444444446,
      "grad_norm": 0.00028019570163451135,
      "learning_rate": 1.6705898291428767e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9889616,
      "step": 35320
    },
    {
      "epoch": 392.5,
      "grad_norm": 0.0053623090498149395,
      "learning_rate": 1.6670630433535395e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9891072,
      "step": 35325
    },
    {
      "epoch": 392.55555555555554,
      "grad_norm": 0.00017449371807742864,
      "learning_rate": 1.6635398557962979e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9892416,
      "step": 35330
    },
    {
      "epoch": 392.6111111111111,
      "grad_norm": 3.715101411216892e-05,
      "learning_rate": 1.660020267014481e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9893840,
      "step": 35335
    },
    {
      "epoch": 392.6666666666667,
      "grad_norm": 4.170248575974256e-05,
      "learning_rate": 1.6565042775508438e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9895200,
      "step": 35340
    },
    {
      "epoch": 392.72222222222223,
      "grad_norm": 2.2911641281098127e-05,
      "learning_rate": 1.6529918879475997e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9896640,
      "step": 35345
    },
    {
      "epoch": 392.77777777777777,
      "grad_norm": 0.00010032747377408668,
      "learning_rate": 1.6494830987464043e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9898000,
      "step": 35350
    },
    {
      "epoch": 392.8333333333333,
      "grad_norm": 0.00021736013877671212,
      "learning_rate": 1.6459779104883555e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9899440,
      "step": 35355
    },
    {
      "epoch": 392.8888888888889,
      "grad_norm": 0.00010181667312281206,
      "learning_rate": 1.6424763237140013e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9900864,
      "step": 35360
    },
    {
      "epoch": 392.94444444444446,
      "grad_norm": 1.8248876585857943e-05,
      "learning_rate": 1.6389783389633207e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9902272,
      "step": 35365
    },
    {
      "epoch": 393.0,
      "grad_norm": 0.00011680614261422306,
      "learning_rate": 1.6354839567757546e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9903712,
      "step": 35370
    },
    {
      "epoch": 393.05555555555554,
      "grad_norm": 7.860914047341794e-05,
      "learning_rate": 1.6319931776901831e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9905072,
      "step": 35375
    },
    {
      "epoch": 393.1111111111111,
      "grad_norm": 2.065260378003586e-05,
      "learning_rate": 1.6285060022449229e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9906480,
      "step": 35380
    },
    {
      "epoch": 393.1666666666667,
      "grad_norm": 4.6226985432440415e-05,
      "learning_rate": 1.6250224309777434e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9907856,
      "step": 35385
    },
    {
      "epoch": 393.22222222222223,
      "grad_norm": 4.576309584081173e-05,
      "learning_rate": 1.6215424644258515e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9909280,
      "step": 35390
    },
    {
      "epoch": 393.27777777777777,
      "grad_norm": 0.00025645748246461153,
      "learning_rate": 1.6180661031259036e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9910672,
      "step": 35395
    },
    {
      "epoch": 393.3333333333333,
      "grad_norm": 0.00010986937559209764,
      "learning_rate": 1.614593347613999e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9912112,
      "step": 35400
    },
    {
      "epoch": 393.3333333333333,
      "eval_loss": 0.9180132150650024,
      "eval_runtime": 1.2371,
      "eval_samples_per_second": 32.334,
      "eval_steps_per_second": 16.167,
      "num_input_tokens_seen": 9912112,
      "step": 35400
    },
    {
      "epoch": 393.3888888888889,
      "grad_norm": 1.666731077420991e-05,
      "learning_rate": 1.6111241984256758e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9913472,
      "step": 35405
    },
    {
      "epoch": 393.44444444444446,
      "grad_norm": 2.2110607460490428e-05,
      "learning_rate": 1.6076586560959257e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9914864,
      "step": 35410
    },
    {
      "epoch": 393.5,
      "grad_norm": 4.0887942304834723e-05,
      "learning_rate": 1.604196721159182e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9916240,
      "step": 35415
    },
    {
      "epoch": 393.55555555555554,
      "grad_norm": 0.0002973595983348787,
      "learning_rate": 1.6007383941493092e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9917664,
      "step": 35420
    },
    {
      "epoch": 393.6111111111111,
      "grad_norm": 0.0033519184216856956,
      "learning_rate": 1.5972836755996285e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9919056,
      "step": 35425
    },
    {
      "epoch": 393.6666666666667,
      "grad_norm": 0.00012671701551880687,
      "learning_rate": 1.5938325660429076e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9920448,
      "step": 35430
    },
    {
      "epoch": 393.72222222222223,
      "grad_norm": 0.0003096048894803971,
      "learning_rate": 1.5903850660113378e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9921872,
      "step": 35435
    },
    {
      "epoch": 393.77777777777777,
      "grad_norm": 1.1750234989449382e-05,
      "learning_rate": 1.5869411760365826e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9923280,
      "step": 35440
    },
    {
      "epoch": 393.8333333333333,
      "grad_norm": 8.967069879872724e-05,
      "learning_rate": 1.58350089664972e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9924688,
      "step": 35445
    },
    {
      "epoch": 393.8888888888889,
      "grad_norm": 0.00015544920461252332,
      "learning_rate": 1.5800642283812865e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9926064,
      "step": 35450
    },
    {
      "epoch": 393.94444444444446,
      "grad_norm": 0.00015839746629353613,
      "learning_rate": 1.5766311717612698e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9927456,
      "step": 35455
    },
    {
      "epoch": 394.0,
      "grad_norm": 3.394778468646109e-05,
      "learning_rate": 1.5732017273190818e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9928832,
      "step": 35460
    },
    {
      "epoch": 394.05555555555554,
      "grad_norm": 7.730546349193901e-05,
      "learning_rate": 1.5697758955835806e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9930256,
      "step": 35465
    },
    {
      "epoch": 394.1111111111111,
      "grad_norm": 2.721514283621218e-05,
      "learning_rate": 1.566353677083085e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9931664,
      "step": 35470
    },
    {
      "epoch": 394.1666666666667,
      "grad_norm": 5.267161031952128e-05,
      "learning_rate": 1.562935072345334e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9933056,
      "step": 35475
    },
    {
      "epoch": 394.22222222222223,
      "grad_norm": 6.35895921732299e-05,
      "learning_rate": 1.5595200818975281e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9934432,
      "step": 35480
    },
    {
      "epoch": 394.27777777777777,
      "grad_norm": 9.376520029036328e-05,
      "learning_rate": 1.5561087062662905e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9935808,
      "step": 35485
    },
    {
      "epoch": 394.3333333333333,
      "grad_norm": 0.0002540414279792458,
      "learning_rate": 1.5527009459777087e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9937216,
      "step": 35490
    },
    {
      "epoch": 394.3888888888889,
      "grad_norm": 0.00026928336592391133,
      "learning_rate": 1.5492968015572984e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9938672,
      "step": 35495
    },
    {
      "epoch": 394.44444444444446,
      "grad_norm": 4.1565213905414566e-05,
      "learning_rate": 1.5458962735300203e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9940080,
      "step": 35500
    },
    {
      "epoch": 394.5,
      "grad_norm": 6.932626274647191e-05,
      "learning_rate": 1.54249936242028e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9941488,
      "step": 35505
    },
    {
      "epoch": 394.55555555555554,
      "grad_norm": 0.00047564899432472885,
      "learning_rate": 1.5391060687519222e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9942832,
      "step": 35510
    },
    {
      "epoch": 394.6111111111111,
      "grad_norm": 0.00016669479373376817,
      "learning_rate": 1.5357163930482367e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9944224,
      "step": 35515
    },
    {
      "epoch": 394.6666666666667,
      "grad_norm": 2.6787012757267803e-05,
      "learning_rate": 1.532330335831955e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9945648,
      "step": 35520
    },
    {
      "epoch": 394.72222222222223,
      "grad_norm": 3.116711013717577e-05,
      "learning_rate": 1.5289478976252491e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9947040,
      "step": 35525
    },
    {
      "epoch": 394.77777777777777,
      "grad_norm": 0.00027468500775285065,
      "learning_rate": 1.5255690789497345e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9948480,
      "step": 35530
    },
    {
      "epoch": 394.8333333333333,
      "grad_norm": 1.9893499484169297e-05,
      "learning_rate": 1.5221938803264641e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9949888,
      "step": 35535
    },
    {
      "epoch": 394.8888888888889,
      "grad_norm": 2.5510362320346758e-05,
      "learning_rate": 1.518822302275938e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9951312,
      "step": 35540
    },
    {
      "epoch": 394.94444444444446,
      "grad_norm": 0.00016274859081022441,
      "learning_rate": 1.5154543453180958e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9952720,
      "step": 35545
    },
    {
      "epoch": 395.0,
      "grad_norm": 0.00023278634762391448,
      "learning_rate": 1.5120900099723167e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9954096,
      "step": 35550
    },
    {
      "epoch": 395.05555555555554,
      "grad_norm": 0.00018478214042261243,
      "learning_rate": 1.5087292967574273e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9955504,
      "step": 35555
    },
    {
      "epoch": 395.1111111111111,
      "grad_norm": 9.959455928765237e-05,
      "learning_rate": 1.5053722061916908e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9956864,
      "step": 35560
    },
    {
      "epoch": 395.1666666666667,
      "grad_norm": 6.305195711320266e-05,
      "learning_rate": 1.5020187387928124e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9958240,
      "step": 35565
    },
    {
      "epoch": 395.22222222222223,
      "grad_norm": 1.585685822647065e-05,
      "learning_rate": 1.4986688950779343e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9959696,
      "step": 35570
    },
    {
      "epoch": 395.27777777777777,
      "grad_norm": 0.00010194830974796787,
      "learning_rate": 1.495322675563654e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9961104,
      "step": 35575
    },
    {
      "epoch": 395.3333333333333,
      "grad_norm": 6.558540917467326e-05,
      "learning_rate": 1.4919800807659922e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9962496,
      "step": 35580
    },
    {
      "epoch": 395.3888888888889,
      "grad_norm": 4.7555306082358584e-05,
      "learning_rate": 1.4886411112004255e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9963888,
      "step": 35585
    },
    {
      "epoch": 395.44444444444446,
      "grad_norm": 3.176899554091506e-05,
      "learning_rate": 1.4853057673818588e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9965312,
      "step": 35590
    },
    {
      "epoch": 395.5,
      "grad_norm": 6.02439213253092e-05,
      "learning_rate": 1.481974049824647e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9966704,
      "step": 35595
    },
    {
      "epoch": 395.55555555555554,
      "grad_norm": 0.00048802857054397464,
      "learning_rate": 1.4786459590425849e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9968112,
      "step": 35600
    },
    {
      "epoch": 395.55555555555554,
      "eval_loss": 0.8993112444877625,
      "eval_runtime": 1.2377,
      "eval_samples_per_second": 32.318,
      "eval_steps_per_second": 16.159,
      "num_input_tokens_seen": 9968112,
      "step": 35600
    },
    {
      "epoch": 395.6111111111111,
      "grad_norm": 2.726096863625571e-05,
      "learning_rate": 1.4753214955489036e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9969520,
      "step": 35605
    },
    {
      "epoch": 395.6666666666667,
      "grad_norm": 2.5774212190299295e-05,
      "learning_rate": 1.4720006598562737e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9970912,
      "step": 35610
    },
    {
      "epoch": 395.72222222222223,
      "grad_norm": 0.0002180615410907194,
      "learning_rate": 1.4686834524768185e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9972304,
      "step": 35615
    },
    {
      "epoch": 395.77777777777777,
      "grad_norm": 0.00028669487801380455,
      "learning_rate": 1.4653698739220844e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9973712,
      "step": 35620
    },
    {
      "epoch": 395.8333333333333,
      "grad_norm": 6.7464257881511e-05,
      "learning_rate": 1.4620599247030715e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9975072,
      "step": 35625
    },
    {
      "epoch": 395.8888888888889,
      "grad_norm": 3.965350697399117e-05,
      "learning_rate": 1.4587536053302125e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9976480,
      "step": 35630
    },
    {
      "epoch": 395.94444444444446,
      "grad_norm": 5.223303378443234e-05,
      "learning_rate": 1.4554509163133862e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9977888,
      "step": 35635
    },
    {
      "epoch": 396.0,
      "grad_norm": 5.559599230764434e-05,
      "learning_rate": 1.4521518581619098e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9979264,
      "step": 35640
    },
    {
      "epoch": 396.05555555555554,
      "grad_norm": 1.3065024177194573e-05,
      "learning_rate": 1.4488564313845348e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9980640,
      "step": 35645
    },
    {
      "epoch": 396.1111111111111,
      "grad_norm": 3.338102760608308e-05,
      "learning_rate": 1.4455646364894603e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9982048,
      "step": 35650
    },
    {
      "epoch": 396.1666666666667,
      "grad_norm": 2.6774134312290698e-05,
      "learning_rate": 1.4422764739843247e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9983424,
      "step": 35655
    },
    {
      "epoch": 396.22222222222223,
      "grad_norm": 9.661626245360821e-05,
      "learning_rate": 1.4389919443762e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9984816,
      "step": 35660
    },
    {
      "epoch": 396.27777777777777,
      "grad_norm": 2.62152316281572e-05,
      "learning_rate": 1.4357110481716063e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9986256,
      "step": 35665
    },
    {
      "epoch": 396.3333333333333,
      "grad_norm": 2.580858017608989e-05,
      "learning_rate": 1.4324337858764941e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9987664,
      "step": 35670
    },
    {
      "epoch": 396.3888888888889,
      "grad_norm": 0.00014738141908310354,
      "learning_rate": 1.4291601579962622e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9989120,
      "step": 35675
    },
    {
      "epoch": 396.44444444444446,
      "grad_norm": 0.008701824583113194,
      "learning_rate": 1.42589016503574e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9990576,
      "step": 35680
    },
    {
      "epoch": 396.5,
      "grad_norm": 0.001143320114351809,
      "learning_rate": 1.4226238074992099e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9991984,
      "step": 35685
    },
    {
      "epoch": 396.55555555555554,
      "grad_norm": 3.3538184652570635e-05,
      "learning_rate": 1.4193610858903778e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9993392,
      "step": 35690
    },
    {
      "epoch": 396.6111111111111,
      "grad_norm": 0.00027679867343977094,
      "learning_rate": 1.416102000712402e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9994784,
      "step": 35695
    },
    {
      "epoch": 396.6666666666667,
      "grad_norm": 5.227497240412049e-05,
      "learning_rate": 1.4128465524678668e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9996160,
      "step": 35700
    },
    {
      "epoch": 396.72222222222223,
      "grad_norm": 0.00026089619495905936,
      "learning_rate": 1.4095947416588124e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9997584,
      "step": 35705
    },
    {
      "epoch": 396.77777777777777,
      "grad_norm": 0.00017438286158721894,
      "learning_rate": 1.4063465687866983e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 9998944,
      "step": 35710
    },
    {
      "epoch": 396.8333333333333,
      "grad_norm": 9.715796477394179e-05,
      "learning_rate": 1.4031020343524438e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10000368,
      "step": 35715
    },
    {
      "epoch": 396.8888888888889,
      "grad_norm": 0.00012735111522488296,
      "learning_rate": 1.3998611388563926e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10001792,
      "step": 35720
    },
    {
      "epoch": 396.94444444444446,
      "grad_norm": 2.9794529837090522e-05,
      "learning_rate": 1.3966238827983314e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10003184,
      "step": 35725
    },
    {
      "epoch": 397.0,
      "grad_norm": 0.0029128212481737137,
      "learning_rate": 1.393390266677483e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10004528,
      "step": 35730
    },
    {
      "epoch": 397.05555555555554,
      "grad_norm": 2.2676556909573264e-05,
      "learning_rate": 1.3901602909925204e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10005936,
      "step": 35735
    },
    {
      "epoch": 397.1111111111111,
      "grad_norm": 0.000471745734103024,
      "learning_rate": 1.3869339562415373e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10007344,
      "step": 35740
    },
    {
      "epoch": 397.1666666666667,
      "grad_norm": 4.5539614802692086e-05,
      "learning_rate": 1.38371126292208e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10008752,
      "step": 35745
    },
    {
      "epoch": 397.22222222222223,
      "grad_norm": 0.00013790074444841594,
      "learning_rate": 1.3804922115311286e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10010176,
      "step": 35750
    },
    {
      "epoch": 397.27777777777777,
      "grad_norm": 0.00013432632840704173,
      "learning_rate": 1.3772768025650945e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10011536,
      "step": 35755
    },
    {
      "epoch": 397.3333333333333,
      "grad_norm": 6.80218799971044e-05,
      "learning_rate": 1.3740650365198448e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10012928,
      "step": 35760
    },
    {
      "epoch": 397.3888888888889,
      "grad_norm": 0.00010904538066824898,
      "learning_rate": 1.3708569138906612e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10014304,
      "step": 35765
    },
    {
      "epoch": 397.44444444444446,
      "grad_norm": 2.7027725082007237e-05,
      "learning_rate": 1.367652435172287e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10015728,
      "step": 35770
    },
    {
      "epoch": 397.5,
      "grad_norm": 8.456620707875118e-05,
      "learning_rate": 1.364451600858893e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10017152,
      "step": 35775
    },
    {
      "epoch": 397.55555555555554,
      "grad_norm": 7.360564632108435e-05,
      "learning_rate": 1.3612544114440823e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10018560,
      "step": 35780
    },
    {
      "epoch": 397.6111111111111,
      "grad_norm": 5.7133125665131956e-05,
      "learning_rate": 1.3580608674209072e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10019920,
      "step": 35785
    },
    {
      "epoch": 397.6666666666667,
      "grad_norm": 0.00013306761684361845,
      "learning_rate": 1.3548709692818434e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10021296,
      "step": 35790
    },
    {
      "epoch": 397.72222222222223,
      "grad_norm": 4.501829971559346e-05,
      "learning_rate": 1.3516847175188223e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10022704,
      "step": 35795
    },
    {
      "epoch": 397.77777777777777,
      "grad_norm": 0.0001760109153110534,
      "learning_rate": 1.348502112623204e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10024160,
      "step": 35800
    },
    {
      "epoch": 397.77777777777777,
      "eval_loss": 0.8826104998588562,
      "eval_runtime": 1.2491,
      "eval_samples_per_second": 32.022,
      "eval_steps_per_second": 16.011,
      "num_input_tokens_seen": 10024160,
      "step": 35800
    },
    {
      "epoch": 397.8333333333333,
      "grad_norm": 0.0006551297847181559,
      "learning_rate": 1.3453231550857787e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10025536,
      "step": 35805
    },
    {
      "epoch": 397.8888888888889,
      "grad_norm": 4.337451173341833e-05,
      "learning_rate": 1.3421478453967878e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10026912,
      "step": 35810
    },
    {
      "epoch": 397.94444444444446,
      "grad_norm": 0.0005788753041997552,
      "learning_rate": 1.3389761840459065e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10028336,
      "step": 35815
    },
    {
      "epoch": 398.0,
      "grad_norm": 2.8944972655153833e-05,
      "learning_rate": 1.3358081715222376e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10029728,
      "step": 35820
    },
    {
      "epoch": 398.05555555555554,
      "grad_norm": 0.0002914493961725384,
      "learning_rate": 1.3326438083143295e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10031120,
      "step": 35825
    },
    {
      "epoch": 398.1111111111111,
      "grad_norm": 0.0006922989268787205,
      "learning_rate": 1.3294830949101723e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10032544,
      "step": 35830
    },
    {
      "epoch": 398.1666666666667,
      "grad_norm": 1.1425245247664861e-05,
      "learning_rate": 1.3263260317971815e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10033984,
      "step": 35835
    },
    {
      "epoch": 398.22222222222223,
      "grad_norm": 0.00022430758690461516,
      "learning_rate": 1.3231726194622208e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10035376,
      "step": 35840
    },
    {
      "epoch": 398.27777777777777,
      "grad_norm": 7.905001984909177e-05,
      "learning_rate": 1.3200228583915814e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10036800,
      "step": 35845
    },
    {
      "epoch": 398.3333333333333,
      "grad_norm": 0.00012797735689673573,
      "learning_rate": 1.3168767490709971e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10038176,
      "step": 35850
    },
    {
      "epoch": 398.3888888888889,
      "grad_norm": 0.00012827727186959237,
      "learning_rate": 1.3137342919856437e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10039600,
      "step": 35855
    },
    {
      "epoch": 398.44444444444446,
      "grad_norm": 0.00036010160692967474,
      "learning_rate": 1.310595487620117e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10040992,
      "step": 35860
    },
    {
      "epoch": 398.5,
      "grad_norm": 0.0005874878843314946,
      "learning_rate": 1.3074603364584715e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10042416,
      "step": 35865
    },
    {
      "epoch": 398.55555555555554,
      "grad_norm": 0.0001563179976074025,
      "learning_rate": 1.3043288389841758e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10043760,
      "step": 35870
    },
    {
      "epoch": 398.6111111111111,
      "grad_norm": 2.3846281692385674e-05,
      "learning_rate": 1.3012009956801546e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10045216,
      "step": 35875
    },
    {
      "epoch": 398.6666666666667,
      "grad_norm": 0.0006584068760275841,
      "learning_rate": 1.2980768070287586e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10046592,
      "step": 35880
    },
    {
      "epoch": 398.72222222222223,
      "grad_norm": 0.00023367458197753876,
      "learning_rate": 1.2949562735117716e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10047968,
      "step": 35885
    },
    {
      "epoch": 398.77777777777777,
      "grad_norm": 5.428149597719312e-05,
      "learning_rate": 1.291839395610428e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10049392,
      "step": 35890
    },
    {
      "epoch": 398.8333333333333,
      "grad_norm": 3.9612885302631184e-05,
      "learning_rate": 1.2887261738053852e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10050832,
      "step": 35895
    },
    {
      "epoch": 398.8888888888889,
      "grad_norm": 0.00012670371506828815,
      "learning_rate": 1.2856166085767396e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10052176,
      "step": 35900
    },
    {
      "epoch": 398.94444444444446,
      "grad_norm": 0.00011614764662226662,
      "learning_rate": 1.2825107004040272e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10053600,
      "step": 35905
    },
    {
      "epoch": 399.0,
      "grad_norm": 0.0005033689667470753,
      "learning_rate": 1.2794084497662146e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10055008,
      "step": 35910
    },
    {
      "epoch": 399.05555555555554,
      "grad_norm": 1.701677501841914e-05,
      "learning_rate": 1.276309857141711e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10056432,
      "step": 35915
    },
    {
      "epoch": 399.1111111111111,
      "grad_norm": 4.470532076084055e-05,
      "learning_rate": 1.273214923008359e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10057840,
      "step": 35920
    },
    {
      "epoch": 399.1666666666667,
      "grad_norm": 2.8196443963679485e-05,
      "learning_rate": 1.2701236478434352e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10059248,
      "step": 35925
    },
    {
      "epoch": 399.22222222222223,
      "grad_norm": 0.0003903048054780811,
      "learning_rate": 1.2670360321236502e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10060656,
      "step": 35930
    },
    {
      "epoch": 399.27777777777777,
      "grad_norm": 8.80683001014404e-05,
      "learning_rate": 1.2639520763251617e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10062080,
      "step": 35935
    },
    {
      "epoch": 399.3333333333333,
      "grad_norm": 7.925494719529524e-05,
      "learning_rate": 1.2608717809235448e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10063456,
      "step": 35940
    },
    {
      "epoch": 399.3888888888889,
      "grad_norm": 2.6184639864368364e-05,
      "learning_rate": 1.2577951463938282e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10064896,
      "step": 35945
    },
    {
      "epoch": 399.44444444444446,
      "grad_norm": 5.8767731388797984e-05,
      "learning_rate": 1.2547221732104569e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10066320,
      "step": 35950
    },
    {
      "epoch": 399.5,
      "grad_norm": 0.00018138441373594105,
      "learning_rate": 1.25165286184733e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10067696,
      "step": 35955
    },
    {
      "epoch": 399.55555555555554,
      "grad_norm": 0.00010911056597251445,
      "learning_rate": 1.248587212777777e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10069120,
      "step": 35960
    },
    {
      "epoch": 399.6111111111111,
      "grad_norm": 0.00013873162970412523,
      "learning_rate": 1.2455252264745532e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10070464,
      "step": 35965
    },
    {
      "epoch": 399.6666666666667,
      "grad_norm": 4.1802060877671465e-05,
      "learning_rate": 1.2424669034098528e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10071840,
      "step": 35970
    },
    {
      "epoch": 399.72222222222223,
      "grad_norm": 4.867379175266251e-05,
      "learning_rate": 1.2394122440553185e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10073248,
      "step": 35975
    },
    {
      "epoch": 399.77777777777777,
      "grad_norm": 6.288744771154597e-05,
      "learning_rate": 1.2363612488820037e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10074656,
      "step": 35980
    },
    {
      "epoch": 399.8333333333333,
      "grad_norm": 0.000261312146903947,
      "learning_rate": 1.2333139183604208e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10076064,
      "step": 35985
    },
    {
      "epoch": 399.8888888888889,
      "grad_norm": 0.0001221274142153561,
      "learning_rate": 1.2302702529604998e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10077456,
      "step": 35990
    },
    {
      "epoch": 399.94444444444446,
      "grad_norm": 2.247602242277935e-05,
      "learning_rate": 1.227230253151615e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10078848,
      "step": 35995
    },
    {
      "epoch": 400.0,
      "grad_norm": 0.00017160688003059477,
      "learning_rate": 1.2241939194025748e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10080240,
      "step": 36000
    },
    {
      "epoch": 400.0,
      "eval_loss": 0.9163358807563782,
      "eval_runtime": 1.2356,
      "eval_samples_per_second": 32.372,
      "eval_steps_per_second": 16.186,
      "num_input_tokens_seen": 10080240,
      "step": 36000
    },
    {
      "epoch": 400.05555555555554,
      "grad_norm": 4.473425360629335e-05,
      "learning_rate": 1.2211612521816156e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10081712,
      "step": 36005
    },
    {
      "epoch": 400.1111111111111,
      "grad_norm": 0.0027196237351745367,
      "learning_rate": 1.2181322519564137e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10083104,
      "step": 36010
    },
    {
      "epoch": 400.1666666666667,
      "grad_norm": 0.0014291884144768119,
      "learning_rate": 1.2151069191940839e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10084528,
      "step": 36015
    },
    {
      "epoch": 400.22222222222223,
      "grad_norm": 0.001192153780721128,
      "learning_rate": 1.2120852543611644e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10085984,
      "step": 36020
    },
    {
      "epoch": 400.27777777777777,
      "grad_norm": 1.3287904039316345e-05,
      "learning_rate": 1.2090672579236379e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10087392,
      "step": 36025
    },
    {
      "epoch": 400.3333333333333,
      "grad_norm": 9.430239151697606e-05,
      "learning_rate": 1.2060529303469126e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10088784,
      "step": 36030
    },
    {
      "epoch": 400.3888888888889,
      "grad_norm": 4.145306229474954e-05,
      "learning_rate": 1.2030422720958445e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10090160,
      "step": 36035
    },
    {
      "epoch": 400.44444444444446,
      "grad_norm": 2.6452304155100137e-05,
      "learning_rate": 1.200035283634704e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10091536,
      "step": 36040
    },
    {
      "epoch": 400.5,
      "grad_norm": 6.740607204847038e-05,
      "learning_rate": 1.1970319654272144e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10092928,
      "step": 36045
    },
    {
      "epoch": 400.55555555555554,
      "grad_norm": 0.00018299309886060655,
      "learning_rate": 1.1940323179365192e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10094304,
      "step": 36050
    },
    {
      "epoch": 400.6111111111111,
      "grad_norm": 8.237712609115988e-05,
      "learning_rate": 1.1910363416252095e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10095680,
      "step": 36055
    },
    {
      "epoch": 400.6666666666667,
      "grad_norm": 0.00010211868357146159,
      "learning_rate": 1.1880440369552964e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10097072,
      "step": 36060
    },
    {
      "epoch": 400.72222222222223,
      "grad_norm": 0.00029719420126639307,
      "learning_rate": 1.1850554043882328e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10098432,
      "step": 36065
    },
    {
      "epoch": 400.77777777777777,
      "grad_norm": 4.4940734369447455e-05,
      "learning_rate": 1.1820704443849028e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10099856,
      "step": 36070
    },
    {
      "epoch": 400.8333333333333,
      "grad_norm": 1.6634394341963343e-05,
      "learning_rate": 1.1790891574056219e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10101264,
      "step": 36075
    },
    {
      "epoch": 400.8888888888889,
      "grad_norm": 3.097754597547464e-05,
      "learning_rate": 1.1761115439101523e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10102608,
      "step": 36080
    },
    {
      "epoch": 400.94444444444446,
      "grad_norm": 0.001381000387482345,
      "learning_rate": 1.1731376043576659e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10104032,
      "step": 36085
    },
    {
      "epoch": 401.0,
      "grad_norm": 1.8879296476370655e-05,
      "learning_rate": 1.1701673392067875e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10105376,
      "step": 36090
    },
    {
      "epoch": 401.05555555555554,
      "grad_norm": 2.6266863642376848e-05,
      "learning_rate": 1.1672007489155757e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10106800,
      "step": 36095
    },
    {
      "epoch": 401.1111111111111,
      "grad_norm": 8.159526623785496e-05,
      "learning_rate": 1.164237833941506e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10108240,
      "step": 36100
    },
    {
      "epoch": 401.1666666666667,
      "grad_norm": 0.00010113044118043035,
      "learning_rate": 1.1612785947415022e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10109648,
      "step": 36105
    },
    {
      "epoch": 401.22222222222223,
      "grad_norm": 0.000474222149932757,
      "learning_rate": 1.1583230317719185e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10111072,
      "step": 36110
    },
    {
      "epoch": 401.27777777777777,
      "grad_norm": 6.67953718220815e-05,
      "learning_rate": 1.1553711454885318e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10112448,
      "step": 36115
    },
    {
      "epoch": 401.3333333333333,
      "grad_norm": 3.9607206417713314e-05,
      "learning_rate": 1.152422936346567e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10113824,
      "step": 36120
    },
    {
      "epoch": 401.3888888888889,
      "grad_norm": 2.7883846996701322e-05,
      "learning_rate": 1.1494784048006718e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10115232,
      "step": 36125
    },
    {
      "epoch": 401.44444444444446,
      "grad_norm": 8.691389666637406e-05,
      "learning_rate": 1.1465375513049326e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10116656,
      "step": 36130
    },
    {
      "epoch": 401.5,
      "grad_norm": 3.9616901631234214e-05,
      "learning_rate": 1.1436003763128616e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10118096,
      "step": 36135
    },
    {
      "epoch": 401.55555555555554,
      "grad_norm": 0.00011702206393238157,
      "learning_rate": 1.1406668802774106e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10119488,
      "step": 36140
    },
    {
      "epoch": 401.6111111111111,
      "grad_norm": 8.321051427628845e-05,
      "learning_rate": 1.137737063650965e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10120896,
      "step": 36145
    },
    {
      "epoch": 401.6666666666667,
      "grad_norm": 8.226616046158597e-05,
      "learning_rate": 1.1348109268853323e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10122304,
      "step": 36150
    },
    {
      "epoch": 401.72222222222223,
      "grad_norm": 0.0003643007657956332,
      "learning_rate": 1.1318884704317634e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10123712,
      "step": 36155
    },
    {
      "epoch": 401.77777777777777,
      "grad_norm": 6.032051533111371e-05,
      "learning_rate": 1.1289696947409417e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10125104,
      "step": 36160
    },
    {
      "epoch": 401.8333333333333,
      "grad_norm": 6.993365968810394e-05,
      "learning_rate": 1.126054600262974e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10126496,
      "step": 36165
    },
    {
      "epoch": 401.8888888888889,
      "grad_norm": 0.0003155953891109675,
      "learning_rate": 1.1231431874474064e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10127840,
      "step": 36170
    },
    {
      "epoch": 401.94444444444446,
      "grad_norm": 0.00028454980929382145,
      "learning_rate": 1.12023545674321e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10129216,
      "step": 36175
    },
    {
      "epoch": 402.0,
      "grad_norm": 0.0001025590390781872,
      "learning_rate": 1.117331408598804e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10130624,
      "step": 36180
    },
    {
      "epoch": 402.05555555555554,
      "grad_norm": 0.001596470596268773,
      "learning_rate": 1.1144310434620191e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10132032,
      "step": 36185
    },
    {
      "epoch": 402.1111111111111,
      "grad_norm": 0.00020030196174047887,
      "learning_rate": 1.1115343617801365e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10133392,
      "step": 36190
    },
    {
      "epoch": 402.1666666666667,
      "grad_norm": 0.0005208866205066442,
      "learning_rate": 1.1086413639998515e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10134800,
      "step": 36195
    },
    {
      "epoch": 402.22222222222223,
      "grad_norm": 0.0001702748704701662,
      "learning_rate": 1.1057520505673103e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10136208,
      "step": 36200
    },
    {
      "epoch": 402.22222222222223,
      "eval_loss": 0.9173675775527954,
      "eval_runtime": 1.2618,
      "eval_samples_per_second": 31.701,
      "eval_steps_per_second": 15.85,
      "num_input_tokens_seen": 10136208,
      "step": 36200
    },
    {
      "epoch": 402.27777777777777,
      "grad_norm": 3.1129111448535696e-05,
      "learning_rate": 1.1028664219280727e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10137648,
      "step": 36205
    },
    {
      "epoch": 402.3333333333333,
      "grad_norm": 0.00377520639449358,
      "learning_rate": 1.0999844785271468e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10139056,
      "step": 36210
    },
    {
      "epoch": 402.3888888888889,
      "grad_norm": 0.0028080407064408064,
      "learning_rate": 1.097106220808955e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10140432,
      "step": 36215
    },
    {
      "epoch": 402.44444444444446,
      "grad_norm": 8.97493664524518e-05,
      "learning_rate": 1.0942316492173698e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10141824,
      "step": 36220
    },
    {
      "epoch": 402.5,
      "grad_norm": 3.9359652873827145e-05,
      "learning_rate": 1.0913607641956841e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10143216,
      "step": 36225
    },
    {
      "epoch": 402.55555555555554,
      "grad_norm": 0.000211759761441499,
      "learning_rate": 1.0884935661866213e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10144624,
      "step": 36230
    },
    {
      "epoch": 402.6111111111111,
      "grad_norm": 3.0432905987254344e-05,
      "learning_rate": 1.0856300556323418e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10146032,
      "step": 36235
    },
    {
      "epoch": 402.6666666666667,
      "grad_norm": 9.586104170011822e-06,
      "learning_rate": 1.0827702329744365e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10147408,
      "step": 36240
    },
    {
      "epoch": 402.72222222222223,
      "grad_norm": 6.253443280002102e-05,
      "learning_rate": 1.0799140986539197e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10148816,
      "step": 36245
    },
    {
      "epoch": 402.77777777777777,
      "grad_norm": 3.83961305487901e-05,
      "learning_rate": 1.0770616531112526e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10150224,
      "step": 36250
    },
    {
      "epoch": 402.8333333333333,
      "grad_norm": 4.7923796955728903e-05,
      "learning_rate": 1.0742128967863085e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10151648,
      "step": 36255
    },
    {
      "epoch": 402.8888888888889,
      "grad_norm": 0.0005657515139319003,
      "learning_rate": 1.071367830118411e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10152992,
      "step": 36260
    },
    {
      "epoch": 402.94444444444446,
      "grad_norm": 2.6876412448473275e-05,
      "learning_rate": 1.068526453546298e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10154384,
      "step": 36265
    },
    {
      "epoch": 403.0,
      "grad_norm": 6.533823034260422e-05,
      "learning_rate": 1.0656887675081467e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10155744,
      "step": 36270
    },
    {
      "epoch": 403.05555555555554,
      "grad_norm": 2.4475692043779418e-05,
      "learning_rate": 1.0628547724415628e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10157088,
      "step": 36275
    },
    {
      "epoch": 403.1111111111111,
      "grad_norm": 8.281229384010658e-05,
      "learning_rate": 1.0600244687835881e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10158512,
      "step": 36280
    },
    {
      "epoch": 403.1666666666667,
      "grad_norm": 9.186462557408959e-05,
      "learning_rate": 1.0571978569706876e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10159920,
      "step": 36285
    },
    {
      "epoch": 403.22222222222223,
      "grad_norm": 0.0034431645181030035,
      "learning_rate": 1.0543749374387652e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10161296,
      "step": 36290
    },
    {
      "epoch": 403.27777777777777,
      "grad_norm": 0.0002258020977023989,
      "learning_rate": 1.051555710623142e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10162704,
      "step": 36295
    },
    {
      "epoch": 403.3333333333333,
      "grad_norm": 0.0003401144058443606,
      "learning_rate": 1.0487401769585847e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10164128,
      "step": 36300
    },
    {
      "epoch": 403.3888888888889,
      "grad_norm": 6.502346514025703e-05,
      "learning_rate": 1.0459283368792845e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10165552,
      "step": 36305
    },
    {
      "epoch": 403.44444444444446,
      "grad_norm": 0.0003623896627686918,
      "learning_rate": 1.043120190818858e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10166976,
      "step": 36310
    },
    {
      "epoch": 403.5,
      "grad_norm": 0.0001511124282842502,
      "learning_rate": 1.0403157392103596e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10168384,
      "step": 36315
    },
    {
      "epoch": 403.55555555555554,
      "grad_norm": 0.00010132434545084834,
      "learning_rate": 1.0375149824862735e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10169760,
      "step": 36320
    },
    {
      "epoch": 403.6111111111111,
      "grad_norm": 0.0001535055780550465,
      "learning_rate": 1.034717921078507e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10171200,
      "step": 36325
    },
    {
      "epoch": 403.6666666666667,
      "grad_norm": 4.654085569200106e-05,
      "learning_rate": 1.0319245554184009e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10172624,
      "step": 36330
    },
    {
      "epoch": 403.72222222222223,
      "grad_norm": 0.0010514961322769523,
      "learning_rate": 1.0291348859367361e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10174000,
      "step": 36335
    },
    {
      "epoch": 403.77777777777777,
      "grad_norm": 0.00010289818601449952,
      "learning_rate": 1.0263489130637016e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10175408,
      "step": 36340
    },
    {
      "epoch": 403.8333333333333,
      "grad_norm": 1.4878984075039625e-05,
      "learning_rate": 1.0235666372289427e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10176816,
      "step": 36345
    },
    {
      "epoch": 403.8888888888889,
      "grad_norm": 7.989614096004516e-05,
      "learning_rate": 1.0207880588615076e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10178208,
      "step": 36350
    },
    {
      "epoch": 403.94444444444446,
      "grad_norm": 8.400574006373063e-05,
      "learning_rate": 1.0180131783898984e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10179600,
      "step": 36355
    },
    {
      "epoch": 404.0,
      "grad_norm": 9.83211793936789e-06,
      "learning_rate": 1.0152419962420362e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10180976,
      "step": 36360
    },
    {
      "epoch": 404.05555555555554,
      "grad_norm": 0.00016769427747931331,
      "learning_rate": 1.0124745128452685e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10182384,
      "step": 36365
    },
    {
      "epoch": 404.1111111111111,
      "grad_norm": 3.577375173335895e-05,
      "learning_rate": 1.0097107286263758e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10183728,
      "step": 36370
    },
    {
      "epoch": 404.1666666666667,
      "grad_norm": 4.669305053539574e-05,
      "learning_rate": 1.00695064401157e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10185152,
      "step": 36375
    },
    {
      "epoch": 404.22222222222223,
      "grad_norm": 0.00023580336710438132,
      "learning_rate": 1.0041942594264886e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10186560,
      "step": 36380
    },
    {
      "epoch": 404.27777777777777,
      "grad_norm": 0.00016345041512977332,
      "learning_rate": 1.001441575296208e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 10187984,
      "step": 36385
    },
    {
      "epoch": 404.3333333333333,
      "grad_norm": 0.0001452247961424291,
      "learning_rate": 9.986925920452139e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10189392,
      "step": 36390
    },
    {
      "epoch": 404.3888888888889,
      "grad_norm": 0.0001462560467189178,
      "learning_rate": 9.959473100974475e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10190800,
      "step": 36395
    },
    {
      "epoch": 404.44444444444446,
      "grad_norm": 2.9305663701961748e-05,
      "learning_rate": 9.932057298762564e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10192208,
      "step": 36400
    },
    {
      "epoch": 404.44444444444446,
      "eval_loss": 0.8984518051147461,
      "eval_runtime": 1.2379,
      "eval_samples_per_second": 32.312,
      "eval_steps_per_second": 16.156,
      "num_input_tokens_seen": 10192208,
      "step": 36400
    },
    {
      "epoch": 404.5,
      "grad_norm": 0.0004931341973133385,
      "learning_rate": 9.90467851804433e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10193616,
      "step": 36405
    },
    {
      "epoch": 404.55555555555554,
      "grad_norm": 0.00028837742866016924,
      "learning_rate": 9.877336763041895e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10195008,
      "step": 36410
    },
    {
      "epoch": 404.6111111111111,
      "grad_norm": 2.8419543014024384e-05,
      "learning_rate": 9.850032037971662e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10196432,
      "step": 36415
    },
    {
      "epoch": 404.6666666666667,
      "grad_norm": 0.00020258456061128527,
      "learning_rate": 9.822764347044406e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10197856,
      "step": 36420
    },
    {
      "epoch": 404.72222222222223,
      "grad_norm": 0.0004504794196691364,
      "learning_rate": 9.795533694465175e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10199248,
      "step": 36425
    },
    {
      "epoch": 404.77777777777777,
      "grad_norm": 2.9682063541258685e-05,
      "learning_rate": 9.768340084433197e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10200704,
      "step": 36430
    },
    {
      "epoch": 404.8333333333333,
      "grad_norm": 0.00405932916328311,
      "learning_rate": 9.741183521142143e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10202128,
      "step": 36435
    },
    {
      "epoch": 404.8888888888889,
      "grad_norm": 8.592248923378065e-05,
      "learning_rate": 9.714064008779889e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10203488,
      "step": 36440
    },
    {
      "epoch": 404.94444444444446,
      "grad_norm": 6.6286331275478e-05,
      "learning_rate": 9.686981551528584e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10204880,
      "step": 36445
    },
    {
      "epoch": 405.0,
      "grad_norm": 3.056635250686668e-05,
      "learning_rate": 9.65993615356467e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10206272,
      "step": 36450
    },
    {
      "epoch": 405.05555555555554,
      "grad_norm": 2.452197441016324e-05,
      "learning_rate": 9.632927819058917e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10207648,
      "step": 36455
    },
    {
      "epoch": 405.1111111111111,
      "grad_norm": 0.0001197605233755894,
      "learning_rate": 9.605956552176305e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10209040,
      "step": 36460
    },
    {
      "epoch": 405.1666666666667,
      "grad_norm": 9.43851046031341e-05,
      "learning_rate": 9.579022357076223e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10210416,
      "step": 36465
    },
    {
      "epoch": 405.22222222222223,
      "grad_norm": 3.421895962674171e-05,
      "learning_rate": 9.552125237912158e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10211792,
      "step": 36470
    },
    {
      "epoch": 405.27777777777777,
      "grad_norm": 9.268521534977481e-05,
      "learning_rate": 9.525265198832096e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10213200,
      "step": 36475
    },
    {
      "epoch": 405.3333333333333,
      "grad_norm": 3.51066337316297e-05,
      "learning_rate": 9.498442243978112e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10214608,
      "step": 36480
    },
    {
      "epoch": 405.3888888888889,
      "grad_norm": 0.0003038204158656299,
      "learning_rate": 9.471656377486649e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10215968,
      "step": 36485
    },
    {
      "epoch": 405.44444444444446,
      "grad_norm": 1.801157122827135e-05,
      "learning_rate": 9.444907603488456e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10217376,
      "step": 36490
    },
    {
      "epoch": 405.5,
      "grad_norm": 0.00010364938498241827,
      "learning_rate": 9.418195926108514e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10218784,
      "step": 36495
    },
    {
      "epoch": 405.55555555555554,
      "grad_norm": 1.9364968466106802e-05,
      "learning_rate": 9.391521349466053e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10220224,
      "step": 36500
    },
    {
      "epoch": 405.6111111111111,
      "grad_norm": 0.0006882601301185787,
      "learning_rate": 9.364883877674758e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10221648,
      "step": 36505
    },
    {
      "epoch": 405.6666666666667,
      "grad_norm": 0.0002546471368987113,
      "learning_rate": 9.33828351484231e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10223056,
      "step": 36510
    },
    {
      "epoch": 405.72222222222223,
      "grad_norm": 5.019306263420731e-05,
      "learning_rate": 9.311720265070906e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10224464,
      "step": 36515
    },
    {
      "epoch": 405.77777777777777,
      "grad_norm": 8.39408312458545e-05,
      "learning_rate": 9.285194132456931e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10225840,
      "step": 36520
    },
    {
      "epoch": 405.8333333333333,
      "grad_norm": 0.001729911076836288,
      "learning_rate": 9.258705121091032e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10227200,
      "step": 36525
    },
    {
      "epoch": 405.8888888888889,
      "grad_norm": 0.00010826505604200065,
      "learning_rate": 9.232253235058136e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10228624,
      "step": 36530
    },
    {
      "epoch": 405.94444444444446,
      "grad_norm": 7.800128514645621e-05,
      "learning_rate": 9.205838478437478e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10230000,
      "step": 36535
    },
    {
      "epoch": 406.0,
      "grad_norm": 3.7914698623353615e-05,
      "learning_rate": 9.179460855302524e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10231360,
      "step": 36540
    },
    {
      "epoch": 406.05555555555554,
      "grad_norm": 0.00031867195502854884,
      "learning_rate": 9.153120369721046e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10232736,
      "step": 36545
    },
    {
      "epoch": 406.1111111111111,
      "grad_norm": 0.0051446035504341125,
      "learning_rate": 9.126817025755103e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10234144,
      "step": 36550
    },
    {
      "epoch": 406.1666666666667,
      "grad_norm": 0.0001431901182513684,
      "learning_rate": 9.100550827460947e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10235520,
      "step": 36555
    },
    {
      "epoch": 406.22222222222223,
      "grad_norm": 3.52347778971307e-05,
      "learning_rate": 9.0743217788892e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10236960,
      "step": 36560
    },
    {
      "epoch": 406.27777777777777,
      "grad_norm": 4.03887897846289e-05,
      "learning_rate": 9.048129884084683e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10238384,
      "step": 36565
    },
    {
      "epoch": 406.3333333333333,
      "grad_norm": 0.000874845078215003,
      "learning_rate": 9.021975147086553e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10239776,
      "step": 36570
    },
    {
      "epoch": 406.3888888888889,
      "grad_norm": 2.2267380700213835e-05,
      "learning_rate": 8.995857571928141e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10241184,
      "step": 36575
    },
    {
      "epoch": 406.44444444444446,
      "grad_norm": 1.0898841537709814e-05,
      "learning_rate": 8.969777162637139e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10242576,
      "step": 36580
    },
    {
      "epoch": 406.5,
      "grad_norm": 4.997730502509512e-05,
      "learning_rate": 8.943733923235525e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10244000,
      "step": 36585
    },
    {
      "epoch": 406.55555555555554,
      "grad_norm": 6.42339582554996e-05,
      "learning_rate": 8.917727857739394e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10245392,
      "step": 36590
    },
    {
      "epoch": 406.6111111111111,
      "grad_norm": 0.0003234323812648654,
      "learning_rate": 8.891758970159258e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10246816,
      "step": 36595
    },
    {
      "epoch": 406.6666666666667,
      "grad_norm": 1.3272071555547882e-05,
      "learning_rate": 8.86582726449986e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10248192,
      "step": 36600
    },
    {
      "epoch": 406.6666666666667,
      "eval_loss": 0.9003810882568359,
      "eval_runtime": 1.2376,
      "eval_samples_per_second": 32.319,
      "eval_steps_per_second": 16.16,
      "num_input_tokens_seen": 10248192,
      "step": 36600
    },
    {
      "epoch": 406.72222222222223,
      "grad_norm": 2.237497756141238e-05,
      "learning_rate": 8.839932744760165e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10249600,
      "step": 36605
    },
    {
      "epoch": 406.77777777777777,
      "grad_norm": 6.631980068050325e-05,
      "learning_rate": 8.814075414933482e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10250992,
      "step": 36610
    },
    {
      "epoch": 406.8333333333333,
      "grad_norm": 7.803671178407967e-05,
      "learning_rate": 8.788255279007257e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10252384,
      "step": 36615
    },
    {
      "epoch": 406.8888888888889,
      "grad_norm": 5.220812454354018e-05,
      "learning_rate": 8.762472340963362e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10253776,
      "step": 36620
    },
    {
      "epoch": 406.94444444444446,
      "grad_norm": 2.148199018847663e-05,
      "learning_rate": 8.736726604777811e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10255168,
      "step": 36625
    },
    {
      "epoch": 407.0,
      "grad_norm": 0.0004067301342729479,
      "learning_rate": 8.711018074420901e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10256544,
      "step": 36630
    },
    {
      "epoch": 407.05555555555554,
      "grad_norm": 4.981032543582842e-05,
      "learning_rate": 8.685346753857209e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10257952,
      "step": 36635
    },
    {
      "epoch": 407.1111111111111,
      "grad_norm": 0.00027558114379644394,
      "learning_rate": 8.659712647045654e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10259328,
      "step": 36640
    },
    {
      "epoch": 407.1666666666667,
      "grad_norm": 1.295846959692426e-05,
      "learning_rate": 8.634115757939209e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10260704,
      "step": 36645
    },
    {
      "epoch": 407.22222222222223,
      "grad_norm": 8.793826418695971e-05,
      "learning_rate": 8.608556090485387e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10262096,
      "step": 36650
    },
    {
      "epoch": 407.27777777777777,
      "grad_norm": 0.00022850664390716702,
      "learning_rate": 8.583033648625671e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10263488,
      "step": 36655
    },
    {
      "epoch": 407.3333333333333,
      "grad_norm": 0.0001908485864987597,
      "learning_rate": 8.557548436295998e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10264864,
      "step": 36660
    },
    {
      "epoch": 407.3888888888889,
      "grad_norm": 0.00041826412780210376,
      "learning_rate": 8.532100457426556e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10266320,
      "step": 36665
    },
    {
      "epoch": 407.44444444444446,
      "grad_norm": 0.00010752851085271686,
      "learning_rate": 8.506689715941679e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10267728,
      "step": 36670
    },
    {
      "epoch": 407.5,
      "grad_norm": 0.0004595934587996453,
      "learning_rate": 8.481316215760011e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10269104,
      "step": 36675
    },
    {
      "epoch": 407.55555555555554,
      "grad_norm": 1.3008881978748832e-05,
      "learning_rate": 8.455979960794558e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10270528,
      "step": 36680
    },
    {
      "epoch": 407.6111111111111,
      "grad_norm": 6.720084638800472e-05,
      "learning_rate": 8.430680954952364e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10271920,
      "step": 36685
    },
    {
      "epoch": 407.6666666666667,
      "grad_norm": 0.00011164851457579061,
      "learning_rate": 8.405419202134974e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10273296,
      "step": 36690
    },
    {
      "epoch": 407.72222222222223,
      "grad_norm": 2.0390782083268277e-05,
      "learning_rate": 8.380194706237993e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10274688,
      "step": 36695
    },
    {
      "epoch": 407.77777777777777,
      "grad_norm": 2.4620472686365247e-05,
      "learning_rate": 8.355007471151366e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10276128,
      "step": 36700
    },
    {
      "epoch": 407.8333333333333,
      "grad_norm": 0.004484888166189194,
      "learning_rate": 8.329857500759292e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10277520,
      "step": 36705
    },
    {
      "epoch": 407.8888888888889,
      "grad_norm": 0.0008458244847133756,
      "learning_rate": 8.304744798940194e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10278912,
      "step": 36710
    },
    {
      "epoch": 407.94444444444446,
      "grad_norm": 0.0001281843869946897,
      "learning_rate": 8.279669369566756e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10280288,
      "step": 36715
    },
    {
      "epoch": 408.0,
      "grad_norm": 6.803870928706601e-05,
      "learning_rate": 8.254631216505993e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10281680,
      "step": 36720
    },
    {
      "epoch": 408.05555555555554,
      "grad_norm": 0.0001804322237148881,
      "learning_rate": 8.229630343619038e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10283104,
      "step": 36725
    },
    {
      "epoch": 408.1111111111111,
      "grad_norm": 0.0001304416509810835,
      "learning_rate": 8.204666754761392e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10284544,
      "step": 36730
    },
    {
      "epoch": 408.1666666666667,
      "grad_norm": 0.00017762988863978535,
      "learning_rate": 8.179740453782669e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10285952,
      "step": 36735
    },
    {
      "epoch": 408.22222222222223,
      "grad_norm": 0.0002312192809768021,
      "learning_rate": 8.154851444526907e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10287328,
      "step": 36740
    },
    {
      "epoch": 408.27777777777777,
      "grad_norm": 0.0007663167198188603,
      "learning_rate": 8.129999730832283e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10288688,
      "step": 36745
    },
    {
      "epoch": 408.3333333333333,
      "grad_norm": 0.0001541415840620175,
      "learning_rate": 8.105185316531178e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10290064,
      "step": 36750
    },
    {
      "epoch": 408.3888888888889,
      "grad_norm": 8.596305997343734e-05,
      "learning_rate": 8.08040820545039e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10291472,
      "step": 36755
    },
    {
      "epoch": 408.44444444444446,
      "grad_norm": 0.000106465769931674,
      "learning_rate": 8.055668401410782e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10292896,
      "step": 36760
    },
    {
      "epoch": 408.5,
      "grad_norm": 8.20678033051081e-05,
      "learning_rate": 8.030965908227578e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10294352,
      "step": 36765
    },
    {
      "epoch": 408.55555555555554,
      "grad_norm": 0.0007041179342195392,
      "learning_rate": 8.006300729710203e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10295760,
      "step": 36770
    },
    {
      "epoch": 408.6111111111111,
      "grad_norm": 0.00036149818333797157,
      "learning_rate": 7.981672869662337e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10297168,
      "step": 36775
    },
    {
      "epoch": 408.6666666666667,
      "grad_norm": 1.985848939511925e-05,
      "learning_rate": 7.957082331881888e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10298592,
      "step": 36780
    },
    {
      "epoch": 408.72222222222223,
      "grad_norm": 8.388448623009026e-05,
      "learning_rate": 7.932529120161069e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10299952,
      "step": 36785
    },
    {
      "epoch": 408.77777777777777,
      "grad_norm": 2.7862810384249315e-05,
      "learning_rate": 7.908013238286243e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10301392,
      "step": 36790
    },
    {
      "epoch": 408.8333333333333,
      "grad_norm": 0.0001588627928867936,
      "learning_rate": 7.883534690038136e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10302768,
      "step": 36795
    },
    {
      "epoch": 408.8888888888889,
      "grad_norm": 2.1998903321218677e-05,
      "learning_rate": 7.859093479191559e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10304144,
      "step": 36800
    },
    {
      "epoch": 408.8888888888889,
      "eval_loss": 0.940893828868866,
      "eval_runtime": 1.2416,
      "eval_samples_per_second": 32.215,
      "eval_steps_per_second": 16.108,
      "num_input_tokens_seen": 10304144,
      "step": 36800
    },
    {
      "epoch": 408.94444444444446,
      "grad_norm": 2.7012632926926017e-05,
      "learning_rate": 7.834689609515722e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10305552,
      "step": 36805
    },
    {
      "epoch": 409.0,
      "grad_norm": 5.4108386393636465e-05,
      "learning_rate": 7.810323084774002e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10306960,
      "step": 36810
    },
    {
      "epoch": 409.05555555555554,
      "grad_norm": 0.00019857005099765956,
      "learning_rate": 7.785993908723976e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10308320,
      "step": 36815
    },
    {
      "epoch": 409.1111111111111,
      "grad_norm": 0.00021301404922269285,
      "learning_rate": 7.761702085117534e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10309728,
      "step": 36820
    },
    {
      "epoch": 409.1666666666667,
      "grad_norm": 0.00012200960918562487,
      "learning_rate": 7.737447617700844e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10311136,
      "step": 36825
    },
    {
      "epoch": 409.22222222222223,
      "grad_norm": 4.6541837946278974e-05,
      "learning_rate": 7.713230510214136e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10312544,
      "step": 36830
    },
    {
      "epoch": 409.27777777777777,
      "grad_norm": 2.3164913727669045e-05,
      "learning_rate": 7.689050766392092e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10313936,
      "step": 36835
    },
    {
      "epoch": 409.3333333333333,
      "grad_norm": 5.150716606294736e-05,
      "learning_rate": 7.664908389963477e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10315376,
      "step": 36840
    },
    {
      "epoch": 409.3888888888889,
      "grad_norm": 2.595536716398783e-05,
      "learning_rate": 7.64080338465134e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10316752,
      "step": 36845
    },
    {
      "epoch": 409.44444444444446,
      "grad_norm": 5.2848630730295554e-05,
      "learning_rate": 7.616735754173043e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10318112,
      "step": 36850
    },
    {
      "epoch": 409.5,
      "grad_norm": 0.000149013168993406,
      "learning_rate": 7.592705502240005e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10319520,
      "step": 36855
    },
    {
      "epoch": 409.55555555555554,
      "grad_norm": 0.0008954342338256538,
      "learning_rate": 7.568712632558095e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10320880,
      "step": 36860
    },
    {
      "epoch": 409.6111111111111,
      "grad_norm": 2.9315464416868053e-05,
      "learning_rate": 7.544757148827297e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10322304,
      "step": 36865
    },
    {
      "epoch": 409.6666666666667,
      "grad_norm": 0.0005318865296430886,
      "learning_rate": 7.520839054741797e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10323760,
      "step": 36870
    },
    {
      "epoch": 409.72222222222223,
      "grad_norm": 5.246501314104535e-05,
      "learning_rate": 7.496958353990113e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10325168,
      "step": 36875
    },
    {
      "epoch": 409.77777777777777,
      "grad_norm": 7.8529556049034e-05,
      "learning_rate": 7.473115050254941e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10326608,
      "step": 36880
    },
    {
      "epoch": 409.8333333333333,
      "grad_norm": 6.439149001380429e-05,
      "learning_rate": 7.449309147213173e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10327984,
      "step": 36885
    },
    {
      "epoch": 409.8888888888889,
      "grad_norm": 0.0019128350540995598,
      "learning_rate": 7.425540648536067e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10329360,
      "step": 36890
    },
    {
      "epoch": 409.94444444444446,
      "grad_norm": 5.673895066138357e-05,
      "learning_rate": 7.40180955788894e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10330752,
      "step": 36895
    },
    {
      "epoch": 410.0,
      "grad_norm": 4.7134115447988734e-05,
      "learning_rate": 7.378115878931474e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10332176,
      "step": 36900
    },
    {
      "epoch": 410.05555555555554,
      "grad_norm": 3.37303827109281e-05,
      "learning_rate": 7.354459615317527e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10333600,
      "step": 36905
    },
    {
      "epoch": 410.1111111111111,
      "grad_norm": 3.467262285994366e-05,
      "learning_rate": 7.33084077069518e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10335024,
      "step": 36910
    },
    {
      "epoch": 410.1666666666667,
      "grad_norm": 1.0095575817103963e-05,
      "learning_rate": 7.307259348706768e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10336448,
      "step": 36915
    },
    {
      "epoch": 410.22222222222223,
      "grad_norm": 8.168540080077946e-05,
      "learning_rate": 7.283715352988801e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10337856,
      "step": 36920
    },
    {
      "epoch": 410.27777777777777,
      "grad_norm": 0.00012749448069371283,
      "learning_rate": 7.260208787172068e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10339200,
      "step": 36925
    },
    {
      "epoch": 410.3333333333333,
      "grad_norm": 0.0007744770846329629,
      "learning_rate": 7.23673965488167e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10340608,
      "step": 36930
    },
    {
      "epoch": 410.3888888888889,
      "grad_norm": 0.00012803426943719387,
      "learning_rate": 7.213307959736709e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10342000,
      "step": 36935
    },
    {
      "epoch": 410.44444444444446,
      "grad_norm": 0.0006821781280450523,
      "learning_rate": 7.189913705350715e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10343424,
      "step": 36940
    },
    {
      "epoch": 410.5,
      "grad_norm": 0.0001222990540554747,
      "learning_rate": 7.166556895331411e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10344784,
      "step": 36945
    },
    {
      "epoch": 410.55555555555554,
      "grad_norm": 0.00047486883704550564,
      "learning_rate": 7.143237533280639e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10346128,
      "step": 36950
    },
    {
      "epoch": 410.6111111111111,
      "grad_norm": 0.00013266528549138457,
      "learning_rate": 7.119955622794578e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10347488,
      "step": 36955
    },
    {
      "epoch": 410.6666666666667,
      "grad_norm": 2.7033327569370158e-05,
      "learning_rate": 7.096711167463577e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10348912,
      "step": 36960
    },
    {
      "epoch": 410.72222222222223,
      "grad_norm": 0.00022168911527842283,
      "learning_rate": 7.073504170872213e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10350288,
      "step": 36965
    },
    {
      "epoch": 410.77777777777777,
      "grad_norm": 1.524709841760341e-05,
      "learning_rate": 7.05033463659932e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10351712,
      "step": 36970
    },
    {
      "epoch": 410.8333333333333,
      "grad_norm": 0.00010986644338117912,
      "learning_rate": 7.027202568217928e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10353136,
      "step": 36975
    },
    {
      "epoch": 410.8888888888889,
      "grad_norm": 0.0005914726643823087,
      "learning_rate": 7.004107969295293e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10354576,
      "step": 36980
    },
    {
      "epoch": 410.94444444444446,
      "grad_norm": 2.005253372772131e-05,
      "learning_rate": 6.9810508433929e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10356000,
      "step": 36985
    },
    {
      "epoch": 411.0,
      "grad_norm": 5.6556888011982664e-05,
      "learning_rate": 6.958031194066406e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10357408,
      "step": 36990
    },
    {
      "epoch": 411.05555555555554,
      "grad_norm": 0.0001430385309504345,
      "learning_rate": 6.935049024865776e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10358784,
      "step": 36995
    },
    {
      "epoch": 411.1111111111111,
      "grad_norm": 0.000735629175323993,
      "learning_rate": 6.912104339335118e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10360192,
      "step": 37000
    },
    {
      "epoch": 411.1111111111111,
      "eval_loss": 0.926528811454773,
      "eval_runtime": 1.3013,
      "eval_samples_per_second": 30.74,
      "eval_steps_per_second": 15.37,
      "num_input_tokens_seen": 10360192,
      "step": 37000
    },
    {
      "epoch": 411.1666666666667,
      "grad_norm": 1.839276410464663e-05,
      "learning_rate": 6.889197141012799e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10361616,
      "step": 37005
    },
    {
      "epoch": 411.22222222222223,
      "grad_norm": 2.2454709323938005e-05,
      "learning_rate": 6.866327433431435e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10363040,
      "step": 37010
    },
    {
      "epoch": 411.27777777777777,
      "grad_norm": 8.048405288718641e-05,
      "learning_rate": 6.843495220117735e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10364432,
      "step": 37015
    },
    {
      "epoch": 411.3333333333333,
      "grad_norm": 6.60957011859864e-05,
      "learning_rate": 6.820700504592798e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10365808,
      "step": 37020
    },
    {
      "epoch": 411.3888888888889,
      "grad_norm": 0.0014261420583352447,
      "learning_rate": 6.797943290371839e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10367200,
      "step": 37025
    },
    {
      "epoch": 411.44444444444446,
      "grad_norm": 0.00031931407283991575,
      "learning_rate": 6.775223580964274e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10368592,
      "step": 37030
    },
    {
      "epoch": 411.5,
      "grad_norm": 0.00012236715701874346,
      "learning_rate": 6.7525413798738e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10370016,
      "step": 37035
    },
    {
      "epoch": 411.55555555555554,
      "grad_norm": 7.875201117713004e-05,
      "learning_rate": 6.729896690598259e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10371424,
      "step": 37040
    },
    {
      "epoch": 411.6111111111111,
      "grad_norm": 5.158688509254716e-05,
      "learning_rate": 6.707289516629772e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10372832,
      "step": 37045
    },
    {
      "epoch": 411.6666666666667,
      "grad_norm": 0.0001394408755004406,
      "learning_rate": 6.684719861454692e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10374192,
      "step": 37050
    },
    {
      "epoch": 411.72222222222223,
      "grad_norm": 0.00010391778050689027,
      "learning_rate": 6.662187728553481e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10375600,
      "step": 37055
    },
    {
      "epoch": 411.77777777777777,
      "grad_norm": 1.832687485148199e-05,
      "learning_rate": 6.639693121400892e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10377008,
      "step": 37060
    },
    {
      "epoch": 411.8333333333333,
      "grad_norm": 0.0013483816292136908,
      "learning_rate": 6.617236043465868e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10378432,
      "step": 37065
    },
    {
      "epoch": 411.8888888888889,
      "grad_norm": 8.687270019436255e-05,
      "learning_rate": 6.594816498211587e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10379840,
      "step": 37070
    },
    {
      "epoch": 411.94444444444446,
      "grad_norm": 0.00014009741425979882,
      "learning_rate": 6.572434489095447e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10381264,
      "step": 37075
    },
    {
      "epoch": 412.0,
      "grad_norm": 0.00024750243755988777,
      "learning_rate": 6.550090019568994e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10382640,
      "step": 37080
    },
    {
      "epoch": 412.05555555555554,
      "grad_norm": 5.5975106079131365e-05,
      "learning_rate": 6.527783093078027e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10384016,
      "step": 37085
    },
    {
      "epoch": 412.1111111111111,
      "grad_norm": 7.37653172109276e-05,
      "learning_rate": 6.5055137130626e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10385440,
      "step": 37090
    },
    {
      "epoch": 412.1666666666667,
      "grad_norm": 4.079476639162749e-05,
      "learning_rate": 6.483281882956854e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10386848,
      "step": 37095
    },
    {
      "epoch": 412.22222222222223,
      "grad_norm": 5.91682837693952e-05,
      "learning_rate": 6.461087606189298e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10388240,
      "step": 37100
    },
    {
      "epoch": 412.27777777777777,
      "grad_norm": 0.0007167523726820946,
      "learning_rate": 6.438930886182554e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10389648,
      "step": 37105
    },
    {
      "epoch": 412.3333333333333,
      "grad_norm": 4.217441164655611e-05,
      "learning_rate": 6.416811726353417e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10391072,
      "step": 37110
    },
    {
      "epoch": 412.3888888888889,
      "grad_norm": 4.533588435151614e-05,
      "learning_rate": 6.394730130112991e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10392544,
      "step": 37115
    },
    {
      "epoch": 412.44444444444446,
      "grad_norm": 0.0004391594266053289,
      "learning_rate": 6.372686100866471e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10393936,
      "step": 37120
    },
    {
      "epoch": 412.5,
      "grad_norm": 4.094285759492777e-05,
      "learning_rate": 6.350679642013413e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10395312,
      "step": 37125
    },
    {
      "epoch": 412.55555555555554,
      "grad_norm": 0.00025096526951529086,
      "learning_rate": 6.328710756947437e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10396704,
      "step": 37130
    },
    {
      "epoch": 412.6111111111111,
      "grad_norm": 2.1889092749916017e-05,
      "learning_rate": 6.306779449056416e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10398096,
      "step": 37135
    },
    {
      "epoch": 412.6666666666667,
      "grad_norm": 0.00025822268798947334,
      "learning_rate": 6.284885721722422e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10399520,
      "step": 37140
    },
    {
      "epoch": 412.72222222222223,
      "grad_norm": 0.0003706989227794111,
      "learning_rate": 6.26302957832181e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10400944,
      "step": 37145
    },
    {
      "epoch": 412.77777777777777,
      "grad_norm": 0.0017626748885959387,
      "learning_rate": 6.241211022224997e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10402336,
      "step": 37150
    },
    {
      "epoch": 412.8333333333333,
      "grad_norm": 3.232256494811736e-05,
      "learning_rate": 6.219430056796732e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10403712,
      "step": 37155
    },
    {
      "epoch": 412.8888888888889,
      "grad_norm": 5.163656533113681e-05,
      "learning_rate": 6.19768668539586e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10405104,
      "step": 37160
    },
    {
      "epoch": 412.94444444444446,
      "grad_norm": 0.0001545131963212043,
      "learning_rate": 6.175980911375528e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10406528,
      "step": 37165
    },
    {
      "epoch": 413.0,
      "grad_norm": 0.00019147625425830483,
      "learning_rate": 6.154312738083034e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10407920,
      "step": 37170
    },
    {
      "epoch": 413.05555555555554,
      "grad_norm": 0.00016860259347595274,
      "learning_rate": 6.132682168859843e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10409312,
      "step": 37175
    },
    {
      "epoch": 413.1111111111111,
      "grad_norm": 0.00023855979088693857,
      "learning_rate": 6.111089207041704e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10410640,
      "step": 37180
    },
    {
      "epoch": 413.1666666666667,
      "grad_norm": 4.223341602482833e-05,
      "learning_rate": 6.089533855958507e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10412016,
      "step": 37185
    },
    {
      "epoch": 413.22222222222223,
      "grad_norm": 0.0001302469609072432,
      "learning_rate": 6.068016118934372e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10413472,
      "step": 37190
    },
    {
      "epoch": 413.27777777777777,
      "grad_norm": 2.5046581868082285e-05,
      "learning_rate": 6.04653599928759e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10414848,
      "step": 37195
    },
    {
      "epoch": 413.3333333333333,
      "grad_norm": 2.365854561503511e-05,
      "learning_rate": 6.025093500330675e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10416288,
      "step": 37200
    },
    {
      "epoch": 413.3333333333333,
      "eval_loss": 0.885576605796814,
      "eval_runtime": 1.2492,
      "eval_samples_per_second": 32.022,
      "eval_steps_per_second": 16.011,
      "num_input_tokens_seen": 10416288,
      "step": 37200
    },
    {
      "epoch": 413.3888888888889,
      "grad_norm": 2.5029939934029244e-05,
      "learning_rate": 6.003688625370291e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10417696,
      "step": 37205
    },
    {
      "epoch": 413.44444444444446,
      "grad_norm": 7.263821316882968e-05,
      "learning_rate": 5.982321377707406e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10419136,
      "step": 37210
    },
    {
      "epoch": 413.5,
      "grad_norm": 0.00010285489406669512,
      "learning_rate": 5.96099176063708e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10420560,
      "step": 37215
    },
    {
      "epoch": 413.55555555555554,
      "grad_norm": 8.794508903520182e-05,
      "learning_rate": 5.93969977744857e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10421984,
      "step": 37220
    },
    {
      "epoch": 413.6111111111111,
      "grad_norm": 0.0005496313096955419,
      "learning_rate": 5.918445431425445e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10423408,
      "step": 37225
    },
    {
      "epoch": 413.6666666666667,
      "grad_norm": 0.0003808319743257016,
      "learning_rate": 5.897228725845333e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10424800,
      "step": 37230
    },
    {
      "epoch": 413.72222222222223,
      "grad_norm": 3.192485746694729e-05,
      "learning_rate": 5.876049663980171e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10426208,
      "step": 37235
    },
    {
      "epoch": 413.77777777777777,
      "grad_norm": 1.7255051716347225e-05,
      "learning_rate": 5.854908249095959e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10427616,
      "step": 37240
    },
    {
      "epoch": 413.8333333333333,
      "grad_norm": 1.669188895903062e-05,
      "learning_rate": 5.833804484453031e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10429008,
      "step": 37245
    },
    {
      "epoch": 413.8888888888889,
      "grad_norm": 0.0005995164392516017,
      "learning_rate": 5.81273837330587e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10430416,
      "step": 37250
    },
    {
      "epoch": 413.94444444444446,
      "grad_norm": 0.00012093944678781554,
      "learning_rate": 5.791709918903071e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10431824,
      "step": 37255
    },
    {
      "epoch": 414.0,
      "grad_norm": 0.0005942285060882568,
      "learning_rate": 5.770719124487483e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10433200,
      "step": 37260
    },
    {
      "epoch": 414.05555555555554,
      "grad_norm": 7.168781303334981e-05,
      "learning_rate": 5.749765993296241e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10434624,
      "step": 37265
    },
    {
      "epoch": 414.1111111111111,
      "grad_norm": 1.4073155398364179e-05,
      "learning_rate": 5.728850528560509e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10436016,
      "step": 37270
    },
    {
      "epoch": 414.1666666666667,
      "grad_norm": 0.0038030906580388546,
      "learning_rate": 5.707972733505707e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10437408,
      "step": 37275
    },
    {
      "epoch": 414.22222222222223,
      "grad_norm": 0.0002801019581966102,
      "learning_rate": 5.687132611351509e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10438784,
      "step": 37280
    },
    {
      "epoch": 414.27777777777777,
      "grad_norm": 9.069636871572584e-05,
      "learning_rate": 5.666330165311651e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10440176,
      "step": 37285
    },
    {
      "epoch": 414.3333333333333,
      "grad_norm": 0.0004362966865301132,
      "learning_rate": 5.645565398594204e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10441600,
      "step": 37290
    },
    {
      "epoch": 414.3888888888889,
      "grad_norm": 0.0001441653148503974,
      "learning_rate": 5.624838314401304e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10442976,
      "step": 37295
    },
    {
      "epoch": 414.44444444444446,
      "grad_norm": 0.00222682929597795,
      "learning_rate": 5.604148915929336e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10444384,
      "step": 37300
    },
    {
      "epoch": 414.5,
      "grad_norm": 6.031082011759281e-05,
      "learning_rate": 5.583497206368887e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10445728,
      "step": 37305
    },
    {
      "epoch": 414.55555555555554,
      "grad_norm": 0.00014558325347024947,
      "learning_rate": 5.562883188904688e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10447168,
      "step": 37310
    },
    {
      "epoch": 414.6111111111111,
      "grad_norm": 2.0092871636734344e-05,
      "learning_rate": 5.542306866715724e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10448560,
      "step": 37315
    },
    {
      "epoch": 414.6666666666667,
      "grad_norm": 1.8847855244530365e-05,
      "learning_rate": 5.52176824297504e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10449968,
      "step": 37320
    },
    {
      "epoch": 414.72222222222223,
      "grad_norm": 0.004729277919977903,
      "learning_rate": 5.501267320850018e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10451376,
      "step": 37325
    },
    {
      "epoch": 414.77777777777777,
      "grad_norm": 1.9620767488959245e-05,
      "learning_rate": 5.480804103502157e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10452800,
      "step": 37330
    },
    {
      "epoch": 414.8333333333333,
      "grad_norm": 0.0005005698767490685,
      "learning_rate": 5.460378594087101e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10454192,
      "step": 37335
    },
    {
      "epoch": 414.8888888888889,
      "grad_norm": 6.811622006352991e-05,
      "learning_rate": 5.439990795754773e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10455600,
      "step": 37340
    },
    {
      "epoch": 414.94444444444446,
      "grad_norm": 3.9188078517327085e-05,
      "learning_rate": 5.419640711649188e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10456992,
      "step": 37345
    },
    {
      "epoch": 415.0,
      "grad_norm": 0.00027044545277021825,
      "learning_rate": 5.399328344908583e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10458416,
      "step": 37350
    },
    {
      "epoch": 415.05555555555554,
      "grad_norm": 3.056618879782036e-05,
      "learning_rate": 5.379053698665399e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10459856,
      "step": 37355
    },
    {
      "epoch": 415.1111111111111,
      "grad_norm": 3.771080810111016e-05,
      "learning_rate": 5.358816776046216e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10461248,
      "step": 37360
    },
    {
      "epoch": 415.1666666666667,
      "grad_norm": 4.855382576351985e-05,
      "learning_rate": 5.338617580171817e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10462608,
      "step": 37365
    },
    {
      "epoch": 415.22222222222223,
      "grad_norm": 0.0015749912708997726,
      "learning_rate": 5.318456114157239e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10464016,
      "step": 37370
    },
    {
      "epoch": 415.27777777777777,
      "grad_norm": 6.592028512386605e-05,
      "learning_rate": 5.298332381111576e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10465392,
      "step": 37375
    },
    {
      "epoch": 415.3333333333333,
      "grad_norm": 0.0003908084472641349,
      "learning_rate": 5.27824638413818e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10466800,
      "step": 37380
    },
    {
      "epoch": 415.3888888888889,
      "grad_norm": 0.00010845263750525191,
      "learning_rate": 5.258198126334546e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10468192,
      "step": 37385
    },
    {
      "epoch": 415.44444444444446,
      "grad_norm": 6.658331403741613e-05,
      "learning_rate": 5.238187610792367e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10469520,
      "step": 37390
    },
    {
      "epoch": 415.5,
      "grad_norm": 6.142907659523189e-05,
      "learning_rate": 5.218214840597563e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10470912,
      "step": 37395
    },
    {
      "epoch": 415.55555555555554,
      "grad_norm": 4.4016680476488546e-05,
      "learning_rate": 5.198279818830115e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10472368,
      "step": 37400
    },
    {
      "epoch": 415.55555555555554,
      "eval_loss": 0.8769880533218384,
      "eval_runtime": 1.2389,
      "eval_samples_per_second": 32.287,
      "eval_steps_per_second": 16.144,
      "num_input_tokens_seen": 10472368,
      "step": 37400
    },
    {
      "epoch": 415.6111111111111,
      "grad_norm": 3.055676643270999e-05,
      "learning_rate": 5.178382548564287e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10473776,
      "step": 37405
    },
    {
      "epoch": 415.6666666666667,
      "grad_norm": 0.0001915755565278232,
      "learning_rate": 5.15852303286854e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10475216,
      "step": 37410
    },
    {
      "epoch": 415.72222222222223,
      "grad_norm": 4.274285674910061e-05,
      "learning_rate": 5.138701274805396e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10476608,
      "step": 37415
    },
    {
      "epoch": 415.77777777777777,
      "grad_norm": 9.974245040211827e-05,
      "learning_rate": 5.118917277431606e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10477984,
      "step": 37420
    },
    {
      "epoch": 415.8333333333333,
      "grad_norm": 0.0001906537654576823,
      "learning_rate": 5.099171043798145e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10479360,
      "step": 37425
    },
    {
      "epoch": 415.8888888888889,
      "grad_norm": 0.0006268461002036929,
      "learning_rate": 5.079462576950133e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10480752,
      "step": 37430
    },
    {
      "epoch": 415.94444444444446,
      "grad_norm": 2.201302777393721e-05,
      "learning_rate": 5.059791879926862e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10482160,
      "step": 37435
    },
    {
      "epoch": 416.0,
      "grad_norm": 4.220685514155775e-05,
      "learning_rate": 5.040158955761793e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10483600,
      "step": 37440
    },
    {
      "epoch": 416.05555555555554,
      "grad_norm": 0.0008727462845854461,
      "learning_rate": 5.020563807482559e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10485024,
      "step": 37445
    },
    {
      "epoch": 416.1111111111111,
      "grad_norm": 2.9001608709222637e-05,
      "learning_rate": 5.001006438110995e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10486432,
      "step": 37450
    },
    {
      "epoch": 416.1666666666667,
      "grad_norm": 0.0006982215563766658,
      "learning_rate": 4.981486850663075e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10487792,
      "step": 37455
    },
    {
      "epoch": 416.22222222222223,
      "grad_norm": 1.1527018614287954e-05,
      "learning_rate": 4.962005048149005e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10489136,
      "step": 37460
    },
    {
      "epoch": 416.27777777777777,
      "grad_norm": 2.9713684853049926e-05,
      "learning_rate": 4.942561033573073e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10490528,
      "step": 37465
    },
    {
      "epoch": 416.3333333333333,
      "grad_norm": 0.0003448808565735817,
      "learning_rate": 4.923154809933827e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10491920,
      "step": 37470
    },
    {
      "epoch": 416.3888888888889,
      "grad_norm": 3.7681656976928934e-05,
      "learning_rate": 4.903786380223957e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10493360,
      "step": 37475
    },
    {
      "epoch": 416.44444444444446,
      "grad_norm": 5.233665797277354e-05,
      "learning_rate": 4.884455747430266e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10494784,
      "step": 37480
    },
    {
      "epoch": 416.5,
      "grad_norm": 8.655522105982527e-05,
      "learning_rate": 4.865162914533816e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10496144,
      "step": 37485
    },
    {
      "epoch": 416.55555555555554,
      "grad_norm": 5.067757228971459e-05,
      "learning_rate": 4.845907884509809e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10497520,
      "step": 37490
    },
    {
      "epoch": 416.6111111111111,
      "grad_norm": 0.0017255214042961597,
      "learning_rate": 4.82669066032762e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10498928,
      "step": 37495
    },
    {
      "epoch": 416.6666666666667,
      "grad_norm": 6.124899664428085e-05,
      "learning_rate": 4.807511244950768e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10500336,
      "step": 37500
    },
    {
      "epoch": 416.72222222222223,
      "grad_norm": 4.422999336384237e-05,
      "learning_rate": 4.788369641336943e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10501696,
      "step": 37505
    },
    {
      "epoch": 416.77777777777777,
      "grad_norm": 0.0008458346710540354,
      "learning_rate": 4.769265852438032e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10503136,
      "step": 37510
    },
    {
      "epoch": 416.8333333333333,
      "grad_norm": 0.00011434422776801512,
      "learning_rate": 4.750199881200124e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10504544,
      "step": 37515
    },
    {
      "epoch": 416.8888888888889,
      "grad_norm": 0.0022867084480822086,
      "learning_rate": 4.7311717305633664e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10505984,
      "step": 37520
    },
    {
      "epoch": 416.94444444444446,
      "grad_norm": 3.5647968616103753e-05,
      "learning_rate": 4.7121814034621623e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10507408,
      "step": 37525
    },
    {
      "epoch": 417.0,
      "grad_norm": 0.00021031495998613536,
      "learning_rate": 4.693228902825114e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10508816,
      "step": 37530
    },
    {
      "epoch": 417.05555555555554,
      "grad_norm": 2.0985064111300744e-05,
      "learning_rate": 4.6743142315748277e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10510176,
      "step": 37535
    },
    {
      "epoch": 417.1111111111111,
      "grad_norm": 0.0018941637827083468,
      "learning_rate": 4.655437392628276e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10511552,
      "step": 37540
    },
    {
      "epoch": 417.1666666666667,
      "grad_norm": 3.5126802686136216e-05,
      "learning_rate": 4.636598388896463e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10512928,
      "step": 37545
    },
    {
      "epoch": 417.22222222222223,
      "grad_norm": 0.00010966337868012488,
      "learning_rate": 4.6177972232845925e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10514304,
      "step": 37550
    },
    {
      "epoch": 417.27777777777777,
      "grad_norm": 5.578529817285016e-05,
      "learning_rate": 4.5990338986920953e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10515696,
      "step": 37555
    },
    {
      "epoch": 417.3333333333333,
      "grad_norm": 1.4576442481484264e-05,
      "learning_rate": 4.5803084180124633e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10517072,
      "step": 37560
    },
    {
      "epoch": 417.3888888888889,
      "grad_norm": 2.636886711115949e-05,
      "learning_rate": 4.561620784133386e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10518432,
      "step": 37565
    },
    {
      "epoch": 417.44444444444446,
      "grad_norm": 2.2204043489182368e-05,
      "learning_rate": 4.5429709999367796e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10519872,
      "step": 37570
    },
    {
      "epoch": 417.5,
      "grad_norm": 1.258299653272843e-05,
      "learning_rate": 4.5243590682986223e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10521296,
      "step": 37575
    },
    {
      "epoch": 417.55555555555554,
      "grad_norm": 4.87497600261122e-05,
      "learning_rate": 4.5057849920891735e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10522736,
      "step": 37580
    },
    {
      "epoch": 417.6111111111111,
      "grad_norm": 1.826825609896332e-05,
      "learning_rate": 4.487248774172698e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10524128,
      "step": 37585
    },
    {
      "epoch": 417.6666666666667,
      "grad_norm": 3.1571318686474115e-05,
      "learning_rate": 4.4687504174077965e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10525536,
      "step": 37590
    },
    {
      "epoch": 417.72222222222223,
      "grad_norm": 0.0033412310294806957,
      "learning_rate": 4.450289924647133e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10526960,
      "step": 37595
    },
    {
      "epoch": 417.77777777777777,
      "grad_norm": 0.0001683498703641817,
      "learning_rate": 4.431867298737513e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10528352,
      "step": 37600
    },
    {
      "epoch": 417.77777777777777,
      "eval_loss": 0.8980686068534851,
      "eval_runtime": 1.2635,
      "eval_samples_per_second": 31.659,
      "eval_steps_per_second": 15.829,
      "num_input_tokens_seen": 10528352,
      "step": 37600
    },
    {
      "epoch": 417.8333333333333,
      "grad_norm": 0.0005799854989163578,
      "learning_rate": 4.41348254251997e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10529792,
      "step": 37605
    },
    {
      "epoch": 417.8888888888889,
      "grad_norm": 7.247907342389226e-05,
      "learning_rate": 4.395135658829652e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10531184,
      "step": 37610
    },
    {
      "epoch": 417.94444444444446,
      "grad_norm": 0.00034185044933110476,
      "learning_rate": 4.376826650495852e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10532624,
      "step": 37615
    },
    {
      "epoch": 418.0,
      "grad_norm": 3.173912045895122e-05,
      "learning_rate": 4.358555520342117e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10534016,
      "step": 37620
    },
    {
      "epoch": 418.05555555555554,
      "grad_norm": 0.00046705707791261375,
      "learning_rate": 4.3403222711860257e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10535456,
      "step": 37625
    },
    {
      "epoch": 418.1111111111111,
      "grad_norm": 5.237438017502427e-05,
      "learning_rate": 4.3221269058394133e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10536848,
      "step": 37630
    },
    {
      "epoch": 418.1666666666667,
      "grad_norm": 1.1152982551720925e-05,
      "learning_rate": 4.303969427108173e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10538320,
      "step": 37635
    },
    {
      "epoch": 418.22222222222223,
      "grad_norm": 0.006788759957998991,
      "learning_rate": 4.2858498377924825e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10539728,
      "step": 37640
    },
    {
      "epoch": 418.27777777777777,
      "grad_norm": 0.0009890858782455325,
      "learning_rate": 4.267768140686579e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10541104,
      "step": 37645
    },
    {
      "epoch": 418.3333333333333,
      "grad_norm": 4.930675640935078e-05,
      "learning_rate": 4.2497243385788975e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10542480,
      "step": 37650
    },
    {
      "epoch": 418.3888888888889,
      "grad_norm": 2.965684689115733e-05,
      "learning_rate": 4.231718434251991e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10543824,
      "step": 37655
    },
    {
      "epoch": 418.44444444444446,
      "grad_norm": 0.00013460857735481113,
      "learning_rate": 4.213750430482666e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10545200,
      "step": 37660
    },
    {
      "epoch": 418.5,
      "grad_norm": 8.40078282635659e-05,
      "learning_rate": 4.1958203300417054e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10546656,
      "step": 37665
    },
    {
      "epoch": 418.55555555555554,
      "grad_norm": 3.1032632250571623e-05,
      "learning_rate": 4.177928135694259e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10548048,
      "step": 37670
    },
    {
      "epoch": 418.6111111111111,
      "grad_norm": 0.001092295628041029,
      "learning_rate": 4.1600738501994807e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10549472,
      "step": 37675
    },
    {
      "epoch": 418.6666666666667,
      "grad_norm": 0.00011152176011819392,
      "learning_rate": 4.1422574763107237e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10550848,
      "step": 37680
    },
    {
      "epoch": 418.72222222222223,
      "grad_norm": 1.3031654816586524e-05,
      "learning_rate": 4.124479016775512e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10552224,
      "step": 37685
    },
    {
      "epoch": 418.77777777777777,
      "grad_norm": 1.4271677173383068e-05,
      "learning_rate": 4.106738474335514e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10553632,
      "step": 37690
    },
    {
      "epoch": 418.8333333333333,
      "grad_norm": 0.0003902766911778599,
      "learning_rate": 4.089035851726486e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10555008,
      "step": 37695
    },
    {
      "epoch": 418.8888888888889,
      "grad_norm": 0.0005470916512422264,
      "learning_rate": 4.0713711516784937e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10556368,
      "step": 37700
    },
    {
      "epoch": 418.94444444444446,
      "grad_norm": 4.912279837299138e-05,
      "learning_rate": 4.05374437691558e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10557776,
      "step": 37705
    },
    {
      "epoch": 419.0,
      "grad_norm": 3.025291152880527e-05,
      "learning_rate": 4.036155530156044e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10559152,
      "step": 37710
    },
    {
      "epoch": 419.05555555555554,
      "grad_norm": 0.0013419956667348742,
      "learning_rate": 4.018604614112298e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10560560,
      "step": 37715
    },
    {
      "epoch": 419.1111111111111,
      "grad_norm": 2.0911640604026616e-05,
      "learning_rate": 4.0010916314908996e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10562016,
      "step": 37720
    },
    {
      "epoch": 419.1666666666667,
      "grad_norm": 0.0007103204261511564,
      "learning_rate": 3.983616584992578e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10563424,
      "step": 37725
    },
    {
      "epoch": 419.22222222222223,
      "grad_norm": 0.007772424723953009,
      "learning_rate": 3.9661794773122595e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10564800,
      "step": 37730
    },
    {
      "epoch": 419.27777777777777,
      "grad_norm": 9.312957990914583e-05,
      "learning_rate": 3.9487803111388777e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10566192,
      "step": 37735
    },
    {
      "epoch": 419.3333333333333,
      "grad_norm": 5.004136255593039e-05,
      "learning_rate": 3.9314190891556747e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10567600,
      "step": 37740
    },
    {
      "epoch": 419.3888888888889,
      "grad_norm": 1.897176480269991e-05,
      "learning_rate": 3.914095814039925e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10568976,
      "step": 37745
    },
    {
      "epoch": 419.44444444444446,
      "grad_norm": 0.000240124121773988,
      "learning_rate": 3.896810488463104e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10570368,
      "step": 37750
    },
    {
      "epoch": 419.5,
      "grad_norm": 0.0011544987792149186,
      "learning_rate": 3.8795631150908565e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10571808,
      "step": 37755
    },
    {
      "epoch": 419.55555555555554,
      "grad_norm": 8.298635657411069e-05,
      "learning_rate": 3.862353696582888e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10573248,
      "step": 37760
    },
    {
      "epoch": 419.6111111111111,
      "grad_norm": 3.213441596017219e-05,
      "learning_rate": 3.8451822355931313e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10574624,
      "step": 37765
    },
    {
      "epoch": 419.6666666666667,
      "grad_norm": 0.0017300164327025414,
      "learning_rate": 3.82804873476969e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10576016,
      "step": 37770
    },
    {
      "epoch": 419.72222222222223,
      "grad_norm": 5.885109203518368e-05,
      "learning_rate": 3.810953196754702e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10577408,
      "step": 37775
    },
    {
      "epoch": 419.77777777777777,
      "grad_norm": 5.6842905905796215e-05,
      "learning_rate": 3.793895624184529e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10578768,
      "step": 37780
    },
    {
      "epoch": 419.8333333333333,
      "grad_norm": 0.00014738990284968168,
      "learning_rate": 3.776876019689679e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10580128,
      "step": 37785
    },
    {
      "epoch": 419.8888888888889,
      "grad_norm": 0.0008721620542928576,
      "learning_rate": 3.7598943858947743e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10581552,
      "step": 37790
    },
    {
      "epoch": 419.94444444444446,
      "grad_norm": 0.00022358304704539478,
      "learning_rate": 3.742950725418637e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10582960,
      "step": 37795
    },
    {
      "epoch": 420.0,
      "grad_norm": 0.0012029341887682676,
      "learning_rate": 3.726045040874093e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10584384,
      "step": 37800
    },
    {
      "epoch": 420.0,
      "eval_loss": 0.9222637414932251,
      "eval_runtime": 1.237,
      "eval_samples_per_second": 32.336,
      "eval_steps_per_second": 16.168,
      "num_input_tokens_seen": 10584384,
      "step": 37800
    },
    {
      "epoch": 420.05555555555554,
      "grad_norm": 0.0009463416645303369,
      "learning_rate": 3.709177334868308e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10585808,
      "step": 37805
    },
    {
      "epoch": 420.1111111111111,
      "grad_norm": 5.5706022976664826e-05,
      "learning_rate": 3.692347610002478e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10587216,
      "step": 37810
    },
    {
      "epoch": 420.1666666666667,
      "grad_norm": 2.437851799186319e-05,
      "learning_rate": 3.675555868871916e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10588656,
      "step": 37815
    },
    {
      "epoch": 420.22222222222223,
      "grad_norm": 0.000315415469231084,
      "learning_rate": 3.658802114066162e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10590032,
      "step": 37820
    },
    {
      "epoch": 420.27777777777777,
      "grad_norm": 0.0004107991117052734,
      "learning_rate": 3.6420863481688437e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10591424,
      "step": 37825
    },
    {
      "epoch": 420.3333333333333,
      "grad_norm": 0.00020875806512776762,
      "learning_rate": 3.625408573757705e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10592816,
      "step": 37830
    },
    {
      "epoch": 420.3888888888889,
      "grad_norm": 1.2660922948271036e-05,
      "learning_rate": 3.608768793404743e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10594272,
      "step": 37835
    },
    {
      "epoch": 420.44444444444446,
      "grad_norm": 9.059137664735317e-05,
      "learning_rate": 3.592167009675934e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10595648,
      "step": 37840
    },
    {
      "epoch": 420.5,
      "grad_norm": 0.00010792047396535054,
      "learning_rate": 3.575603225131563e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10597056,
      "step": 37845
    },
    {
      "epoch": 420.55555555555554,
      "grad_norm": 6.706933345412835e-05,
      "learning_rate": 3.55907744232592e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10598480,
      "step": 37850
    },
    {
      "epoch": 420.6111111111111,
      "grad_norm": 0.00021369181922636926,
      "learning_rate": 3.5425896638075217e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10599856,
      "step": 37855
    },
    {
      "epoch": 420.6666666666667,
      "grad_norm": 0.0004129761946387589,
      "learning_rate": 3.5261398921189736e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10601248,
      "step": 37860
    },
    {
      "epoch": 420.72222222222223,
      "grad_norm": 0.002767619676887989,
      "learning_rate": 3.509728129797024e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10602672,
      "step": 37865
    },
    {
      "epoch": 420.77777777777777,
      "grad_norm": 0.00017263491463381797,
      "learning_rate": 3.4933543793725656e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10604064,
      "step": 37870
    },
    {
      "epoch": 420.8333333333333,
      "grad_norm": 4.951716255163774e-05,
      "learning_rate": 3.4770186433707163e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10605440,
      "step": 37875
    },
    {
      "epoch": 420.8888888888889,
      "grad_norm": 8.993692608783022e-05,
      "learning_rate": 3.4607209243105453e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10606848,
      "step": 37880
    },
    {
      "epoch": 420.94444444444446,
      "grad_norm": 4.196747249807231e-05,
      "learning_rate": 3.444461224705431e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10608224,
      "step": 37885
    },
    {
      "epoch": 421.0,
      "grad_norm": 0.00027092074742540717,
      "learning_rate": 3.4282395470628116e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10609616,
      "step": 37890
    },
    {
      "epoch": 421.05555555555554,
      "grad_norm": 0.00010378468869021162,
      "learning_rate": 3.4120558938842417e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10611008,
      "step": 37895
    },
    {
      "epoch": 421.1111111111111,
      "grad_norm": 0.00039930225466378033,
      "learning_rate": 3.395910267665503e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10612400,
      "step": 37900
    },
    {
      "epoch": 421.1666666666667,
      "grad_norm": 0.00031153071904554963,
      "learning_rate": 3.3798026708964094e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10613792,
      "step": 37905
    },
    {
      "epoch": 421.22222222222223,
      "grad_norm": 2.3998718461371027e-05,
      "learning_rate": 3.3637331060609456e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10615200,
      "step": 37910
    },
    {
      "epoch": 421.27777777777777,
      "grad_norm": 4.542790338746272e-05,
      "learning_rate": 3.3477015756372966e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10616576,
      "step": 37915
    },
    {
      "epoch": 421.3333333333333,
      "grad_norm": 6.890566146466881e-05,
      "learning_rate": 3.3317080820976785e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10618000,
      "step": 37920
    },
    {
      "epoch": 421.3888888888889,
      "grad_norm": 5.3355895943241194e-05,
      "learning_rate": 3.315752627908508e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10619408,
      "step": 37925
    },
    {
      "epoch": 421.44444444444446,
      "grad_norm": 2.1220834241830744e-05,
      "learning_rate": 3.299835215530317e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10620800,
      "step": 37930
    },
    {
      "epoch": 421.5,
      "grad_norm": 5.491131742019206e-05,
      "learning_rate": 3.2839558474177245e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10622160,
      "step": 37935
    },
    {
      "epoch": 421.55555555555554,
      "grad_norm": 0.00019388904911465943,
      "learning_rate": 3.2681145260196056e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10623520,
      "step": 37940
    },
    {
      "epoch": 421.6111111111111,
      "grad_norm": 6.250177102629095e-05,
      "learning_rate": 3.252311253778839e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10624928,
      "step": 37945
    },
    {
      "epoch": 421.6666666666667,
      "grad_norm": 3.3254818845307454e-05,
      "learning_rate": 3.2365460331325034e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10626384,
      "step": 37950
    },
    {
      "epoch": 421.72222222222223,
      "grad_norm": 4.577215077006258e-05,
      "learning_rate": 3.2208188665117934e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10627776,
      "step": 37955
    },
    {
      "epoch": 421.77777777777777,
      "grad_norm": 0.00015536950377281755,
      "learning_rate": 3.205129756342018e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10629216,
      "step": 37960
    },
    {
      "epoch": 421.8333333333333,
      "grad_norm": 2.246474650746677e-05,
      "learning_rate": 3.189478705042659e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10630608,
      "step": 37965
    },
    {
      "epoch": 421.8888888888889,
      "grad_norm": 0.0026398806367069483,
      "learning_rate": 3.173865715027341e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10632000,
      "step": 37970
    },
    {
      "epoch": 421.94444444444446,
      "grad_norm": 3.4468281228328124e-05,
      "learning_rate": 3.158290788703694e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10633440,
      "step": 37975
    },
    {
      "epoch": 422.0,
      "grad_norm": 2.2103746232460253e-05,
      "learning_rate": 3.1427539284736297e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10634864,
      "step": 37980
    },
    {
      "epoch": 422.05555555555554,
      "grad_norm": 6.09341477684211e-05,
      "learning_rate": 3.127255136733093e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10636256,
      "step": 37985
    },
    {
      "epoch": 422.1111111111111,
      "grad_norm": 1.7228348951903172e-05,
      "learning_rate": 3.1117944158722544e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10637680,
      "step": 37990
    },
    {
      "epoch": 422.1666666666667,
      "grad_norm": 4.453952715266496e-05,
      "learning_rate": 3.0963717682752635e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10639104,
      "step": 37995
    },
    {
      "epoch": 422.22222222222223,
      "grad_norm": 1.7302114429185167e-05,
      "learning_rate": 3.080987196320578e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10640496,
      "step": 38000
    },
    {
      "epoch": 422.22222222222223,
      "eval_loss": 0.9098904728889465,
      "eval_runtime": 1.2457,
      "eval_samples_per_second": 32.11,
      "eval_steps_per_second": 16.055,
      "num_input_tokens_seen": 10640496,
      "step": 38000
    },
    {
      "epoch": 422.27777777777777,
      "grad_norm": 0.0002167279162677005,
      "learning_rate": 3.065640702380607e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10641936,
      "step": 38005
    },
    {
      "epoch": 422.3333333333333,
      "grad_norm": 0.00012152138515375555,
      "learning_rate": 3.050332288822011e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10643344,
      "step": 38010
    },
    {
      "epoch": 422.3888888888889,
      "grad_norm": 3.6806239222642034e-05,
      "learning_rate": 3.035061958005542e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10644768,
      "step": 38015
    },
    {
      "epoch": 422.44444444444446,
      "grad_norm": 4.7608718887204304e-05,
      "learning_rate": 3.019829712286093e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10646144,
      "step": 38020
    },
    {
      "epoch": 422.5,
      "grad_norm": 0.0006110900430940092,
      "learning_rate": 3.004635554012647e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10647536,
      "step": 38025
    },
    {
      "epoch": 422.55555555555554,
      "grad_norm": 0.00012584080104716122,
      "learning_rate": 2.9894794855283017e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10648896,
      "step": 38030
    },
    {
      "epoch": 422.6111111111111,
      "grad_norm": 3.179736086167395e-05,
      "learning_rate": 2.9743615091703816e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10650320,
      "step": 38035
    },
    {
      "epoch": 422.6666666666667,
      "grad_norm": 0.0004598747764248401,
      "learning_rate": 2.959281627270216e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10651760,
      "step": 38040
    },
    {
      "epoch": 422.72222222222223,
      "grad_norm": 0.00019938584591727704,
      "learning_rate": 2.944239842153362e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10653168,
      "step": 38045
    },
    {
      "epoch": 422.77777777777777,
      "grad_norm": 2.2091013306635432e-05,
      "learning_rate": 2.929236156139381e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10654592,
      "step": 38050
    },
    {
      "epoch": 422.8333333333333,
      "grad_norm": 6.442163430619985e-05,
      "learning_rate": 2.9142705715420883e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10655968,
      "step": 38055
    },
    {
      "epoch": 422.8888888888889,
      "grad_norm": 5.038083327235654e-05,
      "learning_rate": 2.8993430906693595e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10657344,
      "step": 38060
    },
    {
      "epoch": 422.94444444444446,
      "grad_norm": 0.00010343788017053157,
      "learning_rate": 2.88445371582316e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10658688,
      "step": 38065
    },
    {
      "epoch": 423.0,
      "grad_norm": 0.0009456478874199092,
      "learning_rate": 2.8696024492996796e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10660080,
      "step": 38070
    },
    {
      "epoch": 423.05555555555554,
      "grad_norm": 0.0016258868854492903,
      "learning_rate": 2.854789293389115e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10661472,
      "step": 38075
    },
    {
      "epoch": 423.1111111111111,
      "grad_norm": 0.000213996201637201,
      "learning_rate": 2.8400142503758606e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10662848,
      "step": 38080
    },
    {
      "epoch": 423.1666666666667,
      "grad_norm": 6.29016212769784e-05,
      "learning_rate": 2.8252773225384276e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10664256,
      "step": 38085
    },
    {
      "epoch": 423.22222222222223,
      "grad_norm": 0.00022202139371074736,
      "learning_rate": 2.8105785121494143e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10665728,
      "step": 38090
    },
    {
      "epoch": 423.27777777777777,
      "grad_norm": 3.253879549447447e-05,
      "learning_rate": 2.795917821475563e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10667120,
      "step": 38095
    },
    {
      "epoch": 423.3333333333333,
      "grad_norm": 3.374801963218488e-05,
      "learning_rate": 2.78129525277776e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10668512,
      "step": 38100
    },
    {
      "epoch": 423.3888888888889,
      "grad_norm": 1.0928527444775682e-05,
      "learning_rate": 2.766710808310952e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10669904,
      "step": 38105
    },
    {
      "epoch": 423.44444444444446,
      "grad_norm": 3.36599005095195e-05,
      "learning_rate": 2.7521644903242827e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10671328,
      "step": 38110
    },
    {
      "epoch": 423.5,
      "grad_norm": 0.00013913684233557433,
      "learning_rate": 2.7376563010609593e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10672752,
      "step": 38115
    },
    {
      "epoch": 423.55555555555554,
      "grad_norm": 0.001044287346303463,
      "learning_rate": 2.72318624275833e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10674144,
      "step": 38120
    },
    {
      "epoch": 423.6111111111111,
      "grad_norm": 0.00020957752713002264,
      "learning_rate": 2.7087543176478324e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10675520,
      "step": 38125
    },
    {
      "epoch": 423.6666666666667,
      "grad_norm": 8.812881424091756e-05,
      "learning_rate": 2.694360527955103e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10676912,
      "step": 38130
    },
    {
      "epoch": 423.72222222222223,
      "grad_norm": 7.951731095090508e-05,
      "learning_rate": 2.680004875899811e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10678320,
      "step": 38135
    },
    {
      "epoch": 423.77777777777777,
      "grad_norm": 0.00010966153786284849,
      "learning_rate": 2.665687363695768e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10679744,
      "step": 38140
    },
    {
      "epoch": 423.8333333333333,
      "grad_norm": 2.5589828510419466e-05,
      "learning_rate": 2.6514079935509584e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10681168,
      "step": 38145
    },
    {
      "epoch": 423.8888888888889,
      "grad_norm": 7.001830090302974e-05,
      "learning_rate": 2.6371667676673983e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10682576,
      "step": 38150
    },
    {
      "epoch": 423.94444444444446,
      "grad_norm": 1.3584826774604153e-05,
      "learning_rate": 2.6229636882412755e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10683952,
      "step": 38155
    },
    {
      "epoch": 424.0,
      "grad_norm": 5.968068307993235e-06,
      "learning_rate": 2.6087987574628935e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10685344,
      "step": 38160
    },
    {
      "epoch": 424.05555555555554,
      "grad_norm": 4.1305116610601544e-05,
      "learning_rate": 2.5946719775166437e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10686720,
      "step": 38165
    },
    {
      "epoch": 424.1111111111111,
      "grad_norm": 0.0002955904055852443,
      "learning_rate": 2.5805833505810616e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10688112,
      "step": 38170
    },
    {
      "epoch": 424.1666666666667,
      "grad_norm": 0.0003661791852209717,
      "learning_rate": 2.566532878828798e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10689552,
      "step": 38175
    },
    {
      "epoch": 424.22222222222223,
      "grad_norm": 2.033383134403266e-05,
      "learning_rate": 2.552520564426619e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10690928,
      "step": 38180
    },
    {
      "epoch": 424.27777777777777,
      "grad_norm": 1.6894111467991024e-05,
      "learning_rate": 2.5385464095353803e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10692336,
      "step": 38185
    },
    {
      "epoch": 424.3333333333333,
      "grad_norm": 4.404038190841675e-05,
      "learning_rate": 2.5246104163100804e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10693728,
      "step": 38190
    },
    {
      "epoch": 424.3888888888889,
      "grad_norm": 0.0001800139289116487,
      "learning_rate": 2.510712586899833e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10695104,
      "step": 38195
    },
    {
      "epoch": 424.44444444444446,
      "grad_norm": 0.000687855645082891,
      "learning_rate": 2.4968529234478124e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10696528,
      "step": 38200
    },
    {
      "epoch": 424.44444444444446,
      "eval_loss": 0.9754177331924438,
      "eval_runtime": 1.2481,
      "eval_samples_per_second": 32.048,
      "eval_steps_per_second": 16.024,
      "num_input_tokens_seen": 10696528,
      "step": 38200
    },
    {
      "epoch": 424.5,
      "grad_norm": 0.000557744933757931,
      "learning_rate": 2.483031428091448e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10697920,
      "step": 38205
    },
    {
      "epoch": 424.55555555555554,
      "grad_norm": 0.0006701360689476132,
      "learning_rate": 2.469248102962091e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10699392,
      "step": 38210
    },
    {
      "epoch": 424.6111111111111,
      "grad_norm": 3.899264629581012e-05,
      "learning_rate": 2.4555029501853455e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10700768,
      "step": 38215
    },
    {
      "epoch": 424.6666666666667,
      "grad_norm": 4.360627644928172e-05,
      "learning_rate": 2.441795971880906e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10702176,
      "step": 38220
    },
    {
      "epoch": 424.72222222222223,
      "grad_norm": 0.00017393880989402533,
      "learning_rate": 2.4281271701625255e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10703536,
      "step": 38225
    },
    {
      "epoch": 424.77777777777777,
      "grad_norm": 3.177636972395703e-05,
      "learning_rate": 2.4144965471381007e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10704928,
      "step": 38230
    },
    {
      "epoch": 424.8333333333333,
      "grad_norm": 0.00010520671639824286,
      "learning_rate": 2.400904104909674e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10706416,
      "step": 38235
    },
    {
      "epoch": 424.8888888888889,
      "grad_norm": 0.0012364210560917854,
      "learning_rate": 2.3873498455733725e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10707808,
      "step": 38240
    },
    {
      "epoch": 424.94444444444446,
      "grad_norm": 0.0006503243348561227,
      "learning_rate": 2.3738337712194137e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10709200,
      "step": 38245
    },
    {
      "epoch": 425.0,
      "grad_norm": 0.0002178973372792825,
      "learning_rate": 2.3603558839321305e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10710608,
      "step": 38250
    },
    {
      "epoch": 425.05555555555554,
      "grad_norm": 3.614370871218853e-05,
      "learning_rate": 2.3469161857900267e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10712016,
      "step": 38255
    },
    {
      "epoch": 425.1111111111111,
      "grad_norm": 4.749328945763409e-05,
      "learning_rate": 2.3335146788656393e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10713424,
      "step": 38260
    },
    {
      "epoch": 425.1666666666667,
      "grad_norm": 0.0011057084193453193,
      "learning_rate": 2.3201513652256757e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10714832,
      "step": 38265
    },
    {
      "epoch": 425.22222222222223,
      "grad_norm": 2.4697996195754968e-05,
      "learning_rate": 2.3068262469308766e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10716224,
      "step": 38270
    },
    {
      "epoch": 425.27777777777777,
      "grad_norm": 0.0001533287577331066,
      "learning_rate": 2.2935393260362093e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10717632,
      "step": 38275
    },
    {
      "epoch": 425.3333333333333,
      "grad_norm": 0.0022361513692885637,
      "learning_rate": 2.2802906045906458e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10718992,
      "step": 38280
    },
    {
      "epoch": 425.3888888888889,
      "grad_norm": 4.691627327702008e-05,
      "learning_rate": 2.2670800846373018e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10720416,
      "step": 38285
    },
    {
      "epoch": 425.44444444444446,
      "grad_norm": 0.0003190096467733383,
      "learning_rate": 2.2539077682134367e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10721808,
      "step": 38290
    },
    {
      "epoch": 425.5,
      "grad_norm": 0.00023795024026185274,
      "learning_rate": 2.2407736573503423e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10723248,
      "step": 38295
    },
    {
      "epoch": 425.55555555555554,
      "grad_norm": 2.4426957679679617e-05,
      "learning_rate": 2.2276777540735093e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10724656,
      "step": 38300
    },
    {
      "epoch": 425.6111111111111,
      "grad_norm": 1.8676368199521676e-05,
      "learning_rate": 2.2146200604024613e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10726096,
      "step": 38305
    },
    {
      "epoch": 425.6666666666667,
      "grad_norm": 0.00011380443174857646,
      "learning_rate": 2.2016005783508375e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10727536,
      "step": 38310
    },
    {
      "epoch": 425.72222222222223,
      "grad_norm": 0.000987204723060131,
      "learning_rate": 2.1886193099264763e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10728912,
      "step": 38315
    },
    {
      "epoch": 425.77777777777777,
      "grad_norm": 0.00020173334633000195,
      "learning_rate": 2.175676257131165e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10730256,
      "step": 38320
    },
    {
      "epoch": 425.8333333333333,
      "grad_norm": 0.0025582569651305676,
      "learning_rate": 2.162771421960974e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10731648,
      "step": 38325
    },
    {
      "epoch": 425.8888888888889,
      "grad_norm": 0.00010957834456348792,
      "learning_rate": 2.1499048064059224e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10733024,
      "step": 38330
    },
    {
      "epoch": 425.94444444444446,
      "grad_norm": 2.12638005905319e-05,
      "learning_rate": 2.1370764124502285e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10734416,
      "step": 38335
    },
    {
      "epoch": 426.0,
      "grad_norm": 0.0008949681068770587,
      "learning_rate": 2.1242862420721988e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10735840,
      "step": 38340
    },
    {
      "epoch": 426.05555555555554,
      "grad_norm": 0.0003217467456124723,
      "learning_rate": 2.1115342972442276e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10737200,
      "step": 38345
    },
    {
      "epoch": 426.1111111111111,
      "grad_norm": 0.00014899369853083044,
      "learning_rate": 2.0988205799328252e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10738608,
      "step": 38350
    },
    {
      "epoch": 426.1666666666667,
      "grad_norm": 0.0015850034542381763,
      "learning_rate": 2.0861450920986182e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10740032,
      "step": 38355
    },
    {
      "epoch": 426.22222222222223,
      "grad_norm": 3.13929449475836e-05,
      "learning_rate": 2.07350783569632e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10741408,
      "step": 38360
    },
    {
      "epoch": 426.27777777777777,
      "grad_norm": 5.2093222620896995e-05,
      "learning_rate": 2.060908812674761e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10742864,
      "step": 38365
    },
    {
      "epoch": 426.3333333333333,
      "grad_norm": 0.0009726042626425624,
      "learning_rate": 2.0483480249768317e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10744224,
      "step": 38370
    },
    {
      "epoch": 426.3888888888889,
      "grad_norm": 7.459349581040442e-05,
      "learning_rate": 2.035825474539621e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10745616,
      "step": 38375
    },
    {
      "epoch": 426.44444444444446,
      "grad_norm": 2.4847222448443063e-05,
      "learning_rate": 2.0233411632942235e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10747056,
      "step": 38380
    },
    {
      "epoch": 426.5,
      "grad_norm": 0.00020978867541998625,
      "learning_rate": 2.0108950931658764e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10748416,
      "step": 38385
    },
    {
      "epoch": 426.55555555555554,
      "grad_norm": 6.668042624369264e-05,
      "learning_rate": 1.998487266073934e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10749824,
      "step": 38390
    },
    {
      "epoch": 426.6111111111111,
      "grad_norm": 2.055562799796462e-05,
      "learning_rate": 1.986117683931865e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10751248,
      "step": 38395
    },
    {
      "epoch": 426.6666666666667,
      "grad_norm": 3.566519808373414e-05,
      "learning_rate": 1.9737863486471442e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10752640,
      "step": 38400
    },
    {
      "epoch": 426.6666666666667,
      "eval_loss": 0.9143226742744446,
      "eval_runtime": 1.2399,
      "eval_samples_per_second": 32.261,
      "eval_steps_per_second": 16.131,
      "num_input_tokens_seen": 10752640,
      "step": 38400
    },
    {
      "epoch": 426.72222222222223,
      "grad_norm": 0.0004006259550806135,
      "learning_rate": 1.9614932621215e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10754096,
      "step": 38405
    },
    {
      "epoch": 426.77777777777777,
      "grad_norm": 0.00018365787400398403,
      "learning_rate": 1.9492384262506102e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10755456,
      "step": 38410
    },
    {
      "epoch": 426.8333333333333,
      "grad_norm": 0.0004007855895906687,
      "learning_rate": 1.9370218429243524e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10756880,
      "step": 38415
    },
    {
      "epoch": 426.8888888888889,
      "grad_norm": 3.0947569030104205e-05,
      "learning_rate": 1.9248435140267197e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10758288,
      "step": 38420
    },
    {
      "epoch": 426.94444444444446,
      "grad_norm": 5.623493507300736e-06,
      "learning_rate": 1.9127034414356814e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10759696,
      "step": 38425
    },
    {
      "epoch": 427.0,
      "grad_norm": 6.366827437886968e-05,
      "learning_rate": 1.9006016270234627e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10761104,
      "step": 38430
    },
    {
      "epoch": 427.05555555555554,
      "grad_norm": 0.000138316405354999,
      "learning_rate": 1.888538072656293e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10762496,
      "step": 38435
    },
    {
      "epoch": 427.1111111111111,
      "grad_norm": 0.00034151901490986347,
      "learning_rate": 1.8765127801944893e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10763936,
      "step": 38440
    },
    {
      "epoch": 427.1666666666667,
      "grad_norm": 0.0002336371981073171,
      "learning_rate": 1.8645257514925406e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10765360,
      "step": 38445
    },
    {
      "epoch": 427.22222222222223,
      "grad_norm": 6.137421587482095e-05,
      "learning_rate": 1.8525769883989685e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10766752,
      "step": 38450
    },
    {
      "epoch": 427.27777777777777,
      "grad_norm": 0.0001393890706822276,
      "learning_rate": 1.8406664927564654e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10768176,
      "step": 38455
    },
    {
      "epoch": 427.3333333333333,
      "grad_norm": 0.000406396109610796,
      "learning_rate": 1.8287942664017566e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10769584,
      "step": 38460
    },
    {
      "epoch": 427.3888888888889,
      "grad_norm": 2.59096450463403e-05,
      "learning_rate": 1.8169603111656552e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10770976,
      "step": 38465
    },
    {
      "epoch": 427.44444444444446,
      "grad_norm": 9.43387858569622e-05,
      "learning_rate": 1.805164628873146e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10772336,
      "step": 38470
    },
    {
      "epoch": 427.5,
      "grad_norm": 0.0001578674273332581,
      "learning_rate": 1.793407221343274e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10773744,
      "step": 38475
    },
    {
      "epoch": 427.55555555555554,
      "grad_norm": 0.004389948211610317,
      "learning_rate": 1.781688090389172e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10775136,
      "step": 38480
    },
    {
      "epoch": 427.6111111111111,
      "grad_norm": 1.3068675798422191e-05,
      "learning_rate": 1.770007237818061e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10776512,
      "step": 38485
    },
    {
      "epoch": 427.6666666666667,
      "grad_norm": 0.0004188684979453683,
      "learning_rate": 1.7583646654313059e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10777888,
      "step": 38490
    },
    {
      "epoch": 427.72222222222223,
      "grad_norm": 4.062096559209749e-05,
      "learning_rate": 1.7467603750242757e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10779264,
      "step": 38495
    },
    {
      "epoch": 427.77777777777777,
      "grad_norm": 0.0039236717857420444,
      "learning_rate": 1.7351943683865944e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10780688,
      "step": 38500
    },
    {
      "epoch": 427.8333333333333,
      "grad_norm": 0.00010598233347991481,
      "learning_rate": 1.723666647301808e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10782112,
      "step": 38505
    },
    {
      "epoch": 427.8888888888889,
      "grad_norm": 6.160380144137889e-05,
      "learning_rate": 1.712177213547661e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10783472,
      "step": 38510
    },
    {
      "epoch": 427.94444444444446,
      "grad_norm": 0.00017213253886438906,
      "learning_rate": 1.7007260688959581e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10784896,
      "step": 38515
    },
    {
      "epoch": 428.0,
      "grad_norm": 0.00041129044257104397,
      "learning_rate": 1.68931321511262e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10786256,
      "step": 38520
    },
    {
      "epoch": 428.05555555555554,
      "grad_norm": 2.6838908524950966e-05,
      "learning_rate": 1.6779386539576835e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10787632,
      "step": 38525
    },
    {
      "epoch": 428.1111111111111,
      "grad_norm": 0.00010720748832682148,
      "learning_rate": 1.666602387185162e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10789072,
      "step": 38530
    },
    {
      "epoch": 428.1666666666667,
      "grad_norm": 0.00016433843120466918,
      "learning_rate": 1.655304416543352e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10790432,
      "step": 38535
    },
    {
      "epoch": 428.22222222222223,
      "grad_norm": 0.0001522581878816709,
      "learning_rate": 1.6440447437744698e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10791824,
      "step": 38540
    },
    {
      "epoch": 428.27777777777777,
      "grad_norm": 6.503782788058743e-05,
      "learning_rate": 1.6328233706149332e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10793248,
      "step": 38545
    },
    {
      "epoch": 428.3333333333333,
      "grad_norm": 0.002042145701125264,
      "learning_rate": 1.6216402987951906e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10794624,
      "step": 38550
    },
    {
      "epoch": 428.3888888888889,
      "grad_norm": 0.0009935525013133883,
      "learning_rate": 1.6104955300398627e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10796048,
      "step": 38555
    },
    {
      "epoch": 428.44444444444446,
      "grad_norm": 3.286356513854116e-05,
      "learning_rate": 1.5993890660675748e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10797424,
      "step": 38560
    },
    {
      "epoch": 428.5,
      "grad_norm": 0.0003598551847971976,
      "learning_rate": 1.5883209085910678e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10798816,
      "step": 38565
    },
    {
      "epoch": 428.55555555555554,
      "grad_norm": 4.7120716772042215e-05,
      "learning_rate": 1.5772910593172264e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10800224,
      "step": 38570
    },
    {
      "epoch": 428.6111111111111,
      "grad_norm": 0.00039077011751942337,
      "learning_rate": 1.5662995199469954e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10801616,
      "step": 38575
    },
    {
      "epoch": 428.6666666666667,
      "grad_norm": 4.405475192470476e-05,
      "learning_rate": 1.5553462921753802e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10803024,
      "step": 38580
    },
    {
      "epoch": 428.72222222222223,
      "grad_norm": 5.321991557138972e-05,
      "learning_rate": 1.544431377691502e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10804432,
      "step": 38585
    },
    {
      "epoch": 428.77777777777777,
      "grad_norm": 0.0012051782105118036,
      "learning_rate": 1.5335547781785975e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10805840,
      "step": 38590
    },
    {
      "epoch": 428.8333333333333,
      "grad_norm": 7.266960892593488e-05,
      "learning_rate": 1.5227164953139917e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10807264,
      "step": 38595
    },
    {
      "epoch": 428.8888888888889,
      "grad_norm": 0.00027919336571358144,
      "learning_rate": 1.511916530769042e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10808672,
      "step": 38600
    },
    {
      "epoch": 428.8888888888889,
      "eval_loss": 0.8928903341293335,
      "eval_runtime": 1.2674,
      "eval_samples_per_second": 31.561,
      "eval_steps_per_second": 15.781,
      "num_input_tokens_seen": 10808672,
      "step": 38600
    },
    {
      "epoch": 428.94444444444446,
      "grad_norm": 3.191328141838312e-05,
      "learning_rate": 1.5011548862092773e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10810048,
      "step": 38605
    },
    {
      "epoch": 429.0,
      "grad_norm": 2.3143074940890074e-05,
      "learning_rate": 1.490431563294231e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10811408,
      "step": 38610
    },
    {
      "epoch": 429.05555555555554,
      "grad_norm": 1.8714437828748487e-05,
      "learning_rate": 1.4797465636776365e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10812800,
      "step": 38615
    },
    {
      "epoch": 429.1111111111111,
      "grad_norm": 0.0005132727092131972,
      "learning_rate": 1.4690998890072027e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10814160,
      "step": 38620
    },
    {
      "epoch": 429.1666666666667,
      "grad_norm": 6.58379722153768e-05,
      "learning_rate": 1.4584915409248112e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10815568,
      "step": 38625
    },
    {
      "epoch": 429.22222222222223,
      "grad_norm": 0.00031387346098199487,
      "learning_rate": 1.4479215210663754e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10816976,
      "step": 38630
    },
    {
      "epoch": 429.27777777777777,
      "grad_norm": 5.2117091399850324e-05,
      "learning_rate": 1.4373898310619528e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10818352,
      "step": 38635
    },
    {
      "epoch": 429.3333333333333,
      "grad_norm": 1.7874157492769882e-05,
      "learning_rate": 1.4268964725356604e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10819712,
      "step": 38640
    },
    {
      "epoch": 429.3888888888889,
      "grad_norm": 4.0926068322733045e-05,
      "learning_rate": 1.4164414471056764e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10821168,
      "step": 38645
    },
    {
      "epoch": 429.44444444444446,
      "grad_norm": 0.0006752730114385486,
      "learning_rate": 1.4060247563843497e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10822560,
      "step": 38650
    },
    {
      "epoch": 429.5,
      "grad_norm": 3.247567656217143e-05,
      "learning_rate": 1.3956464019780068e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10823952,
      "step": 38655
    },
    {
      "epoch": 429.55555555555554,
      "grad_norm": 0.0008803570526652038,
      "learning_rate": 1.385306385487145e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10825376,
      "step": 38660
    },
    {
      "epoch": 429.6111111111111,
      "grad_norm": 0.00012843812874052674,
      "learning_rate": 1.3750047085063222e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10826784,
      "step": 38665
    },
    {
      "epoch": 429.6666666666667,
      "grad_norm": 0.00014753294817637652,
      "learning_rate": 1.3647413726242119e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10828208,
      "step": 38670
    },
    {
      "epoch": 429.72222222222223,
      "grad_norm": 0.00044331050594337285,
      "learning_rate": 1.3545163794235205e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10829616,
      "step": 38675
    },
    {
      "epoch": 429.77777777777777,
      "grad_norm": 0.00031602263334207237,
      "learning_rate": 1.3443297304810698e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10830976,
      "step": 38680
    },
    {
      "epoch": 429.8333333333333,
      "grad_norm": 4.61035524494946e-05,
      "learning_rate": 1.3341814273677977e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10832384,
      "step": 38685
    },
    {
      "epoch": 429.8888888888889,
      "grad_norm": 9.65313520282507e-05,
      "learning_rate": 1.324071471648647e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10833808,
      "step": 38690
    },
    {
      "epoch": 429.94444444444446,
      "grad_norm": 0.0003113678831141442,
      "learning_rate": 1.3139998648827312e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10835168,
      "step": 38695
    },
    {
      "epoch": 430.0,
      "grad_norm": 1.7027865396812558e-05,
      "learning_rate": 1.3039666086232526e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10836592,
      "step": 38700
    },
    {
      "epoch": 430.05555555555554,
      "grad_norm": 0.009428427554666996,
      "learning_rate": 1.2939717044174183e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10838032,
      "step": 38705
    },
    {
      "epoch": 430.1111111111111,
      "grad_norm": 2.4884082449716516e-05,
      "learning_rate": 1.284015153806578e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10839408,
      "step": 38710
    },
    {
      "epoch": 430.1666666666667,
      "grad_norm": 0.00012423945008777082,
      "learning_rate": 1.274096958326171e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10840768,
      "step": 38715
    },
    {
      "epoch": 430.22222222222223,
      "grad_norm": 3.4466051147319376e-05,
      "learning_rate": 1.2642171195056952e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10842176,
      "step": 38720
    },
    {
      "epoch": 430.27777777777777,
      "grad_norm": 0.00013256249076221138,
      "learning_rate": 1.2543756388687377e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10843552,
      "step": 38725
    },
    {
      "epoch": 430.3333333333333,
      "grad_norm": 9.137613960774615e-05,
      "learning_rate": 1.2445725179330014e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10844976,
      "step": 38730
    },
    {
      "epoch": 430.3888888888889,
      "grad_norm": 0.00014552066568285227,
      "learning_rate": 1.2348077582102212e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10846400,
      "step": 38735
    },
    {
      "epoch": 430.44444444444446,
      "grad_norm": 2.7584437702898867e-05,
      "learning_rate": 1.2250813612062762e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10847744,
      "step": 38740
    },
    {
      "epoch": 430.5,
      "grad_norm": 0.002561961067840457,
      "learning_rate": 1.215393328421105e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10849136,
      "step": 38745
    },
    {
      "epoch": 430.55555555555554,
      "grad_norm": 2.6792933567776345e-05,
      "learning_rate": 1.2057436613486796e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10850496,
      "step": 38750
    },
    {
      "epoch": 430.6111111111111,
      "grad_norm": 3.075748827541247e-05,
      "learning_rate": 1.1961323614771424e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10851920,
      "step": 38755
    },
    {
      "epoch": 430.6666666666667,
      "grad_norm": 0.00026477122446522117,
      "learning_rate": 1.1865594302886418e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10853328,
      "step": 38760
    },
    {
      "epoch": 430.72222222222223,
      "grad_norm": 8.081705163931474e-05,
      "learning_rate": 1.1770248692594687e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10854688,
      "step": 38765
    },
    {
      "epoch": 430.77777777777777,
      "grad_norm": 0.0011922511039301753,
      "learning_rate": 1.167528679859975e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10856112,
      "step": 38770
    },
    {
      "epoch": 430.8333333333333,
      "grad_norm": 0.0002091580827254802,
      "learning_rate": 1.1580708635545446e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10857552,
      "step": 38775
    },
    {
      "epoch": 430.8888888888889,
      "grad_norm": 4.0295275539392605e-05,
      "learning_rate": 1.1486514218017885e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10858976,
      "step": 38780
    },
    {
      "epoch": 430.94444444444446,
      "grad_norm": 0.00045565603068098426,
      "learning_rate": 1.1392703560542117e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10860352,
      "step": 38785
    },
    {
      "epoch": 431.0,
      "grad_norm": 9.408823098056018e-05,
      "learning_rate": 1.129927667758518e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10861760,
      "step": 38790
    },
    {
      "epoch": 431.05555555555554,
      "grad_norm": 8.154916577041149e-05,
      "learning_rate": 1.1206233583554992e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10863152,
      "step": 38795
    },
    {
      "epoch": 431.1111111111111,
      "grad_norm": 0.0008567755576223135,
      "learning_rate": 1.1113574292799523e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10864512,
      "step": 38800
    },
    {
      "epoch": 431.1111111111111,
      "eval_loss": 0.9007051587104797,
      "eval_runtime": 1.2342,
      "eval_samples_per_second": 32.41,
      "eval_steps_per_second": 16.205,
      "num_input_tokens_seen": 10864512,
      "step": 38800
    },
    {
      "epoch": 431.1666666666667,
      "grad_norm": 4.918787089991383e-05,
      "learning_rate": 1.1021298819608449e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10865968,
      "step": 38805
    },
    {
      "epoch": 431.22222222222223,
      "grad_norm": 8.10059136711061e-05,
      "learning_rate": 1.0929407178211226e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10867392,
      "step": 38810
    },
    {
      "epoch": 431.27777777777777,
      "grad_norm": 9.51734182308428e-05,
      "learning_rate": 1.0837899382779293e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10868768,
      "step": 38815
    },
    {
      "epoch": 431.3333333333333,
      "grad_norm": 3.0417084417422302e-05,
      "learning_rate": 1.0746775447423862e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10870176,
      "step": 38820
    },
    {
      "epoch": 431.3888888888889,
      "grad_norm": 0.0006017524865455925,
      "learning_rate": 1.0656035386197583e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10871536,
      "step": 38825
    },
    {
      "epoch": 431.44444444444446,
      "grad_norm": 2.4225069864769466e-05,
      "learning_rate": 1.0565679213093982e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10872880,
      "step": 38830
    },
    {
      "epoch": 431.5,
      "grad_norm": 6.359663530020043e-05,
      "learning_rate": 1.0475706942046638e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10874208,
      "step": 38835
    },
    {
      "epoch": 431.55555555555554,
      "grad_norm": 0.00010004473006119952,
      "learning_rate": 1.0386118586930282e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10875632,
      "step": 38840
    },
    {
      "epoch": 431.6111111111111,
      "grad_norm": 7.1510971793031786e-06,
      "learning_rate": 1.0296914161561367e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10877072,
      "step": 38845
    },
    {
      "epoch": 431.6666666666667,
      "grad_norm": 4.618897946784273e-05,
      "learning_rate": 1.0208093679695552e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10878432,
      "step": 38850
    },
    {
      "epoch": 431.72222222222223,
      "grad_norm": 0.00023813688312657177,
      "learning_rate": 1.0119657155030493e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10879776,
      "step": 38855
    },
    {
      "epoch": 431.77777777777777,
      "grad_norm": 8.421376696787775e-05,
      "learning_rate": 1.003160460120417e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 10881200,
      "step": 38860
    },
    {
      "epoch": 431.8333333333333,
      "grad_norm": 1.1401107258279808e-05,
      "learning_rate": 9.943936031795165e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10882624,
      "step": 38865
    },
    {
      "epoch": 431.8888888888889,
      "grad_norm": 0.0002414925693301484,
      "learning_rate": 9.856651460323219e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10884080,
      "step": 38870
    },
    {
      "epoch": 431.94444444444446,
      "grad_norm": 7.249905320350081e-05,
      "learning_rate": 9.769750900248953e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10885504,
      "step": 38875
    },
    {
      "epoch": 432.0,
      "grad_norm": 3.465071495156735e-05,
      "learning_rate": 9.683234364973038e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10886912,
      "step": 38880
    },
    {
      "epoch": 432.05555555555554,
      "grad_norm": 0.0020921954419463873,
      "learning_rate": 9.597101867837854e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10888304,
      "step": 38885
    },
    {
      "epoch": 432.1111111111111,
      "grad_norm": 0.00047418568283319473,
      "learning_rate": 9.511353422125835e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10889696,
      "step": 38890
    },
    {
      "epoch": 432.1666666666667,
      "grad_norm": 1.9859609892591834e-05,
      "learning_rate": 9.42598904106029e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10891072,
      "step": 38895
    },
    {
      "epoch": 432.22222222222223,
      "grad_norm": 8.280642941826954e-05,
      "learning_rate": 9.341008737806245e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10892480,
      "step": 38900
    },
    {
      "epoch": 432.27777777777777,
      "grad_norm": 5.339485142030753e-05,
      "learning_rate": 9.256412525467661e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10893888,
      "step": 38905
    },
    {
      "epoch": 432.3333333333333,
      "grad_norm": 0.00011790099233621731,
      "learning_rate": 9.172200417091326e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10895328,
      "step": 38910
    },
    {
      "epoch": 432.3888888888889,
      "grad_norm": 4.9553556891623884e-05,
      "learning_rate": 9.088372425663239e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10896736,
      "step": 38915
    },
    {
      "epoch": 432.44444444444446,
      "grad_norm": 1.5797440937603824e-05,
      "learning_rate": 9.004928564110837e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10898096,
      "step": 38920
    },
    {
      "epoch": 432.5,
      "grad_norm": 6.734178168699145e-05,
      "learning_rate": 8.92186884530244e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10899472,
      "step": 38925
    },
    {
      "epoch": 432.55555555555554,
      "grad_norm": 0.00016599969239905477,
      "learning_rate": 8.83919328204641e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10900864,
      "step": 38930
    },
    {
      "epoch": 432.6111111111111,
      "grad_norm": 1.1920299584744498e-05,
      "learning_rate": 8.756901887093105e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10902224,
      "step": 38935
    },
    {
      "epoch": 432.6666666666667,
      "grad_norm": 0.00020244474580977112,
      "learning_rate": 8.674994673132098e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10903616,
      "step": 38940
    },
    {
      "epoch": 432.72222222222223,
      "grad_norm": 0.00012465333566069603,
      "learning_rate": 8.593471652794949e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10905024,
      "step": 38945
    },
    {
      "epoch": 432.77777777777777,
      "grad_norm": 9.161544585367665e-05,
      "learning_rate": 8.512332838653548e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10906464,
      "step": 38950
    },
    {
      "epoch": 432.8333333333333,
      "grad_norm": 0.0012030956568196416,
      "learning_rate": 8.431578243220106e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10907856,
      "step": 38955
    },
    {
      "epoch": 432.8888888888889,
      "grad_norm": 0.0006631132564507425,
      "learning_rate": 8.351207878948552e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10909280,
      "step": 38960
    },
    {
      "epoch": 432.94444444444446,
      "grad_norm": 4.76955137855839e-05,
      "learning_rate": 8.271221758232583e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10910688,
      "step": 38965
    },
    {
      "epoch": 433.0,
      "grad_norm": 1.984108894248493e-05,
      "learning_rate": 8.191619893407332e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10912096,
      "step": 38970
    },
    {
      "epoch": 433.05555555555554,
      "grad_norm": 4.565375274978578e-05,
      "learning_rate": 8.112402296748534e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10913536,
      "step": 38975
    },
    {
      "epoch": 433.1111111111111,
      "grad_norm": 4.822477058041841e-05,
      "learning_rate": 8.033568980471973e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10914960,
      "step": 38980
    },
    {
      "epoch": 433.1666666666667,
      "grad_norm": 0.00048072662320919335,
      "learning_rate": 7.955119956735146e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10916352,
      "step": 38985
    },
    {
      "epoch": 433.22222222222223,
      "grad_norm": 8.747407991904765e-05,
      "learning_rate": 7.877055237636155e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10917776,
      "step": 38990
    },
    {
      "epoch": 433.27777777777777,
      "grad_norm": 4.287358024157584e-05,
      "learning_rate": 7.79937483521287e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10919200,
      "step": 38995
    },
    {
      "epoch": 433.3333333333333,
      "grad_norm": 1.9145743863191456e-05,
      "learning_rate": 7.722078761444873e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10920608,
      "step": 39000
    },
    {
      "epoch": 433.3333333333333,
      "eval_loss": 0.9196301698684692,
      "eval_runtime": 1.2411,
      "eval_samples_per_second": 32.23,
      "eval_steps_per_second": 16.115,
      "num_input_tokens_seen": 10920608,
      "step": 39000
    },
    {
      "epoch": 433.3888888888889,
      "grad_norm": 0.00014664549962617457,
      "learning_rate": 7.645167028252631e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10921968,
      "step": 39005
    },
    {
      "epoch": 433.44444444444446,
      "grad_norm": 6.619089253945276e-05,
      "learning_rate": 7.568639647496379e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10923360,
      "step": 39010
    },
    {
      "epoch": 433.5,
      "grad_norm": 0.0010888816323131323,
      "learning_rate": 7.492496630977508e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10924768,
      "step": 39015
    },
    {
      "epoch": 433.55555555555554,
      "grad_norm": 0.0011362639488652349,
      "learning_rate": 7.416737990438571e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10926160,
      "step": 39020
    },
    {
      "epoch": 433.6111111111111,
      "grad_norm": 3.1494888389715925e-05,
      "learning_rate": 7.341363737562445e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10927600,
      "step": 39025
    },
    {
      "epoch": 433.6666666666667,
      "grad_norm": 0.00034046670771203935,
      "learning_rate": 7.266373883972887e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10928992,
      "step": 39030
    },
    {
      "epoch": 433.72222222222223,
      "grad_norm": 2.2389222067431547e-05,
      "learning_rate": 7.191768441233981e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10930400,
      "step": 39035
    },
    {
      "epoch": 433.77777777777777,
      "grad_norm": 2.178894283133559e-05,
      "learning_rate": 7.11754742085069e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10931824,
      "step": 39040
    },
    {
      "epoch": 433.8333333333333,
      "grad_norm": 0.00010421498154755682,
      "learning_rate": 7.043710834269413e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10933232,
      "step": 39045
    },
    {
      "epoch": 433.8888888888889,
      "grad_norm": 0.0001330515369772911,
      "learning_rate": 6.970258692876319e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10934608,
      "step": 39050
    },
    {
      "epoch": 433.94444444444446,
      "grad_norm": 4.354095653980039e-05,
      "learning_rate": 6.897191007998738e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10935984,
      "step": 39055
    },
    {
      "epoch": 434.0,
      "grad_norm": 0.000766275217756629,
      "learning_rate": 6.824507790904599e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10937392,
      "step": 39060
    },
    {
      "epoch": 434.05555555555554,
      "grad_norm": 6.926076457602903e-05,
      "learning_rate": 6.752209052802439e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10938800,
      "step": 39065
    },
    {
      "epoch": 434.1111111111111,
      "grad_norm": 5.3422365454025567e-05,
      "learning_rate": 6.680294804841946e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10940160,
      "step": 39070
    },
    {
      "epoch": 434.1666666666667,
      "grad_norm": 4.101446393178776e-05,
      "learning_rate": 6.608765058112865e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10941552,
      "step": 39075
    },
    {
      "epoch": 434.22222222222223,
      "grad_norm": 0.00019625527784228325,
      "learning_rate": 6.537619823646368e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10942960,
      "step": 39080
    },
    {
      "epoch": 434.27777777777777,
      "grad_norm": 1.0816926078405231e-05,
      "learning_rate": 6.466859112413404e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10944416,
      "step": 39085
    },
    {
      "epoch": 434.3333333333333,
      "grad_norm": 6.954642594791949e-05,
      "learning_rate": 6.39648293532663e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10945808,
      "step": 39090
    },
    {
      "epoch": 434.3888888888889,
      "grad_norm": 3.4666634746827185e-05,
      "learning_rate": 6.32649130323848e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10947216,
      "step": 39095
    },
    {
      "epoch": 434.44444444444446,
      "grad_norm": 7.968006684677675e-05,
      "learning_rate": 6.256884226943094e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10948608,
      "step": 39100
    },
    {
      "epoch": 434.5,
      "grad_norm": 3.968253804487176e-05,
      "learning_rate": 6.187661717174386e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10950016,
      "step": 39105
    },
    {
      "epoch": 434.55555555555554,
      "grad_norm": 0.00015261372027453035,
      "learning_rate": 6.118823784607708e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10951360,
      "step": 39110
    },
    {
      "epoch": 434.6111111111111,
      "grad_norm": 0.0001156169455498457,
      "learning_rate": 6.050370439858178e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10952752,
      "step": 39115
    },
    {
      "epoch": 434.6666666666667,
      "grad_norm": 0.00024057697737589478,
      "learning_rate": 5.98230169348235e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10954144,
      "step": 39120
    },
    {
      "epoch": 434.72222222222223,
      "grad_norm": 3.68232540495228e-05,
      "learning_rate": 5.914617555977664e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10955520,
      "step": 39125
    },
    {
      "epoch": 434.77777777777777,
      "grad_norm": 0.00043279759120196104,
      "learning_rate": 5.8473180377816017e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10956912,
      "step": 39130
    },
    {
      "epoch": 434.8333333333333,
      "grad_norm": 4.033400909975171e-05,
      "learning_rate": 5.780403149272251e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10958352,
      "step": 39135
    },
    {
      "epoch": 434.8888888888889,
      "grad_norm": 2.1251285943435505e-05,
      "learning_rate": 5.7138729007694126e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10959744,
      "step": 39140
    },
    {
      "epoch": 434.94444444444446,
      "grad_norm": 5.834844705532305e-05,
      "learning_rate": 5.64772730253238e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10961200,
      "step": 39145
    },
    {
      "epoch": 435.0,
      "grad_norm": 1.0922580258920789e-05,
      "learning_rate": 5.5819663647618814e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10962608,
      "step": 39150
    },
    {
      "epoch": 435.05555555555554,
      "grad_norm": 7.302786252694204e-05,
      "learning_rate": 5.5165900975989723e-08,
      "loss": 0.0002,
      "num_input_tokens_seen": 10964000,
      "step": 39155
    },
    {
      "epoch": 435.1111111111111,
      "grad_norm": 0.0005250805406831205,
      "learning_rate": 5.451598511125311e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10965456,
      "step": 39160
    },
    {
      "epoch": 435.1666666666667,
      "grad_norm": 0.0003097826265729964,
      "learning_rate": 5.3869916153637124e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10966864,
      "step": 39165
    },
    {
      "epoch": 435.22222222222223,
      "grad_norm": 0.00020646839402616024,
      "learning_rate": 5.322769420277318e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10968272,
      "step": 39170
    },
    {
      "epoch": 435.27777777777777,
      "grad_norm": 0.00010823841876117513,
      "learning_rate": 5.258931935769873e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10969648,
      "step": 39175
    },
    {
      "epoch": 435.3333333333333,
      "grad_norm": 3.324795397929847e-05,
      "learning_rate": 5.19547917168628e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10970992,
      "step": 39180
    },
    {
      "epoch": 435.3888888888889,
      "grad_norm": 3.913236650987528e-05,
      "learning_rate": 5.13241113781121e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10972400,
      "step": 39185
    },
    {
      "epoch": 435.44444444444446,
      "grad_norm": 4.770653322339058e-05,
      "learning_rate": 5.0697278438707755e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10973824,
      "step": 39190
    },
    {
      "epoch": 435.5,
      "grad_norm": 9.153965947916731e-05,
      "learning_rate": 5.0074292995316854e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10975216,
      "step": 39195
    },
    {
      "epoch": 435.55555555555554,
      "grad_norm": 0.00010770757216960192,
      "learning_rate": 4.945515514400978e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10976624,
      "step": 39200
    },
    {
      "epoch": 435.55555555555554,
      "eval_loss": 0.907706618309021,
      "eval_runtime": 1.2474,
      "eval_samples_per_second": 32.066,
      "eval_steps_per_second": 16.033,
      "num_input_tokens_seen": 10976624,
      "step": 39200
    },
    {
      "epoch": 435.6111111111111,
      "grad_norm": 3.592646316974424e-05,
      "learning_rate": 4.883986498026571e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10977936,
      "step": 39205
    },
    {
      "epoch": 435.6666666666667,
      "grad_norm": 7.807107613189146e-05,
      "learning_rate": 4.822842259896987e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10979312,
      "step": 39210
    },
    {
      "epoch": 435.72222222222223,
      "grad_norm": 0.000790843041613698,
      "learning_rate": 4.762082809441626e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10980704,
      "step": 39215
    },
    {
      "epoch": 435.77777777777777,
      "grad_norm": 3.390727215446532e-05,
      "learning_rate": 4.7017081560302156e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10982064,
      "step": 39220
    },
    {
      "epoch": 435.8333333333333,
      "grad_norm": 1.729592986521311e-05,
      "learning_rate": 4.6417183089730866e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10983520,
      "step": 39225
    },
    {
      "epoch": 435.8888888888889,
      "grad_norm": 4.753356915898621e-05,
      "learning_rate": 4.5821132775217265e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10984960,
      "step": 39230
    },
    {
      "epoch": 435.94444444444446,
      "grad_norm": 2.456351649016142e-05,
      "learning_rate": 4.5228930708679504e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10986368,
      "step": 39235
    },
    {
      "epoch": 436.0,
      "grad_norm": 0.0012001299764961004,
      "learning_rate": 4.464057698144175e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10987808,
      "step": 39240
    },
    {
      "epoch": 436.05555555555554,
      "grad_norm": 0.00012307989527471364,
      "learning_rate": 4.4056071684236974e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10989216,
      "step": 39245
    },
    {
      "epoch": 436.1111111111111,
      "grad_norm": 1.370244899590034e-05,
      "learning_rate": 4.347541490719864e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10990608,
      "step": 39250
    },
    {
      "epoch": 436.1666666666667,
      "grad_norm": 0.0011488660238683224,
      "learning_rate": 4.2898606739877336e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10992048,
      "step": 39255
    },
    {
      "epoch": 436.22222222222223,
      "grad_norm": 0.00015285908011719584,
      "learning_rate": 4.232564727122135e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10993424,
      "step": 39260
    },
    {
      "epoch": 436.27777777777777,
      "grad_norm": 0.0010775894625112414,
      "learning_rate": 4.1756536589585004e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10994800,
      "step": 39265
    },
    {
      "epoch": 436.3333333333333,
      "grad_norm": 3.892730819643475e-05,
      "learning_rate": 4.119127478273976e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10996240,
      "step": 39270
    },
    {
      "epoch": 436.3888888888889,
      "grad_norm": 4.6912333345972e-05,
      "learning_rate": 4.062986193784923e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10997616,
      "step": 39275
    },
    {
      "epoch": 436.44444444444446,
      "grad_norm": 7.591420580865815e-05,
      "learning_rate": 4.007229814149416e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 10998960,
      "step": 39280
    },
    {
      "epoch": 436.5,
      "grad_norm": 9.680656512500718e-05,
      "learning_rate": 3.951858347965576e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11000336,
      "step": 39285
    },
    {
      "epoch": 436.55555555555554,
      "grad_norm": 0.00015262416854966432,
      "learning_rate": 3.896871803772684e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11001744,
      "step": 39290
    },
    {
      "epoch": 436.6111111111111,
      "grad_norm": 0.0002108633198076859,
      "learning_rate": 3.842270190050068e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11003136,
      "step": 39295
    },
    {
      "epoch": 436.6666666666667,
      "grad_norm": 1.742511085467413e-05,
      "learning_rate": 3.7880535152179376e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11004512,
      "step": 39300
    },
    {
      "epoch": 436.72222222222223,
      "grad_norm": 3.7910143873887137e-05,
      "learning_rate": 3.734221787637382e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11005888,
      "step": 39305
    },
    {
      "epoch": 436.77777777777777,
      "grad_norm": 3.858908894471824e-05,
      "learning_rate": 3.680775015609817e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11007280,
      "step": 39310
    },
    {
      "epoch": 436.8333333333333,
      "grad_norm": 0.0008771634893491864,
      "learning_rate": 3.627713207377537e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11008704,
      "step": 39315
    },
    {
      "epoch": 436.8888888888889,
      "grad_norm": 0.00010729900532169268,
      "learning_rate": 3.575036371123164e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11010112,
      "step": 39320
    },
    {
      "epoch": 436.94444444444446,
      "grad_norm": 0.00012031920778099447,
      "learning_rate": 3.5227445149704776e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11011552,
      "step": 39325
    },
    {
      "epoch": 437.0,
      "grad_norm": 1.6485360902152024e-05,
      "learning_rate": 3.470837646983027e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11012960,
      "step": 39330
    },
    {
      "epoch": 437.05555555555554,
      "grad_norm": 0.000168670027051121,
      "learning_rate": 3.419315775165799e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11014352,
      "step": 39335
    },
    {
      "epoch": 437.1111111111111,
      "grad_norm": 4.968360372004099e-05,
      "learning_rate": 3.368178907464103e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11015728,
      "step": 39340
    },
    {
      "epoch": 437.1666666666667,
      "grad_norm": 0.00028681065305136144,
      "learning_rate": 3.317427051763855e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11017184,
      "step": 39345
    },
    {
      "epoch": 437.22222222222223,
      "grad_norm": 1.7763106370694004e-05,
      "learning_rate": 3.267060215891571e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11018544,
      "step": 39350
    },
    {
      "epoch": 437.27777777777777,
      "grad_norm": 9.394308290211484e-05,
      "learning_rate": 3.217078407614649e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11019936,
      "step": 39355
    },
    {
      "epoch": 437.3333333333333,
      "grad_norm": 0.0001421880879206583,
      "learning_rate": 3.1674816346405345e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11021376,
      "step": 39360
    },
    {
      "epoch": 437.3888888888889,
      "grad_norm": 2.3602531655342318e-05,
      "learning_rate": 3.11826990461811e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11022720,
      "step": 39365
    },
    {
      "epoch": 437.44444444444446,
      "grad_norm": 0.00035053968895226717,
      "learning_rate": 3.069443225136304e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11024128,
      "step": 39370
    },
    {
      "epoch": 437.5,
      "grad_norm": 3.2590414775768295e-05,
      "learning_rate": 3.021001603724372e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11025504,
      "step": 39375
    },
    {
      "epoch": 437.55555555555554,
      "grad_norm": 0.0029606977477669716,
      "learning_rate": 2.9729450478532818e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11026928,
      "step": 39380
    },
    {
      "epoch": 437.6111111111111,
      "grad_norm": 1.8129247109754942e-05,
      "learning_rate": 2.9252735649337726e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11028368,
      "step": 39385
    },
    {
      "epoch": 437.6666666666667,
      "grad_norm": 9.273592149838805e-05,
      "learning_rate": 2.8779871623171863e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11029776,
      "step": 39390
    },
    {
      "epoch": 437.72222222222223,
      "grad_norm": 0.00011964162695221603,
      "learning_rate": 2.8310858472957448e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11031168,
      "step": 39395
    },
    {
      "epoch": 437.77777777777777,
      "grad_norm": 2.7599968234426342e-05,
      "learning_rate": 2.784569627101996e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11032608,
      "step": 39400
    },
    {
      "epoch": 437.77777777777777,
      "eval_loss": 0.907706618309021,
      "eval_runtime": 1.2389,
      "eval_samples_per_second": 32.286,
      "eval_steps_per_second": 16.143,
      "num_input_tokens_seen": 11032608,
      "step": 39400
    },
    {
      "epoch": 437.8333333333333,
      "grad_norm": 0.00011216085113119334,
      "learning_rate": 2.738438508909924e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11034032,
      "step": 39405
    },
    {
      "epoch": 437.8888888888889,
      "grad_norm": 6.145250517874956e-05,
      "learning_rate": 2.692692499833005e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11035424,
      "step": 39410
    },
    {
      "epoch": 437.94444444444446,
      "grad_norm": 2.598065293568652e-05,
      "learning_rate": 2.647331606926151e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11036832,
      "step": 39415
    },
    {
      "epoch": 438.0,
      "grad_norm": 0.00010689787450246513,
      "learning_rate": 2.6023558371843225e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11038224,
      "step": 39420
    },
    {
      "epoch": 438.05555555555554,
      "grad_norm": 3.303512494312599e-05,
      "learning_rate": 2.557765197543638e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11039664,
      "step": 39425
    },
    {
      "epoch": 438.1111111111111,
      "grad_norm": 0.00012678820348810405,
      "learning_rate": 2.513559694880263e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11041104,
      "step": 39430
    },
    {
      "epoch": 438.1666666666667,
      "grad_norm": 1.0310171091987286e-05,
      "learning_rate": 2.469739336011523e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11042512,
      "step": 39435
    },
    {
      "epoch": 438.22222222222223,
      "grad_norm": 7.991270831553265e-05,
      "learning_rate": 2.4263041276947894e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11043920,
      "step": 39440
    },
    {
      "epoch": 438.27777777777777,
      "grad_norm": 0.00017813252634368837,
      "learning_rate": 2.3832540766283164e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11045344,
      "step": 39445
    },
    {
      "epoch": 438.3333333333333,
      "grad_norm": 6.330938049359247e-05,
      "learning_rate": 2.3405891894512366e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11046784,
      "step": 39450
    },
    {
      "epoch": 438.3888888888889,
      "grad_norm": 0.0007727244519628584,
      "learning_rate": 2.29830947274301e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11048208,
      "step": 39455
    },
    {
      "epoch": 438.44444444444446,
      "grad_norm": 0.00030575503478758037,
      "learning_rate": 2.2564149330231432e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11049584,
      "step": 39460
    },
    {
      "epoch": 438.5,
      "grad_norm": 0.002053870353847742,
      "learning_rate": 2.2149055767528572e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11050992,
      "step": 39465
    },
    {
      "epoch": 438.55555555555554,
      "grad_norm": 0.0001629105390748009,
      "learning_rate": 2.1737814103334197e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11052368,
      "step": 39470
    },
    {
      "epoch": 438.6111111111111,
      "grad_norm": 0.00012975563004147261,
      "learning_rate": 2.1330424401064253e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11053760,
      "step": 39475
    },
    {
      "epoch": 438.6666666666667,
      "grad_norm": 1.737856109684799e-05,
      "learning_rate": 2.092688672354348e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11055152,
      "step": 39480
    },
    {
      "epoch": 438.72222222222223,
      "grad_norm": 2.0426714399945922e-05,
      "learning_rate": 2.0527201133005435e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11056528,
      "step": 39485
    },
    {
      "epoch": 438.77777777777777,
      "grad_norm": 0.00012958522711414844,
      "learning_rate": 2.0131367691084148e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11057920,
      "step": 39490
    },
    {
      "epoch": 438.8333333333333,
      "grad_norm": 1.2657300430873875e-05,
      "learning_rate": 1.9739386458819675e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11059344,
      "step": 39495
    },
    {
      "epoch": 438.8888888888889,
      "grad_norm": 0.0002336602919967845,
      "learning_rate": 1.9351257496666442e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11060736,
      "step": 39500
    },
    {
      "epoch": 438.94444444444446,
      "grad_norm": 2.7335867343936116e-05,
      "learning_rate": 1.896698086447657e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11062128,
      "step": 39505
    },
    {
      "epoch": 439.0,
      "grad_norm": 0.00011742032074835151,
      "learning_rate": 1.8586556621505436e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11063488,
      "step": 39510
    },
    {
      "epoch": 439.05555555555554,
      "grad_norm": 4.278277629055083e-05,
      "learning_rate": 1.820998482642833e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11064864,
      "step": 39515
    },
    {
      "epoch": 439.1111111111111,
      "grad_norm": 6.845409370725974e-05,
      "learning_rate": 1.7837265537309912e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11066288,
      "step": 39520
    },
    {
      "epoch": 439.1666666666667,
      "grad_norm": 0.00019736694230232388,
      "learning_rate": 1.7468398811629206e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11067664,
      "step": 39525
    },
    {
      "epoch": 439.22222222222223,
      "grad_norm": 0.00011343092046445236,
      "learning_rate": 1.710338470627404e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11069056,
      "step": 39530
    },
    {
      "epoch": 439.27777777777777,
      "grad_norm": 0.0001094950785045512,
      "learning_rate": 1.6742223277529945e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11070480,
      "step": 39535
    },
    {
      "epoch": 439.3333333333333,
      "grad_norm": 0.00020848019630648196,
      "learning_rate": 1.6384914581094036e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11071824,
      "step": 39540
    },
    {
      "epoch": 439.3888888888889,
      "grad_norm": 0.00026313416310586035,
      "learning_rate": 1.6031458672069455e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11073200,
      "step": 39545
    },
    {
      "epoch": 439.44444444444446,
      "grad_norm": 3.942142939195037e-05,
      "learning_rate": 1.5681855604962602e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11074576,
      "step": 39550
    },
    {
      "epoch": 439.5,
      "grad_norm": 1.7286338334088214e-05,
      "learning_rate": 1.5336105433683135e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11076032,
      "step": 39555
    },
    {
      "epoch": 439.55555555555554,
      "grad_norm": 2.880411375372205e-05,
      "learning_rate": 1.499420821155506e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11077408,
      "step": 39560
    },
    {
      "epoch": 439.6111111111111,
      "grad_norm": 0.002062394516542554,
      "learning_rate": 1.4656163991302874e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11078800,
      "step": 39565
    },
    {
      "epoch": 439.6666666666667,
      "grad_norm": 3.387023389223032e-05,
      "learning_rate": 1.4321972825051544e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11080224,
      "step": 39570
    },
    {
      "epoch": 439.72222222222223,
      "grad_norm": 0.0012313618790358305,
      "learning_rate": 1.3991634764345951e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11081664,
      "step": 39575
    },
    {
      "epoch": 439.77777777777777,
      "grad_norm": 3.483521504676901e-05,
      "learning_rate": 1.3665149860120352e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11083104,
      "step": 39580
    },
    {
      "epoch": 439.8333333333333,
      "grad_norm": 0.0002588585193734616,
      "learning_rate": 1.3342518162728912e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11084512,
      "step": 39585
    },
    {
      "epoch": 439.8888888888889,
      "grad_norm": 5.313479050528258e-05,
      "learning_rate": 1.30237397219235e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11085888,
      "step": 39590
    },
    {
      "epoch": 439.94444444444446,
      "grad_norm": 0.00024037282855715603,
      "learning_rate": 1.2708814586862016e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11087296,
      "step": 39595
    },
    {
      "epoch": 440.0,
      "grad_norm": 6.862504233140498e-05,
      "learning_rate": 1.2397742806111168e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11088720,
      "step": 39600
    },
    {
      "epoch": 440.0,
      "eval_loss": 0.907706618309021,
      "eval_runtime": 1.2346,
      "eval_samples_per_second": 32.4,
      "eval_steps_per_second": 16.2,
      "num_input_tokens_seen": 11088720,
      "step": 39600
    },
    {
      "epoch": 440.05555555555554,
      "grad_norm": 3.406175528652966e-05,
      "learning_rate": 1.209052442764369e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11090144,
      "step": 39605
    },
    {
      "epoch": 440.1111111111111,
      "grad_norm": 1.4704497516504489e-05,
      "learning_rate": 1.17871594988328e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11091552,
      "step": 39610
    },
    {
      "epoch": 440.1666666666667,
      "grad_norm": 6.019916781951906e-06,
      "learning_rate": 1.1487648066466072e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11092912,
      "step": 39615
    },
    {
      "epoch": 440.22222222222223,
      "grad_norm": 0.0001840910699684173,
      "learning_rate": 1.1191990176728784e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11094320,
      "step": 39620
    },
    {
      "epoch": 440.27777777777777,
      "grad_norm": 1.968267315533012e-05,
      "learning_rate": 1.0900185875215018e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11095696,
      "step": 39625
    },
    {
      "epoch": 440.3333333333333,
      "grad_norm": 0.002931024180725217,
      "learning_rate": 1.0612235206924891e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11097088,
      "step": 39630
    },
    {
      "epoch": 440.3888888888889,
      "grad_norm": 0.00025293894577771425,
      "learning_rate": 1.0328138216264549e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11098544,
      "step": 39635
    },
    {
      "epoch": 440.44444444444446,
      "grad_norm": 0.00015389877080451697,
      "learning_rate": 1.004789494704339e-08,
      "loss": 0.0,
      "num_input_tokens_seen": 11099952,
      "step": 39640
    },
    {
      "epoch": 440.5,
      "grad_norm": 6.240647780941799e-05,
      "learning_rate": 9.771505442482397e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11101376,
      "step": 39645
    },
    {
      "epoch": 440.55555555555554,
      "grad_norm": 0.00010550618753768504,
      "learning_rate": 9.498969745200259e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11102752,
      "step": 39650
    },
    {
      "epoch": 440.6111111111111,
      "grad_norm": 0.00026632912340573967,
      "learning_rate": 9.230287897230017e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11104128,
      "step": 39655
    },
    {
      "epoch": 440.6666666666667,
      "grad_norm": 0.0010806320933625102,
      "learning_rate": 8.965459940002419e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11105504,
      "step": 39660
    },
    {
      "epoch": 440.72222222222223,
      "grad_norm": 4.9677593779051676e-05,
      "learning_rate": 8.704485914357019e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11106928,
      "step": 39665
    },
    {
      "epoch": 440.77777777777777,
      "grad_norm": 0.0009064407786354423,
      "learning_rate": 8.447365860539402e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11108304,
      "step": 39670
    },
    {
      "epoch": 440.8333333333333,
      "grad_norm": 0.00035869303974322975,
      "learning_rate": 8.194099818201184e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11109696,
      "step": 39675
    },
    {
      "epoch": 440.8888888888889,
      "grad_norm": 1.7640502846916206e-05,
      "learning_rate": 7.944687826400011e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11111104,
      "step": 39680
    },
    {
      "epoch": 440.94444444444446,
      "grad_norm": 0.0002351640723645687,
      "learning_rate": 7.699129923599557e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11112496,
      "step": 39685
    },
    {
      "epoch": 441.0,
      "grad_norm": 5.408766810433008e-05,
      "learning_rate": 7.457426147663982e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11113888,
      "step": 39690
    },
    {
      "epoch": 441.05555555555554,
      "grad_norm": 0.00012875099491793662,
      "learning_rate": 7.219576535871797e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11115264,
      "step": 39695
    },
    {
      "epoch": 441.1111111111111,
      "grad_norm": 7.125645788619295e-05,
      "learning_rate": 6.985581124896445e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11116672,
      "step": 39700
    },
    {
      "epoch": 441.1666666666667,
      "grad_norm": 4.142922261962667e-05,
      "learning_rate": 6.755439950828501e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11118096,
      "step": 39705
    },
    {
      "epoch": 441.22222222222223,
      "grad_norm": 5.225919449003413e-05,
      "learning_rate": 6.5291530491562444e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11119520,
      "step": 39710
    },
    {
      "epoch": 441.27777777777777,
      "grad_norm": 0.00018761996761895716,
      "learning_rate": 6.3067204547739845e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11120912,
      "step": 39715
    },
    {
      "epoch": 441.3333333333333,
      "grad_norm": 4.673034345614724e-05,
      "learning_rate": 6.088142201987612e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11122320,
      "step": 39720
    },
    {
      "epoch": 441.3888888888889,
      "grad_norm": 1.833057831390761e-05,
      "learning_rate": 5.873418324503499e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11123728,
      "step": 39725
    },
    {
      "epoch": 441.44444444444446,
      "grad_norm": 7.636808004463091e-05,
      "learning_rate": 5.6625488554340465e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11125168,
      "step": 39730
    },
    {
      "epoch": 441.5,
      "grad_norm": 0.00020893392502330244,
      "learning_rate": 5.455533827297688e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11126576,
      "step": 39735
    },
    {
      "epoch": 441.55555555555554,
      "grad_norm": 0.000676287745591253,
      "learning_rate": 5.252373272018885e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11127952,
      "step": 39740
    },
    {
      "epoch": 441.6111111111111,
      "grad_norm": 3.578179530450143e-05,
      "learning_rate": 5.053067220925356e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11129360,
      "step": 39745
    },
    {
      "epoch": 441.6666666666667,
      "grad_norm": 0.0010837690206244588,
      "learning_rate": 4.857615704759177e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11130736,
      "step": 39750
    },
    {
      "epoch": 441.72222222222223,
      "grad_norm": 7.58307141950354e-05,
      "learning_rate": 4.666018753654577e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11132144,
      "step": 39755
    },
    {
      "epoch": 441.77777777777777,
      "grad_norm": 9.674943430582061e-05,
      "learning_rate": 4.478276397162917e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11133552,
      "step": 39760
    },
    {
      "epoch": 441.8333333333333,
      "grad_norm": 0.00025672389892861247,
      "learning_rate": 4.294388664233262e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11134912,
      "step": 39765
    },
    {
      "epoch": 441.8888888888889,
      "grad_norm": 7.163186091929674e-05,
      "learning_rate": 4.114355583223484e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11136336,
      "step": 39770
    },
    {
      "epoch": 441.94444444444446,
      "grad_norm": 0.00017971267516259104,
      "learning_rate": 3.9381771818974845e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11137744,
      "step": 39775
    },
    {
      "epoch": 442.0,
      "grad_norm": 2.0016748749185354e-05,
      "learning_rate": 3.765853487427973e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11139120,
      "step": 39780
    },
    {
      "epoch": 442.05555555555554,
      "grad_norm": 7.577335054520518e-05,
      "learning_rate": 3.5973845263825857e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11140544,
      "step": 39785
    },
    {
      "epoch": 442.1111111111111,
      "grad_norm": 0.0005931092309765518,
      "learning_rate": 3.4327703247488684e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11141920,
      "step": 39790
    },
    {
      "epoch": 442.1666666666667,
      "grad_norm": 0.000154532230226323,
      "learning_rate": 3.2720109079037443e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11143312,
      "step": 39795
    },
    {
      "epoch": 442.22222222222223,
      "grad_norm": 3.045353150810115e-05,
      "learning_rate": 3.1151063006468193e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11144688,
      "step": 39800
    },
    {
      "epoch": 442.22222222222223,
      "eval_loss": 0.907706618309021,
      "eval_runtime": 1.2319,
      "eval_samples_per_second": 32.471,
      "eval_steps_per_second": 16.236,
      "num_input_tokens_seen": 11144688,
      "step": 39800
    },
    {
      "epoch": 442.27777777777777,
      "grad_norm": 0.00019136510672979057,
      "learning_rate": 2.962056527169854e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11146048,
      "step": 39805
    },
    {
      "epoch": 442.3333333333333,
      "grad_norm": 7.562056271126494e-05,
      "learning_rate": 2.8128616110761898e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11147488,
      "step": 39810
    },
    {
      "epoch": 442.3888888888889,
      "grad_norm": 8.436920325038955e-05,
      "learning_rate": 2.6675215753724223e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11148912,
      "step": 39815
    },
    {
      "epoch": 442.44444444444446,
      "grad_norm": 8.886105933925137e-05,
      "learning_rate": 2.5260364424739557e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11150288,
      "step": 39820
    },
    {
      "epoch": 442.5,
      "grad_norm": 0.00019241380505263805,
      "learning_rate": 2.3884062341994475e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11151664,
      "step": 39825
    },
    {
      "epoch": 442.55555555555554,
      "grad_norm": 0.0001467284164391458,
      "learning_rate": 2.25463097177081e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11152992,
      "step": 39830
    },
    {
      "epoch": 442.6111111111111,
      "grad_norm": 0.0005767904222011566,
      "learning_rate": 2.1247106758215397e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11154416,
      "step": 39835
    },
    {
      "epoch": 442.6666666666667,
      "grad_norm": 1.4416797057492658e-05,
      "learning_rate": 1.998645366382834e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11155824,
      "step": 39840
    },
    {
      "epoch": 442.72222222222223,
      "grad_norm": 0.00010341243614675477,
      "learning_rate": 1.876435062897475e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11157216,
      "step": 39845
    },
    {
      "epoch": 442.77777777777777,
      "grad_norm": 3.6215398722561076e-05,
      "learning_rate": 1.758079784211497e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11158624,
      "step": 39850
    },
    {
      "epoch": 442.8333333333333,
      "grad_norm": 0.001542398938909173,
      "learning_rate": 1.6435795485797434e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11160000,
      "step": 39855
    },
    {
      "epoch": 442.8888888888889,
      "grad_norm": 0.00030676371534354985,
      "learning_rate": 1.5329343736547596e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11161456,
      "step": 39860
    },
    {
      "epoch": 442.94444444444446,
      "grad_norm": 0.00010307446063961834,
      "learning_rate": 1.4261442765006739e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11162896,
      "step": 39865
    },
    {
      "epoch": 443.0,
      "grad_norm": 0.0001444169320166111,
      "learning_rate": 1.3232092735876445e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11164336,
      "step": 39870
    },
    {
      "epoch": 443.05555555555554,
      "grad_norm": 0.0029588784091174603,
      "learning_rate": 1.2241293807918607e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11165712,
      "step": 39875
    },
    {
      "epoch": 443.1111111111111,
      "grad_norm": 0.0007632976630702615,
      "learning_rate": 1.128904613387216e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11167104,
      "step": 39880
    },
    {
      "epoch": 443.1666666666667,
      "grad_norm": 2.0959620087523945e-05,
      "learning_rate": 1.0375349860591853e-09,
      "loss": 0.0,
      "num_input_tokens_seen": 11168560,
      "step": 39885
    },
    {
      "epoch": 443.22222222222223,
      "grad_norm": 0.0004609647439792752,
      "learning_rate": 9.5002051290205e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11169968,
      "step": 39890
    },
    {
      "epoch": 443.27777777777777,
      "grad_norm": 0.0007526787812821567,
      "learning_rate": 8.663612074077954e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11171376,
      "step": 39895
    },
    {
      "epoch": 443.3333333333333,
      "grad_norm": 0.0002191399544244632,
      "learning_rate": 7.865570824799884e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11172768,
      "step": 39900
    },
    {
      "epoch": 443.3888888888889,
      "grad_norm": 7.540896331192926e-05,
      "learning_rate": 7.106081504254514e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11174160,
      "step": 39905
    },
    {
      "epoch": 443.44444444444446,
      "grad_norm": 0.0001558668736834079,
      "learning_rate": 6.385144229570372e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11175584,
      "step": 39910
    },
    {
      "epoch": 443.5,
      "grad_norm": 0.0002537294349167496,
      "learning_rate": 5.70275911190854e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11176928,
      "step": 39915
    },
    {
      "epoch": 443.55555555555554,
      "grad_norm": 1.700544817140326e-05,
      "learning_rate": 5.058926256490403e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11178304,
      "step": 39920
    },
    {
      "epoch": 443.6111111111111,
      "grad_norm": 0.00029819904011674225,
      "learning_rate": 4.4536457626254134e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11179728,
      "step": 39925
    },
    {
      "epoch": 443.6666666666667,
      "grad_norm": 0.00205252505838871,
      "learning_rate": 3.88691772365557e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11181104,
      "step": 39930
    },
    {
      "epoch": 443.72222222222223,
      "grad_norm": 1.6264924852293916e-05,
      "learning_rate": 3.358742226955425e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11182512,
      "step": 39935
    },
    {
      "epoch": 443.77777777777777,
      "grad_norm": 4.5560442231362686e-05,
      "learning_rate": 2.8691193539875925e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11183952,
      "step": 39940
    },
    {
      "epoch": 443.8333333333333,
      "grad_norm": 7.998326327651739e-05,
      "learning_rate": 2.418049180274995e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11185344,
      "step": 39945
    },
    {
      "epoch": 443.8888888888889,
      "grad_norm": 6.220262730494142e-05,
      "learning_rate": 2.005531775373104e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11186752,
      "step": 39950
    },
    {
      "epoch": 443.94444444444446,
      "grad_norm": 0.00014842391829006374,
      "learning_rate": 1.6315672028699435e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11188176,
      "step": 39955
    },
    {
      "epoch": 444.0,
      "grad_norm": 4.18050040025264e-05,
      "learning_rate": 1.2961555204693555e-10,
      "loss": 0.0,
      "num_input_tokens_seen": 11189552,
      "step": 39960
    },
    {
      "epoch": 444.05555555555554,
      "grad_norm": 8.89392049430171e-06,
      "learning_rate": 9.992967798799768e-11,
      "loss": 0.0,
      "num_input_tokens_seen": 11190960,
      "step": 39965
    },
    {
      "epoch": 444.1111111111111,
      "grad_norm": 5.35631479579024e-05,
      "learning_rate": 7.409910268707521e-11,
      "loss": 0.0,
      "num_input_tokens_seen": 11192368,
      "step": 39970
    },
    {
      "epoch": 444.1666666666667,
      "grad_norm": 5.0295828259550035e-05,
      "learning_rate": 5.212383012986877e-11,
      "loss": 0.0,
      "num_input_tokens_seen": 11193776,
      "step": 39975
    },
    {
      "epoch": 444.22222222222223,
      "grad_norm": 0.00011101820564363152,
      "learning_rate": 3.400386370533415e-11,
      "loss": 0.0,
      "num_input_tokens_seen": 11195216,
      "step": 39980
    },
    {
      "epoch": 444.27777777777777,
      "grad_norm": 7.796719728503376e-05,
      "learning_rate": 1.9739206205682258e-11,
      "loss": 0.0,
      "num_input_tokens_seen": 11196624,
      "step": 39985
    },
    {
      "epoch": 444.3333333333333,
      "grad_norm": 0.00019137452181894332,
      "learning_rate": 9.329859829154685e-12,
      "loss": 0.0,
      "num_input_tokens_seen": 11198032,
      "step": 39990
    },
    {
      "epoch": 444.3888888888889,
      "grad_norm": 0.0001133825135184452,
      "learning_rate": 2.7758261855748148e-12,
      "loss": 0.0,
      "num_input_tokens_seen": 11199424,
      "step": 39995
    },
    {
      "epoch": 444.44444444444446,
      "grad_norm": 6.102303814259358e-06,
      "learning_rate": 7.710628524559838e-14,
      "loss": 0.0,
      "num_input_tokens_seen": 11200800,
      "step": 40000
    },
    {
      "epoch": 444.44444444444446,
      "eval_loss": 0.907706618309021,
      "eval_runtime": 1.2385,
      "eval_samples_per_second": 32.297,
      "eval_steps_per_second": 16.149,
      "num_input_tokens_seen": 11200800,
      "step": 40000
    },
    {
      "epoch": 444.44444444444446,
      "num_input_tokens_seen": 11200800,
      "step": 40000,
      "total_flos": 4.69048729503744e+16,
      "train_loss": 0.041804404937768734,
      "train_runtime": 14360.258,
      "train_samples_per_second": 11.142,
      "train_steps_per_second": 2.785
    }
  ],
  "logging_steps": 5,
  "max_steps": 40000,
  "num_input_tokens_seen": 11200800,
  "num_train_epochs": 445,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.69048729503744e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}