{
  "best_global_step": 5306,
  "best_metric": 0.09362584352493286,
  "best_model_checkpoint": "saves_bts_preliminary/base/llama-3.2-1b-instruct/train_sst2_42_1779354538/checkpoint-5306",
  "epoch": 1.0,
  "eval_steps": 379,
  "global_step": 7577,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006598917777484492,
      "grad_norm": 438.2236328125,
      "learning_rate": 1.0554089709762531e-08,
      "loss": 1.4317,
      "num_input_tokens_seen": 2240,
      "step": 5
    },
    {
      "epoch": 0.0013197835554968984,
      "grad_norm": 445.3807678222656,
      "learning_rate": 2.3746701846965696e-08,
      "loss": 1.5404,
      "num_input_tokens_seen": 4672,
      "step": 10
    },
    {
      "epoch": 0.0019796753332453477,
      "grad_norm": 458.7563781738281,
      "learning_rate": 3.6939313984168866e-08,
      "loss": 1.4488,
      "num_input_tokens_seen": 7040,
      "step": 15
    },
    {
      "epoch": 0.002639567110993797,
      "grad_norm": 498.70074462890625,
      "learning_rate": 5.013192612137203e-08,
      "loss": 1.469,
      "num_input_tokens_seen": 9600,
      "step": 20
    },
    {
      "epoch": 0.0032994588887422464,
      "grad_norm": 452.42120361328125,
      "learning_rate": 6.33245382585752e-08,
      "loss": 1.3406,
      "num_input_tokens_seen": 12160,
      "step": 25
    },
    {
      "epoch": 0.0039593506664906955,
      "grad_norm": 394.526123046875,
      "learning_rate": 7.651715039577835e-08,
      "loss": 1.421,
      "num_input_tokens_seen": 14528,
      "step": 30
    },
    {
      "epoch": 0.004619242444239145,
      "grad_norm": 373.5523376464844,
      "learning_rate": 8.970976253298153e-08,
      "loss": 1.1582,
      "num_input_tokens_seen": 16768,
      "step": 35
    },
    {
      "epoch": 0.005279134221987594,
      "grad_norm": 333.46490478515625,
      "learning_rate": 1.0290237467018468e-07,
      "loss": 1.1864,
      "num_input_tokens_seen": 19264,
      "step": 40
    },
    {
      "epoch": 0.005939025999736044,
      "grad_norm": 272.6488342285156,
      "learning_rate": 1.1609498680738786e-07,
      "loss": 0.8458,
      "num_input_tokens_seen": 21632,
      "step": 45
    },
    {
      "epoch": 0.006598917777484493,
      "grad_norm": 193.0662384033203,
      "learning_rate": 1.29287598944591e-07,
      "loss": 0.7602,
      "num_input_tokens_seen": 24000,
      "step": 50
    },
    {
      "epoch": 0.007258809555232942,
      "grad_norm": 167.6585235595703,
      "learning_rate": 1.424802110817942e-07,
      "loss": 0.7195,
      "num_input_tokens_seen": 26496,
      "step": 55
    },
    {
      "epoch": 0.007918701332981391,
      "grad_norm": 51.53580856323242,
      "learning_rate": 1.5567282321899736e-07,
      "loss": 0.3928,
      "num_input_tokens_seen": 29120,
      "step": 60
    },
    {
      "epoch": 0.008578593110729841,
      "grad_norm": 29.78285026550293,
      "learning_rate": 1.688654353562005e-07,
      "loss": 0.3093,
      "num_input_tokens_seen": 31744,
      "step": 65
    },
    {
      "epoch": 0.00923848488847829,
      "grad_norm": 54.6888313293457,
      "learning_rate": 1.820580474934037e-07,
      "loss": 0.2986,
      "num_input_tokens_seen": 34176,
      "step": 70
    },
    {
      "epoch": 0.009898376666226739,
      "grad_norm": 53.20941925048828,
      "learning_rate": 1.9525065963060686e-07,
      "loss": 0.2998,
      "num_input_tokens_seen": 36864,
      "step": 75
    },
    {
      "epoch": 0.010558268443975187,
      "grad_norm": 28.97679328918457,
      "learning_rate": 2.0844327176781002e-07,
      "loss": 0.2555,
      "num_input_tokens_seen": 39424,
      "step": 80
    },
    {
      "epoch": 0.011218160221723637,
      "grad_norm": 37.55499267578125,
      "learning_rate": 2.2163588390501316e-07,
      "loss": 0.2595,
      "num_input_tokens_seen": 42112,
      "step": 85
    },
    {
      "epoch": 0.011878051999472087,
      "grad_norm": 32.20515060424805,
      "learning_rate": 2.3482849604221635e-07,
      "loss": 0.2567,
      "num_input_tokens_seen": 44544,
      "step": 90
    },
    {
      "epoch": 0.012537943777220536,
      "grad_norm": 45.711021423339844,
      "learning_rate": 2.480211081794195e-07,
      "loss": 0.1981,
      "num_input_tokens_seen": 47104,
      "step": 95
    },
    {
      "epoch": 0.013197835554968985,
      "grad_norm": 23.666234970092773,
      "learning_rate": 2.612137203166227e-07,
      "loss": 0.1626,
      "num_input_tokens_seen": 49664,
      "step": 100
    },
    {
      "epoch": 0.013857727332717434,
      "grad_norm": 21.503978729248047,
      "learning_rate": 2.744063324538258e-07,
      "loss": 0.1359,
      "num_input_tokens_seen": 52352,
      "step": 105
    },
    {
      "epoch": 0.014517619110465884,
      "grad_norm": 84.25057220458984,
      "learning_rate": 2.8759894459102903e-07,
      "loss": 0.1025,
      "num_input_tokens_seen": 54720,
      "step": 110
    },
    {
      "epoch": 0.015177510888214334,
      "grad_norm": 49.4913444519043,
      "learning_rate": 3.007915567282322e-07,
      "loss": 0.1435,
      "num_input_tokens_seen": 57152,
      "step": 115
    },
    {
      "epoch": 0.015837402665962782,
      "grad_norm": 98.1129379272461,
      "learning_rate": 3.139841688654353e-07,
      "loss": 0.2027,
      "num_input_tokens_seen": 59776,
      "step": 120
    },
    {
      "epoch": 0.01649729444371123,
      "grad_norm": 26.11796760559082,
      "learning_rate": 3.271767810026385e-07,
      "loss": 0.1344,
      "num_input_tokens_seen": 62464,
      "step": 125
    },
    {
      "epoch": 0.017157186221459682,
      "grad_norm": 80.43219757080078,
      "learning_rate": 3.403693931398417e-07,
      "loss": 0.1188,
      "num_input_tokens_seen": 65088,
      "step": 130
    },
    {
      "epoch": 0.01781707799920813,
      "grad_norm": 75.85595703125,
      "learning_rate": 3.5356200527704485e-07,
      "loss": 0.2073,
      "num_input_tokens_seen": 67776,
      "step": 135
    },
    {
      "epoch": 0.01847696977695658,
      "grad_norm": 69.01596069335938,
      "learning_rate": 3.66754617414248e-07,
      "loss": 0.2589,
      "num_input_tokens_seen": 70400,
      "step": 140
    },
    {
      "epoch": 0.01913686155470503,
      "grad_norm": 48.234092712402344,
      "learning_rate": 3.7994722955145113e-07,
      "loss": 0.1435,
      "num_input_tokens_seen": 72704,
      "step": 145
    },
    {
      "epoch": 0.019796753332453478,
      "grad_norm": 63.58314514160156,
      "learning_rate": 3.9313984168865435e-07,
      "loss": 0.3408,
      "num_input_tokens_seen": 75136,
      "step": 150
    },
    {
      "epoch": 0.020456645110201926,
      "grad_norm": 22.535675048828125,
      "learning_rate": 4.063324538258575e-07,
      "loss": 0.0612,
      "num_input_tokens_seen": 77632,
      "step": 155
    },
    {
      "epoch": 0.021116536887950375,
      "grad_norm": 69.68572998046875,
      "learning_rate": 4.195250659630606e-07,
      "loss": 0.2462,
      "num_input_tokens_seen": 80320,
      "step": 160
    },
    {
      "epoch": 0.021776428665698826,
      "grad_norm": 110.42927551269531,
      "learning_rate": 4.3271767810026384e-07,
      "loss": 0.1542,
      "num_input_tokens_seen": 82752,
      "step": 165
    },
    {
      "epoch": 0.022436320443447275,
      "grad_norm": 151.29769897460938,
      "learning_rate": 4.45910290237467e-07,
      "loss": 0.3088,
      "num_input_tokens_seen": 85248,
      "step": 170
    },
    {
      "epoch": 0.023096212221195723,
      "grad_norm": 170.58566284179688,
      "learning_rate": 4.5910290237467017e-07,
      "loss": 0.2887,
      "num_input_tokens_seen": 87872,
      "step": 175
    },
    {
      "epoch": 0.023756103998944175,
      "grad_norm": 51.48556137084961,
      "learning_rate": 4.7229551451187333e-07,
      "loss": 0.1364,
      "num_input_tokens_seen": 90368,
      "step": 180
    },
    {
      "epoch": 0.024415995776692623,
      "grad_norm": 31.3718204498291,
      "learning_rate": 4.854881266490765e-07,
      "loss": 0.0589,
      "num_input_tokens_seen": 92928,
      "step": 185
    },
    {
      "epoch": 0.02507588755444107,
      "grad_norm": 66.79593658447266,
      "learning_rate": 4.986807387862796e-07,
      "loss": 0.1639,
      "num_input_tokens_seen": 95296,
      "step": 190
    },
    {
      "epoch": 0.02573577933218952,
      "grad_norm": 57.898441314697266,
      "learning_rate": 5.118733509234829e-07,
      "loss": 0.1103,
      "num_input_tokens_seen": 97984,
      "step": 195
    },
    {
      "epoch": 0.02639567110993797,
      "grad_norm": 106.46971130371094,
      "learning_rate": 5.250659630606859e-07,
      "loss": 0.2017,
      "num_input_tokens_seen": 100352,
      "step": 200
    },
    {
      "epoch": 0.02705556288768642,
      "grad_norm": 17.6440372467041,
      "learning_rate": 5.382585751978892e-07,
      "loss": 0.1383,
      "num_input_tokens_seen": 102464,
      "step": 205
    },
    {
      "epoch": 0.027715454665434867,
      "grad_norm": 14.92352294921875,
      "learning_rate": 5.514511873350924e-07,
      "loss": 0.1266,
      "num_input_tokens_seen": 105088,
      "step": 210
    },
    {
      "epoch": 0.02837534644318332,
      "grad_norm": 16.06985092163086,
      "learning_rate": 5.646437994722954e-07,
      "loss": 0.3019,
      "num_input_tokens_seen": 107648,
      "step": 215
    },
    {
      "epoch": 0.029035238220931767,
      "grad_norm": 108.87126159667969,
      "learning_rate": 5.778364116094987e-07,
      "loss": 0.2375,
      "num_input_tokens_seen": 110144,
      "step": 220
    },
    {
      "epoch": 0.029695129998680216,
      "grad_norm": 33.7409553527832,
      "learning_rate": 5.910290237467019e-07,
      "loss": 0.1332,
      "num_input_tokens_seen": 112576,
      "step": 225
    },
    {
      "epoch": 0.030355021776428667,
      "grad_norm": 58.12620162963867,
      "learning_rate": 6.042216358839049e-07,
      "loss": 0.2314,
      "num_input_tokens_seen": 115264,
      "step": 230
    },
    {
      "epoch": 0.031014913554177116,
      "grad_norm": 19.326017379760742,
      "learning_rate": 6.174142480211082e-07,
      "loss": 0.0848,
      "num_input_tokens_seen": 117888,
      "step": 235
    },
    {
      "epoch": 0.031674805331925564,
      "grad_norm": 34.517738342285156,
      "learning_rate": 6.306068601583114e-07,
      "loss": 0.0189,
      "num_input_tokens_seen": 120320,
      "step": 240
    },
    {
      "epoch": 0.032334697109674015,
      "grad_norm": 89.7800521850586,
      "learning_rate": 6.437994722955144e-07,
      "loss": 0.1826,
      "num_input_tokens_seen": 122688,
      "step": 245
    },
    {
      "epoch": 0.03299458888742246,
      "grad_norm": 81.10912322998047,
      "learning_rate": 6.569920844327177e-07,
      "loss": 0.2716,
      "num_input_tokens_seen": 125248,
      "step": 250
    },
    {
      "epoch": 0.03365448066517091,
      "grad_norm": 89.25950622558594,
      "learning_rate": 6.701846965699208e-07,
      "loss": 0.203,
      "num_input_tokens_seen": 127680,
      "step": 255
    },
    {
      "epoch": 0.034314372442919364,
      "grad_norm": 19.84525489807129,
      "learning_rate": 6.833773087071239e-07,
      "loss": 0.083,
      "num_input_tokens_seen": 130496,
      "step": 260
    },
    {
      "epoch": 0.03497426422066781,
      "grad_norm": 34.019805908203125,
      "learning_rate": 6.965699208443272e-07,
      "loss": 0.1837,
      "num_input_tokens_seen": 132992,
      "step": 265
    },
    {
      "epoch": 0.03563415599841626,
      "grad_norm": 63.352012634277344,
      "learning_rate": 7.097625329815303e-07,
      "loss": 0.1372,
      "num_input_tokens_seen": 135040,
      "step": 270
    },
    {
      "epoch": 0.03629404777616471,
      "grad_norm": 70.96591186523438,
      "learning_rate": 7.229551451187335e-07,
      "loss": 0.258,
      "num_input_tokens_seen": 137600,
      "step": 275
    },
    {
      "epoch": 0.03695393955391316,
      "grad_norm": 57.32151412963867,
      "learning_rate": 7.361477572559367e-07,
      "loss": 0.1103,
      "num_input_tokens_seen": 139904,
      "step": 280
    },
    {
      "epoch": 0.03761383133166161,
      "grad_norm": 16.635353088378906,
      "learning_rate": 7.493403693931398e-07,
      "loss": 0.1476,
      "num_input_tokens_seen": 142016,
      "step": 285
    },
    {
      "epoch": 0.03827372310941006,
      "grad_norm": 9.866713523864746,
      "learning_rate": 7.62532981530343e-07,
      "loss": 0.0087,
      "num_input_tokens_seen": 144576,
      "step": 290
    },
    {
      "epoch": 0.038933614887158505,
      "grad_norm": 118.24456024169922,
      "learning_rate": 7.757255936675461e-07,
      "loss": 0.085,
      "num_input_tokens_seen": 146880,
      "step": 295
    },
    {
      "epoch": 0.039593506664906956,
      "grad_norm": 92.64739990234375,
      "learning_rate": 7.889182058047493e-07,
      "loss": 0.3963,
      "num_input_tokens_seen": 149184,
      "step": 300
    },
    {
      "epoch": 0.0402533984426554,
      "grad_norm": 0.17220743000507355,
      "learning_rate": 8.021108179419525e-07,
      "loss": 0.221,
      "num_input_tokens_seen": 151296,
      "step": 305
    },
    {
      "epoch": 0.04091329022040385,
      "grad_norm": 89.11080932617188,
      "learning_rate": 8.153034300791555e-07,
      "loss": 0.2076,
      "num_input_tokens_seen": 153664,
      "step": 310
    },
    {
      "epoch": 0.041573181998152305,
      "grad_norm": 35.5322265625,
      "learning_rate": 8.284960422163588e-07,
      "loss": 0.2588,
      "num_input_tokens_seen": 156032,
      "step": 315
    },
    {
      "epoch": 0.04223307377590075,
      "grad_norm": 0.28457769751548767,
      "learning_rate": 8.41688654353562e-07,
      "loss": 0.227,
      "num_input_tokens_seen": 158528,
      "step": 320
    },
    {
      "epoch": 0.0428929655536492,
      "grad_norm": 14.495915412902832,
      "learning_rate": 8.54881266490765e-07,
      "loss": 0.12,
      "num_input_tokens_seen": 160704,
      "step": 325
    },
    {
      "epoch": 0.04355285733139765,
      "grad_norm": 5.979187965393066,
      "learning_rate": 8.680738786279683e-07,
      "loss": 0.1931,
      "num_input_tokens_seen": 163072,
      "step": 330
    },
    {
      "epoch": 0.0442127491091461,
      "grad_norm": 65.1185302734375,
      "learning_rate": 8.812664907651715e-07,
      "loss": 0.1389,
      "num_input_tokens_seen": 165568,
      "step": 335
    },
    {
      "epoch": 0.04487264088689455,
      "grad_norm": 0.6913341283798218,
      "learning_rate": 8.944591029023745e-07,
      "loss": 0.2857,
      "num_input_tokens_seen": 167936,
      "step": 340
    },
    {
      "epoch": 0.045532532664643,
      "grad_norm": 178.3228759765625,
      "learning_rate": 9.076517150395778e-07,
      "loss": 0.0616,
      "num_input_tokens_seen": 170176,
      "step": 345
    },
    {
      "epoch": 0.046192424442391446,
      "grad_norm": 140.8086700439453,
      "learning_rate": 9.20844327176781e-07,
      "loss": 0.149,
      "num_input_tokens_seen": 172352,
      "step": 350
    },
    {
      "epoch": 0.0468523162201399,
      "grad_norm": 42.391597747802734,
      "learning_rate": 9.340369393139841e-07,
      "loss": 0.2579,
      "num_input_tokens_seen": 175168,
      "step": 355
    },
    {
      "epoch": 0.04751220799788835,
      "grad_norm": 2.5807368755340576,
      "learning_rate": 9.472295514511873e-07,
      "loss": 0.1596,
      "num_input_tokens_seen": 177856,
      "step": 360
    },
    {
      "epoch": 0.048172099775636794,
      "grad_norm": 41.364768981933594,
      "learning_rate": 9.604221635883904e-07,
      "loss": 0.2923,
      "num_input_tokens_seen": 180160,
      "step": 365
    },
    {
      "epoch": 0.048831991553385246,
      "grad_norm": 139.357177734375,
      "learning_rate": 9.736147757255936e-07,
      "loss": 0.2137,
      "num_input_tokens_seen": 182784,
      "step": 370
    },
    {
      "epoch": 0.0494918833311337,
      "grad_norm": 1.2181538343429565,
      "learning_rate": 9.86807387862797e-07,
      "loss": 0.0578,
      "num_input_tokens_seen": 185024,
      "step": 375
    },
    {
      "epoch": 0.05001979675333245,
      "eval_loss": 0.16884121298789978,
      "eval_runtime": 7.8728,
      "eval_samples_per_second": 855.48,
      "eval_steps_per_second": 106.951,
      "num_input_tokens_seen": 187072,
      "step": 379
    },
    {
      "epoch": 0.05015177510888214,
      "grad_norm": 64.27156066894531,
      "learning_rate": 1e-06,
      "loss": 0.1527,
      "num_input_tokens_seen": 187712,
      "step": 380
    },
    {
      "epoch": 0.050811666886630594,
      "grad_norm": 49.03231430053711,
      "learning_rate": 1.0131926121372032e-06,
      "loss": 0.2528,
      "num_input_tokens_seen": 190400,
      "step": 385
    },
    {
      "epoch": 0.05147155866437904,
      "grad_norm": 53.97622299194336,
      "learning_rate": 1.0263852242744063e-06,
      "loss": 0.1124,
      "num_input_tokens_seen": 193280,
      "step": 390
    },
    {
      "epoch": 0.05213145044212749,
      "grad_norm": 38.25093078613281,
      "learning_rate": 1.0395778364116096e-06,
      "loss": 0.1545,
      "num_input_tokens_seen": 195584,
      "step": 395
    },
    {
      "epoch": 0.05279134221987594,
      "grad_norm": 23.770427703857422,
      "learning_rate": 1.0527704485488126e-06,
      "loss": 0.0824,
      "num_input_tokens_seen": 198208,
      "step": 400
    },
    {
      "epoch": 0.05345123399762439,
      "grad_norm": 109.20024871826172,
      "learning_rate": 1.0659630606860157e-06,
      "loss": 0.255,
      "num_input_tokens_seen": 200704,
      "step": 405
    },
    {
      "epoch": 0.05411112577537284,
      "grad_norm": 86.4642333984375,
      "learning_rate": 1.079155672823219e-06,
      "loss": 0.2055,
      "num_input_tokens_seen": 203136,
      "step": 410
    },
    {
      "epoch": 0.05477101755312129,
      "grad_norm": 77.17768859863281,
      "learning_rate": 1.0923482849604222e-06,
      "loss": 0.1496,
      "num_input_tokens_seen": 205504,
      "step": 415
    },
    {
      "epoch": 0.055430909330869735,
      "grad_norm": 90.73381042480469,
      "learning_rate": 1.1055408970976253e-06,
      "loss": 0.2539,
      "num_input_tokens_seen": 208000,
      "step": 420
    },
    {
      "epoch": 0.056090801108618187,
      "grad_norm": 227.53492736816406,
      "learning_rate": 1.1187335092348285e-06,
      "loss": 0.1106,
      "num_input_tokens_seen": 210560,
      "step": 425
    },
    {
      "epoch": 0.05675069288636664,
      "grad_norm": 39.24838638305664,
      "learning_rate": 1.1319261213720316e-06,
      "loss": 0.1771,
      "num_input_tokens_seen": 213056,
      "step": 430
    },
    {
      "epoch": 0.05741058466411508,
      "grad_norm": 54.15324020385742,
      "learning_rate": 1.1451187335092347e-06,
      "loss": 0.1155,
      "num_input_tokens_seen": 215296,
      "step": 435
    },
    {
      "epoch": 0.058070476441863535,
      "grad_norm": 0.09268882125616074,
      "learning_rate": 1.158311345646438e-06,
      "loss": 0.0203,
      "num_input_tokens_seen": 217472,
      "step": 440
    },
    {
      "epoch": 0.058730368219611986,
      "grad_norm": 126.56912994384766,
      "learning_rate": 1.1715039577836412e-06,
      "loss": 0.0346,
      "num_input_tokens_seen": 219968,
      "step": 445
    },
    {
      "epoch": 0.05939025999736043,
      "grad_norm": 0.09806288778781891,
      "learning_rate": 1.1846965699208443e-06,
      "loss": 0.0218,
      "num_input_tokens_seen": 222592,
      "step": 450
    },
    {
      "epoch": 0.06005015177510888,
      "grad_norm": 183.16590881347656,
      "learning_rate": 1.1978891820580475e-06,
      "loss": 0.522,
      "num_input_tokens_seen": 224768,
      "step": 455
    },
    {
      "epoch": 0.060710043552857335,
      "grad_norm": 3.036947250366211,
      "learning_rate": 1.2110817941952508e-06,
      "loss": 0.2349,
      "num_input_tokens_seen": 227264,
      "step": 460
    },
    {
      "epoch": 0.06136993533060578,
      "grad_norm": 18.994815826416016,
      "learning_rate": 1.2242744063324536e-06,
      "loss": 0.2363,
      "num_input_tokens_seen": 229760,
      "step": 465
    },
    {
      "epoch": 0.06202982710835423,
      "grad_norm": 97.5959243774414,
      "learning_rate": 1.237467018469657e-06,
      "loss": 0.1719,
      "num_input_tokens_seen": 232000,
      "step": 470
    },
    {
      "epoch": 0.06268971888610268,
      "grad_norm": 2.0452661514282227,
      "learning_rate": 1.2506596306068602e-06,
      "loss": 0.1564,
      "num_input_tokens_seen": 234176,
      "step": 475
    },
    {
      "epoch": 0.06334961066385113,
      "grad_norm": 0.3037824034690857,
      "learning_rate": 1.2638522427440632e-06,
      "loss": 0.0993,
      "num_input_tokens_seen": 236736,
      "step": 480
    },
    {
      "epoch": 0.06400950244159957,
      "grad_norm": 0.12219711393117905,
      "learning_rate": 1.2770448548812665e-06,
      "loss": 0.006,
      "num_input_tokens_seen": 239104,
      "step": 485
    },
    {
      "epoch": 0.06466939421934803,
      "grad_norm": 166.3263702392578,
      "learning_rate": 1.2902374670184698e-06,
      "loss": 0.3751,
      "num_input_tokens_seen": 241600,
      "step": 490
    },
    {
      "epoch": 0.06532928599709648,
      "grad_norm": 131.08555603027344,
      "learning_rate": 1.3034300791556726e-06,
      "loss": 0.31,
      "num_input_tokens_seen": 244160,
      "step": 495
    },
    {
      "epoch": 0.06598917777484492,
      "grad_norm": 1.1034032106399536,
      "learning_rate": 1.316622691292876e-06,
      "loss": 0.1775,
      "num_input_tokens_seen": 246464,
      "step": 500
    },
    {
      "epoch": 0.06664906955259338,
      "grad_norm": 1.8536452054977417,
      "learning_rate": 1.3298153034300792e-06,
      "loss": 0.0849,
      "num_input_tokens_seen": 248960,
      "step": 505
    },
    {
      "epoch": 0.06730896133034182,
      "grad_norm": 91.25196075439453,
      "learning_rate": 1.3430079155672822e-06,
      "loss": 0.1716,
      "num_input_tokens_seen": 251392,
      "step": 510
    },
    {
      "epoch": 0.06796885310809027,
      "grad_norm": 30.315645217895508,
      "learning_rate": 1.3562005277044855e-06,
      "loss": 0.2321,
      "num_input_tokens_seen": 253888,
      "step": 515
    },
    {
      "epoch": 0.06862874488583873,
      "grad_norm": 20.200546264648438,
      "learning_rate": 1.3693931398416888e-06,
      "loss": 0.1398,
      "num_input_tokens_seen": 256384,
      "step": 520
    },
    {
      "epoch": 0.06928863666358717,
      "grad_norm": 70.2831039428711,
      "learning_rate": 1.3825857519788916e-06,
      "loss": 0.1866,
      "num_input_tokens_seen": 258496,
      "step": 525
    },
    {
      "epoch": 0.06994852844133562,
      "grad_norm": 0.08841226994991302,
      "learning_rate": 1.3957783641160949e-06,
      "loss": 0.1035,
      "num_input_tokens_seen": 260864,
      "step": 530
    },
    {
      "epoch": 0.07060842021908408,
      "grad_norm": 16.285932540893555,
      "learning_rate": 1.4089709762532982e-06,
      "loss": 0.0497,
      "num_input_tokens_seen": 263360,
      "step": 535
    },
    {
      "epoch": 0.07126831199683252,
      "grad_norm": 291.2605285644531,
      "learning_rate": 1.4221635883905012e-06,
      "loss": 0.3041,
      "num_input_tokens_seen": 266112,
      "step": 540
    },
    {
      "epoch": 0.07192820377458096,
      "grad_norm": 41.77130889892578,
      "learning_rate": 1.4353562005277045e-06,
      "loss": 0.2543,
      "num_input_tokens_seen": 268544,
      "step": 545
    },
    {
      "epoch": 0.07258809555232942,
      "grad_norm": 0.281996488571167,
      "learning_rate": 1.4485488126649078e-06,
      "loss": 0.4179,
      "num_input_tokens_seen": 270912,
      "step": 550
    },
    {
      "epoch": 0.07324798733007787,
      "grad_norm": 0.33394086360931396,
      "learning_rate": 1.4617414248021108e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 273664,
      "step": 555
    },
    {
      "epoch": 0.07390787910782631,
      "grad_norm": 92.83995819091797,
      "learning_rate": 1.4749340369393139e-06,
      "loss": 0.1789,
      "num_input_tokens_seen": 276160,
      "step": 560
    },
    {
      "epoch": 0.07456777088557477,
      "grad_norm": 0.3077968955039978,
      "learning_rate": 1.4881266490765171e-06,
      "loss": 0.0054,
      "num_input_tokens_seen": 278784,
      "step": 565
    },
    {
      "epoch": 0.07522766266332322,
      "grad_norm": 28.540903091430664,
      "learning_rate": 1.5013192612137202e-06,
      "loss": 0.2245,
      "num_input_tokens_seen": 281152,
      "step": 570
    },
    {
      "epoch": 0.07588755444107166,
      "grad_norm": 75.34148406982422,
      "learning_rate": 1.5145118733509235e-06,
      "loss": 0.2359,
      "num_input_tokens_seen": 283456,
      "step": 575
    },
    {
      "epoch": 0.07654744621882012,
      "grad_norm": 2.8185484409332275,
      "learning_rate": 1.5277044854881265e-06,
      "loss": 0.0789,
      "num_input_tokens_seen": 286016,
      "step": 580
    },
    {
      "epoch": 0.07720733799656856,
      "grad_norm": 0.16082888841629028,
      "learning_rate": 1.5408970976253298e-06,
      "loss": 0.1872,
      "num_input_tokens_seen": 288448,
      "step": 585
    },
    {
      "epoch": 0.07786722977431701,
      "grad_norm": 0.02060079760849476,
      "learning_rate": 1.5540897097625329e-06,
      "loss": 0.1543,
      "num_input_tokens_seen": 290816,
      "step": 590
    },
    {
      "epoch": 0.07852712155206547,
      "grad_norm": 66.97010803222656,
      "learning_rate": 1.567282321899736e-06,
      "loss": 0.2174,
      "num_input_tokens_seen": 293504,
      "step": 595
    },
    {
      "epoch": 0.07918701332981391,
      "grad_norm": 0.04142444580793381,
      "learning_rate": 1.5804749340369392e-06,
      "loss": 0.1323,
      "num_input_tokens_seen": 295744,
      "step": 600
    },
    {
      "epoch": 0.07984690510756236,
      "grad_norm": 0.039387296885252,
      "learning_rate": 1.5936675461741425e-06,
      "loss": 0.0865,
      "num_input_tokens_seen": 298112,
      "step": 605
    },
    {
      "epoch": 0.0805067968853108,
      "grad_norm": 93.4574203491211,
      "learning_rate": 1.6068601583113455e-06,
      "loss": 0.1034,
      "num_input_tokens_seen": 300608,
      "step": 610
    },
    {
      "epoch": 0.08116668866305926,
      "grad_norm": 0.054611898958683014,
      "learning_rate": 1.6200527704485488e-06,
      "loss": 0.0048,
      "num_input_tokens_seen": 303360,
      "step": 615
    },
    {
      "epoch": 0.0818265804408077,
      "grad_norm": 4.433136940002441,
      "learning_rate": 1.633245382585752e-06,
      "loss": 0.0325,
      "num_input_tokens_seen": 305920,
      "step": 620
    },
    {
      "epoch": 0.08248647221855615,
      "grad_norm": 1.7903838157653809,
      "learning_rate": 1.646437994722955e-06,
      "loss": 0.0698,
      "num_input_tokens_seen": 308416,
      "step": 625
    },
    {
      "epoch": 0.08314636399630461,
      "grad_norm": 0.04191221296787262,
      "learning_rate": 1.6596306068601582e-06,
      "loss": 0.1542,
      "num_input_tokens_seen": 310848,
      "step": 630
    },
    {
      "epoch": 0.08380625577405305,
      "grad_norm": 60.00360107421875,
      "learning_rate": 1.6728232189973614e-06,
      "loss": 0.225,
      "num_input_tokens_seen": 313408,
      "step": 635
    },
    {
      "epoch": 0.0844661475518015,
      "grad_norm": 77.6712417602539,
      "learning_rate": 1.6860158311345645e-06,
      "loss": 0.0946,
      "num_input_tokens_seen": 315904,
      "step": 640
    },
    {
      "epoch": 0.08512603932954996,
      "grad_norm": 63.06571578979492,
      "learning_rate": 1.6992084432717678e-06,
      "loss": 0.2463,
      "num_input_tokens_seen": 318080,
      "step": 645
    },
    {
      "epoch": 0.0857859311072984,
      "grad_norm": 3.3637969493865967,
      "learning_rate": 1.712401055408971e-06,
      "loss": 0.0751,
      "num_input_tokens_seen": 320256,
      "step": 650
    },
    {
      "epoch": 0.08644582288504685,
      "grad_norm": 0.38727056980133057,
      "learning_rate": 1.7255936675461739e-06,
      "loss": 0.0116,
      "num_input_tokens_seen": 322496,
      "step": 655
    },
    {
      "epoch": 0.0871057146627953,
      "grad_norm": 62.513736724853516,
      "learning_rate": 1.7387862796833772e-06,
      "loss": 0.1616,
      "num_input_tokens_seen": 325120,
      "step": 660
    },
    {
      "epoch": 0.08776560644054375,
      "grad_norm": 270.4856872558594,
      "learning_rate": 1.7519788918205804e-06,
      "loss": 0.0802,
      "num_input_tokens_seen": 327296,
      "step": 665
    },
    {
      "epoch": 0.0884254982182922,
      "grad_norm": 42.6474723815918,
      "learning_rate": 1.7651715039577835e-06,
      "loss": 0.4002,
      "num_input_tokens_seen": 329664,
      "step": 670
    },
    {
      "epoch": 0.08908538999604065,
      "grad_norm": 24.991104125976562,
      "learning_rate": 1.7783641160949868e-06,
      "loss": 0.0953,
      "num_input_tokens_seen": 332416,
      "step": 675
    },
    {
      "epoch": 0.0897452817737891,
      "grad_norm": 0.23048889636993408,
      "learning_rate": 1.79155672823219e-06,
      "loss": 0.0779,
      "num_input_tokens_seen": 334976,
      "step": 680
    },
    {
      "epoch": 0.09040517355153754,
      "grad_norm": 0.02817719615995884,
      "learning_rate": 1.8047493403693929e-06,
      "loss": 0.0366,
      "num_input_tokens_seen": 337472,
      "step": 685
    },
    {
      "epoch": 0.091065065329286,
      "grad_norm": 6.201309680938721,
      "learning_rate": 1.8179419525065961e-06,
      "loss": 0.1887,
      "num_input_tokens_seen": 339904,
      "step": 690
    },
    {
      "epoch": 0.09172495710703445,
      "grad_norm": 85.3891830444336,
      "learning_rate": 1.8311345646437994e-06,
      "loss": 0.2216,
      "num_input_tokens_seen": 342272,
      "step": 695
    },
    {
      "epoch": 0.09238484888478289,
      "grad_norm": 0.08057679235935211,
      "learning_rate": 1.8443271767810025e-06,
      "loss": 0.197,
      "num_input_tokens_seen": 344640,
      "step": 700
    },
    {
      "epoch": 0.09304474066253135,
      "grad_norm": 0.027219658717513084,
      "learning_rate": 1.8575197889182057e-06,
      "loss": 0.0457,
      "num_input_tokens_seen": 347072,
      "step": 705
    },
    {
      "epoch": 0.0937046324402798,
      "grad_norm": 0.05592739209532738,
      "learning_rate": 1.870712401055409e-06,
      "loss": 0.1862,
      "num_input_tokens_seen": 349696,
      "step": 710
    },
    {
      "epoch": 0.09436452421802824,
      "grad_norm": 4.520129680633545,
      "learning_rate": 1.883905013192612e-06,
      "loss": 0.0762,
      "num_input_tokens_seen": 352256,
      "step": 715
    },
    {
      "epoch": 0.0950244159957767,
      "grad_norm": 0.11173778027296066,
      "learning_rate": 1.8970976253298151e-06,
      "loss": 0.1502,
      "num_input_tokens_seen": 355008,
      "step": 720
    },
    {
      "epoch": 0.09568430777352514,
      "grad_norm": 173.77919006347656,
      "learning_rate": 1.9102902374670186e-06,
      "loss": 0.2955,
      "num_input_tokens_seen": 357376,
      "step": 725
    },
    {
      "epoch": 0.09634419955127359,
      "grad_norm": 2.091977119445801,
      "learning_rate": 1.9234828496042215e-06,
      "loss": 0.1803,
      "num_input_tokens_seen": 359680,
      "step": 730
    },
    {
      "epoch": 0.09700409132902205,
      "grad_norm": 0.5139044523239136,
      "learning_rate": 1.9366754617414247e-06,
      "loss": 0.1343,
      "num_input_tokens_seen": 362176,
      "step": 735
    },
    {
      "epoch": 0.09766398310677049,
      "grad_norm": 0.8499874472618103,
      "learning_rate": 1.949868073878628e-06,
      "loss": 0.0823,
      "num_input_tokens_seen": 365056,
      "step": 740
    },
    {
      "epoch": 0.09832387488451894,
      "grad_norm": 0.42520672082901,
      "learning_rate": 1.963060686015831e-06,
      "loss": 0.3164,
      "num_input_tokens_seen": 367488,
      "step": 745
    },
    {
      "epoch": 0.0989837666622674,
      "grad_norm": 0.8814378976821899,
      "learning_rate": 1.976253298153034e-06,
      "loss": 0.3437,
      "num_input_tokens_seen": 369792,
      "step": 750
    },
    {
      "epoch": 0.09964365844001584,
      "grad_norm": 114.21378326416016,
      "learning_rate": 1.9894459102902374e-06,
      "loss": 0.1097,
      "num_input_tokens_seen": 372160,
      "step": 755
    },
    {
      "epoch": 0.1000395935066649,
      "eval_loss": 0.13667261600494385,
      "eval_runtime": 7.9292,
      "eval_samples_per_second": 849.39,
      "eval_steps_per_second": 106.19,
      "num_input_tokens_seen": 373504,
      "step": 758
    },
    {
      "epoch": 0.10030355021776428,
      "grad_norm": 2.3556041717529297,
      "learning_rate": 1.9999998938723955e-06,
      "loss": 0.1005,
      "num_input_tokens_seen": 374272,
      "step": 760
    },
    {
      "epoch": 0.10096344199551274,
      "grad_norm": 4.03176212310791,
      "learning_rate": 1.9999961794086063e-06,
      "loss": 0.0727,
      "num_input_tokens_seen": 376704,
      "step": 765
    },
    {
      "epoch": 0.10162333377326119,
      "grad_norm": 250.72093200683594,
      "learning_rate": 1.999987158587122e-06,
      "loss": 0.287,
      "num_input_tokens_seen": 379136,
      "step": 770
    },
    {
      "epoch": 0.10228322555100963,
      "grad_norm": 42.057003021240234,
      "learning_rate": 1.9999728314558114e-06,
      "loss": 0.1199,
      "num_input_tokens_seen": 381568,
      "step": 775
    },
    {
      "epoch": 0.10294311732875808,
      "grad_norm": 0.05412933975458145,
      "learning_rate": 1.9999531980906988e-06,
      "loss": 0.2967,
      "num_input_tokens_seen": 384128,
      "step": 780
    },
    {
      "epoch": 0.10360300910650654,
      "grad_norm": 40.294551849365234,
      "learning_rate": 1.999928258595967e-06,
      "loss": 0.227,
      "num_input_tokens_seen": 386304,
      "step": 785
    },
    {
      "epoch": 0.10426290088425498,
      "grad_norm": 21.820249557495117,
      "learning_rate": 1.9998980131039534e-06,
      "loss": 0.3706,
      "num_input_tokens_seen": 388864,
      "step": 790
    },
    {
      "epoch": 0.10492279266200343,
      "grad_norm": 29.383682250976562,
      "learning_rate": 1.999862461775153e-06,
      "loss": 0.0609,
      "num_input_tokens_seen": 391104,
      "step": 795
    },
    {
      "epoch": 0.10558268443975188,
      "grad_norm": 19.28838539123535,
      "learning_rate": 1.999821604798214e-06,
      "loss": 0.1122,
      "num_input_tokens_seen": 393856,
      "step": 800
    },
    {
      "epoch": 0.10624257621750033,
      "grad_norm": 38.194297790527344,
      "learning_rate": 1.999775442389939e-06,
      "loss": 0.2655,
      "num_input_tokens_seen": 396352,
      "step": 805
    },
    {
      "epoch": 0.10690246799524877,
      "grad_norm": 1.154589295387268,
      "learning_rate": 1.9997239747952843e-06,
      "loss": 0.1068,
      "num_input_tokens_seen": 398592,
      "step": 810
    },
    {
      "epoch": 0.10756235977299723,
      "grad_norm": 97.90460968017578,
      "learning_rate": 1.9996672022873546e-06,
      "loss": 0.0669,
      "num_input_tokens_seen": 401088,
      "step": 815
    },
    {
      "epoch": 0.10822225155074568,
      "grad_norm": 133.7577362060547,
      "learning_rate": 1.9996051251674073e-06,
      "loss": 0.1053,
      "num_input_tokens_seen": 403456,
      "step": 820
    },
    {
      "epoch": 0.10888214332849412,
      "grad_norm": 41.88015365600586,
      "learning_rate": 1.999537743764847e-06,
      "loss": 0.3053,
      "num_input_tokens_seen": 405696,
      "step": 825
    },
    {
      "epoch": 0.10954203510624258,
      "grad_norm": 22.91583251953125,
      "learning_rate": 1.999465058437225e-06,
      "loss": 0.4762,
      "num_input_tokens_seen": 408128,
      "step": 830
    },
    {
      "epoch": 0.11020192688399102,
      "grad_norm": 0.7091745734214783,
      "learning_rate": 1.9993870695702364e-06,
      "loss": 0.0134,
      "num_input_tokens_seen": 411008,
      "step": 835
    },
    {
      "epoch": 0.11086181866173947,
      "grad_norm": 23.488370895385742,
      "learning_rate": 1.9993037775777206e-06,
      "loss": 0.3147,
      "num_input_tokens_seen": 413312,
      "step": 840
    },
    {
      "epoch": 0.11152171043948793,
      "grad_norm": 2.423909902572632,
      "learning_rate": 1.999215182901656e-06,
      "loss": 0.1734,
      "num_input_tokens_seen": 415616,
      "step": 845
    },
    {
      "epoch": 0.11218160221723637,
      "grad_norm": 156.0017852783203,
      "learning_rate": 1.9991212860121587e-06,
      "loss": 0.23,
      "num_input_tokens_seen": 418368,
      "step": 850
    },
    {
      "epoch": 0.11284149399498482,
      "grad_norm": 0.895738959312439,
      "learning_rate": 1.999022087407482e-06,
      "loss": 0.0658,
      "num_input_tokens_seen": 420864,
      "step": 855
    },
    {
      "epoch": 0.11350138577273328,
      "grad_norm": 57.05610275268555,
      "learning_rate": 1.998917587614011e-06,
      "loss": 0.2781,
      "num_input_tokens_seen": 423040,
      "step": 860
    },
    {
      "epoch": 0.11416127755048172,
      "grad_norm": 34.64775848388672,
      "learning_rate": 1.9988077871862615e-06,
      "loss": 0.3602,
      "num_input_tokens_seen": 425344,
      "step": 865
    },
    {
      "epoch": 0.11482116932823017,
      "grad_norm": 0.43117237091064453,
      "learning_rate": 1.9986926867068752e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 427968,
      "step": 870
    },
    {
      "epoch": 0.11548106110597862,
      "grad_norm": 0.1273239254951477,
      "learning_rate": 1.998572286786619e-06,
      "loss": 0.3594,
      "num_input_tokens_seen": 430592,
      "step": 875
    },
    {
      "epoch": 0.11614095288372707,
      "grad_norm": 9.043426513671875,
      "learning_rate": 1.9984465880643807e-06,
      "loss": 0.2425,
      "num_input_tokens_seen": 433152,
      "step": 880
    },
    {
      "epoch": 0.11680084466147551,
      "grad_norm": 65.71261596679688,
      "learning_rate": 1.998315591207165e-06,
      "loss": 0.0489,
      "num_input_tokens_seen": 435456,
      "step": 885
    },
    {
      "epoch": 0.11746073643922397,
      "grad_norm": 88.8459701538086,
      "learning_rate": 1.9981792969100912e-06,
      "loss": 0.1876,
      "num_input_tokens_seen": 438080,
      "step": 890
    },
    {
      "epoch": 0.11812062821697242,
      "grad_norm": 124.02445983886719,
      "learning_rate": 1.9980377058963875e-06,
      "loss": 0.2044,
      "num_input_tokens_seen": 440640,
      "step": 895
    },
    {
      "epoch": 0.11878051999472086,
      "grad_norm": 1.4761052131652832,
      "learning_rate": 1.99789081891739e-06,
      "loss": 0.028,
      "num_input_tokens_seen": 443008,
      "step": 900
    },
    {
      "epoch": 0.11944041177246932,
      "grad_norm": 110.05660247802734,
      "learning_rate": 1.997738636752536e-06,
      "loss": 0.1773,
      "num_input_tokens_seen": 445312,
      "step": 905
    },
    {
      "epoch": 0.12010030355021777,
      "grad_norm": 0.8630380630493164,
      "learning_rate": 1.9975811602093624e-06,
      "loss": 0.014,
      "num_input_tokens_seen": 447680,
      "step": 910
    },
    {
      "epoch": 0.12076019532796621,
      "grad_norm": 53.049163818359375,
      "learning_rate": 1.9974183901234984e-06,
      "loss": 0.2574,
      "num_input_tokens_seen": 450368,
      "step": 915
    },
    {
      "epoch": 0.12142008710571467,
      "grad_norm": 44.74892044067383,
      "learning_rate": 1.997250327358664e-06,
      "loss": 0.1599,
      "num_input_tokens_seen": 452800,
      "step": 920
    },
    {
      "epoch": 0.12207997888346311,
      "grad_norm": 45.7139892578125,
      "learning_rate": 1.997076972806664e-06,
      "loss": 0.023,
      "num_input_tokens_seen": 455744,
      "step": 925
    },
    {
      "epoch": 0.12273987066121156,
      "grad_norm": 35.15928268432617,
      "learning_rate": 1.9968983273873827e-06,
      "loss": 0.201,
      "num_input_tokens_seen": 458176,
      "step": 930
    },
    {
      "epoch": 0.12339976243896002,
      "grad_norm": 0.17246577143669128,
      "learning_rate": 1.99671439204878e-06,
      "loss": 0.0808,
      "num_input_tokens_seen": 460480,
      "step": 935
    },
    {
      "epoch": 0.12405965421670846,
      "grad_norm": 1.2687770128250122,
      "learning_rate": 1.9965251677668873e-06,
      "loss": 0.1766,
      "num_input_tokens_seen": 462656,
      "step": 940
    },
    {
      "epoch": 0.1247195459944569,
      "grad_norm": 24.687292098999023,
      "learning_rate": 1.9963306555458e-06,
      "loss": 0.3062,
      "num_input_tokens_seen": 465344,
      "step": 945
    },
    {
      "epoch": 0.12537943777220537,
      "grad_norm": 252.84188842773438,
      "learning_rate": 1.9961308564176723e-06,
      "loss": 0.4272,
      "num_input_tokens_seen": 467712,
      "step": 950
    },
    {
      "epoch": 0.1260393295499538,
      "grad_norm": 36.122955322265625,
      "learning_rate": 1.9959257714427147e-06,
      "loss": 0.223,
      "num_input_tokens_seen": 470080,
      "step": 955
    },
    {
      "epoch": 0.12669922132770225,
      "grad_norm": 11.395262718200684,
      "learning_rate": 1.995715401709186e-06,
      "loss": 0.1762,
      "num_input_tokens_seen": 472512,
      "step": 960
    },
    {
      "epoch": 0.1273591131054507,
      "grad_norm": 25.220579147338867,
      "learning_rate": 1.995499748333387e-06,
      "loss": 0.1271,
      "num_input_tokens_seen": 474752,
      "step": 965
    },
    {
      "epoch": 0.12801900488319914,
      "grad_norm": 46.467193603515625,
      "learning_rate": 1.9952788124596555e-06,
      "loss": 0.1233,
      "num_input_tokens_seen": 477440,
      "step": 970
    },
    {
      "epoch": 0.12867889666094762,
      "grad_norm": 19.14841651916504,
      "learning_rate": 1.9950525952603617e-06,
      "loss": 0.1272,
      "num_input_tokens_seen": 480000,
      "step": 975
    },
    {
      "epoch": 0.12933878843869606,
      "grad_norm": 51.99488067626953,
      "learning_rate": 1.994821097935899e-06,
      "loss": 0.1456,
      "num_input_tokens_seen": 482368,
      "step": 980
    },
    {
      "epoch": 0.1299986802164445,
      "grad_norm": 1.0634769201278687,
      "learning_rate": 1.9945843217146804e-06,
      "loss": 0.0077,
      "num_input_tokens_seen": 484544,
      "step": 985
    },
    {
      "epoch": 0.13065857199419295,
      "grad_norm": 0.03618798032402992,
      "learning_rate": 1.9943422678531293e-06,
      "loss": 0.0852,
      "num_input_tokens_seen": 486720,
      "step": 990
    },
    {
      "epoch": 0.1313184637719414,
      "grad_norm": 0.015866678208112717,
      "learning_rate": 1.994094937635675e-06,
      "loss": 0.0741,
      "num_input_tokens_seen": 489344,
      "step": 995
    },
    {
      "epoch": 0.13197835554968984,
      "grad_norm": 197.39834594726562,
      "learning_rate": 1.9938423323747457e-06,
      "loss": 0.2262,
      "num_input_tokens_seen": 491776,
      "step": 1000
    },
    {
      "epoch": 0.1326382473274383,
      "grad_norm": 147.0883026123047,
      "learning_rate": 1.99358445341076e-06,
      "loss": 0.1184,
      "num_input_tokens_seen": 493952,
      "step": 1005
    },
    {
      "epoch": 0.13329813910518676,
      "grad_norm": 78.60313415527344,
      "learning_rate": 1.993321302112121e-06,
      "loss": 0.2204,
      "num_input_tokens_seen": 496320,
      "step": 1010
    },
    {
      "epoch": 0.1339580308829352,
      "grad_norm": 6.261760711669922,
      "learning_rate": 1.993052879875209e-06,
      "loss": 0.0399,
      "num_input_tokens_seen": 498496,
      "step": 1015
    },
    {
      "epoch": 0.13461792266068365,
      "grad_norm": 131.00633239746094,
      "learning_rate": 1.992779188124374e-06,
      "loss": 0.2953,
      "num_input_tokens_seen": 501056,
      "step": 1020
    },
    {
      "epoch": 0.1352778144384321,
      "grad_norm": 81.73246002197266,
      "learning_rate": 1.992500228311928e-06,
      "loss": 0.1176,
      "num_input_tokens_seen": 503296,
      "step": 1025
    },
    {
      "epoch": 0.13593770621618054,
      "grad_norm": 24.69017219543457,
      "learning_rate": 1.9922160019181372e-06,
      "loss": 0.2695,
      "num_input_tokens_seen": 505856,
      "step": 1030
    },
    {
      "epoch": 0.13659759799392898,
      "grad_norm": 0.1977771818637848,
      "learning_rate": 1.9919265104512138e-06,
      "loss": 0.0981,
      "num_input_tokens_seen": 508416,
      "step": 1035
    },
    {
      "epoch": 0.13725748977167745,
      "grad_norm": 66.1187515258789,
      "learning_rate": 1.9916317554473094e-06,
      "loss": 0.2986,
      "num_input_tokens_seen": 511040,
      "step": 1040
    },
    {
      "epoch": 0.1379173815494259,
      "grad_norm": 67.56658935546875,
      "learning_rate": 1.9913317384705052e-06,
      "loss": 0.115,
      "num_input_tokens_seen": 513216,
      "step": 1045
    },
    {
      "epoch": 0.13857727332717434,
      "grad_norm": 32.97856903076172,
      "learning_rate": 1.991026461112805e-06,
      "loss": 0.0893,
      "num_input_tokens_seen": 515456,
      "step": 1050
    },
    {
      "epoch": 0.1392371651049228,
      "grad_norm": 2.204728841781616,
      "learning_rate": 1.9907159249941257e-06,
      "loss": 0.1591,
      "num_input_tokens_seen": 517824,
      "step": 1055
    },
    {
      "epoch": 0.13989705688267123,
      "grad_norm": 0.19428718090057373,
      "learning_rate": 1.990400131762289e-06,
      "loss": 0.0627,
      "num_input_tokens_seen": 520320,
      "step": 1060
    },
    {
      "epoch": 0.14055694866041968,
      "grad_norm": 189.14625549316406,
      "learning_rate": 1.9900790830930134e-06,
      "loss": 0.1068,
      "num_input_tokens_seen": 522752,
      "step": 1065
    },
    {
      "epoch": 0.14121684043816815,
      "grad_norm": 39.105308532714844,
      "learning_rate": 1.9897527806899047e-06,
      "loss": 0.1066,
      "num_input_tokens_seen": 525376,
      "step": 1070
    },
    {
      "epoch": 0.1418767322159166,
      "grad_norm": 0.7390586733818054,
      "learning_rate": 1.9894212262844465e-06,
      "loss": 0.2544,
      "num_input_tokens_seen": 527808,
      "step": 1075
    },
    {
      "epoch": 0.14253662399366504,
      "grad_norm": 0.13609175384044647,
      "learning_rate": 1.989084421635992e-06,
      "loss": 0.1578,
      "num_input_tokens_seen": 530304,
      "step": 1080
    },
    {
      "epoch": 0.14319651577141349,
      "grad_norm": 13.118515968322754,
      "learning_rate": 1.988742368531754e-06,
      "loss": 0.2733,
      "num_input_tokens_seen": 532480,
      "step": 1085
    },
    {
      "epoch": 0.14385640754916193,
      "grad_norm": 2.5871739387512207,
      "learning_rate": 1.9883950687867947e-06,
      "loss": 0.0745,
      "num_input_tokens_seen": 535168,
      "step": 1090
    },
    {
      "epoch": 0.14451629932691037,
      "grad_norm": 127.09754180908203,
      "learning_rate": 1.9880425242440187e-06,
      "loss": 0.1377,
      "num_input_tokens_seen": 537600,
      "step": 1095
    },
    {
      "epoch": 0.14517619110465885,
      "grad_norm": 485.19842529296875,
      "learning_rate": 1.9876847367741607e-06,
      "loss": 0.0585,
      "num_input_tokens_seen": 540096,
      "step": 1100
    },
    {
      "epoch": 0.1458360828824073,
      "grad_norm": 0.17331336438655853,
      "learning_rate": 1.987321708275776e-06,
      "loss": 0.1719,
      "num_input_tokens_seen": 542592,
      "step": 1105
    },
    {
      "epoch": 0.14649597466015574,
      "grad_norm": 0.26521632075309753,
      "learning_rate": 1.986953440675231e-06,
      "loss": 0.0895,
      "num_input_tokens_seen": 544960,
      "step": 1110
    },
    {
      "epoch": 0.14715586643790418,
      "grad_norm": 0.27949467301368713,
      "learning_rate": 1.9865799359266925e-06,
      "loss": 0.1572,
      "num_input_tokens_seen": 547136,
      "step": 1115
    },
    {
      "epoch": 0.14781575821565263,
      "grad_norm": 159.8411102294922,
      "learning_rate": 1.986201196012118e-06,
      "loss": 0.1272,
      "num_input_tokens_seen": 549440,
      "step": 1120
    },
    {
      "epoch": 0.14847564999340107,
      "grad_norm": 87.8428726196289,
      "learning_rate": 1.985817222941245e-06,
      "loss": 0.2212,
      "num_input_tokens_seen": 552064,
      "step": 1125
    },
    {
      "epoch": 0.14913554177114954,
      "grad_norm": 14.358078956604004,
      "learning_rate": 1.9854280187515794e-06,
      "loss": 0.0754,
      "num_input_tokens_seen": 554432,
      "step": 1130
    },
    {
      "epoch": 0.149795433548898,
      "grad_norm": 1.0141674280166626,
      "learning_rate": 1.985033585508386e-06,
      "loss": 0.0564,
      "num_input_tokens_seen": 556800,
      "step": 1135
    },
    {
      "epoch": 0.15005939025999737,
      "eval_loss": 0.16564705967903137,
      "eval_runtime": 7.8399,
      "eval_samples_per_second": 859.067,
      "eval_steps_per_second": 107.399,
      "num_input_tokens_seen": 557824,
      "step": 1137
    },
    {
      "epoch": 0.15045532532664643,
      "grad_norm": 257.7950744628906,
      "learning_rate": 1.9846339253046766e-06,
      "loss": 0.5119,
      "num_input_tokens_seen": 559296,
      "step": 1140
    },
    {
      "epoch": 0.15111521710439488,
      "grad_norm": 31.520845413208008,
      "learning_rate": 1.984229040261199e-06,
      "loss": 0.163,
      "num_input_tokens_seen": 562112,
      "step": 1145
    },
    {
      "epoch": 0.15177510888214332,
      "grad_norm": 0.35355716943740845,
      "learning_rate": 1.9838189325264263e-06,
      "loss": 0.1812,
      "num_input_tokens_seen": 564288,
      "step": 1150
    },
    {
      "epoch": 0.15243500065989177,
      "grad_norm": 70.49393463134766,
      "learning_rate": 1.983403604276546e-06,
      "loss": 0.033,
      "num_input_tokens_seen": 566848,
      "step": 1155
    },
    {
      "epoch": 0.15309489243764024,
      "grad_norm": 18.635469436645508,
      "learning_rate": 1.9829830577154457e-06,
      "loss": 0.3099,
      "num_input_tokens_seen": 569152,
      "step": 1160
    },
    {
      "epoch": 0.15375478421538868,
      "grad_norm": 34.90991973876953,
      "learning_rate": 1.982557295074705e-06,
      "loss": 0.0496,
      "num_input_tokens_seen": 571456,
      "step": 1165
    },
    {
      "epoch": 0.15441467599313713,
      "grad_norm": 0.17582400143146515,
      "learning_rate": 1.982126318613581e-06,
      "loss": 0.0902,
      "num_input_tokens_seen": 573824,
      "step": 1170
    },
    {
      "epoch": 0.15507456777088557,
      "grad_norm": 230.71153259277344,
      "learning_rate": 1.9816901306189977e-06,
      "loss": 0.0556,
      "num_input_tokens_seen": 576128,
      "step": 1175
    },
    {
      "epoch": 0.15573445954863402,
      "grad_norm": 0.5229840278625488,
      "learning_rate": 1.9812487334055342e-06,
      "loss": 0.1048,
      "num_input_tokens_seen": 578432,
      "step": 1180
    },
    {
      "epoch": 0.15639435132638246,
      "grad_norm": 0.39622417092323303,
      "learning_rate": 1.98080212931541e-06,
      "loss": 0.1522,
      "num_input_tokens_seen": 580736,
      "step": 1185
    },
    {
      "epoch": 0.15705424310413094,
      "grad_norm": 0.09377877414226532,
      "learning_rate": 1.980350320718476e-06,
      "loss": 0.0609,
      "num_input_tokens_seen": 583040,
      "step": 1190
    },
    {
      "epoch": 0.15771413488187938,
      "grad_norm": 1.1462613344192505,
      "learning_rate": 1.9798933100121985e-06,
      "loss": 0.0032,
      "num_input_tokens_seen": 585344,
      "step": 1195
    },
    {
      "epoch": 0.15837402665962783,
      "grad_norm": 1.4526684284210205,
      "learning_rate": 1.97943109962165e-06,
      "loss": 0.0896,
      "num_input_tokens_seen": 587904,
      "step": 1200
    },
    {
      "epoch": 0.15903391843737627,
      "grad_norm": 173.52330017089844,
      "learning_rate": 1.978963691999493e-06,
      "loss": 0.1423,
      "num_input_tokens_seen": 590208,
      "step": 1205
    },
    {
      "epoch": 0.15969381021512472,
      "grad_norm": 126.84811401367188,
      "learning_rate": 1.978491089625969e-06,
      "loss": 0.2118,
      "num_input_tokens_seen": 592512,
      "step": 1210
    },
    {
      "epoch": 0.16035370199287316,
      "grad_norm": 222.47836303710938,
      "learning_rate": 1.9780132950088854e-06,
      "loss": 0.2427,
      "num_input_tokens_seen": 595072,
      "step": 1215
    },
    {
      "epoch": 0.1610135937706216,
      "grad_norm": 19.530790328979492,
      "learning_rate": 1.9775303106836e-06,
      "loss": 0.3134,
      "num_input_tokens_seen": 597632,
      "step": 1220
    },
    {
      "epoch": 0.16167348554837008,
      "grad_norm": 0.11579433083534241,
      "learning_rate": 1.977042139213011e-06,
      "loss": 0.117,
      "num_input_tokens_seen": 600192,
      "step": 1225
    },
    {
      "epoch": 0.16233337732611852,
      "grad_norm": 0.30019301176071167,
      "learning_rate": 1.9765487831875404e-06,
      "loss": 0.1205,
      "num_input_tokens_seen": 602304,
      "step": 1230
    },
    {
      "epoch": 0.16299326910386697,
      "grad_norm": 1.5647317171096802,
      "learning_rate": 1.9760502452251217e-06,
      "loss": 0.1485,
      "num_input_tokens_seen": 604608,
      "step": 1235
    },
    {
      "epoch": 0.1636531608816154,
      "grad_norm": 12.981266021728516,
      "learning_rate": 1.975546527971186e-06,
      "loss": 0.1371,
      "num_input_tokens_seen": 606976,
      "step": 1240
    },
    {
      "epoch": 0.16431305265936386,
      "grad_norm": 7.285887241363525,
      "learning_rate": 1.9750376340986472e-06,
      "loss": 0.0801,
      "num_input_tokens_seen": 609600,
      "step": 1245
    },
    {
      "epoch": 0.1649729444371123,
      "grad_norm": 49.19084167480469,
      "learning_rate": 1.974523566307889e-06,
      "loss": 0.2009,
      "num_input_tokens_seen": 611840,
      "step": 1250
    },
    {
      "epoch": 0.16563283621486077,
      "grad_norm": 110.00751495361328,
      "learning_rate": 1.9740043273267487e-06,
      "loss": 0.1902,
      "num_input_tokens_seen": 614528,
      "step": 1255
    },
    {
      "epoch": 0.16629272799260922,
      "grad_norm": 0.1264021098613739,
      "learning_rate": 1.973479919910505e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 617024,
      "step": 1260
    },
    {
      "epoch": 0.16695261977035766,
      "grad_norm": 58.373043060302734,
      "learning_rate": 1.972950346841862e-06,
      "loss": 0.1153,
      "num_input_tokens_seen": 619392,
      "step": 1265
    },
    {
      "epoch": 0.1676125115481061,
      "grad_norm": 7.883070468902588,
      "learning_rate": 1.972415610930934e-06,
      "loss": 0.1353,
      "num_input_tokens_seen": 621888,
      "step": 1270
    },
    {
      "epoch": 0.16827240332585455,
      "grad_norm": 60.600013732910156,
      "learning_rate": 1.9718757150152324e-06,
      "loss": 0.1096,
      "num_input_tokens_seen": 624192,
      "step": 1275
    },
    {
      "epoch": 0.168932295103603,
      "grad_norm": 5.226385116577148,
      "learning_rate": 1.9713306619596488e-06,
      "loss": 0.0658,
      "num_input_tokens_seen": 626624,
      "step": 1280
    },
    {
      "epoch": 0.16959218688135147,
      "grad_norm": 10.980446815490723,
      "learning_rate": 1.9707804546564407e-06,
      "loss": 0.0575,
      "num_input_tokens_seen": 628928,
      "step": 1285
    },
    {
      "epoch": 0.17025207865909991,
      "grad_norm": 0.39055877923965454,
      "learning_rate": 1.9702250960252164e-06,
      "loss": 0.1254,
      "num_input_tokens_seen": 631616,
      "step": 1290
    },
    {
      "epoch": 0.17091197043684836,
      "grad_norm": 0.10477243363857269,
      "learning_rate": 1.969664589012918e-06,
      "loss": 0.0568,
      "num_input_tokens_seen": 634112,
      "step": 1295
    },
    {
      "epoch": 0.1715718622145968,
      "grad_norm": 273.6374206542969,
      "learning_rate": 1.9690989365938077e-06,
      "loss": 0.1854,
      "num_input_tokens_seen": 636416,
      "step": 1300
    },
    {
      "epoch": 0.17223175399234525,
      "grad_norm": 0.025812866166234016,
      "learning_rate": 1.9685281417694513e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 638848,
      "step": 1305
    },
    {
      "epoch": 0.1728916457700937,
      "grad_norm": 1.0203826427459717,
      "learning_rate": 1.967952207568702e-06,
      "loss": 0.12,
      "num_input_tokens_seen": 641216,
      "step": 1310
    },
    {
      "epoch": 0.17355153754784217,
      "grad_norm": 1.9090206623077393,
      "learning_rate": 1.967371137047685e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 644032,
      "step": 1315
    },
    {
      "epoch": 0.1742114293255906,
      "grad_norm": 0.15419095754623413,
      "learning_rate": 1.966784933289778e-06,
      "loss": 0.0887,
      "num_input_tokens_seen": 646528,
      "step": 1320
    },
    {
      "epoch": 0.17487132110333906,
      "grad_norm": 0.09177876263856888,
      "learning_rate": 1.9661935994056014e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 649088,
      "step": 1325
    },
    {
      "epoch": 0.1755312128810875,
      "grad_norm": 0.9890398383140564,
      "learning_rate": 1.965597138532996e-06,
      "loss": 0.1035,
      "num_input_tokens_seen": 651520,
      "step": 1330
    },
    {
      "epoch": 0.17619110465883595,
      "grad_norm": 90.32373046875,
      "learning_rate": 1.964995553837009e-06,
      "loss": 0.0123,
      "num_input_tokens_seen": 654016,
      "step": 1335
    },
    {
      "epoch": 0.1768509964365844,
      "grad_norm": 0.014015772379934788,
      "learning_rate": 1.964388848509875e-06,
      "loss": 0.0924,
      "num_input_tokens_seen": 656320,
      "step": 1340
    },
    {
      "epoch": 0.17751088821433286,
      "grad_norm": 0.044904567301273346,
      "learning_rate": 1.9637770257710026e-06,
      "loss": 0.1818,
      "num_input_tokens_seen": 658880,
      "step": 1345
    },
    {
      "epoch": 0.1781707799920813,
      "grad_norm": 0.15751226246356964,
      "learning_rate": 1.9631600888669545e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 661184,
      "step": 1350
    },
    {
      "epoch": 0.17883067176982975,
      "grad_norm": 0.13751578330993652,
      "learning_rate": 1.962538041071431e-06,
      "loss": 0.0885,
      "num_input_tokens_seen": 663680,
      "step": 1355
    },
    {
      "epoch": 0.1794905635475782,
      "grad_norm": 164.34730529785156,
      "learning_rate": 1.961910885685253e-06,
      "loss": 0.0377,
      "num_input_tokens_seen": 666048,
      "step": 1360
    },
    {
      "epoch": 0.18015045532532664,
      "grad_norm": 1.029883623123169,
      "learning_rate": 1.9612786260363436e-06,
      "loss": 0.0748,
      "num_input_tokens_seen": 668480,
      "step": 1365
    },
    {
      "epoch": 0.1808103471030751,
      "grad_norm": 0.1113181859254837,
      "learning_rate": 1.9606412654797116e-06,
      "loss": 0.1292,
      "num_input_tokens_seen": 671488,
      "step": 1370
    },
    {
      "epoch": 0.18147023888082353,
      "grad_norm": 1.6563535928726196,
      "learning_rate": 1.9599988073974332e-06,
      "loss": 0.1406,
      "num_input_tokens_seen": 673920,
      "step": 1375
    },
    {
      "epoch": 0.182130130658572,
      "grad_norm": 0.45396727323532104,
      "learning_rate": 1.959351255198634e-06,
      "loss": 0.0929,
      "num_input_tokens_seen": 676416,
      "step": 1380
    },
    {
      "epoch": 0.18279002243632045,
      "grad_norm": 0.1408662647008896,
      "learning_rate": 1.9586986123194704e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 679040,
      "step": 1385
    },
    {
      "epoch": 0.1834499142140689,
      "grad_norm": 0.10570546239614487,
      "learning_rate": 1.958040882223112e-06,
      "loss": 0.1902,
      "num_input_tokens_seen": 681920,
      "step": 1390
    },
    {
      "epoch": 0.18410980599181734,
      "grad_norm": 0.1604253500699997,
      "learning_rate": 1.9573780683997235e-06,
      "loss": 0.048,
      "num_input_tokens_seen": 684416,
      "step": 1395
    },
    {
      "epoch": 0.18476969776956578,
      "grad_norm": 33.932865142822266,
      "learning_rate": 1.956710174366445e-06,
      "loss": 0.0828,
      "num_input_tokens_seen": 686976,
      "step": 1400
    },
    {
      "epoch": 0.18542958954731423,
      "grad_norm": 0.08173060417175293,
      "learning_rate": 1.9560372036673764e-06,
      "loss": 0.2272,
      "num_input_tokens_seen": 689408,
      "step": 1405
    },
    {
      "epoch": 0.1860894813250627,
      "grad_norm": 0.028357749804854393,
      "learning_rate": 1.955359159873553e-06,
      "loss": 0.0649,
      "num_input_tokens_seen": 691712,
      "step": 1410
    },
    {
      "epoch": 0.18674937310281114,
      "grad_norm": 0.021360548213124275,
      "learning_rate": 1.954676046582932e-06,
      "loss": 0.066,
      "num_input_tokens_seen": 694080,
      "step": 1415
    },
    {
      "epoch": 0.1874092648805596,
      "grad_norm": 80.15442657470703,
      "learning_rate": 1.9539878674203706e-06,
      "loss": 0.2175,
      "num_input_tokens_seen": 696640,
      "step": 1420
    },
    {
      "epoch": 0.18806915665830803,
      "grad_norm": 0.01937083899974823,
      "learning_rate": 1.9532946260376076e-06,
      "loss": 0.0003,
      "num_input_tokens_seen": 699136,
      "step": 1425
    },
    {
      "epoch": 0.18872904843605648,
      "grad_norm": 18.852214813232422,
      "learning_rate": 1.952596326113244e-06,
      "loss": 0.4092,
      "num_input_tokens_seen": 701696,
      "step": 1430
    },
    {
      "epoch": 0.18938894021380492,
      "grad_norm": 12.444201469421387,
      "learning_rate": 1.9518929713527226e-06,
      "loss": 0.2349,
      "num_input_tokens_seen": 704384,
      "step": 1435
    },
    {
      "epoch": 0.1900488319915534,
      "grad_norm": 2.0845847129821777,
      "learning_rate": 1.9511845654883097e-06,
      "loss": 0.0082,
      "num_input_tokens_seen": 706560,
      "step": 1440
    },
    {
      "epoch": 0.19070872376930184,
      "grad_norm": 0.3514378070831299,
      "learning_rate": 1.9504711122790754e-06,
      "loss": 0.1221,
      "num_input_tokens_seen": 709248,
      "step": 1445
    },
    {
      "epoch": 0.19136861554705029,
      "grad_norm": 79.33631896972656,
      "learning_rate": 1.949752615510871e-06,
      "loss": 0.08,
      "num_input_tokens_seen": 711296,
      "step": 1450
    },
    {
      "epoch": 0.19202850732479873,
      "grad_norm": 0.0773974359035492,
      "learning_rate": 1.949029078996313e-06,
      "loss": 0.0655,
      "num_input_tokens_seen": 713728,
      "step": 1455
    },
    {
      "epoch": 0.19268839910254718,
      "grad_norm": 15.809504508972168,
      "learning_rate": 1.9483005065747584e-06,
      "loss": 0.1876,
      "num_input_tokens_seen": 716224,
      "step": 1460
    },
    {
      "epoch": 0.19334829088029562,
      "grad_norm": 307.3660583496094,
      "learning_rate": 1.947566902112289e-06,
      "loss": 0.3791,
      "num_input_tokens_seen": 718528,
      "step": 1465
    },
    {
      "epoch": 0.1940081826580441,
      "grad_norm": 17.658618927001953,
      "learning_rate": 1.9468282695016863e-06,
      "loss": 0.2271,
      "num_input_tokens_seen": 720960,
      "step": 1470
    },
    {
      "epoch": 0.19466807443579254,
      "grad_norm": 1.276165246963501,
      "learning_rate": 1.946084612662415e-06,
      "loss": 0.1805,
      "num_input_tokens_seen": 723200,
      "step": 1475
    },
    {
      "epoch": 0.19532796621354098,
      "grad_norm": 104.93605041503906,
      "learning_rate": 1.9453359355405987e-06,
      "loss": 0.2018,
      "num_input_tokens_seen": 725888,
      "step": 1480
    },
    {
      "epoch": 0.19598785799128943,
      "grad_norm": 0.48601382970809937,
      "learning_rate": 1.944582242109002e-06,
      "loss": 0.0364,
      "num_input_tokens_seen": 728256,
      "step": 1485
    },
    {
      "epoch": 0.19664774976903787,
      "grad_norm": 24.217121124267578,
      "learning_rate": 1.943823536367006e-06,
      "loss": 0.1441,
      "num_input_tokens_seen": 730688,
      "step": 1490
    },
    {
      "epoch": 0.19730764154678632,
      "grad_norm": 1.9565227031707764,
      "learning_rate": 1.9430598223405913e-06,
      "loss": 0.1269,
      "num_input_tokens_seen": 732992,
      "step": 1495
    },
    {
      "epoch": 0.1979675333245348,
      "grad_norm": 1.4035013914108276,
      "learning_rate": 1.9422911040823125e-06,
      "loss": 0.1163,
      "num_input_tokens_seen": 735424,
      "step": 1500
    },
    {
      "epoch": 0.19862742510228323,
      "grad_norm": 28.88270378112793,
      "learning_rate": 1.941517385671279e-06,
      "loss": 0.1995,
      "num_input_tokens_seen": 737664,
      "step": 1505
    },
    {
      "epoch": 0.19928731688003168,
      "grad_norm": 0.40199992060661316,
      "learning_rate": 1.940738671213134e-06,
      "loss": 0.0888,
      "num_input_tokens_seen": 740096,
      "step": 1510
    },
    {
      "epoch": 0.19994720865778012,
      "grad_norm": 104.9893798828125,
      "learning_rate": 1.93995496484003e-06,
      "loss": 0.1883,
      "num_input_tokens_seen": 742912,
      "step": 1515
    },
    {
      "epoch": 0.2000791870133298,
      "eval_loss": 0.10974650084972382,
      "eval_runtime": 7.9505,
      "eval_samples_per_second": 847.113,
      "eval_steps_per_second": 105.905,
      "num_input_tokens_seen": 743424,
      "step": 1516
    },
    {
      "epoch": 0.20060710043552857,
      "grad_norm": 20.379575729370117,
      "learning_rate": 1.9391662707106092e-06,
      "loss": 0.1544,
      "num_input_tokens_seen": 745536,
      "step": 1520
    },
    {
      "epoch": 0.201266992213277,
      "grad_norm": 0.3665579557418823,
      "learning_rate": 1.9383725930099814e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 747968,
      "step": 1525
    },
    {
      "epoch": 0.20192688399102549,
      "grad_norm": 207.8310089111328,
      "learning_rate": 1.9375739359497e-06,
      "loss": 0.2261,
      "num_input_tokens_seen": 750464,
      "step": 1530
    },
    {
      "epoch": 0.20258677576877393,
      "grad_norm": 0.281808078289032,
      "learning_rate": 1.936770303767741e-06,
      "loss": 0.2511,
      "num_input_tokens_seen": 752896,
      "step": 1535
    },
    {
      "epoch": 0.20324666754652237,
      "grad_norm": 16.87381362915039,
      "learning_rate": 1.9359617007284815e-06,
      "loss": 0.1788,
      "num_input_tokens_seen": 755648,
      "step": 1540
    },
    {
      "epoch": 0.20390655932427082,
      "grad_norm": 18.007219314575195,
      "learning_rate": 1.9351481311226738e-06,
      "loss": 0.2098,
      "num_input_tokens_seen": 758144,
      "step": 1545
    },
    {
      "epoch": 0.20456645110201926,
      "grad_norm": 3.9224605560302734,
      "learning_rate": 1.934329599267426e-06,
      "loss": 0.0831,
      "num_input_tokens_seen": 760704,
      "step": 1550
    },
    {
      "epoch": 0.2052263428797677,
      "grad_norm": 4.013158798217773,
      "learning_rate": 1.933506109506178e-06,
      "loss": 0.0039,
      "num_input_tokens_seen": 763136,
      "step": 1555
    },
    {
      "epoch": 0.20588623465751615,
      "grad_norm": 8.997451782226562,
      "learning_rate": 1.9326776662086765e-06,
      "loss": 0.0802,
      "num_input_tokens_seen": 766016,
      "step": 1560
    },
    {
      "epoch": 0.20654612643526463,
      "grad_norm": 47.96493911743164,
      "learning_rate": 1.9318442737709565e-06,
      "loss": 0.2901,
      "num_input_tokens_seen": 768512,
      "step": 1565
    },
    {
      "epoch": 0.20720601821301307,
      "grad_norm": 0.5580891966819763,
      "learning_rate": 1.9310059366153116e-06,
      "loss": 0.1689,
      "num_input_tokens_seen": 770816,
      "step": 1570
    },
    {
      "epoch": 0.20786590999076152,
      "grad_norm": 3.661959648132324,
      "learning_rate": 1.930162659190277e-06,
      "loss": 0.248,
      "num_input_tokens_seen": 773312,
      "step": 1575
    },
    {
      "epoch": 0.20852580176850996,
      "grad_norm": 167.700927734375,
      "learning_rate": 1.9293144459706007e-06,
      "loss": 0.0688,
      "num_input_tokens_seen": 775680,
      "step": 1580
    },
    {
      "epoch": 0.2091856935462584,
      "grad_norm": 1471.7938232421875,
      "learning_rate": 1.928461301457223e-06,
      "loss": 0.1075,
      "num_input_tokens_seen": 778048,
      "step": 1585
    },
    {
      "epoch": 0.20984558532400685,
      "grad_norm": 0.19680626690387726,
      "learning_rate": 1.92760323017725e-06,
      "loss": 0.2795,
      "num_input_tokens_seen": 780672,
      "step": 1590
    },
    {
      "epoch": 0.21050547710175532,
      "grad_norm": 0.7471222281455994,
      "learning_rate": 1.9267402366839338e-06,
      "loss": 0.2294,
      "num_input_tokens_seen": 783360,
      "step": 1595
    },
    {
      "epoch": 0.21116536887950377,
      "grad_norm": 105.85987854003906,
      "learning_rate": 1.9258723255566433e-06,
      "loss": 0.1843,
      "num_input_tokens_seen": 785856,
      "step": 1600
    },
    {
      "epoch": 0.2118252606572522,
      "grad_norm": 0.5641558170318604,
      "learning_rate": 1.924999501400843e-06,
      "loss": 0.1321,
      "num_input_tokens_seen": 788480,
      "step": 1605
    },
    {
      "epoch": 0.21248515243500066,
      "grad_norm": 0.07420846074819565,
      "learning_rate": 1.924121768848068e-06,
      "loss": 0.1504,
      "num_input_tokens_seen": 791040,
      "step": 1610
    },
    {
      "epoch": 0.2131450442127491,
      "grad_norm": 0.1806056797504425,
      "learning_rate": 1.923239132555899e-06,
      "loss": 0.1462,
      "num_input_tokens_seen": 793600,
      "step": 1615
    },
    {
      "epoch": 0.21380493599049755,
      "grad_norm": 86.85829162597656,
      "learning_rate": 1.9223515972079378e-06,
      "loss": 0.1516,
      "num_input_tokens_seen": 795968,
      "step": 1620
    },
    {
      "epoch": 0.21446482776824602,
      "grad_norm": 79.25112915039062,
      "learning_rate": 1.9214591675137813e-06,
      "loss": 0.1019,
      "num_input_tokens_seen": 798272,
      "step": 1625
    },
    {
      "epoch": 0.21512471954599446,
      "grad_norm": 345.95281982421875,
      "learning_rate": 1.9205618482090003e-06,
      "loss": 0.1059,
      "num_input_tokens_seen": 801024,
      "step": 1630
    },
    {
      "epoch": 0.2157846113237429,
      "grad_norm": 24.057254791259766,
      "learning_rate": 1.91965964405511e-06,
      "loss": 0.1862,
      "num_input_tokens_seen": 803584,
      "step": 1635
    },
    {
      "epoch": 0.21644450310149135,
      "grad_norm": 159.59852600097656,
      "learning_rate": 1.9187525598395457e-06,
      "loss": 0.0438,
      "num_input_tokens_seen": 805952,
      "step": 1640
    },
    {
      "epoch": 0.2171043948792398,
      "grad_norm": 64.74698638916016,
      "learning_rate": 1.9178406003756396e-06,
      "loss": 0.1326,
      "num_input_tokens_seen": 808512,
      "step": 1645
    },
    {
      "epoch": 0.21776428665698824,
      "grad_norm": 79.24069213867188,
      "learning_rate": 1.9169237705025936e-06,
      "loss": 0.1071,
      "num_input_tokens_seen": 811136,
      "step": 1650
    },
    {
      "epoch": 0.21842417843473672,
      "grad_norm": 0.03825217857956886,
      "learning_rate": 1.9160020750854533e-06,
      "loss": 0.0347,
      "num_input_tokens_seen": 813376,
      "step": 1655
    },
    {
      "epoch": 0.21908407021248516,
      "grad_norm": 208.05564880371094,
      "learning_rate": 1.915075519015083e-06,
      "loss": 0.1731,
      "num_input_tokens_seen": 815872,
      "step": 1660
    },
    {
      "epoch": 0.2197439619902336,
      "grad_norm": 108.54914855957031,
      "learning_rate": 1.914144107208139e-06,
      "loss": 0.0142,
      "num_input_tokens_seen": 818240,
      "step": 1665
    },
    {
      "epoch": 0.22040385376798205,
      "grad_norm": 0.16779685020446777,
      "learning_rate": 1.913207844607045e-06,
      "loss": 0.0476,
      "num_input_tokens_seen": 820736,
      "step": 1670
    },
    {
      "epoch": 0.2210637455457305,
      "grad_norm": 33.12580108642578,
      "learning_rate": 1.912266736179964e-06,
      "loss": 0.3054,
      "num_input_tokens_seen": 823616,
      "step": 1675
    },
    {
      "epoch": 0.22172363732347894,
      "grad_norm": 13.499357223510742,
      "learning_rate": 1.9113207869207727e-06,
      "loss": 0.2016,
      "num_input_tokens_seen": 826112,
      "step": 1680
    },
    {
      "epoch": 0.2223835291012274,
      "grad_norm": 0.33417344093322754,
      "learning_rate": 1.9103700018490365e-06,
      "loss": 0.1342,
      "num_input_tokens_seen": 828672,
      "step": 1685
    },
    {
      "epoch": 0.22304342087897586,
      "grad_norm": 0.9366137385368347,
      "learning_rate": 1.9094143860099787e-06,
      "loss": 0.163,
      "num_input_tokens_seen": 831296,
      "step": 1690
    },
    {
      "epoch": 0.2237033126567243,
      "grad_norm": 101.39453887939453,
      "learning_rate": 1.9084539444744594e-06,
      "loss": 0.1052,
      "num_input_tokens_seen": 833856,
      "step": 1695
    },
    {
      "epoch": 0.22436320443447275,
      "grad_norm": 0.10196679830551147,
      "learning_rate": 1.907488682338944e-06,
      "loss": 0.1483,
      "num_input_tokens_seen": 836480,
      "step": 1700
    },
    {
      "epoch": 0.2250230962122212,
      "grad_norm": 32.67580032348633,
      "learning_rate": 1.9065186047254782e-06,
      "loss": 0.0834,
      "num_input_tokens_seen": 838976,
      "step": 1705
    },
    {
      "epoch": 0.22568298798996964,
      "grad_norm": 0.4758983552455902,
      "learning_rate": 1.9055437167816604e-06,
      "loss": 0.1598,
      "num_input_tokens_seen": 841728,
      "step": 1710
    },
    {
      "epoch": 0.22634287976771808,
      "grad_norm": 0.09636105597019196,
      "learning_rate": 1.9045640236806149e-06,
      "loss": 0.0044,
      "num_input_tokens_seen": 843968,
      "step": 1715
    },
    {
      "epoch": 0.22700277154546655,
      "grad_norm": 265.4130859375,
      "learning_rate": 1.903579530620963e-06,
      "loss": 0.4454,
      "num_input_tokens_seen": 846464,
      "step": 1720
    },
    {
      "epoch": 0.227662663323215,
      "grad_norm": 0.19204720854759216,
      "learning_rate": 1.9025902428267975e-06,
      "loss": 0.1272,
      "num_input_tokens_seen": 849088,
      "step": 1725
    },
    {
      "epoch": 0.22832255510096344,
      "grad_norm": 0.07899138331413269,
      "learning_rate": 1.901596165547653e-06,
      "loss": 0.0506,
      "num_input_tokens_seen": 851712,
      "step": 1730
    },
    {
      "epoch": 0.2289824468787119,
      "grad_norm": 0.2107008397579193,
      "learning_rate": 1.9005973040584796e-06,
      "loss": 0.1537,
      "num_input_tokens_seen": 854208,
      "step": 1735
    },
    {
      "epoch": 0.22964233865646033,
      "grad_norm": 0.8915808796882629,
      "learning_rate": 1.8995936636596138e-06,
      "loss": 0.1147,
      "num_input_tokens_seen": 856576,
      "step": 1740
    },
    {
      "epoch": 0.23030223043420878,
      "grad_norm": 0.37453988194465637,
      "learning_rate": 1.8985852496767504e-06,
      "loss": 0.1347,
      "num_input_tokens_seen": 859008,
      "step": 1745
    },
    {
      "epoch": 0.23096212221195725,
      "grad_norm": 13.8438720703125,
      "learning_rate": 1.897572067460916e-06,
      "loss": 0.146,
      "num_input_tokens_seen": 861440,
      "step": 1750
    },
    {
      "epoch": 0.2316220139897057,
      "grad_norm": 0.8792198300361633,
      "learning_rate": 1.8965541223884377e-06,
      "loss": 0.1079,
      "num_input_tokens_seen": 863936,
      "step": 1755
    },
    {
      "epoch": 0.23228190576745414,
      "grad_norm": 0.31022199988365173,
      "learning_rate": 1.8955314198609171e-06,
      "loss": 0.0547,
      "num_input_tokens_seen": 866176,
      "step": 1760
    },
    {
      "epoch": 0.23294179754520258,
      "grad_norm": 0.3442046344280243,
      "learning_rate": 1.8945039653052005e-06,
      "loss": 0.0747,
      "num_input_tokens_seen": 868480,
      "step": 1765
    },
    {
      "epoch": 0.23360168932295103,
      "grad_norm": 0.8551641702651978,
      "learning_rate": 1.8934717641733498e-06,
      "loss": 0.1454,
      "num_input_tokens_seen": 870976,
      "step": 1770
    },
    {
      "epoch": 0.23426158110069947,
      "grad_norm": 0.31109607219696045,
      "learning_rate": 1.8924348219426143e-06,
      "loss": 0.2544,
      "num_input_tokens_seen": 873088,
      "step": 1775
    },
    {
      "epoch": 0.23492147287844795,
      "grad_norm": 16.480606079101562,
      "learning_rate": 1.8913931441154016e-06,
      "loss": 0.2463,
      "num_input_tokens_seen": 875520,
      "step": 1780
    },
    {
      "epoch": 0.2355813646561964,
      "grad_norm": 45.99843978881836,
      "learning_rate": 1.8903467362192482e-06,
      "loss": 0.0807,
      "num_input_tokens_seen": 877632,
      "step": 1785
    },
    {
      "epoch": 0.23624125643394484,
      "grad_norm": 92.43214416503906,
      "learning_rate": 1.8892956038067895e-06,
      "loss": 0.1002,
      "num_input_tokens_seen": 880000,
      "step": 1790
    },
    {
      "epoch": 0.23690114821169328,
      "grad_norm": 0.566708505153656,
      "learning_rate": 1.8882397524557317e-06,
      "loss": 0.0187,
      "num_input_tokens_seen": 882176,
      "step": 1795
    },
    {
      "epoch": 0.23756103998944172,
      "grad_norm": 0.06913850456476212,
      "learning_rate": 1.8871791877688208e-06,
      "loss": 0.0792,
      "num_input_tokens_seen": 884800,
      "step": 1800
    },
    {
      "epoch": 0.23822093176719017,
      "grad_norm": 19.2426700592041,
      "learning_rate": 1.8861139153738143e-06,
      "loss": 0.1225,
      "num_input_tokens_seen": 887104,
      "step": 1805
    },
    {
      "epoch": 0.23888082354493864,
      "grad_norm": 153.66583251953125,
      "learning_rate": 1.8850439409234498e-06,
      "loss": 0.0388,
      "num_input_tokens_seen": 889408,
      "step": 1810
    },
    {
      "epoch": 0.2395407153226871,
      "grad_norm": 91.53621673583984,
      "learning_rate": 1.8839692700954161e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 891648,
      "step": 1815
    },
    {
      "epoch": 0.24020060710043553,
      "grad_norm": 456.29779052734375,
      "learning_rate": 1.8828899085923234e-06,
      "loss": 0.2359,
      "num_input_tokens_seen": 894208,
      "step": 1820
    },
    {
      "epoch": 0.24086049887818398,
      "grad_norm": 0.05232629179954529,
      "learning_rate": 1.881805862141671e-06,
      "loss": 0.0859,
      "num_input_tokens_seen": 896704,
      "step": 1825
    },
    {
      "epoch": 0.24152039065593242,
      "grad_norm": 0.039349839091300964,
      "learning_rate": 1.8807171364958196e-06,
      "loss": 0.0579,
      "num_input_tokens_seen": 899264,
      "step": 1830
    },
    {
      "epoch": 0.24218028243368087,
      "grad_norm": 0.08498682081699371,
      "learning_rate": 1.879623737431959e-06,
      "loss": 0.0186,
      "num_input_tokens_seen": 901760,
      "step": 1835
    },
    {
      "epoch": 0.24284017421142934,
      "grad_norm": 7.6966872215271,
      "learning_rate": 1.8785256707520778e-06,
      "loss": 0.1353,
      "num_input_tokens_seen": 903872,
      "step": 1840
    },
    {
      "epoch": 0.24350006598917778,
      "grad_norm": 0.09622366726398468,
      "learning_rate": 1.8774229422829325e-06,
      "loss": 0.0463,
      "num_input_tokens_seen": 906368,
      "step": 1845
    },
    {
      "epoch": 0.24415995776692623,
      "grad_norm": 0.03518239036202431,
      "learning_rate": 1.8763155578760181e-06,
      "loss": 0.0658,
      "num_input_tokens_seen": 908864,
      "step": 1850
    },
    {
      "epoch": 0.24481984954467467,
      "grad_norm": 0.2804614305496216,
      "learning_rate": 1.8752035234075336e-06,
      "loss": 0.163,
      "num_input_tokens_seen": 911040,
      "step": 1855
    },
    {
      "epoch": 0.24547974132242312,
      "grad_norm": 34.17547607421875,
      "learning_rate": 1.8740868447783554e-06,
      "loss": 0.1979,
      "num_input_tokens_seen": 913408,
      "step": 1860
    },
    {
      "epoch": 0.24613963310017156,
      "grad_norm": 65.69808197021484,
      "learning_rate": 1.8729655279140012e-06,
      "loss": 0.2382,
      "num_input_tokens_seen": 915968,
      "step": 1865
    },
    {
      "epoch": 0.24679952487792003,
      "grad_norm": 154.89785766601562,
      "learning_rate": 1.8718395787646029e-06,
      "loss": 0.1639,
      "num_input_tokens_seen": 918528,
      "step": 1870
    },
    {
      "epoch": 0.24745941665566848,
      "grad_norm": 0.06571167707443237,
      "learning_rate": 1.870709003304872e-06,
      "loss": 0.0099,
      "num_input_tokens_seen": 921152,
      "step": 1875
    },
    {
      "epoch": 0.24811930843341692,
      "grad_norm": 0.09134677052497864,
      "learning_rate": 1.8695738075340693e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 923520,
      "step": 1880
    },
    {
      "epoch": 0.24877920021116537,
      "grad_norm": 0.3240724503993988,
      "learning_rate": 1.8684339974759723e-06,
      "loss": 0.0253,
      "num_input_tokens_seen": 925888,
      "step": 1885
    },
    {
      "epoch": 0.2494390919889138,
      "grad_norm": 18.70522689819336,
      "learning_rate": 1.8672895791788445e-06,
      "loss": 0.0925,
      "num_input_tokens_seen": 928704,
      "step": 1890
    },
    {
      "epoch": 0.2500989837666623,
      "grad_norm": 134.16648864746094,
      "learning_rate": 1.8661405587154017e-06,
      "loss": 0.2857,
      "num_input_tokens_seen": 930944,
      "step": 1895
    },
    {
      "epoch": 0.2500989837666623,
      "eval_loss": 0.13187885284423828,
      "eval_runtime": 7.8602,
      "eval_samples_per_second": 856.843,
      "eval_steps_per_second": 107.121,
      "num_input_tokens_seen": 930944,
      "step": 1895
    },
    {
      "epoch": 0.25075887554441073,
      "grad_norm": 169.28016662597656,
      "learning_rate": 1.8649869421827808e-06,
      "loss": 0.3692,
      "num_input_tokens_seen": 933376,
      "step": 1900
    },
    {
      "epoch": 0.2514187673221592,
      "grad_norm": 2.5725350379943848,
      "learning_rate": 1.863828735702507e-06,
      "loss": 0.0896,
      "num_input_tokens_seen": 936000,
      "step": 1905
    },
    {
      "epoch": 0.2520786590999076,
      "grad_norm": 99.35297393798828,
      "learning_rate": 1.862665945420462e-06,
      "loss": 0.1121,
      "num_input_tokens_seen": 938432,
      "step": 1910
    },
    {
      "epoch": 0.25273855087765607,
      "grad_norm": 29.4595947265625,
      "learning_rate": 1.8614985775068498e-06,
      "loss": 0.1862,
      "num_input_tokens_seen": 941312,
      "step": 1915
    },
    {
      "epoch": 0.2533984426554045,
      "grad_norm": 2.1499125957489014,
      "learning_rate": 1.860326638156167e-06,
      "loss": 0.0727,
      "num_input_tokens_seen": 943488,
      "step": 1920
    },
    {
      "epoch": 0.25405833443315295,
      "grad_norm": 32.39081573486328,
      "learning_rate": 1.8591501335871653e-06,
      "loss": 0.083,
      "num_input_tokens_seen": 945856,
      "step": 1925
    },
    {
      "epoch": 0.2547182262109014,
      "grad_norm": 132.70303344726562,
      "learning_rate": 1.857969070042824e-06,
      "loss": 0.2212,
      "num_input_tokens_seen": 948352,
      "step": 1930
    },
    {
      "epoch": 0.25537811798864984,
      "grad_norm": 0.30987945199012756,
      "learning_rate": 1.8567834537903116e-06,
      "loss": 0.08,
      "num_input_tokens_seen": 950976,
      "step": 1935
    },
    {
      "epoch": 0.2560380097663983,
      "grad_norm": 26.0654296875,
      "learning_rate": 1.8555932911209565e-06,
      "loss": 0.1289,
      "num_input_tokens_seen": 953216,
      "step": 1940
    },
    {
      "epoch": 0.25669790154414673,
      "grad_norm": 0.553709089756012,
      "learning_rate": 1.8543985883502119e-06,
      "loss": 0.0221,
      "num_input_tokens_seen": 955648,
      "step": 1945
    },
    {
      "epoch": 0.25735779332189523,
      "grad_norm": 0.7263230681419373,
      "learning_rate": 1.8531993518176216e-06,
      "loss": 0.0906,
      "num_input_tokens_seen": 957888,
      "step": 1950
    },
    {
      "epoch": 0.2580176850996437,
      "grad_norm": 19.012561798095703,
      "learning_rate": 1.8519955878867889e-06,
      "loss": 0.3163,
      "num_input_tokens_seen": 960128,
      "step": 1955
    },
    {
      "epoch": 0.2586775768773921,
      "grad_norm": 0.9245926141738892,
      "learning_rate": 1.8507873029453392e-06,
      "loss": 0.1495,
      "num_input_tokens_seen": 962496,
      "step": 1960
    },
    {
      "epoch": 0.25933746865514057,
      "grad_norm": 16.104307174682617,
      "learning_rate": 1.8495745034048896e-06,
      "loss": 0.1628,
      "num_input_tokens_seen": 965120,
      "step": 1965
    },
    {
      "epoch": 0.259997360432889,
      "grad_norm": 0.15961866080760956,
      "learning_rate": 1.8483571957010127e-06,
      "loss": 0.045,
      "num_input_tokens_seen": 967616,
      "step": 1970
    },
    {
      "epoch": 0.26065725221063746,
      "grad_norm": 0.15467530488967896,
      "learning_rate": 1.8471353862932035e-06,
      "loss": 0.0574,
      "num_input_tokens_seen": 970240,
      "step": 1975
    },
    {
      "epoch": 0.2613171439883859,
      "grad_norm": 5.751570224761963,
      "learning_rate": 1.8459090816648444e-06,
      "loss": 0.0719,
      "num_input_tokens_seen": 972544,
      "step": 1980
    },
    {
      "epoch": 0.26197703576613435,
      "grad_norm": 0.7070101499557495,
      "learning_rate": 1.8446782883231713e-06,
      "loss": 0.2544,
      "num_input_tokens_seen": 974912,
      "step": 1985
    },
    {
      "epoch": 0.2626369275438828,
      "grad_norm": 21.680789947509766,
      "learning_rate": 1.8434430127992387e-06,
      "loss": 0.4055,
      "num_input_tokens_seen": 977088,
      "step": 1990
    },
    {
      "epoch": 0.26329681932163124,
      "grad_norm": 218.16334533691406,
      "learning_rate": 1.8422032616478857e-06,
      "loss": 0.1845,
      "num_input_tokens_seen": 979648,
      "step": 1995
    },
    {
      "epoch": 0.2639567110993797,
      "grad_norm": 2.263739585876465,
      "learning_rate": 1.8409590414477001e-06,
      "loss": 0.1184,
      "num_input_tokens_seen": 982336,
      "step": 2000
    },
    {
      "epoch": 0.2646166028771281,
      "grad_norm": 4.402483940124512,
      "learning_rate": 1.839710358800985e-06,
      "loss": 0.0071,
      "num_input_tokens_seen": 984768,
      "step": 2005
    },
    {
      "epoch": 0.2652764946548766,
      "grad_norm": 1.3311794996261597,
      "learning_rate": 1.8384572203337224e-06,
      "loss": 0.049,
      "num_input_tokens_seen": 987136,
      "step": 2010
    },
    {
      "epoch": 0.26593638643262507,
      "grad_norm": 0.14126527309417725,
      "learning_rate": 1.837199632695538e-06,
      "loss": 0.1358,
      "num_input_tokens_seen": 989824,
      "step": 2015
    },
    {
      "epoch": 0.2665962782103735,
      "grad_norm": 113.8551254272461,
      "learning_rate": 1.8359376025596682e-06,
      "loss": 0.2944,
      "num_input_tokens_seen": 992064,
      "step": 2020
    },
    {
      "epoch": 0.26725616998812196,
      "grad_norm": 35.01311111450195,
      "learning_rate": 1.8346711366229215e-06,
      "loss": 0.1288,
      "num_input_tokens_seen": 994368,
      "step": 2025
    },
    {
      "epoch": 0.2679160617658704,
      "grad_norm": 17.396068572998047,
      "learning_rate": 1.8334002416056442e-06,
      "loss": 0.181,
      "num_input_tokens_seen": 996864,
      "step": 2030
    },
    {
      "epoch": 0.26857595354361885,
      "grad_norm": 0.1548883020877838,
      "learning_rate": 1.8321249242516865e-06,
      "loss": 0.2285,
      "num_input_tokens_seen": 999360,
      "step": 2035
    },
    {
      "epoch": 0.2692358453213673,
      "grad_norm": 51.23111343383789,
      "learning_rate": 1.8308451913283638e-06,
      "loss": 0.0873,
      "num_input_tokens_seen": 1001920,
      "step": 2040
    },
    {
      "epoch": 0.26989573709911574,
      "grad_norm": 0.2864314913749695,
      "learning_rate": 1.8295610496264229e-06,
      "loss": 0.0305,
      "num_input_tokens_seen": 1004224,
      "step": 2045
    },
    {
      "epoch": 0.2705556288768642,
      "grad_norm": 0.4007130563259125,
      "learning_rate": 1.828272505960005e-06,
      "loss": 0.0393,
      "num_input_tokens_seen": 1006528,
      "step": 2050
    },
    {
      "epoch": 0.27121552065461263,
      "grad_norm": 0.32620009779930115,
      "learning_rate": 1.8269795671666098e-06,
      "loss": 0.1813,
      "num_input_tokens_seen": 1008896,
      "step": 2055
    },
    {
      "epoch": 0.2718754124323611,
      "grad_norm": 0.6125988960266113,
      "learning_rate": 1.8256822401070591e-06,
      "loss": 0.1234,
      "num_input_tokens_seen": 1011648,
      "step": 2060
    },
    {
      "epoch": 0.2725353042101095,
      "grad_norm": 31.247455596923828,
      "learning_rate": 1.8243805316654611e-06,
      "loss": 0.0522,
      "num_input_tokens_seen": 1014208,
      "step": 2065
    },
    {
      "epoch": 0.27319519598785796,
      "grad_norm": 155.57479858398438,
      "learning_rate": 1.823074448749172e-06,
      "loss": 0.1725,
      "num_input_tokens_seen": 1016640,
      "step": 2070
    },
    {
      "epoch": 0.27385508776560646,
      "grad_norm": 0.04504437372088432,
      "learning_rate": 1.8217639982887623e-06,
      "loss": 0.063,
      "num_input_tokens_seen": 1019328,
      "step": 2075
    },
    {
      "epoch": 0.2745149795433549,
      "grad_norm": 0.030620688572525978,
      "learning_rate": 1.8204491872379769e-06,
      "loss": 0.0781,
      "num_input_tokens_seen": 1021696,
      "step": 2080
    },
    {
      "epoch": 0.27517487132110335,
      "grad_norm": 0.2540840208530426,
      "learning_rate": 1.8191300225737e-06,
      "loss": 0.09,
      "num_input_tokens_seen": 1024256,
      "step": 2085
    },
    {
      "epoch": 0.2758347630988518,
      "grad_norm": 35.219879150390625,
      "learning_rate": 1.8178065112959184e-06,
      "loss": 0.2074,
      "num_input_tokens_seen": 1026560,
      "step": 2090
    },
    {
      "epoch": 0.27649465487660024,
      "grad_norm": 56.63616943359375,
      "learning_rate": 1.8164786604276832e-06,
      "loss": 0.2426,
      "num_input_tokens_seen": 1029184,
      "step": 2095
    },
    {
      "epoch": 0.2771545466543487,
      "grad_norm": 0.794495701789856,
      "learning_rate": 1.8151464770150727e-06,
      "loss": 0.1464,
      "num_input_tokens_seen": 1031744,
      "step": 2100
    },
    {
      "epoch": 0.27781443843209713,
      "grad_norm": 0.1718120276927948,
      "learning_rate": 1.8138099681271558e-06,
      "loss": 0.1528,
      "num_input_tokens_seen": 1034048,
      "step": 2105
    },
    {
      "epoch": 0.2784743302098456,
      "grad_norm": 0.1188565269112587,
      "learning_rate": 1.8124691408559536e-06,
      "loss": 0.1601,
      "num_input_tokens_seen": 1036544,
      "step": 2110
    },
    {
      "epoch": 0.279134221987594,
      "grad_norm": 0.09950896352529526,
      "learning_rate": 1.8111240023164023e-06,
      "loss": 0.1008,
      "num_input_tokens_seen": 1038848,
      "step": 2115
    },
    {
      "epoch": 0.27979411376534247,
      "grad_norm": 47.676021575927734,
      "learning_rate": 1.809774559646316e-06,
      "loss": 0.0632,
      "num_input_tokens_seen": 1041152,
      "step": 2120
    },
    {
      "epoch": 0.2804540055430909,
      "grad_norm": 19.95380401611328,
      "learning_rate": 1.8084208200063469e-06,
      "loss": 0.0935,
      "num_input_tokens_seen": 1043968,
      "step": 2125
    },
    {
      "epoch": 0.28111389732083936,
      "grad_norm": 20.683349609375,
      "learning_rate": 1.8070627905799496e-06,
      "loss": 0.2149,
      "num_input_tokens_seen": 1046272,
      "step": 2130
    },
    {
      "epoch": 0.28177378909858786,
      "grad_norm": 0.5064588189125061,
      "learning_rate": 1.8057004785733413e-06,
      "loss": 0.0862,
      "num_input_tokens_seen": 1048448,
      "step": 2135
    },
    {
      "epoch": 0.2824336808763363,
      "grad_norm": 2.3065133094787598,
      "learning_rate": 1.8043338912154647e-06,
      "loss": 0.1758,
      "num_input_tokens_seen": 1051072,
      "step": 2140
    },
    {
      "epoch": 0.28309357265408475,
      "grad_norm": 0.986880362033844,
      "learning_rate": 1.8029630357579486e-06,
      "loss": 0.0486,
      "num_input_tokens_seen": 1053312,
      "step": 2145
    },
    {
      "epoch": 0.2837534644318332,
      "grad_norm": 0.09180797636508942,
      "learning_rate": 1.8015879194750702e-06,
      "loss": 0.0795,
      "num_input_tokens_seen": 1055680,
      "step": 2150
    },
    {
      "epoch": 0.28441335620958164,
      "grad_norm": 0.19105878472328186,
      "learning_rate": 1.8002085496637165e-06,
      "loss": 0.187,
      "num_input_tokens_seen": 1057984,
      "step": 2155
    },
    {
      "epoch": 0.2850732479873301,
      "grad_norm": 0.3145781457424164,
      "learning_rate": 1.7988249336433448e-06,
      "loss": 0.1492,
      "num_input_tokens_seen": 1060736,
      "step": 2160
    },
    {
      "epoch": 0.2857331397650785,
      "grad_norm": 0.49677714705467224,
      "learning_rate": 1.7974370787559447e-06,
      "loss": 0.1319,
      "num_input_tokens_seen": 1063424,
      "step": 2165
    },
    {
      "epoch": 0.28639303154282697,
      "grad_norm": 0.031023921445012093,
      "learning_rate": 1.796044992365999e-06,
      "loss": 0.0575,
      "num_input_tokens_seen": 1065728,
      "step": 2170
    },
    {
      "epoch": 0.2870529233205754,
      "grad_norm": 0.07371291518211365,
      "learning_rate": 1.794648681860444e-06,
      "loss": 0.0009,
      "num_input_tokens_seen": 1068160,
      "step": 2175
    },
    {
      "epoch": 0.28771281509832386,
      "grad_norm": 0.030941056087613106,
      "learning_rate": 1.7932481546486312e-06,
      "loss": 0.2734,
      "num_input_tokens_seen": 1070592,
      "step": 2180
    },
    {
      "epoch": 0.2883727068760723,
      "grad_norm": 20.994426727294922,
      "learning_rate": 1.791843418162287e-06,
      "loss": 0.1758,
      "num_input_tokens_seen": 1073280,
      "step": 2185
    },
    {
      "epoch": 0.28903259865382075,
      "grad_norm": 6.4550886154174805,
      "learning_rate": 1.7904344798554748e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 1075584,
      "step": 2190
    },
    {
      "epoch": 0.28969249043156925,
      "grad_norm": 4.786499977111816,
      "learning_rate": 1.789021347204553e-06,
      "loss": 0.0927,
      "num_input_tokens_seen": 1078016,
      "step": 2195
    },
    {
      "epoch": 0.2903523822093177,
      "grad_norm": 2.4327032566070557,
      "learning_rate": 1.7876040277081381e-06,
      "loss": 0.1631,
      "num_input_tokens_seen": 1080512,
      "step": 2200
    },
    {
      "epoch": 0.29101227398706614,
      "grad_norm": 128.65020751953125,
      "learning_rate": 1.7861825288870632e-06,
      "loss": 0.0382,
      "num_input_tokens_seen": 1082752,
      "step": 2205
    },
    {
      "epoch": 0.2916721657648146,
      "grad_norm": 12.312292098999023,
      "learning_rate": 1.7847568582843376e-06,
      "loss": 0.3717,
      "num_input_tokens_seen": 1085184,
      "step": 2210
    },
    {
      "epoch": 0.29233205754256303,
      "grad_norm": 24.310449600219727,
      "learning_rate": 1.7833270234651088e-06,
      "loss": 0.1799,
      "num_input_tokens_seen": 1087360,
      "step": 2215
    },
    {
      "epoch": 0.2929919493203115,
      "grad_norm": 14.611623764038086,
      "learning_rate": 1.781893032016621e-06,
      "loss": 0.0596,
      "num_input_tokens_seen": 1089984,
      "step": 2220
    },
    {
      "epoch": 0.2936518410980599,
      "grad_norm": 0.6433891654014587,
      "learning_rate": 1.7804548915481746e-06,
      "loss": 0.0264,
      "num_input_tokens_seen": 1092608,
      "step": 2225
    },
    {
      "epoch": 0.29431173287580836,
      "grad_norm": 0.19885502755641937,
      "learning_rate": 1.7790126096910865e-06,
      "loss": 0.1052,
      "num_input_tokens_seen": 1095040,
      "step": 2230
    },
    {
      "epoch": 0.2949716246535568,
      "grad_norm": 12.348299980163574,
      "learning_rate": 1.7775661940986492e-06,
      "loss": 0.1063,
      "num_input_tokens_seen": 1097728,
      "step": 2235
    },
    {
      "epoch": 0.29563151643130525,
      "grad_norm": 1.2165111303329468,
      "learning_rate": 1.776115652446091e-06,
      "loss": 0.1857,
      "num_input_tokens_seen": 1100096,
      "step": 2240
    },
    {
      "epoch": 0.2962914082090537,
      "grad_norm": 149.12095642089844,
      "learning_rate": 1.7746609924305336e-06,
      "loss": 0.1076,
      "num_input_tokens_seen": 1102400,
      "step": 2245
    },
    {
      "epoch": 0.29695129998680214,
      "grad_norm": 0.0535690002143383,
      "learning_rate": 1.7732022217709534e-06,
      "loss": 0.0825,
      "num_input_tokens_seen": 1104960,
      "step": 2250
    },
    {
      "epoch": 0.2976111917645506,
      "grad_norm": 12.77878475189209,
      "learning_rate": 1.7717393482081384e-06,
      "loss": 0.1648,
      "num_input_tokens_seen": 1107520,
      "step": 2255
    },
    {
      "epoch": 0.2982710835422991,
      "grad_norm": 37.477237701416016,
      "learning_rate": 1.7702723795046492e-06,
      "loss": 0.1223,
      "num_input_tokens_seen": 1109952,
      "step": 2260
    },
    {
      "epoch": 0.29893097532004753,
      "grad_norm": 0.3493918478488922,
      "learning_rate": 1.7688013234447757e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 1112128,
      "step": 2265
    },
    {
      "epoch": 0.299590867097796,
      "grad_norm": 0.07301745563745499,
      "learning_rate": 1.7673261878344973e-06,
      "loss": 0.1387,
      "num_input_tokens_seen": 1114688,
      "step": 2270
    },
    {
      "epoch": 0.30011878051999474,
      "eval_loss": 0.12411058694124222,
      "eval_runtime": 7.994,
      "eval_samples_per_second": 842.505,
      "eval_steps_per_second": 105.329,
      "num_input_tokens_seen": 1116800,
      "step": 2274
    },
    {
      "epoch": 0.3002507588755444,
      "grad_norm": 0.08390434831380844,
      "learning_rate": 1.7658469805014414e-06,
      "loss": 0.1185,
      "num_input_tokens_seen": 1117248,
      "step": 2275
    },
    {
      "epoch": 0.30091065065329287,
      "grad_norm": 41.85004806518555,
      "learning_rate": 1.7643637092948415e-06,
      "loss": 0.1096,
      "num_input_tokens_seen": 1119808,
      "step": 2280
    },
    {
      "epoch": 0.3015705424310413,
      "grad_norm": 55.66747283935547,
      "learning_rate": 1.7628763820854948e-06,
      "loss": 0.2572,
      "num_input_tokens_seen": 1122112,
      "step": 2285
    },
    {
      "epoch": 0.30223043420878976,
      "grad_norm": 0.3250507712364197,
      "learning_rate": 1.7613850067657216e-06,
      "loss": 0.0884,
      "num_input_tokens_seen": 1124544,
      "step": 2290
    },
    {
      "epoch": 0.3028903259865382,
      "grad_norm": 0.10577468574047089,
      "learning_rate": 1.7598895912493232e-06,
      "loss": 0.1247,
      "num_input_tokens_seen": 1127104,
      "step": 2295
    },
    {
      "epoch": 0.30355021776428665,
      "grad_norm": 0.9884712100028992,
      "learning_rate": 1.7583901434715397e-06,
      "loss": 0.0773,
      "num_input_tokens_seen": 1129536,
      "step": 2300
    },
    {
      "epoch": 0.3042101095420351,
      "grad_norm": 45.45866394042969,
      "learning_rate": 1.7568866713890074e-06,
      "loss": 0.1009,
      "num_input_tokens_seen": 1131840,
      "step": 2305
    },
    {
      "epoch": 0.30487000131978353,
      "grad_norm": 1.1424263715744019,
      "learning_rate": 1.7553791829797175e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 1134336,
      "step": 2310
    },
    {
      "epoch": 0.305529893097532,
      "grad_norm": 28.028568267822266,
      "learning_rate": 1.7538676862429737e-06,
      "loss": 0.3449,
      "num_input_tokens_seen": 1136640,
      "step": 2315
    },
    {
      "epoch": 0.3061897848752805,
      "grad_norm": 0.5848109126091003,
      "learning_rate": 1.7523521891993486e-06,
      "loss": 0.1248,
      "num_input_tokens_seen": 1139136,
      "step": 2320
    },
    {
      "epoch": 0.3068496766530289,
      "grad_norm": 0.08309627324342728,
      "learning_rate": 1.7508326998906422e-06,
      "loss": 0.0367,
      "num_input_tokens_seen": 1141568,
      "step": 2325
    },
    {
      "epoch": 0.30750956843077737,
      "grad_norm": 0.08207709342241287,
      "learning_rate": 1.7493092263798394e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 1143936,
      "step": 2330
    },
    {
      "epoch": 0.3081694602085258,
      "grad_norm": 156.30946350097656,
      "learning_rate": 1.7477817767510664e-06,
      "loss": 0.0188,
      "num_input_tokens_seen": 1146624,
      "step": 2335
    },
    {
      "epoch": 0.30882935198627426,
      "grad_norm": 0.43475231528282166,
      "learning_rate": 1.7462503591095484e-06,
      "loss": 0.0209,
      "num_input_tokens_seen": 1149120,
      "step": 2340
    },
    {
      "epoch": 0.3094892437640227,
      "grad_norm": 0.033984895795583725,
      "learning_rate": 1.7447149815815659e-06,
      "loss": 0.1152,
      "num_input_tokens_seen": 1151488,
      "step": 2345
    },
    {
      "epoch": 0.31014913554177115,
      "grad_norm": 0.023453880101442337,
      "learning_rate": 1.7431756523144126e-06,
      "loss": 0.1426,
      "num_input_tokens_seen": 1153600,
      "step": 2350
    },
    {
      "epoch": 0.3108090273195196,
      "grad_norm": 0.04266361892223358,
      "learning_rate": 1.7416323794763512e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 1156224,
      "step": 2355
    },
    {
      "epoch": 0.31146891909726804,
      "grad_norm": 0.01946975663304329,
      "learning_rate": 1.7400851712565707e-06,
      "loss": 0.234,
      "num_input_tokens_seen": 1158656,
      "step": 2360
    },
    {
      "epoch": 0.3121288108750165,
      "grad_norm": 0.08395219594240189,
      "learning_rate": 1.7385340358651432e-06,
      "loss": 0.1276,
      "num_input_tokens_seen": 1161408,
      "step": 2365
    },
    {
      "epoch": 0.3127887026527649,
      "grad_norm": 105.1695327758789,
      "learning_rate": 1.736978981532979e-06,
      "loss": 0.0223,
      "num_input_tokens_seen": 1163904,
      "step": 2370
    },
    {
      "epoch": 0.31344859443051337,
      "grad_norm": 0.06841170787811279,
      "learning_rate": 1.7354200165117838e-06,
      "loss": 0.2808,
      "num_input_tokens_seen": 1166208,
      "step": 2375
    },
    {
      "epoch": 0.3141084862082619,
      "grad_norm": 0.11680270731449127,
      "learning_rate": 1.733857149074016e-06,
      "loss": 0.2086,
      "num_input_tokens_seen": 1168512,
      "step": 2380
    },
    {
      "epoch": 0.3147683779860103,
      "grad_norm": 0.2520144581794739,
      "learning_rate": 1.7322903875128402e-06,
      "loss": 0.2359,
      "num_input_tokens_seen": 1171072,
      "step": 2385
    },
    {
      "epoch": 0.31542826976375876,
      "grad_norm": 0.23810061812400818,
      "learning_rate": 1.7307197401420858e-06,
      "loss": 0.0071,
      "num_input_tokens_seen": 1173312,
      "step": 2390
    },
    {
      "epoch": 0.3160881615415072,
      "grad_norm": 0.48759013414382935,
      "learning_rate": 1.7291452152962018e-06,
      "loss": 0.1059,
      "num_input_tokens_seen": 1175744,
      "step": 2395
    },
    {
      "epoch": 0.31674805331925565,
      "grad_norm": 45.06856918334961,
      "learning_rate": 1.7275668213302116e-06,
      "loss": 0.2033,
      "num_input_tokens_seen": 1178112,
      "step": 2400
    },
    {
      "epoch": 0.3174079450970041,
      "grad_norm": 44.76649856567383,
      "learning_rate": 1.72598456661967e-06,
      "loss": 0.0488,
      "num_input_tokens_seen": 1180352,
      "step": 2405
    },
    {
      "epoch": 0.31806783687475254,
      "grad_norm": 0.12351297587156296,
      "learning_rate": 1.7243984595606191e-06,
      "loss": 0.1241,
      "num_input_tokens_seen": 1182528,
      "step": 2410
    },
    {
      "epoch": 0.318727728652501,
      "grad_norm": 3.226271152496338,
      "learning_rate": 1.722808508569542e-06,
      "loss": 0.0697,
      "num_input_tokens_seen": 1185280,
      "step": 2415
    },
    {
      "epoch": 0.31938762043024943,
      "grad_norm": 16.573923110961914,
      "learning_rate": 1.72121472208332e-06,
      "loss": 0.0766,
      "num_input_tokens_seen": 1188032,
      "step": 2420
    },
    {
      "epoch": 0.3200475122079979,
      "grad_norm": 92.98533630371094,
      "learning_rate": 1.7196171085591864e-06,
      "loss": 0.1857,
      "num_input_tokens_seen": 1190464,
      "step": 2425
    },
    {
      "epoch": 0.3207074039857463,
      "grad_norm": 37.98392868041992,
      "learning_rate": 1.7180156764746824e-06,
      "loss": 0.2532,
      "num_input_tokens_seen": 1192960,
      "step": 2430
    },
    {
      "epoch": 0.32136729576349476,
      "grad_norm": 0.12210986018180847,
      "learning_rate": 1.7164104343276113e-06,
      "loss": 0.0694,
      "num_input_tokens_seen": 1195072,
      "step": 2435
    },
    {
      "epoch": 0.3220271875412432,
      "grad_norm": 0.015084991231560707,
      "learning_rate": 1.714801390635996e-06,
      "loss": 0.0516,
      "num_input_tokens_seen": 1197376,
      "step": 2440
    },
    {
      "epoch": 0.3226870793189917,
      "grad_norm": 0.035268671810626984,
      "learning_rate": 1.7131885539380297e-06,
      "loss": 0.0735,
      "num_input_tokens_seen": 1199936,
      "step": 2445
    },
    {
      "epoch": 0.32334697109674015,
      "grad_norm": 33.19987106323242,
      "learning_rate": 1.7115719327920335e-06,
      "loss": 0.141,
      "num_input_tokens_seen": 1202368,
      "step": 2450
    },
    {
      "epoch": 0.3240068628744886,
      "grad_norm": 0.026460843160748482,
      "learning_rate": 1.70995153577641e-06,
      "loss": 0.0172,
      "num_input_tokens_seen": 1204800,
      "step": 2455
    },
    {
      "epoch": 0.32466675465223704,
      "grad_norm": 33.02996826171875,
      "learning_rate": 1.7083273714895991e-06,
      "loss": 0.0639,
      "num_input_tokens_seen": 1207552,
      "step": 2460
    },
    {
      "epoch": 0.3253266464299855,
      "grad_norm": 0.04908216744661331,
      "learning_rate": 1.7066994485500298e-06,
      "loss": 0.1031,
      "num_input_tokens_seen": 1209856,
      "step": 2465
    },
    {
      "epoch": 0.32598653820773393,
      "grad_norm": 0.16728860139846802,
      "learning_rate": 1.7050677755960762e-06,
      "loss": 0.1125,
      "num_input_tokens_seen": 1212352,
      "step": 2470
    },
    {
      "epoch": 0.3266464299854824,
      "grad_norm": 0.5578970313072205,
      "learning_rate": 1.7034323612860124e-06,
      "loss": 0.0832,
      "num_input_tokens_seen": 1214912,
      "step": 2475
    },
    {
      "epoch": 0.3273063217632308,
      "grad_norm": 0.029092831537127495,
      "learning_rate": 1.7017932142979645e-06,
      "loss": 0.1275,
      "num_input_tokens_seen": 1217088,
      "step": 2480
    },
    {
      "epoch": 0.32796621354097927,
      "grad_norm": 0.5877671241760254,
      "learning_rate": 1.700150343329866e-06,
      "loss": 0.2478,
      "num_input_tokens_seen": 1219584,
      "step": 2485
    },
    {
      "epoch": 0.3286261053187277,
      "grad_norm": 62.71138000488281,
      "learning_rate": 1.6985037570994113e-06,
      "loss": 0.1182,
      "num_input_tokens_seen": 1222336,
      "step": 2490
    },
    {
      "epoch": 0.32928599709647616,
      "grad_norm": 0.06207147613167763,
      "learning_rate": 1.6968534643440088e-06,
      "loss": 0.0913,
      "num_input_tokens_seen": 1224832,
      "step": 2495
    },
    {
      "epoch": 0.3299458888742246,
      "grad_norm": 23.29184913635254,
      "learning_rate": 1.6951994738207364e-06,
      "loss": 0.1998,
      "num_input_tokens_seen": 1227392,
      "step": 2500
    },
    {
      "epoch": 0.3306057806519731,
      "grad_norm": 14.910649299621582,
      "learning_rate": 1.6935417943062928e-06,
      "loss": 0.2007,
      "num_input_tokens_seen": 1229952,
      "step": 2505
    },
    {
      "epoch": 0.33126567242972155,
      "grad_norm": 1.2473188638687134,
      "learning_rate": 1.6918804345969516e-06,
      "loss": 0.0615,
      "num_input_tokens_seen": 1232640,
      "step": 2510
    },
    {
      "epoch": 0.33192556420747,
      "grad_norm": 121.41654968261719,
      "learning_rate": 1.6902154035085156e-06,
      "loss": 0.0479,
      "num_input_tokens_seen": 1235200,
      "step": 2515
    },
    {
      "epoch": 0.33258545598521844,
      "grad_norm": 0.5897153615951538,
      "learning_rate": 1.688546709876269e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 1237632,
      "step": 2520
    },
    {
      "epoch": 0.3332453477629669,
      "grad_norm": 1.850326418876648,
      "learning_rate": 1.6868743625549314e-06,
      "loss": 0.0662,
      "num_input_tokens_seen": 1239936,
      "step": 2525
    },
    {
      "epoch": 0.3339052395407153,
      "grad_norm": 0.34998998045921326,
      "learning_rate": 1.6851983704186092e-06,
      "loss": 0.0455,
      "num_input_tokens_seen": 1242304,
      "step": 2530
    },
    {
      "epoch": 0.33456513131846377,
      "grad_norm": 0.0726654902100563,
      "learning_rate": 1.6835187423607503e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 1244736,
      "step": 2535
    },
    {
      "epoch": 0.3352250230962122,
      "grad_norm": 0.07731109857559204,
      "learning_rate": 1.681835487294096e-06,
      "loss": 0.1863,
      "num_input_tokens_seen": 1247488,
      "step": 2540
    },
    {
      "epoch": 0.33588491487396066,
      "grad_norm": 0.04418403282761574,
      "learning_rate": 1.6801486141506342e-06,
      "loss": 0.229,
      "num_input_tokens_seen": 1250048,
      "step": 2545
    },
    {
      "epoch": 0.3365448066517091,
      "grad_norm": 133.3740692138672,
      "learning_rate": 1.6784581318815514e-06,
      "loss": 0.3541,
      "num_input_tokens_seen": 1252928,
      "step": 2550
    },
    {
      "epoch": 0.33720469842945755,
      "grad_norm": 0.03932896628975868,
      "learning_rate": 1.6767640494571849e-06,
      "loss": 0.1596,
      "num_input_tokens_seen": 1255488,
      "step": 2555
    },
    {
      "epoch": 0.337864590207206,
      "grad_norm": 33.07372283935547,
      "learning_rate": 1.6750663758669767e-06,
      "loss": 0.3643,
      "num_input_tokens_seen": 1257984,
      "step": 2560
    },
    {
      "epoch": 0.3385244819849545,
      "grad_norm": 0.150083988904953,
      "learning_rate": 1.6733651201194245e-06,
      "loss": 0.1098,
      "num_input_tokens_seen": 1260416,
      "step": 2565
    },
    {
      "epoch": 0.33918437376270294,
      "grad_norm": 19.72997283935547,
      "learning_rate": 1.6716602912420342e-06,
      "loss": 0.1186,
      "num_input_tokens_seen": 1263168,
      "step": 2570
    },
    {
      "epoch": 0.3398442655404514,
      "grad_norm": 0.24268358945846558,
      "learning_rate": 1.6699518982812726e-06,
      "loss": 0.1177,
      "num_input_tokens_seen": 1265600,
      "step": 2575
    },
    {
      "epoch": 0.34050415731819983,
      "grad_norm": 0.06883673369884491,
      "learning_rate": 1.6682399503025183e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 1268032,
      "step": 2580
    },
    {
      "epoch": 0.3411640490959483,
      "grad_norm": 38.50511932373047,
      "learning_rate": 1.666524456390014e-06,
      "loss": 0.1831,
      "num_input_tokens_seen": 1270336,
      "step": 2585
    },
    {
      "epoch": 0.3418239408736967,
      "grad_norm": 25.056659698486328,
      "learning_rate": 1.664805425646819e-06,
      "loss": 0.0639,
      "num_input_tokens_seen": 1273088,
      "step": 2590
    },
    {
      "epoch": 0.34248383265144516,
      "grad_norm": 0.358341783285141,
      "learning_rate": 1.6630828671947606e-06,
      "loss": 0.2223,
      "num_input_tokens_seen": 1275456,
      "step": 2595
    },
    {
      "epoch": 0.3431437244291936,
      "grad_norm": 0.3967565596103668,
      "learning_rate": 1.6613567901743842e-06,
      "loss": 0.0484,
      "num_input_tokens_seen": 1277888,
      "step": 2600
    },
    {
      "epoch": 0.34380361620694205,
      "grad_norm": 0.4620976448059082,
      "learning_rate": 1.6596272037449075e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 1280384,
      "step": 2605
    },
    {
      "epoch": 0.3444635079846905,
      "grad_norm": 111.44522094726562,
      "learning_rate": 1.6578941170841696e-06,
      "loss": 0.0633,
      "num_input_tokens_seen": 1282944,
      "step": 2610
    },
    {
      "epoch": 0.34512339976243894,
      "grad_norm": 10.548276901245117,
      "learning_rate": 1.6561575393885833e-06,
      "loss": 0.0451,
      "num_input_tokens_seen": 1285184,
      "step": 2615
    },
    {
      "epoch": 0.3457832915401874,
      "grad_norm": 0.21035003662109375,
      "learning_rate": 1.6544174798730864e-06,
      "loss": 0.2038,
      "num_input_tokens_seen": 1287808,
      "step": 2620
    },
    {
      "epoch": 0.34644318331793583,
      "grad_norm": 0.9157705307006836,
      "learning_rate": 1.6526739477710923e-06,
      "loss": 0.1932,
      "num_input_tokens_seen": 1290432,
      "step": 2625
    },
    {
      "epoch": 0.34710307509568433,
      "grad_norm": 318.77642822265625,
      "learning_rate": 1.650926952334441e-06,
      "loss": 0.2289,
      "num_input_tokens_seen": 1292736,
      "step": 2630
    },
    {
      "epoch": 0.3477629668734328,
      "grad_norm": 0.13021092116832733,
      "learning_rate": 1.6491765028333516e-06,
      "loss": 0.2947,
      "num_input_tokens_seen": 1295104,
      "step": 2635
    },
    {
      "epoch": 0.3484228586511812,
      "grad_norm": 0.7741267681121826,
      "learning_rate": 1.6474226085563693e-06,
      "loss": 0.0444,
      "num_input_tokens_seen": 1297600,
      "step": 2640
    },
    {
      "epoch": 0.34908275042892967,
      "grad_norm": 0.11857028305530548,
      "learning_rate": 1.6456652788103215e-06,
      "loss": 0.0057,
      "num_input_tokens_seen": 1300224,
      "step": 2645
    },
    {
      "epoch": 0.3497426422066781,
      "grad_norm": 0.3336769640445709,
      "learning_rate": 1.6439045229202631e-06,
      "loss": 0.0494,
      "num_input_tokens_seen": 1302528,
      "step": 2650
    },
    {
      "epoch": 0.3501385772733272,
      "eval_loss": 0.13837853074073792,
      "eval_runtime": 7.8873,
      "eval_samples_per_second": 853.902,
      "eval_steps_per_second": 106.754,
      "num_input_tokens_seen": 1303872,
      "step": 2653
    },
    {
      "epoch": 0.35040253398442656,
      "grad_norm": 0.26019707322120667,
      "learning_rate": 1.6421403502294307e-06,
      "loss": 0.1541,
      "num_input_tokens_seen": 1305024,
      "step": 2655
    },
    {
      "epoch": 0.351062425762175,
      "grad_norm": 0.39249280095100403,
      "learning_rate": 1.6403727700991915e-06,
      "loss": 0.1878,
      "num_input_tokens_seen": 1307392,
      "step": 2660
    },
    {
      "epoch": 0.35172231753992345,
      "grad_norm": 0.15294888615608215,
      "learning_rate": 1.6386017919089933e-06,
      "loss": 0.273,
      "num_input_tokens_seen": 1310016,
      "step": 2665
    },
    {
      "epoch": 0.3523822093176719,
      "grad_norm": 0.07149966061115265,
      "learning_rate": 1.636827425056316e-06,
      "loss": 0.0471,
      "num_input_tokens_seen": 1312576,
      "step": 2670
    },
    {
      "epoch": 0.35304210109542034,
      "grad_norm": 0.1564963310956955,
      "learning_rate": 1.635049678956621e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 1315072,
      "step": 2675
    },
    {
      "epoch": 0.3537019928731688,
      "grad_norm": 3.6665759086608887,
      "learning_rate": 1.633268563043301e-06,
      "loss": 0.125,
      "num_input_tokens_seen": 1317504,
      "step": 2680
    },
    {
      "epoch": 0.3543618846509172,
      "grad_norm": 0.23970353603363037,
      "learning_rate": 1.63148408676763e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 1319680,
      "step": 2685
    },
    {
      "epoch": 0.3550217764286657,
      "grad_norm": 0.059584952890872955,
      "learning_rate": 1.6296962595987141e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 1322240,
      "step": 2690
    },
    {
      "epoch": 0.35568166820641417,
      "grad_norm": 0.21580637991428375,
      "learning_rate": 1.6279050910234392e-06,
      "loss": 0.1102,
      "num_input_tokens_seen": 1324736,
      "step": 2695
    },
    {
      "epoch": 0.3563415599841626,
      "grad_norm": 0.024059053510427475,
      "learning_rate": 1.626110590546423e-06,
      "loss": 0.0617,
      "num_input_tokens_seen": 1327104,
      "step": 2700
    },
    {
      "epoch": 0.35700145176191106,
      "grad_norm": 188.6191864013672,
      "learning_rate": 1.6243127676899635e-06,
      "loss": 0.2021,
      "num_input_tokens_seen": 1329920,
      "step": 2705
    },
    {
      "epoch": 0.3576613435396595,
      "grad_norm": 0.04233895614743233,
      "learning_rate": 1.6225116319939884e-06,
      "loss": 0.2292,
      "num_input_tokens_seen": 1332352,
      "step": 2710
    },
    {
      "epoch": 0.35832123531740795,
      "grad_norm": 79.07254028320312,
      "learning_rate": 1.6207071930160044e-06,
      "loss": 0.1136,
      "num_input_tokens_seen": 1335040,
      "step": 2715
    },
    {
      "epoch": 0.3589811270951564,
      "grad_norm": 0.3499235510826111,
      "learning_rate": 1.6188994603310468e-06,
      "loss": 0.0427,
      "num_input_tokens_seen": 1337472,
      "step": 2720
    },
    {
      "epoch": 0.35964101887290484,
      "grad_norm": 114.59748840332031,
      "learning_rate": 1.617088443531628e-06,
      "loss": 0.158,
      "num_input_tokens_seen": 1339712,
      "step": 2725
    },
    {
      "epoch": 0.3603009106506533,
      "grad_norm": 0.12750740349292755,
      "learning_rate": 1.6152741522276882e-06,
      "loss": 0.0092,
      "num_input_tokens_seen": 1342144,
      "step": 2730
    },
    {
      "epoch": 0.36096080242840173,
      "grad_norm": 0.6062318682670593,
      "learning_rate": 1.6134565960465425e-06,
      "loss": 0.0433,
      "num_input_tokens_seen": 1344512,
      "step": 2735
    },
    {
      "epoch": 0.3616206942061502,
      "grad_norm": 0.28183653950691223,
      "learning_rate": 1.6116357846328312e-06,
      "loss": 0.1562,
      "num_input_tokens_seen": 1346880,
      "step": 2740
    },
    {
      "epoch": 0.3622805859838986,
      "grad_norm": 0.0915948897600174,
      "learning_rate": 1.609811727648468e-06,
      "loss": 0.0123,
      "num_input_tokens_seen": 1349056,
      "step": 2745
    },
    {
      "epoch": 0.36294047776164706,
      "grad_norm": 0.5089015960693359,
      "learning_rate": 1.6079844347725882e-06,
      "loss": 0.0425,
      "num_input_tokens_seen": 1351488,
      "step": 2750
    },
    {
      "epoch": 0.36360036953939556,
      "grad_norm": 0.19435492157936096,
      "learning_rate": 1.6061539157014987e-06,
      "loss": 0.0654,
      "num_input_tokens_seen": 1353920,
      "step": 2755
    },
    {
      "epoch": 0.364260261317144,
      "grad_norm": 0.4347739815711975,
      "learning_rate": 1.6043201801486257e-06,
      "loss": 0.2615,
      "num_input_tokens_seen": 1356352,
      "step": 2760
    },
    {
      "epoch": 0.36492015309489245,
      "grad_norm": 6.802909851074219,
      "learning_rate": 1.6024832378444628e-06,
      "loss": 0.1949,
      "num_input_tokens_seen": 1359104,
      "step": 2765
    },
    {
      "epoch": 0.3655800448726409,
      "grad_norm": 25.862701416015625,
      "learning_rate": 1.6006430985365204e-06,
      "loss": 0.3066,
      "num_input_tokens_seen": 1361536,
      "step": 2770
    },
    {
      "epoch": 0.36623993665038934,
      "grad_norm": 8.995551109313965,
      "learning_rate": 1.5987997719892735e-06,
      "loss": 0.2606,
      "num_input_tokens_seen": 1364160,
      "step": 2775
    },
    {
      "epoch": 0.3668998284281378,
      "grad_norm": 0.2326563000679016,
      "learning_rate": 1.5969532679841088e-06,
      "loss": 0.005,
      "num_input_tokens_seen": 1366656,
      "step": 2780
    },
    {
      "epoch": 0.36755972020588623,
      "grad_norm": 29.3525390625,
      "learning_rate": 1.5951035963192752e-06,
      "loss": 0.0512,
      "num_input_tokens_seen": 1369216,
      "step": 2785
    },
    {
      "epoch": 0.3682196119836347,
      "grad_norm": 3.3923957347869873,
      "learning_rate": 1.593250766809829e-06,
      "loss": 0.2619,
      "num_input_tokens_seen": 1371712,
      "step": 2790
    },
    {
      "epoch": 0.3688795037613831,
      "grad_norm": 47.703060150146484,
      "learning_rate": 1.5913947892875842e-06,
      "loss": 0.1739,
      "num_input_tokens_seen": 1374080,
      "step": 2795
    },
    {
      "epoch": 0.36953939553913157,
      "grad_norm": 0.36688122153282166,
      "learning_rate": 1.589535673601059e-06,
      "loss": 0.1383,
      "num_input_tokens_seen": 1377024,
      "step": 2800
    },
    {
      "epoch": 0.37019928731688,
      "grad_norm": 43.584991455078125,
      "learning_rate": 1.587673429615424e-06,
      "loss": 0.0587,
      "num_input_tokens_seen": 1379392,
      "step": 2805
    },
    {
      "epoch": 0.37085917909462845,
      "grad_norm": 0.4732750356197357,
      "learning_rate": 1.5858080672124495e-06,
      "loss": 0.1913,
      "num_input_tokens_seen": 1381760,
      "step": 2810
    },
    {
      "epoch": 0.37151907087237696,
      "grad_norm": 2.4631268978118896,
      "learning_rate": 1.5839395962904536e-06,
      "loss": 0.0931,
      "num_input_tokens_seen": 1384128,
      "step": 2815
    },
    {
      "epoch": 0.3721789626501254,
      "grad_norm": 0.13429832458496094,
      "learning_rate": 1.5820680267642494e-06,
      "loss": 0.0094,
      "num_input_tokens_seen": 1386496,
      "step": 2820
    },
    {
      "epoch": 0.37283885442787384,
      "grad_norm": 0.12962228059768677,
      "learning_rate": 1.5801933685650917e-06,
      "loss": 0.0489,
      "num_input_tokens_seen": 1388736,
      "step": 2825
    },
    {
      "epoch": 0.3734987462056223,
      "grad_norm": 88.37797546386719,
      "learning_rate": 1.5783156316406259e-06,
      "loss": 0.0478,
      "num_input_tokens_seen": 1391040,
      "step": 2830
    },
    {
      "epoch": 0.37415863798337073,
      "grad_norm": 115.88335418701172,
      "learning_rate": 1.5764348259548334e-06,
      "loss": 0.2375,
      "num_input_tokens_seen": 1393344,
      "step": 2835
    },
    {
      "epoch": 0.3748185297611192,
      "grad_norm": 76.65043640136719,
      "learning_rate": 1.5745509614879806e-06,
      "loss": 0.0726,
      "num_input_tokens_seen": 1395648,
      "step": 2840
    },
    {
      "epoch": 0.3754784215388676,
      "grad_norm": 0.021454842761158943,
      "learning_rate": 1.572664048236564e-06,
      "loss": 0.3055,
      "num_input_tokens_seen": 1398272,
      "step": 2845
    },
    {
      "epoch": 0.37613831331661607,
      "grad_norm": 0.13567005097866058,
      "learning_rate": 1.570774096213259e-06,
      "loss": 0.0405,
      "num_input_tokens_seen": 1400576,
      "step": 2850
    },
    {
      "epoch": 0.3767982050943645,
      "grad_norm": 0.033437322825193405,
      "learning_rate": 1.5688811154468649e-06,
      "loss": 0.0015,
      "num_input_tokens_seen": 1403136,
      "step": 2855
    },
    {
      "epoch": 0.37745809687211296,
      "grad_norm": 0.10280902683734894,
      "learning_rate": 1.5669851159822532e-06,
      "loss": 0.1948,
      "num_input_tokens_seen": 1405504,
      "step": 2860
    },
    {
      "epoch": 0.3781179886498614,
      "grad_norm": 102.37528991699219,
      "learning_rate": 1.5650861078803137e-06,
      "loss": 0.1405,
      "num_input_tokens_seen": 1407808,
      "step": 2865
    },
    {
      "epoch": 0.37877788042760985,
      "grad_norm": 25.63332748413086,
      "learning_rate": 1.5631841012179013e-06,
      "loss": 0.0803,
      "num_input_tokens_seen": 1410304,
      "step": 2870
    },
    {
      "epoch": 0.37943777220535835,
      "grad_norm": 0.2513096034526825,
      "learning_rate": 1.5612791060877818e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 1412736,
      "step": 2875
    },
    {
      "epoch": 0.3800976639831068,
      "grad_norm": 103.188232421875,
      "learning_rate": 1.5593711325985801e-06,
      "loss": 0.0403,
      "num_input_tokens_seen": 1415488,
      "step": 2880
    },
    {
      "epoch": 0.38075755576085524,
      "grad_norm": 0.05830248445272446,
      "learning_rate": 1.5574601908747245e-06,
      "loss": 0.213,
      "num_input_tokens_seen": 1417856,
      "step": 2885
    },
    {
      "epoch": 0.3814174475386037,
      "grad_norm": 0.040386002510786057,
      "learning_rate": 1.5555462910563936e-06,
      "loss": 0.077,
      "num_input_tokens_seen": 1420096,
      "step": 2890
    },
    {
      "epoch": 0.3820773393163521,
      "grad_norm": 21.231006622314453,
      "learning_rate": 1.5536294432994636e-06,
      "loss": 0.2334,
      "num_input_tokens_seen": 1422656,
      "step": 2895
    },
    {
      "epoch": 0.38273723109410057,
      "grad_norm": 2.1260502338409424,
      "learning_rate": 1.5517096577754528e-06,
      "loss": 0.0744,
      "num_input_tokens_seen": 1425152,
      "step": 2900
    },
    {
      "epoch": 0.383397122871849,
      "grad_norm": 0.10484765470027924,
      "learning_rate": 1.5497869446714695e-06,
      "loss": 0.0393,
      "num_input_tokens_seen": 1427840,
      "step": 2905
    },
    {
      "epoch": 0.38405701464959746,
      "grad_norm": 0.09525975584983826,
      "learning_rate": 1.5478613141901558e-06,
      "loss": 0.0065,
      "num_input_tokens_seen": 1430144,
      "step": 2910
    },
    {
      "epoch": 0.3847169064273459,
      "grad_norm": 0.13482730090618134,
      "learning_rate": 1.5459327765496348e-06,
      "loss": 0.1329,
      "num_input_tokens_seen": 1432448,
      "step": 2915
    },
    {
      "epoch": 0.38537679820509435,
      "grad_norm": 0.25041621923446655,
      "learning_rate": 1.5440013419834563e-06,
      "loss": 0.0276,
      "num_input_tokens_seen": 1434752,
      "step": 2920
    },
    {
      "epoch": 0.3860366899828428,
      "grad_norm": 0.17661698162555695,
      "learning_rate": 1.5420670207405419e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 1437184,
      "step": 2925
    },
    {
      "epoch": 0.38669658176059124,
      "grad_norm": 0.07189780473709106,
      "learning_rate": 1.5401298230851314e-06,
      "loss": 0.0935,
      "num_input_tokens_seen": 1440000,
      "step": 2930
    },
    {
      "epoch": 0.3873564735383397,
      "grad_norm": 62.480587005615234,
      "learning_rate": 1.5381897592967275e-06,
      "loss": 0.0716,
      "num_input_tokens_seen": 1442624,
      "step": 2935
    },
    {
      "epoch": 0.3880163653160882,
      "grad_norm": 0.13392320275306702,
      "learning_rate": 1.5362468396700426e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 1445184,
      "step": 2940
    },
    {
      "epoch": 0.38867625709383663,
      "grad_norm": 5.734695911407471,
      "learning_rate": 1.5343010745149418e-06,
      "loss": 0.2755,
      "num_input_tokens_seen": 1447616,
      "step": 2945
    },
    {
      "epoch": 0.3893361488715851,
      "grad_norm": 0.01573280617594719,
      "learning_rate": 1.532352474156391e-06,
      "loss": 0.0004,
      "num_input_tokens_seen": 1450176,
      "step": 2950
    },
    {
      "epoch": 0.3899960406493335,
      "grad_norm": 79.7118148803711,
      "learning_rate": 1.5304010489343995e-06,
      "loss": 0.5725,
      "num_input_tokens_seen": 1452672,
      "step": 2955
    },
    {
      "epoch": 0.39065593242708196,
      "grad_norm": 0.11838354915380478,
      "learning_rate": 1.528446809203968e-06,
      "loss": 0.2012,
      "num_input_tokens_seen": 1455232,
      "step": 2960
    },
    {
      "epoch": 0.3913158242048304,
      "grad_norm": 65.51634216308594,
      "learning_rate": 1.526489765335031e-06,
      "loss": 0.1031,
      "num_input_tokens_seen": 1457792,
      "step": 2965
    },
    {
      "epoch": 0.39197571598257885,
      "grad_norm": 0.2883387804031372,
      "learning_rate": 1.5245299277124026e-06,
      "loss": 0.1237,
      "num_input_tokens_seen": 1460160,
      "step": 2970
    },
    {
      "epoch": 0.3926356077603273,
      "grad_norm": 0.13738037645816803,
      "learning_rate": 1.5225673067357218e-06,
      "loss": 0.1705,
      "num_input_tokens_seen": 1462400,
      "step": 2975
    },
    {
      "epoch": 0.39329549953807574,
      "grad_norm": 9.973140716552734,
      "learning_rate": 1.5206019128193981e-06,
      "loss": 0.1399,
      "num_input_tokens_seen": 1465088,
      "step": 2980
    },
    {
      "epoch": 0.3939553913158242,
      "grad_norm": 0.1793886423110962,
      "learning_rate": 1.5186337563925538e-06,
      "loss": 0.167,
      "num_input_tokens_seen": 1467456,
      "step": 2985
    },
    {
      "epoch": 0.39461528309357263,
      "grad_norm": 0.18464945256710052,
      "learning_rate": 1.516662847898971e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 1470016,
      "step": 2990
    },
    {
      "epoch": 0.3952751748713211,
      "grad_norm": 0.3655398190021515,
      "learning_rate": 1.5146891977970349e-06,
      "loss": 0.0936,
      "num_input_tokens_seen": 1472448,
      "step": 2995
    },
    {
      "epoch": 0.3959350666490696,
      "grad_norm": 0.48678961396217346,
      "learning_rate": 1.5127128165596794e-06,
      "loss": 0.0271,
      "num_input_tokens_seen": 1475072,
      "step": 3000
    },
    {
      "epoch": 0.396594958426818,
      "grad_norm": 0.7548299431800842,
      "learning_rate": 1.51073371467433e-06,
      "loss": 0.0571,
      "num_input_tokens_seen": 1477440,
      "step": 3005
    },
    {
      "epoch": 0.39725485020456647,
      "grad_norm": 1.1893014907836914,
      "learning_rate": 1.5087519026428498e-06,
      "loss": 0.027,
      "num_input_tokens_seen": 1479872,
      "step": 3010
    },
    {
      "epoch": 0.3979147419823149,
      "grad_norm": 24.348011016845703,
      "learning_rate": 1.5067673909814818e-06,
      "loss": 0.2218,
      "num_input_tokens_seen": 1481920,
      "step": 3015
    },
    {
      "epoch": 0.39857463376006336,
      "grad_norm": 0.06322956830263138,
      "learning_rate": 1.5047801902207953e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 1484992,
      "step": 3020
    },
    {
      "epoch": 0.3992345255378118,
      "grad_norm": 0.24212270975112915,
      "learning_rate": 1.5027903109056288e-06,
      "loss": 0.2134,
      "num_input_tokens_seen": 1487232,
      "step": 3025
    },
    {
      "epoch": 0.39989441731556025,
      "grad_norm": 38.114532470703125,
      "learning_rate": 1.5007977635950336e-06,
      "loss": 0.1412,
      "num_input_tokens_seen": 1489728,
      "step": 3030
    },
    {
      "epoch": 0.4001583740266596,
      "eval_loss": 0.13093648850917816,
      "eval_runtime": 7.9673,
      "eval_samples_per_second": 845.332,
      "eval_steps_per_second": 105.682,
      "num_input_tokens_seen": 1490688,
      "step": 3032
    },
    {
      "epoch": 0.4005543090933087,
      "grad_norm": 190.9660186767578,
      "learning_rate": 1.498802558862219e-06,
      "loss": 0.2057,
      "num_input_tokens_seen": 1491968,
      "step": 3035
    },
    {
      "epoch": 0.40121420087105714,
      "grad_norm": 0.09837495535612106,
      "learning_rate": 1.496804707294496e-06,
      "loss": 0.1003,
      "num_input_tokens_seen": 1494336,
      "step": 3040
    },
    {
      "epoch": 0.4018740926488056,
      "grad_norm": 0.6492270827293396,
      "learning_rate": 1.4948042194932195e-06,
      "loss": 0.065,
      "num_input_tokens_seen": 1497472,
      "step": 3045
    },
    {
      "epoch": 0.402533984426554,
      "grad_norm": 0.34846293926239014,
      "learning_rate": 1.4928011060737341e-06,
      "loss": 0.0242,
      "num_input_tokens_seen": 1499968,
      "step": 3050
    },
    {
      "epoch": 0.40319387620430247,
      "grad_norm": 0.09735693782567978,
      "learning_rate": 1.4907953776653171e-06,
      "loss": 0.0811,
      "num_input_tokens_seen": 1502336,
      "step": 3055
    },
    {
      "epoch": 0.40385376798205097,
      "grad_norm": 21.974517822265625,
      "learning_rate": 1.4887870449111206e-06,
      "loss": 0.1462,
      "num_input_tokens_seen": 1504576,
      "step": 3060
    },
    {
      "epoch": 0.4045136597597994,
      "grad_norm": 85.91175079345703,
      "learning_rate": 1.486776118468118e-06,
      "loss": 0.2155,
      "num_input_tokens_seen": 1507136,
      "step": 3065
    },
    {
      "epoch": 0.40517355153754786,
      "grad_norm": 0.19757795333862305,
      "learning_rate": 1.4847626090070451e-06,
      "loss": 0.0479,
      "num_input_tokens_seen": 1509696,
      "step": 3070
    },
    {
      "epoch": 0.4058334433152963,
      "grad_norm": 0.6578057408332825,
      "learning_rate": 1.4827465272123439e-06,
      "loss": 0.3045,
      "num_input_tokens_seen": 1512192,
      "step": 3075
    },
    {
      "epoch": 0.40649333509304475,
      "grad_norm": 0.17303939163684845,
      "learning_rate": 1.4807278837821063e-06,
      "loss": 0.0134,
      "num_input_tokens_seen": 1514752,
      "step": 3080
    },
    {
      "epoch": 0.4071532268707932,
      "grad_norm": 14.41674518585205,
      "learning_rate": 1.4787066894280178e-06,
      "loss": 0.3,
      "num_input_tokens_seen": 1517440,
      "step": 3085
    },
    {
      "epoch": 0.40781311864854164,
      "grad_norm": 0.12199345231056213,
      "learning_rate": 1.476682954875299e-06,
      "loss": 0.0679,
      "num_input_tokens_seen": 1519744,
      "step": 3090
    },
    {
      "epoch": 0.4084730104262901,
      "grad_norm": 29.185821533203125,
      "learning_rate": 1.4746566908626506e-06,
      "loss": 0.1745,
      "num_input_tokens_seen": 1522176,
      "step": 3095
    },
    {
      "epoch": 0.40913290220403853,
      "grad_norm": 1.1170824766159058,
      "learning_rate": 1.4726279081421956e-06,
      "loss": 0.0519,
      "num_input_tokens_seen": 1524352,
      "step": 3100
    },
    {
      "epoch": 0.409792793981787,
      "grad_norm": 35.82356643676758,
      "learning_rate": 1.4705966174794216e-06,
      "loss": 0.2616,
      "num_input_tokens_seen": 1526976,
      "step": 3105
    },
    {
      "epoch": 0.4104526857595354,
      "grad_norm": 0.1983165442943573,
      "learning_rate": 1.4685628296531248e-06,
      "loss": 0.1047,
      "num_input_tokens_seen": 1529152,
      "step": 3110
    },
    {
      "epoch": 0.41111257753728386,
      "grad_norm": 0.7452152371406555,
      "learning_rate": 1.466526555455352e-06,
      "loss": 0.0288,
      "num_input_tokens_seen": 1531648,
      "step": 3115
    },
    {
      "epoch": 0.4117724693150323,
      "grad_norm": 0.30635786056518555,
      "learning_rate": 1.4644878056913432e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 1533952,
      "step": 3120
    },
    {
      "epoch": 0.4124323610927808,
      "grad_norm": 74.26456451416016,
      "learning_rate": 1.4624465911794764e-06,
      "loss": 0.165,
      "num_input_tokens_seen": 1536640,
      "step": 3125
    },
    {
      "epoch": 0.41309225287052925,
      "grad_norm": 0.07176000624895096,
      "learning_rate": 1.4604029227512062e-06,
      "loss": 0.0024,
      "num_input_tokens_seen": 1539200,
      "step": 3130
    },
    {
      "epoch": 0.4137521446482777,
      "grad_norm": 0.4018791615962982,
      "learning_rate": 1.4583568112510108e-06,
      "loss": 0.1984,
      "num_input_tokens_seen": 1541632,
      "step": 3135
    },
    {
      "epoch": 0.41441203642602614,
      "grad_norm": 15.011775016784668,
      "learning_rate": 1.4563082675363302e-06,
      "loss": 0.0697,
      "num_input_tokens_seen": 1544128,
      "step": 3140
    },
    {
      "epoch": 0.4150719282037746,
      "grad_norm": 0.17546619474887848,
      "learning_rate": 1.4542573024775122e-06,
      "loss": 0.0009,
      "num_input_tokens_seen": 1546368,
      "step": 3145
    },
    {
      "epoch": 0.41573181998152303,
      "grad_norm": 2.7632203102111816,
      "learning_rate": 1.4522039269577521e-06,
      "loss": 0.3214,
      "num_input_tokens_seen": 1548736,
      "step": 3150
    },
    {
      "epoch": 0.4163917117592715,
      "grad_norm": 0.050859030336141586,
      "learning_rate": 1.4501481518730372e-06,
      "loss": 0.223,
      "num_input_tokens_seen": 1551168,
      "step": 3155
    },
    {
      "epoch": 0.4170516035370199,
      "grad_norm": 0.2458106428384781,
      "learning_rate": 1.4480899881320868e-06,
      "loss": 0.0751,
      "num_input_tokens_seen": 1553664,
      "step": 3160
    },
    {
      "epoch": 0.41771149531476837,
      "grad_norm": 129.09617614746094,
      "learning_rate": 1.4460294466562956e-06,
      "loss": 0.1681,
      "num_input_tokens_seen": 1555968,
      "step": 3165
    },
    {
      "epoch": 0.4183713870925168,
      "grad_norm": 0.13649986684322357,
      "learning_rate": 1.4439665383796756e-06,
      "loss": 0.0974,
      "num_input_tokens_seen": 1558208,
      "step": 3170
    },
    {
      "epoch": 0.41903127887026526,
      "grad_norm": 1.2176202535629272,
      "learning_rate": 1.4419012742487972e-06,
      "loss": 0.003,
      "num_input_tokens_seen": 1560640,
      "step": 3175
    },
    {
      "epoch": 0.4196911706480137,
      "grad_norm": 0.7410973310470581,
      "learning_rate": 1.4398336652227335e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 1563328,
      "step": 3180
    },
    {
      "epoch": 0.4203510624257622,
      "grad_norm": 0.07748635113239288,
      "learning_rate": 1.4377637222729986e-06,
      "loss": 0.1522,
      "num_input_tokens_seen": 1565696,
      "step": 3185
    },
    {
      "epoch": 0.42101095420351065,
      "grad_norm": 0.073076032102108,
      "learning_rate": 1.435691456383493e-06,
      "loss": 0.1285,
      "num_input_tokens_seen": 1568640,
      "step": 3190
    },
    {
      "epoch": 0.4216708459812591,
      "grad_norm": 0.4473647475242615,
      "learning_rate": 1.433616878550442e-06,
      "loss": 0.128,
      "num_input_tokens_seen": 1571328,
      "step": 3195
    },
    {
      "epoch": 0.42233073775900754,
      "grad_norm": 17.642181396484375,
      "learning_rate": 1.4315399997823403e-06,
      "loss": 0.3784,
      "num_input_tokens_seen": 1574016,
      "step": 3200
    },
    {
      "epoch": 0.422990629536756,
      "grad_norm": 14.036954879760742,
      "learning_rate": 1.429460831099891e-06,
      "loss": 0.2654,
      "num_input_tokens_seen": 1576384,
      "step": 3205
    },
    {
      "epoch": 0.4236505213145044,
      "grad_norm": 0.1135653704404831,
      "learning_rate": 1.4273793835359492e-06,
      "loss": 0.2098,
      "num_input_tokens_seen": 1579200,
      "step": 3210
    },
    {
      "epoch": 0.42431041309225287,
      "grad_norm": 34.17720413208008,
      "learning_rate": 1.4252956681354631e-06,
      "loss": 0.1142,
      "num_input_tokens_seen": 1581632,
      "step": 3215
    },
    {
      "epoch": 0.4249703048700013,
      "grad_norm": 0.9577163457870483,
      "learning_rate": 1.4232096959554135e-06,
      "loss": 0.0092,
      "num_input_tokens_seen": 1584064,
      "step": 3220
    },
    {
      "epoch": 0.42563019664774976,
      "grad_norm": 43.000099182128906,
      "learning_rate": 1.4211214780647572e-06,
      "loss": 0.0457,
      "num_input_tokens_seen": 1586752,
      "step": 3225
    },
    {
      "epoch": 0.4262900884254982,
      "grad_norm": 12.291510581970215,
      "learning_rate": 1.4190310255443676e-06,
      "loss": 0.0764,
      "num_input_tokens_seen": 1589248,
      "step": 3230
    },
    {
      "epoch": 0.42694998020324665,
      "grad_norm": 0.037998050451278687,
      "learning_rate": 1.4169383494869764e-06,
      "loss": 0.0141,
      "num_input_tokens_seen": 1591552,
      "step": 3235
    },
    {
      "epoch": 0.4276098719809951,
      "grad_norm": 19.27412223815918,
      "learning_rate": 1.414843460997113e-06,
      "loss": 0.051,
      "num_input_tokens_seen": 1594048,
      "step": 3240
    },
    {
      "epoch": 0.4282697637587436,
      "grad_norm": 3.373072385787964,
      "learning_rate": 1.4127463711910483e-06,
      "loss": 0.1432,
      "num_input_tokens_seen": 1596544,
      "step": 3245
    },
    {
      "epoch": 0.42892965553649204,
      "grad_norm": 1.571363925933838,
      "learning_rate": 1.410647091196733e-06,
      "loss": 0.1601,
      "num_input_tokens_seen": 1599104,
      "step": 3250
    },
    {
      "epoch": 0.4295895473142405,
      "grad_norm": 0.016967564821243286,
      "learning_rate": 1.4085456321537402e-06,
      "loss": 0.1324,
      "num_input_tokens_seen": 1601344,
      "step": 3255
    },
    {
      "epoch": 0.43024943909198893,
      "grad_norm": 12.01763916015625,
      "learning_rate": 1.4064420052132056e-06,
      "loss": 0.2152,
      "num_input_tokens_seen": 1603968,
      "step": 3260
    },
    {
      "epoch": 0.4309093308697374,
      "grad_norm": 43.05500030517578,
      "learning_rate": 1.4043362215377696e-06,
      "loss": 0.0317,
      "num_input_tokens_seen": 1606400,
      "step": 3265
    },
    {
      "epoch": 0.4315692226474858,
      "grad_norm": 0.42787039279937744,
      "learning_rate": 1.4022282923015158e-06,
      "loss": 0.1258,
      "num_input_tokens_seen": 1608960,
      "step": 3270
    },
    {
      "epoch": 0.43222911442523426,
      "grad_norm": 0.7687868475914001,
      "learning_rate": 1.4001182286899136e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 1611456,
      "step": 3275
    },
    {
      "epoch": 0.4328890062029827,
      "grad_norm": 0.07358451187610626,
      "learning_rate": 1.398006041899758e-06,
      "loss": 0.0702,
      "num_input_tokens_seen": 1613952,
      "step": 3280
    },
    {
      "epoch": 0.43354889798073115,
      "grad_norm": 23.080883026123047,
      "learning_rate": 1.3958917431391102e-06,
      "loss": 0.1748,
      "num_input_tokens_seen": 1616320,
      "step": 3285
    },
    {
      "epoch": 0.4342087897584796,
      "grad_norm": 0.1040918156504631,
      "learning_rate": 1.3937753436272388e-06,
      "loss": 0.1856,
      "num_input_tokens_seen": 1619136,
      "step": 3290
    },
    {
      "epoch": 0.43486868153622804,
      "grad_norm": 174.55502319335938,
      "learning_rate": 1.3916568545945597e-06,
      "loss": 0.118,
      "num_input_tokens_seen": 1621632,
      "step": 3295
    },
    {
      "epoch": 0.4355285733139765,
      "grad_norm": 0.32156044244766235,
      "learning_rate": 1.3895362872825764e-06,
      "loss": 0.1392,
      "num_input_tokens_seen": 1624064,
      "step": 3300
    },
    {
      "epoch": 0.43618846509172493,
      "grad_norm": 0.5658749938011169,
      "learning_rate": 1.3874136529438205e-06,
      "loss": 0.0494,
      "num_input_tokens_seen": 1626496,
      "step": 3305
    },
    {
      "epoch": 0.43684835686947343,
      "grad_norm": 14.565958023071289,
      "learning_rate": 1.3852889628417918e-06,
      "loss": 0.1045,
      "num_input_tokens_seen": 1628800,
      "step": 3310
    },
    {
      "epoch": 0.4375082486472219,
      "grad_norm": 1.4210195541381836,
      "learning_rate": 1.3831622282508994e-06,
      "loss": 0.104,
      "num_input_tokens_seen": 1631232,
      "step": 3315
    },
    {
      "epoch": 0.4381681404249703,
      "grad_norm": 0.14714202284812927,
      "learning_rate": 1.3810334604564007e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 1633728,
      "step": 3320
    },
    {
      "epoch": 0.43882803220271877,
      "grad_norm": 0.5814191102981567,
      "learning_rate": 1.3789026707543423e-06,
      "loss": 0.0735,
      "num_input_tokens_seen": 1636224,
      "step": 3325
    },
    {
      "epoch": 0.4394879239804672,
      "grad_norm": 0.12930363416671753,
      "learning_rate": 1.3767698704514998e-06,
      "loss": 0.063,
      "num_input_tokens_seen": 1638272,
      "step": 3330
    },
    {
      "epoch": 0.44014781575821565,
      "grad_norm": 0.01950267143547535,
      "learning_rate": 1.3746350708653175e-06,
      "loss": 0.1941,
      "num_input_tokens_seen": 1640512,
      "step": 3335
    },
    {
      "epoch": 0.4408077075359641,
      "grad_norm": 30.412141799926758,
      "learning_rate": 1.3724982833238495e-06,
      "loss": 0.2224,
      "num_input_tokens_seen": 1642944,
      "step": 3340
    },
    {
      "epoch": 0.44146759931371254,
      "grad_norm": 0.06361498683691025,
      "learning_rate": 1.370359519165697e-06,
      "loss": 0.1334,
      "num_input_tokens_seen": 1645376,
      "step": 3345
    },
    {
      "epoch": 0.442127491091461,
      "grad_norm": 120.06082916259766,
      "learning_rate": 1.368218789739952e-06,
      "loss": 0.0429,
      "num_input_tokens_seen": 1647936,
      "step": 3350
    },
    {
      "epoch": 0.44278738286920943,
      "grad_norm": 0.1117718368768692,
      "learning_rate": 1.3660761064061337e-06,
      "loss": 0.1404,
      "num_input_tokens_seen": 1650496,
      "step": 3355
    },
    {
      "epoch": 0.4434472746469579,
      "grad_norm": 0.049969542771577835,
      "learning_rate": 1.3639314805341297e-06,
      "loss": 0.1065,
      "num_input_tokens_seen": 1652992,
      "step": 3360
    },
    {
      "epoch": 0.4441071664247063,
      "grad_norm": 66.16351318359375,
      "learning_rate": 1.3617849235041355e-06,
      "loss": 0.0935,
      "num_input_tokens_seen": 1655488,
      "step": 3365
    },
    {
      "epoch": 0.4447670582024548,
      "grad_norm": 0.091923289000988,
      "learning_rate": 1.3596364467065938e-06,
      "loss": 0.1084,
      "num_input_tokens_seen": 1657984,
      "step": 3370
    },
    {
      "epoch": 0.44542694998020327,
      "grad_norm": 31.401330947875977,
      "learning_rate": 1.3574860615421346e-06,
      "loss": 0.2514,
      "num_input_tokens_seen": 1660736,
      "step": 3375
    },
    {
      "epoch": 0.4460868417579517,
      "grad_norm": 14.01511287689209,
      "learning_rate": 1.3553337794215147e-06,
      "loss": 0.1354,
      "num_input_tokens_seen": 1663104,
      "step": 3380
    },
    {
      "epoch": 0.44674673353570016,
      "grad_norm": 32.54231262207031,
      "learning_rate": 1.3531796117655565e-06,
      "loss": 0.1002,
      "num_input_tokens_seen": 1665344,
      "step": 3385
    },
    {
      "epoch": 0.4474066253134486,
      "grad_norm": 123.50605773925781,
      "learning_rate": 1.3510235700050873e-06,
      "loss": 0.156,
      "num_input_tokens_seen": 1668096,
      "step": 3390
    },
    {
      "epoch": 0.44806651709119705,
      "grad_norm": 14.50474739074707,
      "learning_rate": 1.3488656655808801e-06,
      "loss": 0.0959,
      "num_input_tokens_seen": 1670272,
      "step": 3395
    },
    {
      "epoch": 0.4487264088689455,
      "grad_norm": 33.239410400390625,
      "learning_rate": 1.3467059099435912e-06,
      "loss": 0.0913,
      "num_input_tokens_seen": 1672448,
      "step": 3400
    },
    {
      "epoch": 0.44938630064669394,
      "grad_norm": 0.29679805040359497,
      "learning_rate": 1.3445443145537002e-06,
      "loss": 0.0055,
      "num_input_tokens_seen": 1675200,
      "step": 3405
    },
    {
      "epoch": 0.4500461924244424,
      "grad_norm": 3.9020371437072754,
      "learning_rate": 1.3423808908814494e-06,
      "loss": 0.0509,
      "num_input_tokens_seen": 1677696,
      "step": 3410
    },
    {
      "epoch": 0.45017817077999206,
      "eval_loss": 0.11581598222255707,
      "eval_runtime": 7.8583,
      "eval_samples_per_second": 857.06,
      "eval_steps_per_second": 107.148,
      "num_input_tokens_seen": 1678208,
      "step": 3411
    },
    {
      "epoch": 0.4507060842021908,
      "grad_norm": 14.313544273376465,
      "learning_rate": 1.3402156504067826e-06,
      "loss": 0.0982,
      "num_input_tokens_seen": 1680256,
      "step": 3415
    },
    {
      "epoch": 0.45136597597993927,
      "grad_norm": 0.17114846408367157,
      "learning_rate": 1.338048604619284e-06,
      "loss": 0.0816,
      "num_input_tokens_seen": 1682624,
      "step": 3420
    },
    {
      "epoch": 0.4520258677576877,
      "grad_norm": 68.34577178955078,
      "learning_rate": 1.3358797650181178e-06,
      "loss": 0.0548,
      "num_input_tokens_seen": 1685056,
      "step": 3425
    },
    {
      "epoch": 0.45268575953543616,
      "grad_norm": 168.3812713623047,
      "learning_rate": 1.3337091431119662e-06,
      "loss": 0.1743,
      "num_input_tokens_seen": 1687168,
      "step": 3430
    },
    {
      "epoch": 0.45334565131318466,
      "grad_norm": 292.0270690917969,
      "learning_rate": 1.3315367504189698e-06,
      "loss": 0.263,
      "num_input_tokens_seen": 1689216,
      "step": 3435
    },
    {
      "epoch": 0.4540055430909331,
      "grad_norm": 87.19351959228516,
      "learning_rate": 1.3293625984666656e-06,
      "loss": 0.0704,
      "num_input_tokens_seen": 1691776,
      "step": 3440
    },
    {
      "epoch": 0.45466543486868155,
      "grad_norm": 0.12149941176176071,
      "learning_rate": 1.3271866987919254e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 1694336,
      "step": 3445
    },
    {
      "epoch": 0.45532532664643,
      "grad_norm": 0.1733410805463791,
      "learning_rate": 1.325009062940895e-06,
      "loss": 0.2066,
      "num_input_tokens_seen": 1696640,
      "step": 3450
    },
    {
      "epoch": 0.45598521842417844,
      "grad_norm": 16.03499984741211,
      "learning_rate": 1.3228297024689336e-06,
      "loss": 0.0826,
      "num_input_tokens_seen": 1698880,
      "step": 3455
    },
    {
      "epoch": 0.4566451102019269,
      "grad_norm": 2.5192947387695312,
      "learning_rate": 1.3206486289405519e-06,
      "loss": 0.1221,
      "num_input_tokens_seen": 1701312,
      "step": 3460
    },
    {
      "epoch": 0.45730500197967533,
      "grad_norm": 51.37282180786133,
      "learning_rate": 1.3184658539293496e-06,
      "loss": 0.1446,
      "num_input_tokens_seen": 1703808,
      "step": 3465
    },
    {
      "epoch": 0.4579648937574238,
      "grad_norm": 142.88600158691406,
      "learning_rate": 1.3162813890179564e-06,
      "loss": 0.132,
      "num_input_tokens_seen": 1706304,
      "step": 3470
    },
    {
      "epoch": 0.4586247855351722,
      "grad_norm": 13.749252319335938,
      "learning_rate": 1.314095245797969e-06,
      "loss": 0.236,
      "num_input_tokens_seen": 1708736,
      "step": 3475
    },
    {
      "epoch": 0.45928467731292066,
      "grad_norm": 0.7829024791717529,
      "learning_rate": 1.3119074358698891e-06,
      "loss": 0.1477,
      "num_input_tokens_seen": 1711232,
      "step": 3480
    },
    {
      "epoch": 0.4599445690906691,
      "grad_norm": 1.0132334232330322,
      "learning_rate": 1.3097179708430634e-06,
      "loss": 0.003,
      "num_input_tokens_seen": 1713600,
      "step": 3485
    },
    {
      "epoch": 0.46060446086841755,
      "grad_norm": 0.1129785031080246,
      "learning_rate": 1.3075268623356214e-06,
      "loss": 0.1998,
      "num_input_tokens_seen": 1716224,
      "step": 3490
    },
    {
      "epoch": 0.46126435264616605,
      "grad_norm": 41.71617889404297,
      "learning_rate": 1.305334121974412e-06,
      "loss": 0.1462,
      "num_input_tokens_seen": 1718720,
      "step": 3495
    },
    {
      "epoch": 0.4619242444239145,
      "grad_norm": 30.289953231811523,
      "learning_rate": 1.3031397613949448e-06,
      "loss": 0.0858,
      "num_input_tokens_seen": 1721280,
      "step": 3500
    },
    {
      "epoch": 0.46258413620166294,
      "grad_norm": 70.03472137451172,
      "learning_rate": 1.3009437922413266e-06,
      "loss": 0.1099,
      "num_input_tokens_seen": 1723712,
      "step": 3505
    },
    {
      "epoch": 0.4632440279794114,
      "grad_norm": 112.14954376220703,
      "learning_rate": 1.2987462261661994e-06,
      "loss": 0.1115,
      "num_input_tokens_seen": 1725952,
      "step": 3510
    },
    {
      "epoch": 0.46390391975715983,
      "grad_norm": 0.32890069484710693,
      "learning_rate": 1.2965470748306798e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 1728512,
      "step": 3515
    },
    {
      "epoch": 0.4645638115349083,
      "grad_norm": 69.49401092529297,
      "learning_rate": 1.2943463499042957e-06,
      "loss": 0.1338,
      "num_input_tokens_seen": 1731008,
      "step": 3520
    },
    {
      "epoch": 0.4652237033126567,
      "grad_norm": 1.0032602548599243,
      "learning_rate": 1.2921440630649257e-06,
      "loss": 0.0485,
      "num_input_tokens_seen": 1733696,
      "step": 3525
    },
    {
      "epoch": 0.46588359509040517,
      "grad_norm": 221.371826171875,
      "learning_rate": 1.2899402259987355e-06,
      "loss": 0.1013,
      "num_input_tokens_seen": 1736256,
      "step": 3530
    },
    {
      "epoch": 0.4665434868681536,
      "grad_norm": 0.41186392307281494,
      "learning_rate": 1.287734850400118e-06,
      "loss": 0.2473,
      "num_input_tokens_seen": 1738944,
      "step": 3535
    },
    {
      "epoch": 0.46720337864590206,
      "grad_norm": 16.863187789916992,
      "learning_rate": 1.2855279479716297e-06,
      "loss": 0.3237,
      "num_input_tokens_seen": 1741568,
      "step": 3540
    },
    {
      "epoch": 0.4678632704236505,
      "grad_norm": 0.3010137379169464,
      "learning_rate": 1.283319530423929e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 1743808,
      "step": 3545
    },
    {
      "epoch": 0.46852316220139895,
      "grad_norm": 0.09695116430521011,
      "learning_rate": 1.2811096094757144e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 1746176,
      "step": 3550
    },
    {
      "epoch": 0.46918305397914745,
      "grad_norm": 0.07640713453292847,
      "learning_rate": 1.2788981968536612e-06,
      "loss": 0.2019,
      "num_input_tokens_seen": 1748608,
      "step": 3555
    },
    {
      "epoch": 0.4698429457568959,
      "grad_norm": 0.21313799917697906,
      "learning_rate": 1.2766853042923607e-06,
      "loss": 0.1542,
      "num_input_tokens_seen": 1751040,
      "step": 3560
    },
    {
      "epoch": 0.47050283753464434,
      "grad_norm": 0.6278545260429382,
      "learning_rate": 1.2744709435342573e-06,
      "loss": 0.0623,
      "num_input_tokens_seen": 1753280,
      "step": 3565
    },
    {
      "epoch": 0.4711627293123928,
      "grad_norm": 14.035211563110352,
      "learning_rate": 1.2722551263295864e-06,
      "loss": 0.2919,
      "num_input_tokens_seen": 1755712,
      "step": 3570
    },
    {
      "epoch": 0.4718226210901412,
      "grad_norm": 41.04814910888672,
      "learning_rate": 1.2700378644363114e-06,
      "loss": 0.1294,
      "num_input_tokens_seen": 1757952,
      "step": 3575
    },
    {
      "epoch": 0.47248251286788967,
      "grad_norm": 29.484533309936523,
      "learning_rate": 1.2678191696200621e-06,
      "loss": 0.0883,
      "num_input_tokens_seen": 1760384,
      "step": 3580
    },
    {
      "epoch": 0.4731424046456381,
      "grad_norm": 0.2581002712249756,
      "learning_rate": 1.2655990536540717e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 1762944,
      "step": 3585
    },
    {
      "epoch": 0.47380229642338656,
      "grad_norm": 33.701026916503906,
      "learning_rate": 1.2633775283191144e-06,
      "loss": 0.2308,
      "num_input_tokens_seen": 1765504,
      "step": 3590
    },
    {
      "epoch": 0.474462188201135,
      "grad_norm": 0.5347998738288879,
      "learning_rate": 1.2611546054034436e-06,
      "loss": 0.0084,
      "num_input_tokens_seen": 1768128,
      "step": 3595
    },
    {
      "epoch": 0.47512207997888345,
      "grad_norm": 0.7026123404502869,
      "learning_rate": 1.2589302967027285e-06,
      "loss": 0.1521,
      "num_input_tokens_seen": 1770624,
      "step": 3600
    },
    {
      "epoch": 0.4757819717566319,
      "grad_norm": 17.028770446777344,
      "learning_rate": 1.2567046140199914e-06,
      "loss": 0.3261,
      "num_input_tokens_seen": 1773248,
      "step": 3605
    },
    {
      "epoch": 0.47644186353438034,
      "grad_norm": 0.5768800377845764,
      "learning_rate": 1.2544775691655463e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 1775488,
      "step": 3610
    },
    {
      "epoch": 0.4771017553121288,
      "grad_norm": 0.33122268319129944,
      "learning_rate": 1.2522491739569346e-06,
      "loss": 0.138,
      "num_input_tokens_seen": 1777792,
      "step": 3615
    },
    {
      "epoch": 0.4777616470898773,
      "grad_norm": 3.622959852218628,
      "learning_rate": 1.250019440218864e-06,
      "loss": 0.1041,
      "num_input_tokens_seen": 1780352,
      "step": 3620
    },
    {
      "epoch": 0.47842153886762573,
      "grad_norm": 0.5977795124053955,
      "learning_rate": 1.247788379783144e-06,
      "loss": 0.1707,
      "num_input_tokens_seen": 1783168,
      "step": 3625
    },
    {
      "epoch": 0.4790814306453742,
      "grad_norm": 30.13225555419922,
      "learning_rate": 1.2455560044886248e-06,
      "loss": 0.0886,
      "num_input_tokens_seen": 1785920,
      "step": 3630
    },
    {
      "epoch": 0.4797413224231226,
      "grad_norm": 0.14322948455810547,
      "learning_rate": 1.2433223261811337e-06,
      "loss": 0.1456,
      "num_input_tokens_seen": 1788416,
      "step": 3635
    },
    {
      "epoch": 0.48040121420087106,
      "grad_norm": 0.2782398462295532,
      "learning_rate": 1.2410873567134115e-06,
      "loss": 0.0141,
      "num_input_tokens_seen": 1790848,
      "step": 3640
    },
    {
      "epoch": 0.4810611059786195,
      "grad_norm": 10.446566581726074,
      "learning_rate": 1.238851107945051e-06,
      "loss": 0.0342,
      "num_input_tokens_seen": 1793280,
      "step": 3645
    },
    {
      "epoch": 0.48172099775636795,
      "grad_norm": 0.0587058961391449,
      "learning_rate": 1.2366135917424341e-06,
      "loss": 0.1088,
      "num_input_tokens_seen": 1795648,
      "step": 3650
    },
    {
      "epoch": 0.4823808895341164,
      "grad_norm": 18.86115264892578,
      "learning_rate": 1.2343748199786665e-06,
      "loss": 0.1796,
      "num_input_tokens_seen": 1797952,
      "step": 3655
    },
    {
      "epoch": 0.48304078131186484,
      "grad_norm": 0.4261041283607483,
      "learning_rate": 1.2321348045335182e-06,
      "loss": 0.0595,
      "num_input_tokens_seen": 1800192,
      "step": 3660
    },
    {
      "epoch": 0.4837006730896133,
      "grad_norm": 0.32942232489585876,
      "learning_rate": 1.2298935572933575e-06,
      "loss": 0.1465,
      "num_input_tokens_seen": 1802560,
      "step": 3665
    },
    {
      "epoch": 0.48436056486736173,
      "grad_norm": 23.043102264404297,
      "learning_rate": 1.2276510901510892e-06,
      "loss": 0.1584,
      "num_input_tokens_seen": 1805056,
      "step": 3670
    },
    {
      "epoch": 0.4850204566451102,
      "grad_norm": 0.42907005548477173,
      "learning_rate": 1.2254074150060915e-06,
      "loss": 0.1682,
      "num_input_tokens_seen": 1807744,
      "step": 3675
    },
    {
      "epoch": 0.4856803484228587,
      "grad_norm": 46.6706657409668,
      "learning_rate": 1.2231625437641535e-06,
      "loss": 0.1522,
      "num_input_tokens_seen": 1810368,
      "step": 3680
    },
    {
      "epoch": 0.4863402402006071,
      "grad_norm": 0.21515221893787384,
      "learning_rate": 1.2209164883374096e-06,
      "loss": 0.045,
      "num_input_tokens_seen": 1813056,
      "step": 3685
    },
    {
      "epoch": 0.48700013197835557,
      "grad_norm": 0.3806566894054413,
      "learning_rate": 1.2186692606442793e-06,
      "loss": 0.22,
      "num_input_tokens_seen": 1815360,
      "step": 3690
    },
    {
      "epoch": 0.487660023756104,
      "grad_norm": 59.212196350097656,
      "learning_rate": 1.216420872609402e-06,
      "loss": 0.1641,
      "num_input_tokens_seen": 1817920,
      "step": 3695
    },
    {
      "epoch": 0.48831991553385246,
      "grad_norm": 16.957406997680664,
      "learning_rate": 1.2141713361635739e-06,
      "loss": 0.0635,
      "num_input_tokens_seen": 1820288,
      "step": 3700
    },
    {
      "epoch": 0.4889798073116009,
      "grad_norm": 0.02635974995791912,
      "learning_rate": 1.2119206632436864e-06,
      "loss": 0.1703,
      "num_input_tokens_seen": 1822656,
      "step": 3705
    },
    {
      "epoch": 0.48963969908934935,
      "grad_norm": 0.6288211345672607,
      "learning_rate": 1.209668865792661e-06,
      "loss": 0.0631,
      "num_input_tokens_seen": 1824832,
      "step": 3710
    },
    {
      "epoch": 0.4902995908670978,
      "grad_norm": 176.1214599609375,
      "learning_rate": 1.207415955759385e-06,
      "loss": 0.0998,
      "num_input_tokens_seen": 1827200,
      "step": 3715
    },
    {
      "epoch": 0.49095948264484623,
      "grad_norm": 17.406232833862305,
      "learning_rate": 1.2051619450986514e-06,
      "loss": 0.1581,
      "num_input_tokens_seen": 1829632,
      "step": 3720
    },
    {
      "epoch": 0.4916193744225947,
      "grad_norm": 6.454481601715088,
      "learning_rate": 1.2029068457710923e-06,
      "loss": 0.1271,
      "num_input_tokens_seen": 1832192,
      "step": 3725
    },
    {
      "epoch": 0.4922792662003431,
      "grad_norm": 2.918125867843628,
      "learning_rate": 1.200650669743117e-06,
      "loss": 0.091,
      "num_input_tokens_seen": 1834752,
      "step": 3730
    },
    {
      "epoch": 0.49293915797809157,
      "grad_norm": 19.404775619506836,
      "learning_rate": 1.1983934289868488e-06,
      "loss": 0.116,
      "num_input_tokens_seen": 1837248,
      "step": 3735
    },
    {
      "epoch": 0.49359904975584007,
      "grad_norm": 8.872051239013672,
      "learning_rate": 1.1961351354800595e-06,
      "loss": 0.1431,
      "num_input_tokens_seen": 1839680,
      "step": 3740
    },
    {
      "epoch": 0.4942589415335885,
      "grad_norm": 0.526841402053833,
      "learning_rate": 1.193875801206109e-06,
      "loss": 0.1399,
      "num_input_tokens_seen": 1842304,
      "step": 3745
    },
    {
      "epoch": 0.49491883331133696,
      "grad_norm": 0.18063127994537354,
      "learning_rate": 1.1916154381538786e-06,
      "loss": 0.0471,
      "num_input_tokens_seen": 1844480,
      "step": 3750
    },
    {
      "epoch": 0.4955787250890854,
      "grad_norm": 0.19460546970367432,
      "learning_rate": 1.1893540583177083e-06,
      "loss": 0.1858,
      "num_input_tokens_seen": 1846912,
      "step": 3755
    },
    {
      "epoch": 0.49623861686683385,
      "grad_norm": 14.68387508392334,
      "learning_rate": 1.187091673697335e-06,
      "loss": 0.0801,
      "num_input_tokens_seen": 1849024,
      "step": 3760
    },
    {
      "epoch": 0.4968985086445823,
      "grad_norm": 0.8316734433174133,
      "learning_rate": 1.184828296297826e-06,
      "loss": 0.0691,
      "num_input_tokens_seen": 1851712,
      "step": 3765
    },
    {
      "epoch": 0.49755840042233074,
      "grad_norm": 38.36271667480469,
      "learning_rate": 1.182563938129518e-06,
      "loss": 0.0697,
      "num_input_tokens_seen": 1854208,
      "step": 3770
    },
    {
      "epoch": 0.4982182922000792,
      "grad_norm": 0.08225525915622711,
      "learning_rate": 1.1802986112079507e-06,
      "loss": 0.0893,
      "num_input_tokens_seen": 1856704,
      "step": 3775
    },
    {
      "epoch": 0.4988781839778276,
      "grad_norm": 0.5821913480758667,
      "learning_rate": 1.1780323275538056e-06,
      "loss": 0.0831,
      "num_input_tokens_seen": 1858944,
      "step": 3780
    },
    {
      "epoch": 0.49953807575557607,
      "grad_norm": 0.5157123804092407,
      "learning_rate": 1.1757650991928393e-06,
      "loss": 0.2059,
      "num_input_tokens_seen": 1861696,
      "step": 3785
    },
    {
      "epoch": 0.5001979675333246,
      "grad_norm": 0.31598079204559326,
      "learning_rate": 1.1734969381558235e-06,
      "loss": 0.179,
      "num_input_tokens_seen": 1864128,
      "step": 3790
    },
    {
      "epoch": 0.5001979675333246,
      "eval_loss": 0.12760794162750244,
      "eval_runtime": 7.9443,
      "eval_samples_per_second": 847.779,
      "eval_steps_per_second": 105.988,
      "num_input_tokens_seen": 1864128,
      "step": 3790
    },
    {
      "epoch": 0.500857859311073,
      "grad_norm": 0.1666107177734375,
      "learning_rate": 1.1712278564784774e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 1866432,
      "step": 3795
    },
    {
      "epoch": 0.5015177510888215,
      "grad_norm": 20.980514526367188,
      "learning_rate": 1.1689578662014064e-06,
      "loss": 0.0598,
      "num_input_tokens_seen": 1868736,
      "step": 3800
    },
    {
      "epoch": 0.5021776428665699,
      "grad_norm": 14.8715181350708,
      "learning_rate": 1.1666869793700362e-06,
      "loss": 0.3077,
      "num_input_tokens_seen": 1871360,
      "step": 3805
    },
    {
      "epoch": 0.5028375346443184,
      "grad_norm": 0.031641654670238495,
      "learning_rate": 1.1644152080345515e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 1873536,
      "step": 3810
    },
    {
      "epoch": 0.5034974264220667,
      "grad_norm": 28.584814071655273,
      "learning_rate": 1.1621425642498289e-06,
      "loss": 0.2264,
      "num_input_tokens_seen": 1875904,
      "step": 3815
    },
    {
      "epoch": 0.5041573181998152,
      "grad_norm": 405.0102233886719,
      "learning_rate": 1.1598690600753759e-06,
      "loss": 0.2478,
      "num_input_tokens_seen": 1878464,
      "step": 3820
    },
    {
      "epoch": 0.5048172099775636,
      "grad_norm": 34.297298431396484,
      "learning_rate": 1.1575947075752644e-06,
      "loss": 0.2181,
      "num_input_tokens_seen": 1880640,
      "step": 3825
    },
    {
      "epoch": 0.5054771017553121,
      "grad_norm": 145.9020538330078,
      "learning_rate": 1.1553195188180691e-06,
      "loss": 0.1516,
      "num_input_tokens_seen": 1882944,
      "step": 3830
    },
    {
      "epoch": 0.5061369935330606,
      "grad_norm": 1.2247670888900757,
      "learning_rate": 1.1530435058768008e-06,
      "loss": 0.0694,
      "num_input_tokens_seen": 1885248,
      "step": 3835
    },
    {
      "epoch": 0.506796885310809,
      "grad_norm": 19.718626022338867,
      "learning_rate": 1.150766680828845e-06,
      "loss": 0.0841,
      "num_input_tokens_seen": 1887872,
      "step": 3840
    },
    {
      "epoch": 0.5074567770885575,
      "grad_norm": 1.9999217987060547,
      "learning_rate": 1.1484890557558955e-06,
      "loss": 0.0024,
      "num_input_tokens_seen": 1890560,
      "step": 3845
    },
    {
      "epoch": 0.5081166688663059,
      "grad_norm": 19.390962600708008,
      "learning_rate": 1.146210642743892e-06,
      "loss": 0.076,
      "num_input_tokens_seen": 1893056,
      "step": 3850
    },
    {
      "epoch": 0.5087765606440544,
      "grad_norm": 0.07549047470092773,
      "learning_rate": 1.1439314538829554e-06,
      "loss": 0.0353,
      "num_input_tokens_seen": 1895360,
      "step": 3855
    },
    {
      "epoch": 0.5094364524218028,
      "grad_norm": 0.08527141809463501,
      "learning_rate": 1.141651501267323e-06,
      "loss": 0.0793,
      "num_input_tokens_seen": 1897664,
      "step": 3860
    },
    {
      "epoch": 0.5100963441995513,
      "grad_norm": 1.7610729932785034,
      "learning_rate": 1.1393707969952847e-06,
      "loss": 0.1073,
      "num_input_tokens_seen": 1900288,
      "step": 3865
    },
    {
      "epoch": 0.5107562359772997,
      "grad_norm": 0.7622396945953369,
      "learning_rate": 1.13708935316912e-06,
      "loss": 0.0796,
      "num_input_tokens_seen": 1903040,
      "step": 3870
    },
    {
      "epoch": 0.5114161277550482,
      "grad_norm": 0.33791011571884155,
      "learning_rate": 1.134807181895032e-06,
      "loss": 0.0173,
      "num_input_tokens_seen": 1905472,
      "step": 3875
    },
    {
      "epoch": 0.5120760195327966,
      "grad_norm": 0.7325949668884277,
      "learning_rate": 1.132524295283084e-06,
      "loss": 0.0637,
      "num_input_tokens_seen": 1907712,
      "step": 3880
    },
    {
      "epoch": 0.5127359113105451,
      "grad_norm": 0.04774130508303642,
      "learning_rate": 1.1302407054471355e-06,
      "loss": 0.0912,
      "num_input_tokens_seen": 1910080,
      "step": 3885
    },
    {
      "epoch": 0.5133958030882935,
      "grad_norm": 0.2512286305427551,
      "learning_rate": 1.1279564245047767e-06,
      "loss": 0.2374,
      "num_input_tokens_seen": 1912512,
      "step": 3890
    },
    {
      "epoch": 0.514055694866042,
      "grad_norm": 0.08980656415224075,
      "learning_rate": 1.1256714645772662e-06,
      "loss": 0.1336,
      "num_input_tokens_seen": 1914752,
      "step": 3895
    },
    {
      "epoch": 0.5147155866437905,
      "grad_norm": 0.0983128771185875,
      "learning_rate": 1.1233858377894647e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 1917120,
      "step": 3900
    },
    {
      "epoch": 0.5153754784215389,
      "grad_norm": 74.96298217773438,
      "learning_rate": 1.1210995562697722e-06,
      "loss": 0.014,
      "num_input_tokens_seen": 1919232,
      "step": 3905
    },
    {
      "epoch": 0.5160353701992874,
      "grad_norm": 269.811279296875,
      "learning_rate": 1.1188126321500621e-06,
      "loss": 0.0559,
      "num_input_tokens_seen": 1921856,
      "step": 3910
    },
    {
      "epoch": 0.5166952619770357,
      "grad_norm": 32.522220611572266,
      "learning_rate": 1.1165250775656188e-06,
      "loss": 0.1515,
      "num_input_tokens_seen": 1924224,
      "step": 3915
    },
    {
      "epoch": 0.5173551537547842,
      "grad_norm": 0.08811689168214798,
      "learning_rate": 1.1142369046550708e-06,
      "loss": 0.0373,
      "num_input_tokens_seen": 1926464,
      "step": 3920
    },
    {
      "epoch": 0.5180150455325326,
      "grad_norm": 0.0877578929066658,
      "learning_rate": 1.1119481255603289e-06,
      "loss": 0.1665,
      "num_input_tokens_seen": 1928896,
      "step": 3925
    },
    {
      "epoch": 0.5186749373102811,
      "grad_norm": 58.392852783203125,
      "learning_rate": 1.1096587524265197e-06,
      "loss": 0.115,
      "num_input_tokens_seen": 1931200,
      "step": 3930
    },
    {
      "epoch": 0.5193348290880295,
      "grad_norm": 0.029454190284013748,
      "learning_rate": 1.107368797401923e-06,
      "loss": 0.0542,
      "num_input_tokens_seen": 1933632,
      "step": 3935
    },
    {
      "epoch": 0.519994720865778,
      "grad_norm": 0.17615357041358948,
      "learning_rate": 1.1050782726379054e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 1935872,
      "step": 3940
    },
    {
      "epoch": 0.5206546126435264,
      "grad_norm": 0.39868056774139404,
      "learning_rate": 1.1027871902888566e-06,
      "loss": 0.0951,
      "num_input_tokens_seen": 1938048,
      "step": 3945
    },
    {
      "epoch": 0.5213145044212749,
      "grad_norm": 66.22596740722656,
      "learning_rate": 1.1004955625121257e-06,
      "loss": 0.056,
      "num_input_tokens_seen": 1940608,
      "step": 3950
    },
    {
      "epoch": 0.5219743961990233,
      "grad_norm": 24.409608840942383,
      "learning_rate": 1.0982034014679561e-06,
      "loss": 0.2829,
      "num_input_tokens_seen": 1943040,
      "step": 3955
    },
    {
      "epoch": 0.5226342879767718,
      "grad_norm": 0.12048153579235077,
      "learning_rate": 1.0959107193194206e-06,
      "loss": 0.2808,
      "num_input_tokens_seen": 1945664,
      "step": 3960
    },
    {
      "epoch": 0.5232941797545203,
      "grad_norm": 0.041272152215242386,
      "learning_rate": 1.0936175282323575e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 1948032,
      "step": 3965
    },
    {
      "epoch": 0.5239540715322687,
      "grad_norm": 0.2881408929824829,
      "learning_rate": 1.091323840375305e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 1950208,
      "step": 3970
    },
    {
      "epoch": 0.5246139633100172,
      "grad_norm": 57.43034362792969,
      "learning_rate": 1.0890296679194378e-06,
      "loss": 0.2119,
      "num_input_tokens_seen": 1952896,
      "step": 3975
    },
    {
      "epoch": 0.5252738550877656,
      "grad_norm": 0.0778154581785202,
      "learning_rate": 1.086735023038502e-06,
      "loss": 0.068,
      "num_input_tokens_seen": 1955200,
      "step": 3980
    },
    {
      "epoch": 0.5259337468655141,
      "grad_norm": 0.2079089879989624,
      "learning_rate": 1.0844399179087512e-06,
      "loss": 0.0953,
      "num_input_tokens_seen": 1957376,
      "step": 3985
    },
    {
      "epoch": 0.5265936386432625,
      "grad_norm": 0.25119689106941223,
      "learning_rate": 1.0821443647088802e-06,
      "loss": 0.2058,
      "num_input_tokens_seen": 1960064,
      "step": 3990
    },
    {
      "epoch": 0.527253530421011,
      "grad_norm": 0.09684748202562332,
      "learning_rate": 1.0798483756199623e-06,
      "loss": 0.0652,
      "num_input_tokens_seen": 1962624,
      "step": 3995
    },
    {
      "epoch": 0.5279134221987594,
      "grad_norm": 0.08053699135780334,
      "learning_rate": 1.0775519628253833e-06,
      "loss": 0.0704,
      "num_input_tokens_seen": 1965056,
      "step": 4000
    },
    {
      "epoch": 0.5285733139765079,
      "grad_norm": 0.2662527859210968,
      "learning_rate": 1.0752551385107772e-06,
      "loss": 0.1949,
      "num_input_tokens_seen": 1967424,
      "step": 4005
    },
    {
      "epoch": 0.5292332057542563,
      "grad_norm": 18.517011642456055,
      "learning_rate": 1.0729579148639621e-06,
      "loss": 0.1433,
      "num_input_tokens_seen": 1969856,
      "step": 4010
    },
    {
      "epoch": 0.5298930975320048,
      "grad_norm": 0.126367449760437,
      "learning_rate": 1.0706603040748747e-06,
      "loss": 0.0517,
      "num_input_tokens_seen": 1972544,
      "step": 4015
    },
    {
      "epoch": 0.5305529893097533,
      "grad_norm": 0.04751770943403244,
      "learning_rate": 1.0683623183355071e-06,
      "loss": 0.0448,
      "num_input_tokens_seen": 1974912,
      "step": 4020
    },
    {
      "epoch": 0.5312128810875016,
      "grad_norm": 16.610794067382812,
      "learning_rate": 1.0660639698398392e-06,
      "loss": 0.1542,
      "num_input_tokens_seen": 1977216,
      "step": 4025
    },
    {
      "epoch": 0.5318727728652501,
      "grad_norm": 100.28923034667969,
      "learning_rate": 1.0637652707837773e-06,
      "loss": 0.1436,
      "num_input_tokens_seen": 1979648,
      "step": 4030
    },
    {
      "epoch": 0.5325326646429985,
      "grad_norm": 20.682844161987305,
      "learning_rate": 1.0614662333650876e-06,
      "loss": 0.1039,
      "num_input_tokens_seen": 1981888,
      "step": 4035
    },
    {
      "epoch": 0.533192556420747,
      "grad_norm": 19.53425407409668,
      "learning_rate": 1.0591668697833311e-06,
      "loss": 0.2167,
      "num_input_tokens_seen": 1984448,
      "step": 4040
    },
    {
      "epoch": 0.5338524481984954,
      "grad_norm": 67.5270004272461,
      "learning_rate": 1.0568671922398005e-06,
      "loss": 0.1567,
      "num_input_tokens_seen": 1987072,
      "step": 4045
    },
    {
      "epoch": 0.5345123399762439,
      "grad_norm": 1.1474074125289917,
      "learning_rate": 1.054567212937454e-06,
      "loss": 0.1618,
      "num_input_tokens_seen": 1989632,
      "step": 4050
    },
    {
      "epoch": 0.5351722317539923,
      "grad_norm": 0.28706297278404236,
      "learning_rate": 1.0522669440808508e-06,
      "loss": 0.0367,
      "num_input_tokens_seen": 1992192,
      "step": 4055
    },
    {
      "epoch": 0.5358321235317408,
      "grad_norm": 50.1959114074707,
      "learning_rate": 1.0499663978760871e-06,
      "loss": 0.2607,
      "num_input_tokens_seen": 1994624,
      "step": 4060
    },
    {
      "epoch": 0.5364920153094892,
      "grad_norm": 13.985773086547852,
      "learning_rate": 1.0476655865307308e-06,
      "loss": 0.0821,
      "num_input_tokens_seen": 1997056,
      "step": 4065
    },
    {
      "epoch": 0.5371519070872377,
      "grad_norm": 0.9565370082855225,
      "learning_rate": 1.0453645222537556e-06,
      "loss": 0.0918,
      "num_input_tokens_seen": 1999360,
      "step": 4070
    },
    {
      "epoch": 0.5378117988649861,
      "grad_norm": 53.97791290283203,
      "learning_rate": 1.0430632172554796e-06,
      "loss": 0.0861,
      "num_input_tokens_seen": 2001856,
      "step": 4075
    },
    {
      "epoch": 0.5384716906427346,
      "grad_norm": 38.577674865722656,
      "learning_rate": 1.0407616837474963e-06,
      "loss": 0.1321,
      "num_input_tokens_seen": 2004288,
      "step": 4080
    },
    {
      "epoch": 0.5391315824204831,
      "grad_norm": 1.6322799921035767,
      "learning_rate": 1.038459933942612e-06,
      "loss": 0.0784,
      "num_input_tokens_seen": 2006976,
      "step": 4085
    },
    {
      "epoch": 0.5397914741982315,
      "grad_norm": 57.15738296508789,
      "learning_rate": 1.036157980054782e-06,
      "loss": 0.0603,
      "num_input_tokens_seen": 2009280,
      "step": 4090
    },
    {
      "epoch": 0.54045136597598,
      "grad_norm": 201.4349365234375,
      "learning_rate": 1.0338558342990431e-06,
      "loss": 0.0969,
      "num_input_tokens_seen": 2011776,
      "step": 4095
    },
    {
      "epoch": 0.5411112577537284,
      "grad_norm": 15.186023712158203,
      "learning_rate": 1.0315535088914508e-06,
      "loss": 0.2581,
      "num_input_tokens_seen": 2014336,
      "step": 4100
    },
    {
      "epoch": 0.5417711495314769,
      "grad_norm": 22.846664428710938,
      "learning_rate": 1.0292510160490146e-06,
      "loss": 0.1642,
      "num_input_tokens_seen": 2017152,
      "step": 4105
    },
    {
      "epoch": 0.5424310413092253,
      "grad_norm": 21.382556915283203,
      "learning_rate": 1.0269483679896308e-06,
      "loss": 0.0998,
      "num_input_tokens_seen": 2019520,
      "step": 4110
    },
    {
      "epoch": 0.5430909330869738,
      "grad_norm": 50.146427154541016,
      "learning_rate": 1.0246455769320211e-06,
      "loss": 0.162,
      "num_input_tokens_seen": 2021632,
      "step": 4115
    },
    {
      "epoch": 0.5437508248647221,
      "grad_norm": 0.8249346017837524,
      "learning_rate": 1.0223426550956647e-06,
      "loss": 0.1261,
      "num_input_tokens_seen": 2023744,
      "step": 4120
    },
    {
      "epoch": 0.5444107166424706,
      "grad_norm": 1.6365822553634644,
      "learning_rate": 1.0200396147007354e-06,
      "loss": 0.0762,
      "num_input_tokens_seen": 2026048,
      "step": 4125
    },
    {
      "epoch": 0.545070608420219,
      "grad_norm": 0.0482165552675724,
      "learning_rate": 1.0177364679680367e-06,
      "loss": 0.132,
      "num_input_tokens_seen": 2028352,
      "step": 4130
    },
    {
      "epoch": 0.5457305001979675,
      "grad_norm": 0.10031591355800629,
      "learning_rate": 1.015433227118935e-06,
      "loss": 0.0602,
      "num_input_tokens_seen": 2030848,
      "step": 4135
    },
    {
      "epoch": 0.5463903919757159,
      "grad_norm": 0.14653390645980835,
      "learning_rate": 1.0131299043752967e-06,
      "loss": 0.1337,
      "num_input_tokens_seen": 2033344,
      "step": 4140
    },
    {
      "epoch": 0.5470502837534644,
      "grad_norm": 0.3093219995498657,
      "learning_rate": 1.0108265119594233e-06,
      "loss": 0.0548,
      "num_input_tokens_seen": 2035584,
      "step": 4145
    },
    {
      "epoch": 0.5477101755312129,
      "grad_norm": 16.40343475341797,
      "learning_rate": 1.0085230620939853e-06,
      "loss": 0.0713,
      "num_input_tokens_seen": 2038272,
      "step": 4150
    },
    {
      "epoch": 0.5483700673089613,
      "grad_norm": 16.413604736328125,
      "learning_rate": 1.0062195670019583e-06,
      "loss": 0.0909,
      "num_input_tokens_seen": 2040768,
      "step": 4155
    },
    {
      "epoch": 0.5490299590867098,
      "grad_norm": 57.78659439086914,
      "learning_rate": 1.0039160389065582e-06,
      "loss": 0.1428,
      "num_input_tokens_seen": 2043072,
      "step": 4160
    },
    {
      "epoch": 0.5496898508644582,
      "grad_norm": 12.916586875915527,
      "learning_rate": 1.0016124900311755e-06,
      "loss": 0.1555,
      "num_input_tokens_seen": 2045824,
      "step": 4165
    },
    {
      "epoch": 0.550217764286657,
      "eval_loss": 0.09937019646167755,
      "eval_runtime": 7.8852,
      "eval_samples_per_second": 854.136,
      "eval_steps_per_second": 106.783,
      "num_input_tokens_seen": 2047552,
      "step": 4169
    },
    {
      "epoch": 0.5503497426422067,
      "grad_norm": 25.078630447387695,
      "learning_rate": 9.99308932599311e-07,
      "loss": 0.2405,
      "num_input_tokens_seen": 2048064,
      "step": 4170
    },
    {
      "epoch": 0.5510096344199551,
      "grad_norm": 0.6139159798622131,
      "learning_rate": 9.970053788345112e-07,
      "loss": 0.0424,
      "num_input_tokens_seen": 2050432,
      "step": 4175
    },
    {
      "epoch": 0.5516695261977036,
      "grad_norm": 0.7848473191261292,
      "learning_rate": 9.947018409603036e-07,
      "loss": 0.0537,
      "num_input_tokens_seen": 2052928,
      "step": 4180
    },
    {
      "epoch": 0.552329417975452,
      "grad_norm": 0.712689995765686,
      "learning_rate": 9.923983312001304e-07,
      "loss": 0.1485,
      "num_input_tokens_seen": 2055424,
      "step": 4185
    },
    {
      "epoch": 0.5529893097532005,
      "grad_norm": 0.20174185931682587,
      "learning_rate": 9.900948617772846e-07,
      "loss": 0.1457,
      "num_input_tokens_seen": 2057536,
      "step": 4190
    },
    {
      "epoch": 0.5536492015309489,
      "grad_norm": 3.5391955375671387,
      "learning_rate": 9.877914449148462e-07,
      "loss": 0.1713,
      "num_input_tokens_seen": 2059840,
      "step": 4195
    },
    {
      "epoch": 0.5543090933086974,
      "grad_norm": 49.72500228881836,
      "learning_rate": 9.854880928356157e-07,
      "loss": 0.2209,
      "num_input_tokens_seen": 2062656,
      "step": 4200
    },
    {
      "epoch": 0.5549689850864459,
      "grad_norm": 0.5038784742355347,
      "learning_rate": 9.831848177620493e-07,
      "loss": 0.1398,
      "num_input_tokens_seen": 2064960,
      "step": 4205
    },
    {
      "epoch": 0.5556288768641943,
      "grad_norm": 33.89471435546875,
      "learning_rate": 9.808816319161961e-07,
      "loss": 0.2412,
      "num_input_tokens_seen": 2067008,
      "step": 4210
    },
    {
      "epoch": 0.5562887686419428,
      "grad_norm": 0.1182156652212143,
      "learning_rate": 9.785785475196298e-07,
      "loss": 0.0009,
      "num_input_tokens_seen": 2069696,
      "step": 4215
    },
    {
      "epoch": 0.5569486604196912,
      "grad_norm": 0.5332831740379333,
      "learning_rate": 9.76275576793387e-07,
      "loss": 0.0507,
      "num_input_tokens_seen": 2072320,
      "step": 4220
    },
    {
      "epoch": 0.5576085521974397,
      "grad_norm": 3.603602886199951,
      "learning_rate": 9.739727319579007e-07,
      "loss": 0.0016,
      "num_input_tokens_seen": 2074752,
      "step": 4225
    },
    {
      "epoch": 0.558268443975188,
      "grad_norm": 2.214005947113037,
      "learning_rate": 9.716700252329361e-07,
      "loss": 0.1143,
      "num_input_tokens_seen": 2077440,
      "step": 4230
    },
    {
      "epoch": 0.5589283357529365,
      "grad_norm": 29.49093246459961,
      "learning_rate": 9.693674688375254e-07,
      "loss": 0.1856,
      "num_input_tokens_seen": 2080000,
      "step": 4235
    },
    {
      "epoch": 0.5595882275306849,
      "grad_norm": 0.3982854187488556,
      "learning_rate": 9.67065074989903e-07,
      "loss": 0.0667,
      "num_input_tokens_seen": 2082560,
      "step": 4240
    },
    {
      "epoch": 0.5602481193084334,
      "grad_norm": 29.326953887939453,
      "learning_rate": 9.647628559074415e-07,
      "loss": 0.0822,
      "num_input_tokens_seen": 2084864,
      "step": 4245
    },
    {
      "epoch": 0.5609080110861818,
      "grad_norm": 0.06176973879337311,
      "learning_rate": 9.62460823806585e-07,
      "loss": 0.1228,
      "num_input_tokens_seen": 2087424,
      "step": 4250
    },
    {
      "epoch": 0.5615679028639303,
      "grad_norm": 38.027488708496094,
      "learning_rate": 9.601589909027857e-07,
      "loss": 0.2208,
      "num_input_tokens_seen": 2090048,
      "step": 4255
    },
    {
      "epoch": 0.5622277946416787,
      "grad_norm": 0.05623474344611168,
      "learning_rate": 9.578573694104394e-07,
      "loss": 0.0689,
      "num_input_tokens_seen": 2092416,
      "step": 4260
    },
    {
      "epoch": 0.5628876864194272,
      "grad_norm": 34.464744567871094,
      "learning_rate": 9.555559715428199e-07,
      "loss": 0.0788,
      "num_input_tokens_seen": 2094656,
      "step": 4265
    },
    {
      "epoch": 0.5635475781971757,
      "grad_norm": 0.2011699080467224,
      "learning_rate": 9.532548095120134e-07,
      "loss": 0.069,
      "num_input_tokens_seen": 2097024,
      "step": 4270
    },
    {
      "epoch": 0.5642074699749241,
      "grad_norm": 0.07516666501760483,
      "learning_rate": 9.509538955288564e-07,
      "loss": 0.0597,
      "num_input_tokens_seen": 2099392,
      "step": 4275
    },
    {
      "epoch": 0.5648673617526726,
      "grad_norm": 1.4651904106140137,
      "learning_rate": 9.486532418028672e-07,
      "loss": 0.09,
      "num_input_tokens_seen": 2102016,
      "step": 4280
    },
    {
      "epoch": 0.565527253530421,
      "grad_norm": 130.39993286132812,
      "learning_rate": 9.463528605421844e-07,
      "loss": 0.0804,
      "num_input_tokens_seen": 2104320,
      "step": 4285
    },
    {
      "epoch": 0.5661871453081695,
      "grad_norm": 36.727294921875,
      "learning_rate": 9.440527639535004e-07,
      "loss": 0.1016,
      "num_input_tokens_seen": 2107136,
      "step": 4290
    },
    {
      "epoch": 0.5668470370859179,
      "grad_norm": 0.16789622604846954,
      "learning_rate": 9.417529642419971e-07,
      "loss": 0.0533,
      "num_input_tokens_seen": 2109888,
      "step": 4295
    },
    {
      "epoch": 0.5675069288636664,
      "grad_norm": 108.9660873413086,
      "learning_rate": 9.394534736112815e-07,
      "loss": 0.1565,
      "num_input_tokens_seen": 2112192,
      "step": 4300
    },
    {
      "epoch": 0.5681668206414148,
      "grad_norm": 0.09434999525547028,
      "learning_rate": 9.371543042633192e-07,
      "loss": 0.1197,
      "num_input_tokens_seen": 2114752,
      "step": 4305
    },
    {
      "epoch": 0.5688267124191633,
      "grad_norm": 0.1492881327867508,
      "learning_rate": 9.348554683983722e-07,
      "loss": 0.141,
      "num_input_tokens_seen": 2117184,
      "step": 4310
    },
    {
      "epoch": 0.5694866041969117,
      "grad_norm": 0.6214476227760315,
      "learning_rate": 9.325569782149323e-07,
      "loss": 0.1155,
      "num_input_tokens_seen": 2119552,
      "step": 4315
    },
    {
      "epoch": 0.5701464959746602,
      "grad_norm": 35.80785369873047,
      "learning_rate": 9.302588459096574e-07,
      "loss": 0.0744,
      "num_input_tokens_seen": 2121920,
      "step": 4320
    },
    {
      "epoch": 0.5708063877524086,
      "grad_norm": 15.295533180236816,
      "learning_rate": 9.279610836773064e-07,
      "loss": 0.2853,
      "num_input_tokens_seen": 2124096,
      "step": 4325
    },
    {
      "epoch": 0.571466279530157,
      "grad_norm": 28.60223388671875,
      "learning_rate": 9.256637037106735e-07,
      "loss": 0.1079,
      "num_input_tokens_seen": 2126528,
      "step": 4330
    },
    {
      "epoch": 0.5721261713079056,
      "grad_norm": 39.715091705322266,
      "learning_rate": 9.233667182005259e-07,
      "loss": 0.0589,
      "num_input_tokens_seen": 2128576,
      "step": 4335
    },
    {
      "epoch": 0.5727860630856539,
      "grad_norm": 1.4454349279403687,
      "learning_rate": 9.210701393355361e-07,
      "loss": 0.0648,
      "num_input_tokens_seen": 2130688,
      "step": 4340
    },
    {
      "epoch": 0.5734459548634024,
      "grad_norm": 71.0278091430664,
      "learning_rate": 9.187739793022198e-07,
      "loss": 0.1324,
      "num_input_tokens_seen": 2133312,
      "step": 4345
    },
    {
      "epoch": 0.5741058466411508,
      "grad_norm": 0.047647565603256226,
      "learning_rate": 9.164782502848702e-07,
      "loss": 0.0724,
      "num_input_tokens_seen": 2135680,
      "step": 4350
    },
    {
      "epoch": 0.5747657384188993,
      "grad_norm": 0.19318842887878418,
      "learning_rate": 9.141829644654936e-07,
      "loss": 0.1448,
      "num_input_tokens_seen": 2138112,
      "step": 4355
    },
    {
      "epoch": 0.5754256301966477,
      "grad_norm": 110.14704132080078,
      "learning_rate": 9.118881340237432e-07,
      "loss": 0.4271,
      "num_input_tokens_seen": 2140352,
      "step": 4360
    },
    {
      "epoch": 0.5760855219743962,
      "grad_norm": 0.10673969238996506,
      "learning_rate": 9.095937711368573e-07,
      "loss": 0.0894,
      "num_input_tokens_seen": 2143040,
      "step": 4365
    },
    {
      "epoch": 0.5767454137521446,
      "grad_norm": 0.09286520630121231,
      "learning_rate": 9.072998879795923e-07,
      "loss": 0.0716,
      "num_input_tokens_seen": 2145280,
      "step": 4370
    },
    {
      "epoch": 0.5774053055298931,
      "grad_norm": 2.3450353145599365,
      "learning_rate": 9.050064967241596e-07,
      "loss": 0.1048,
      "num_input_tokens_seen": 2147904,
      "step": 4375
    },
    {
      "epoch": 0.5780651973076415,
      "grad_norm": 0.20796601474285126,
      "learning_rate": 9.027136095401598e-07,
      "loss": 0.0838,
      "num_input_tokens_seen": 2150400,
      "step": 4380
    },
    {
      "epoch": 0.57872508908539,
      "grad_norm": 0.1643630564212799,
      "learning_rate": 9.004212385945187e-07,
      "loss": 0.135,
      "num_input_tokens_seen": 2153088,
      "step": 4385
    },
    {
      "epoch": 0.5793849808631385,
      "grad_norm": 0.05613645911216736,
      "learning_rate": 8.981293960514233e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 2155776,
      "step": 4390
    },
    {
      "epoch": 0.5800448726408869,
      "grad_norm": 106.51136779785156,
      "learning_rate": 8.958380940722564e-07,
      "loss": 0.1661,
      "num_input_tokens_seen": 2158400,
      "step": 4395
    },
    {
      "epoch": 0.5807047644186354,
      "grad_norm": 0.41299527883529663,
      "learning_rate": 8.935473448155326e-07,
      "loss": 0.0766,
      "num_input_tokens_seen": 2160704,
      "step": 4400
    },
    {
      "epoch": 0.5813646561963838,
      "grad_norm": 29.65276336669922,
      "learning_rate": 8.912571604368324e-07,
      "loss": 0.0469,
      "num_input_tokens_seen": 2163200,
      "step": 4405
    },
    {
      "epoch": 0.5820245479741323,
      "grad_norm": 49.52018737792969,
      "learning_rate": 8.889675530887404e-07,
      "loss": 0.1421,
      "num_input_tokens_seen": 2165376,
      "step": 4410
    },
    {
      "epoch": 0.5826844397518807,
      "grad_norm": 0.17924949526786804,
      "learning_rate": 8.866785349207786e-07,
      "loss": 0.0994,
      "num_input_tokens_seen": 2167808,
      "step": 4415
    },
    {
      "epoch": 0.5833443315296292,
      "grad_norm": 23.86568260192871,
      "learning_rate": 8.843901180793423e-07,
      "loss": 0.1332,
      "num_input_tokens_seen": 2170112,
      "step": 4420
    },
    {
      "epoch": 0.5840042233073776,
      "grad_norm": 0.38153278827667236,
      "learning_rate": 8.821023147076362e-07,
      "loss": 0.0005,
      "num_input_tokens_seen": 2172480,
      "step": 4425
    },
    {
      "epoch": 0.5846641150851261,
      "grad_norm": 1.3169441223144531,
      "learning_rate": 8.798151369456098e-07,
      "loss": 0.0432,
      "num_input_tokens_seen": 2175104,
      "step": 4430
    },
    {
      "epoch": 0.5853240068628744,
      "grad_norm": 14.025379180908203,
      "learning_rate": 8.775285969298931e-07,
      "loss": 0.1126,
      "num_input_tokens_seen": 2177280,
      "step": 4435
    },
    {
      "epoch": 0.585983898640623,
      "grad_norm": 0.05871421471238136,
      "learning_rate": 8.752427067937312e-07,
      "loss": 0.0735,
      "num_input_tokens_seen": 2179776,
      "step": 4440
    },
    {
      "epoch": 0.5866437904183713,
      "grad_norm": 0.026597043499350548,
      "learning_rate": 8.729574786669214e-07,
      "loss": 0.0949,
      "num_input_tokens_seen": 2182400,
      "step": 4445
    },
    {
      "epoch": 0.5873036821961198,
      "grad_norm": 1.1581672430038452,
      "learning_rate": 8.706729246757477e-07,
      "loss": 0.0655,
      "num_input_tokens_seen": 2185088,
      "step": 4450
    },
    {
      "epoch": 0.5879635739738683,
      "grad_norm": 0.904012143611908,
      "learning_rate": 8.683890569429173e-07,
      "loss": 0.0697,
      "num_input_tokens_seen": 2187776,
      "step": 4455
    },
    {
      "epoch": 0.5886234657516167,
      "grad_norm": 0.04613454267382622,
      "learning_rate": 8.661058875874956e-07,
      "loss": 0.0006,
      "num_input_tokens_seen": 2190016,
      "step": 4460
    },
    {
      "epoch": 0.5892833575293652,
      "grad_norm": 0.06226871907711029,
      "learning_rate": 8.638234287248423e-07,
      "loss": 0.0008,
      "num_input_tokens_seen": 2192320,
      "step": 4465
    },
    {
      "epoch": 0.5899432493071136,
      "grad_norm": 178.02114868164062,
      "learning_rate": 8.615416924665464e-07,
      "loss": 0.0355,
      "num_input_tokens_seen": 2194752,
      "step": 4470
    },
    {
      "epoch": 0.5906031410848621,
      "grad_norm": 0.3306443691253662,
      "learning_rate": 8.592606909203629e-07,
      "loss": 0.0743,
      "num_input_tokens_seen": 2197056,
      "step": 4475
    },
    {
      "epoch": 0.5912630328626105,
      "grad_norm": 0.03277226537466049,
      "learning_rate": 8.569804361901485e-07,
      "loss": 0.0489,
      "num_input_tokens_seen": 2199296,
      "step": 4480
    },
    {
      "epoch": 0.591922924640359,
      "grad_norm": 70.67273712158203,
      "learning_rate": 8.547009403757963e-07,
      "loss": 0.41,
      "num_input_tokens_seen": 2201664,
      "step": 4485
    },
    {
      "epoch": 0.5925828164181074,
      "grad_norm": 15.897239685058594,
      "learning_rate": 8.524222155731731e-07,
      "loss": 0.227,
      "num_input_tokens_seen": 2204288,
      "step": 4490
    },
    {
      "epoch": 0.5932427081958559,
      "grad_norm": 0.5026524662971497,
      "learning_rate": 8.501442738740538e-07,
      "loss": 0.127,
      "num_input_tokens_seen": 2206528,
      "step": 4495
    },
    {
      "epoch": 0.5939025999736043,
      "grad_norm": 0.5848537683486938,
      "learning_rate": 8.47867127366058e-07,
      "loss": 0.0695,
      "num_input_tokens_seen": 2209024,
      "step": 4500
    },
    {
      "epoch": 0.5945624917513528,
      "grad_norm": 0.10000286996364594,
      "learning_rate": 8.455907881325858e-07,
      "loss": 0.0422,
      "num_input_tokens_seen": 2211584,
      "step": 4505
    },
    {
      "epoch": 0.5952223835291012,
      "grad_norm": 19.83750343322754,
      "learning_rate": 8.433152682527533e-07,
      "loss": 0.1233,
      "num_input_tokens_seen": 2213952,
      "step": 4510
    },
    {
      "epoch": 0.5958822753068497,
      "grad_norm": 72.72146606445312,
      "learning_rate": 8.410405798013298e-07,
      "loss": 0.0951,
      "num_input_tokens_seen": 2216192,
      "step": 4515
    },
    {
      "epoch": 0.5965421670845982,
      "grad_norm": 75.5259017944336,
      "learning_rate": 8.387667348486712e-07,
      "loss": 0.0126,
      "num_input_tokens_seen": 2218688,
      "step": 4520
    },
    {
      "epoch": 0.5972020588623466,
      "grad_norm": 0.20401786267757416,
      "learning_rate": 8.364937454606585e-07,
      "loss": 0.0853,
      "num_input_tokens_seen": 2220928,
      "step": 4525
    },
    {
      "epoch": 0.5978619506400951,
      "grad_norm": 0.22834134101867676,
      "learning_rate": 8.342216236986329e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 2223360,
      "step": 4530
    },
    {
      "epoch": 0.5985218424178435,
      "grad_norm": 0.06765160709619522,
      "learning_rate": 8.319503816193305e-07,
      "loss": 0.1511,
      "num_input_tokens_seen": 2225792,
      "step": 4535
    },
    {
      "epoch": 0.599181734195592,
      "grad_norm": 26.021888732910156,
      "learning_rate": 8.296800312748206e-07,
      "loss": 0.1163,
      "num_input_tokens_seen": 2228288,
      "step": 4540
    },
    {
      "epoch": 0.5998416259733403,
      "grad_norm": 0.4052346646785736,
      "learning_rate": 8.274105847124404e-07,
      "loss": 0.1894,
      "num_input_tokens_seen": 2230848,
      "step": 4545
    },
    {
      "epoch": 0.6002375610399895,
      "eval_loss": 0.1099071353673935,
      "eval_runtime": 8.2934,
      "eval_samples_per_second": 812.09,
      "eval_steps_per_second": 101.526,
      "num_input_tokens_seen": 2232448,
      "step": 4548
    },
    {
      "epoch": 0.6005015177510888,
      "grad_norm": 25.49419593811035,
      "learning_rate": 8.251420539747311e-07,
      "loss": 0.233,
      "num_input_tokens_seen": 2233472,
      "step": 4550
    },
    {
      "epoch": 0.6011614095288372,
      "grad_norm": 0.07836712151765823,
      "learning_rate": 8.228744510993742e-07,
      "loss": 0.2218,
      "num_input_tokens_seen": 2236096,
      "step": 4555
    },
    {
      "epoch": 0.6018213013065857,
      "grad_norm": 0.250975102186203,
      "learning_rate": 8.206077881191274e-07,
      "loss": 0.0831,
      "num_input_tokens_seen": 2238720,
      "step": 4560
    },
    {
      "epoch": 0.6024811930843341,
      "grad_norm": 54.536197662353516,
      "learning_rate": 8.183420770617614e-07,
      "loss": 0.1877,
      "num_input_tokens_seen": 2241216,
      "step": 4565
    },
    {
      "epoch": 0.6031410848620826,
      "grad_norm": 28.72020149230957,
      "learning_rate": 8.160773299499955e-07,
      "loss": 0.0651,
      "num_input_tokens_seen": 2243648,
      "step": 4570
    },
    {
      "epoch": 0.6038009766398311,
      "grad_norm": 0.8081544637680054,
      "learning_rate": 8.138135588014339e-07,
      "loss": 0.0543,
      "num_input_tokens_seen": 2246080,
      "step": 4575
    },
    {
      "epoch": 0.6044608684175795,
      "grad_norm": 0.6024172902107239,
      "learning_rate": 8.115507756285017e-07,
      "loss": 0.0016,
      "num_input_tokens_seen": 2248256,
      "step": 4580
    },
    {
      "epoch": 0.605120760195328,
      "grad_norm": 0.1423349827528,
      "learning_rate": 8.092889924383819e-07,
      "loss": 0.07,
      "num_input_tokens_seen": 2250688,
      "step": 4585
    },
    {
      "epoch": 0.6057806519730764,
      "grad_norm": 0.3546024560928345,
      "learning_rate": 8.070282212329508e-07,
      "loss": 0.0872,
      "num_input_tokens_seen": 2253120,
      "step": 4590
    },
    {
      "epoch": 0.6064405437508249,
      "grad_norm": 16.78183937072754,
      "learning_rate": 8.047684740087156e-07,
      "loss": 0.1869,
      "num_input_tokens_seen": 2255360,
      "step": 4595
    },
    {
      "epoch": 0.6071004355285733,
      "grad_norm": 18.29332160949707,
      "learning_rate": 8.025097627567481e-07,
      "loss": 0.2572,
      "num_input_tokens_seen": 2257728,
      "step": 4600
    },
    {
      "epoch": 0.6077603273063218,
      "grad_norm": 0.07837789505720139,
      "learning_rate": 8.002520994626247e-07,
      "loss": 0.0536,
      "num_input_tokens_seen": 2260224,
      "step": 4605
    },
    {
      "epoch": 0.6084202190840702,
      "grad_norm": 0.2844223976135254,
      "learning_rate": 7.979954961063596e-07,
      "loss": 0.0372,
      "num_input_tokens_seen": 2262912,
      "step": 4610
    },
    {
      "epoch": 0.6090801108618187,
      "grad_norm": 64.26956939697266,
      "learning_rate": 7.957399646623436e-07,
      "loss": 0.2998,
      "num_input_tokens_seen": 2265152,
      "step": 4615
    },
    {
      "epoch": 0.6097400026395671,
      "grad_norm": 0.06967580318450928,
      "learning_rate": 7.934855170992788e-07,
      "loss": 0.0008,
      "num_input_tokens_seen": 2267968,
      "step": 4620
    },
    {
      "epoch": 0.6103998944173156,
      "grad_norm": 0.09711959213018417,
      "learning_rate": 7.912321653801161e-07,
      "loss": 0.0407,
      "num_input_tokens_seen": 2270336,
      "step": 4625
    },
    {
      "epoch": 0.611059786195064,
      "grad_norm": 19.1835880279541,
      "learning_rate": 7.889799214619919e-07,
      "loss": 0.135,
      "num_input_tokens_seen": 2273024,
      "step": 4630
    },
    {
      "epoch": 0.6117196779728125,
      "grad_norm": 0.028730077669024467,
      "learning_rate": 7.867287972961629e-07,
      "loss": 0.0678,
      "num_input_tokens_seen": 2275264,
      "step": 4635
    },
    {
      "epoch": 0.612379569750561,
      "grad_norm": 32.42776107788086,
      "learning_rate": 7.844788048279453e-07,
      "loss": 0.1219,
      "num_input_tokens_seen": 2277888,
      "step": 4640
    },
    {
      "epoch": 0.6130394615283093,
      "grad_norm": 0.039022643119096756,
      "learning_rate": 7.822299559966494e-07,
      "loss": 0.0005,
      "num_input_tokens_seen": 2280320,
      "step": 4645
    },
    {
      "epoch": 0.6136993533060578,
      "grad_norm": 0.0474756583571434,
      "learning_rate": 7.799822627355171e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 2282560,
      "step": 4650
    },
    {
      "epoch": 0.6143592450838062,
      "grad_norm": 0.14449454843997955,
      "learning_rate": 7.77735736971659e-07,
      "loss": 0.1505,
      "num_input_tokens_seen": 2284864,
      "step": 4655
    },
    {
      "epoch": 0.6150191368615547,
      "grad_norm": 0.22011378407478333,
      "learning_rate": 7.754903906259889e-07,
      "loss": 0.165,
      "num_input_tokens_seen": 2287168,
      "step": 4660
    },
    {
      "epoch": 0.6156790286393031,
      "grad_norm": 26.424232482910156,
      "learning_rate": 7.732462356131637e-07,
      "loss": 0.0716,
      "num_input_tokens_seen": 2289600,
      "step": 4665
    },
    {
      "epoch": 0.6163389204170516,
      "grad_norm": 1.7064114809036255,
      "learning_rate": 7.710032838415179e-07,
      "loss": 0.0317,
      "num_input_tokens_seen": 2292160,
      "step": 4670
    },
    {
      "epoch": 0.6169988121948,
      "grad_norm": 0.036719437688589096,
      "learning_rate": 7.687615472130016e-07,
      "loss": 0.0943,
      "num_input_tokens_seen": 2294912,
      "step": 4675
    },
    {
      "epoch": 0.6176587039725485,
      "grad_norm": 0.11287426948547363,
      "learning_rate": 7.665210376231165e-07,
      "loss": 0.1314,
      "num_input_tokens_seen": 2297024,
      "step": 4680
    },
    {
      "epoch": 0.6183185957502969,
      "grad_norm": 12.669684410095215,
      "learning_rate": 7.642817669608536e-07,
      "loss": 0.1436,
      "num_input_tokens_seen": 2299456,
      "step": 4685
    },
    {
      "epoch": 0.6189784875280454,
      "grad_norm": 0.13082663714885712,
      "learning_rate": 7.62043747108629e-07,
      "loss": 0.013,
      "num_input_tokens_seen": 2301568,
      "step": 4690
    },
    {
      "epoch": 0.6196383793057938,
      "grad_norm": 265.1707458496094,
      "learning_rate": 7.598069899422221e-07,
      "loss": 0.168,
      "num_input_tokens_seen": 2303936,
      "step": 4695
    },
    {
      "epoch": 0.6202982710835423,
      "grad_norm": 12.800505638122559,
      "learning_rate": 7.575715073307119e-07,
      "loss": 0.1745,
      "num_input_tokens_seen": 2306176,
      "step": 4700
    },
    {
      "epoch": 0.6209581628612908,
      "grad_norm": 0.4455643892288208,
      "learning_rate": 7.55337311136414e-07,
      "loss": 0.0901,
      "num_input_tokens_seen": 2308736,
      "step": 4705
    },
    {
      "epoch": 0.6216180546390392,
      "grad_norm": 18.547971725463867,
      "learning_rate": 7.531044132148183e-07,
      "loss": 0.1379,
      "num_input_tokens_seen": 2311104,
      "step": 4710
    },
    {
      "epoch": 0.6222779464167877,
      "grad_norm": 23.932079315185547,
      "learning_rate": 7.508728254145245e-07,
      "loss": 0.049,
      "num_input_tokens_seen": 2313536,
      "step": 4715
    },
    {
      "epoch": 0.6229378381945361,
      "grad_norm": 18.262922286987305,
      "learning_rate": 7.486425595771817e-07,
      "loss": 0.1208,
      "num_input_tokens_seen": 2316032,
      "step": 4720
    },
    {
      "epoch": 0.6235977299722846,
      "grad_norm": 12.403181076049805,
      "learning_rate": 7.464136275374223e-07,
      "loss": 0.205,
      "num_input_tokens_seen": 2318656,
      "step": 4725
    },
    {
      "epoch": 0.624257621750033,
      "grad_norm": 14.224221229553223,
      "learning_rate": 7.441860411228029e-07,
      "loss": 0.1395,
      "num_input_tokens_seen": 2321216,
      "step": 4730
    },
    {
      "epoch": 0.6249175135277815,
      "grad_norm": 48.01051712036133,
      "learning_rate": 7.419598121537387e-07,
      "loss": 0.1494,
      "num_input_tokens_seen": 2323648,
      "step": 4735
    },
    {
      "epoch": 0.6255774053055299,
      "grad_norm": 16.137556076049805,
      "learning_rate": 7.397349524434424e-07,
      "loss": 0.165,
      "num_input_tokens_seen": 2326080,
      "step": 4740
    },
    {
      "epoch": 0.6262372970832784,
      "grad_norm": 13.062948226928711,
      "learning_rate": 7.375114737978605e-07,
      "loss": 0.0572,
      "num_input_tokens_seen": 2328512,
      "step": 4745
    },
    {
      "epoch": 0.6268971888610267,
      "grad_norm": 7.66116189956665,
      "learning_rate": 7.352893880156106e-07,
      "loss": 0.1462,
      "num_input_tokens_seen": 2331008,
      "step": 4750
    },
    {
      "epoch": 0.6275570806387752,
      "grad_norm": 1.405522346496582,
      "learning_rate": 7.330687068879202e-07,
      "loss": 0.0943,
      "num_input_tokens_seen": 2333376,
      "step": 4755
    },
    {
      "epoch": 0.6282169724165237,
      "grad_norm": 224.89602661132812,
      "learning_rate": 7.308494421985626e-07,
      "loss": 0.2447,
      "num_input_tokens_seen": 2335872,
      "step": 4760
    },
    {
      "epoch": 0.6288768641942721,
      "grad_norm": 0.1741466373205185,
      "learning_rate": 7.286316057237951e-07,
      "loss": 0.0016,
      "num_input_tokens_seen": 2338432,
      "step": 4765
    },
    {
      "epoch": 0.6295367559720206,
      "grad_norm": 9.092262268066406,
      "learning_rate": 7.264152092322963e-07,
      "loss": 0.1152,
      "num_input_tokens_seen": 2340928,
      "step": 4770
    },
    {
      "epoch": 0.630196647749769,
      "grad_norm": 0.2725062370300293,
      "learning_rate": 7.242002644851035e-07,
      "loss": 0.1148,
      "num_input_tokens_seen": 2343680,
      "step": 4775
    },
    {
      "epoch": 0.6308565395275175,
      "grad_norm": 0.1986101120710373,
      "learning_rate": 7.219867832355508e-07,
      "loss": 0.0689,
      "num_input_tokens_seen": 2346240,
      "step": 4780
    },
    {
      "epoch": 0.6315164313052659,
      "grad_norm": 0.1735810786485672,
      "learning_rate": 7.197747772292071e-07,
      "loss": 0.0841,
      "num_input_tokens_seen": 2348544,
      "step": 4785
    },
    {
      "epoch": 0.6321763230830144,
      "grad_norm": 0.06138006970286369,
      "learning_rate": 7.17564258203811e-07,
      "loss": 0.2099,
      "num_input_tokens_seen": 2350976,
      "step": 4790
    },
    {
      "epoch": 0.6328362148607628,
      "grad_norm": 21.949594497680664,
      "learning_rate": 7.153552378892128e-07,
      "loss": 0.115,
      "num_input_tokens_seen": 2353216,
      "step": 4795
    },
    {
      "epoch": 0.6334961066385113,
      "grad_norm": 0.6890119314193726,
      "learning_rate": 7.131477280073091e-07,
      "loss": 0.1212,
      "num_input_tokens_seen": 2355584,
      "step": 4800
    },
    {
      "epoch": 0.6341559984162597,
      "grad_norm": 0.19981062412261963,
      "learning_rate": 7.109417402719813e-07,
      "loss": 0.0817,
      "num_input_tokens_seen": 2358144,
      "step": 4805
    },
    {
      "epoch": 0.6348158901940082,
      "grad_norm": 49.97908401489258,
      "learning_rate": 7.087372863890346e-07,
      "loss": 0.029,
      "num_input_tokens_seen": 2360896,
      "step": 4810
    },
    {
      "epoch": 0.6354757819717566,
      "grad_norm": 1.2881401777267456,
      "learning_rate": 7.065343780561344e-07,
      "loss": 0.2463,
      "num_input_tokens_seen": 2363264,
      "step": 4815
    },
    {
      "epoch": 0.6361356737495051,
      "grad_norm": 13.928070068359375,
      "learning_rate": 7.043330269627448e-07,
      "loss": 0.0744,
      "num_input_tokens_seen": 2365632,
      "step": 4820
    },
    {
      "epoch": 0.6367955655272536,
      "grad_norm": 0.07533244788646698,
      "learning_rate": 7.021332447900671e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 2368000,
      "step": 4825
    },
    {
      "epoch": 0.637455457305002,
      "grad_norm": 199.91444396972656,
      "learning_rate": 6.999350432109766e-07,
      "loss": 0.1116,
      "num_input_tokens_seen": 2370560,
      "step": 4830
    },
    {
      "epoch": 0.6381153490827505,
      "grad_norm": 0.0590510219335556,
      "learning_rate": 6.977384338899617e-07,
      "loss": 0.0011,
      "num_input_tokens_seen": 2373120,
      "step": 4835
    },
    {
      "epoch": 0.6387752408604989,
      "grad_norm": 0.12178171426057816,
      "learning_rate": 6.955434284830619e-07,
      "loss": 0.033,
      "num_input_tokens_seen": 2375872,
      "step": 4840
    },
    {
      "epoch": 0.6394351326382474,
      "grad_norm": 0.07356923073530197,
      "learning_rate": 6.933500386378056e-07,
      "loss": 0.2167,
      "num_input_tokens_seen": 2378432,
      "step": 4845
    },
    {
      "epoch": 0.6400950244159958,
      "grad_norm": 209.99586486816406,
      "learning_rate": 6.911582759931482e-07,
      "loss": 0.213,
      "num_input_tokens_seen": 2380800,
      "step": 4850
    },
    {
      "epoch": 0.6407549161937443,
      "grad_norm": 0.01491602510213852,
      "learning_rate": 6.889681521794109e-07,
      "loss": 0.2747,
      "num_input_tokens_seen": 2383744,
      "step": 4855
    },
    {
      "epoch": 0.6414148079714926,
      "grad_norm": 15.076420783996582,
      "learning_rate": 6.867796788182181e-07,
      "loss": 0.0758,
      "num_input_tokens_seen": 2386112,
      "step": 4860
    },
    {
      "epoch": 0.6420746997492411,
      "grad_norm": 13.008954048156738,
      "learning_rate": 6.845928675224366e-07,
      "loss": 0.1529,
      "num_input_tokens_seen": 2388736,
      "step": 4865
    },
    {
      "epoch": 0.6427345915269895,
      "grad_norm": 0.15090730786323547,
      "learning_rate": 6.82407729896114e-07,
      "loss": 0.0713,
      "num_input_tokens_seen": 2391104,
      "step": 4870
    },
    {
      "epoch": 0.643394483304738,
      "grad_norm": 27.68510627746582,
      "learning_rate": 6.802242775344163e-07,
      "loss": 0.1204,
      "num_input_tokens_seen": 2393728,
      "step": 4875
    },
    {
      "epoch": 0.6440543750824864,
      "grad_norm": 0.20335356891155243,
      "learning_rate": 6.780425220235674e-07,
      "loss": 0.1797,
      "num_input_tokens_seen": 2396480,
      "step": 4880
    },
    {
      "epoch": 0.6447142668602349,
      "grad_norm": 0.1165480837225914,
      "learning_rate": 6.758624749407859e-07,
      "loss": 0.0011,
      "num_input_tokens_seen": 2399104,
      "step": 4885
    },
    {
      "epoch": 0.6453741586379834,
      "grad_norm": 15.720735549926758,
      "learning_rate": 6.736841478542264e-07,
      "loss": 0.0861,
      "num_input_tokens_seen": 2401664,
      "step": 4890
    },
    {
      "epoch": 0.6460340504157318,
      "grad_norm": 2.2438161373138428,
      "learning_rate": 6.715075523229151e-07,
      "loss": 0.0316,
      "num_input_tokens_seen": 2404160,
      "step": 4895
    },
    {
      "epoch": 0.6466939421934803,
      "grad_norm": 77.18789672851562,
      "learning_rate": 6.693326998966909e-07,
      "loss": 0.1035,
      "num_input_tokens_seen": 2406592,
      "step": 4900
    },
    {
      "epoch": 0.6473538339712287,
      "grad_norm": 0.05804076045751572,
      "learning_rate": 6.671596021161431e-07,
      "loss": 0.0014,
      "num_input_tokens_seen": 2409088,
      "step": 4905
    },
    {
      "epoch": 0.6480137257489772,
      "grad_norm": 0.07824068516492844,
      "learning_rate": 6.649882705125494e-07,
      "loss": 0.1158,
      "num_input_tokens_seen": 2411584,
      "step": 4910
    },
    {
      "epoch": 0.6486736175267256,
      "grad_norm": 98.59748077392578,
      "learning_rate": 6.628187166078163e-07,
      "loss": 0.472,
      "num_input_tokens_seen": 2414400,
      "step": 4915
    },
    {
      "epoch": 0.6493335093044741,
      "grad_norm": 54.34969711303711,
      "learning_rate": 6.606509519144166e-07,
      "loss": 0.0382,
      "num_input_tokens_seen": 2416640,
      "step": 4920
    },
    {
      "epoch": 0.6499934010822225,
      "grad_norm": 66.13670349121094,
      "learning_rate": 6.584849879353289e-07,
      "loss": 0.1297,
      "num_input_tokens_seen": 2419136,
      "step": 4925
    },
    {
      "epoch": 0.6502573577933219,
      "eval_loss": 0.09970784932374954,
      "eval_runtime": 7.9122,
      "eval_samples_per_second": 851.219,
      "eval_steps_per_second": 106.418,
      "num_input_tokens_seen": 2420096,
      "step": 4927
    },
    {
      "epoch": 0.650653292859971,
      "grad_norm": 0.60494065284729,
      "learning_rate": 6.563208361639772e-07,
      "loss": 0.0423,
      "num_input_tokens_seen": 2421440,
      "step": 4930
    },
    {
      "epoch": 0.6513131846377194,
      "grad_norm": 0.06657727062702179,
      "learning_rate": 6.541585080841687e-07,
      "loss": 0.0187,
      "num_input_tokens_seen": 2424000,
      "step": 4935
    },
    {
      "epoch": 0.6519730764154679,
      "grad_norm": 159.46224975585938,
      "learning_rate": 6.519980151700332e-07,
      "loss": 0.1785,
      "num_input_tokens_seen": 2426240,
      "step": 4940
    },
    {
      "epoch": 0.6526329681932164,
      "grad_norm": 0.16553577780723572,
      "learning_rate": 6.498393688859629e-07,
      "loss": 0.1136,
      "num_input_tokens_seen": 2428864,
      "step": 4945
    },
    {
      "epoch": 0.6532928599709648,
      "grad_norm": 0.052161458879709244,
      "learning_rate": 6.47682580686551e-07,
      "loss": 0.0011,
      "num_input_tokens_seen": 2431296,
      "step": 4950
    },
    {
      "epoch": 0.6539527517487133,
      "grad_norm": 0.05444502457976341,
      "learning_rate": 6.455276620165307e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 2433984,
      "step": 4955
    },
    {
      "epoch": 0.6546126435264616,
      "grad_norm": 0.030169246718287468,
      "learning_rate": 6.433746243107152e-07,
      "loss": 0.4128,
      "num_input_tokens_seen": 2436224,
      "step": 4960
    },
    {
      "epoch": 0.6552725353042101,
      "grad_norm": 0.1023978590965271,
      "learning_rate": 6.412234789939359e-07,
      "loss": 0.1516,
      "num_input_tokens_seen": 2438720,
      "step": 4965
    },
    {
      "epoch": 0.6559324270819585,
      "grad_norm": 0.13813242316246033,
      "learning_rate": 6.390742374809832e-07,
      "loss": 0.0589,
      "num_input_tokens_seen": 2440960,
      "step": 4970
    },
    {
      "epoch": 0.656592318859707,
      "grad_norm": 0.1587606817483902,
      "learning_rate": 6.369269111765454e-07,
      "loss": 0.0597,
      "num_input_tokens_seen": 2443328,
      "step": 4975
    },
    {
      "epoch": 0.6572522106374554,
      "grad_norm": 0.060975294560194016,
      "learning_rate": 6.347815114751465e-07,
      "loss": 0.1664,
      "num_input_tokens_seen": 2445952,
      "step": 4980
    },
    {
      "epoch": 0.6579121024152039,
      "grad_norm": 0.04723791033029556,
      "learning_rate": 6.326380497610886e-07,
      "loss": 0.1165,
      "num_input_tokens_seen": 2448576,
      "step": 4985
    },
    {
      "epoch": 0.6585719941929523,
      "grad_norm": 35.17012405395508,
      "learning_rate": 6.304965374083899e-07,
      "loss": 0.3173,
      "num_input_tokens_seen": 2451136,
      "step": 4990
    },
    {
      "epoch": 0.6592318859707008,
      "grad_norm": 0.23336689174175262,
      "learning_rate": 6.283569857807245e-07,
      "loss": 0.002,
      "num_input_tokens_seen": 2453632,
      "step": 4995
    },
    {
      "epoch": 0.6598917777484492,
      "grad_norm": 0.17793512344360352,
      "learning_rate": 6.262194062313615e-07,
      "loss": 0.0644,
      "num_input_tokens_seen": 2456192,
      "step": 5000
    },
    {
      "epoch": 0.6605516695261977,
      "grad_norm": 0.052415881305933,
      "learning_rate": 6.240838101031063e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 2458624,
      "step": 5005
    },
    {
      "epoch": 0.6612115613039462,
      "grad_norm": 0.11415770649909973,
      "learning_rate": 6.21950208728239e-07,
      "loss": 0.0896,
      "num_input_tokens_seen": 2460928,
      "step": 5010
    },
    {
      "epoch": 0.6618714530816946,
      "grad_norm": 0.13506917655467987,
      "learning_rate": 6.198186134284554e-07,
      "loss": 0.1054,
      "num_input_tokens_seen": 2463552,
      "step": 5015
    },
    {
      "epoch": 0.6625313448594431,
      "grad_norm": 0.4825565218925476,
      "learning_rate": 6.176890355148049e-07,
      "loss": 0.0028,
      "num_input_tokens_seen": 2465856,
      "step": 5020
    },
    {
      "epoch": 0.6631912366371915,
      "grad_norm": 0.11778873205184937,
      "learning_rate": 6.155614862876335e-07,
      "loss": 0.0981,
      "num_input_tokens_seen": 2468288,
      "step": 5025
    },
    {
      "epoch": 0.66385112841494,
      "grad_norm": 110.11556243896484,
      "learning_rate": 6.134359770365214e-07,
      "loss": 0.1368,
      "num_input_tokens_seen": 2470912,
      "step": 5030
    },
    {
      "epoch": 0.6645110201926884,
      "grad_norm": 0.7118353843688965,
      "learning_rate": 6.11312519040224e-07,
      "loss": 0.0711,
      "num_input_tokens_seen": 2473536,
      "step": 5035
    },
    {
      "epoch": 0.6651709119704369,
      "grad_norm": 0.08325012028217316,
      "learning_rate": 6.091911235666125e-07,
      "loss": 0.0009,
      "num_input_tokens_seen": 2476032,
      "step": 5040
    },
    {
      "epoch": 0.6658308037481853,
      "grad_norm": 0.13275060057640076,
      "learning_rate": 6.070718018726124e-07,
      "loss": 0.1034,
      "num_input_tokens_seen": 2478208,
      "step": 5045
    },
    {
      "epoch": 0.6664906955259338,
      "grad_norm": 0.11056883633136749,
      "learning_rate": 6.049545652041459e-07,
      "loss": 0.1196,
      "num_input_tokens_seen": 2480512,
      "step": 5050
    },
    {
      "epoch": 0.6671505873036822,
      "grad_norm": 0.7402254939079285,
      "learning_rate": 6.028394247960709e-07,
      "loss": 0.1824,
      "num_input_tokens_seen": 2483008,
      "step": 5055
    },
    {
      "epoch": 0.6678104790814307,
      "grad_norm": 0.04158753529191017,
      "learning_rate": 6.007263918721221e-07,
      "loss": 0.1399,
      "num_input_tokens_seen": 2485376,
      "step": 5060
    },
    {
      "epoch": 0.668470370859179,
      "grad_norm": 0.2165919989347458,
      "learning_rate": 5.986154776448507e-07,
      "loss": 0.07,
      "num_input_tokens_seen": 2488064,
      "step": 5065
    },
    {
      "epoch": 0.6691302626369275,
      "grad_norm": 18.02176284790039,
      "learning_rate": 5.965066933155656e-07,
      "loss": 0.0635,
      "num_input_tokens_seen": 2490624,
      "step": 5070
    },
    {
      "epoch": 0.669790154414676,
      "grad_norm": 26.047637939453125,
      "learning_rate": 5.944000500742735e-07,
      "loss": 0.2914,
      "num_input_tokens_seen": 2493248,
      "step": 5075
    },
    {
      "epoch": 0.6704500461924244,
      "grad_norm": 69.23401641845703,
      "learning_rate": 5.922955590996195e-07,
      "loss": 0.2126,
      "num_input_tokens_seen": 2495744,
      "step": 5080
    },
    {
      "epoch": 0.6711099379701729,
      "grad_norm": 0.13096897304058075,
      "learning_rate": 5.901932315588281e-07,
      "loss": 0.0014,
      "num_input_tokens_seen": 2498176,
      "step": 5085
    },
    {
      "epoch": 0.6717698297479213,
      "grad_norm": 19.673627853393555,
      "learning_rate": 5.880930786076441e-07,
      "loss": 0.1766,
      "num_input_tokens_seen": 2500416,
      "step": 5090
    },
    {
      "epoch": 0.6724297215256698,
      "grad_norm": 148.1903533935547,
      "learning_rate": 5.859951113902728e-07,
      "loss": 0.0912,
      "num_input_tokens_seen": 2502848,
      "step": 5095
    },
    {
      "epoch": 0.6730896133034182,
      "grad_norm": 23.524593353271484,
      "learning_rate": 5.83899341039321e-07,
      "loss": 0.1217,
      "num_input_tokens_seen": 2505152,
      "step": 5100
    },
    {
      "epoch": 0.6737495050811667,
      "grad_norm": 23.715028762817383,
      "learning_rate": 5.818057786757386e-07,
      "loss": 0.1365,
      "num_input_tokens_seen": 2507648,
      "step": 5105
    },
    {
      "epoch": 0.6744093968589151,
      "grad_norm": 0.21634319424629211,
      "learning_rate": 5.797144354087588e-07,
      "loss": 0.1131,
      "num_input_tokens_seen": 2510144,
      "step": 5110
    },
    {
      "epoch": 0.6750692886366636,
      "grad_norm": 0.06485975533723831,
      "learning_rate": 5.77625322335839e-07,
      "loss": 0.0198,
      "num_input_tokens_seen": 2513024,
      "step": 5115
    },
    {
      "epoch": 0.675729180414412,
      "grad_norm": 0.3879457116127014,
      "learning_rate": 5.755384505426032e-07,
      "loss": 0.0733,
      "num_input_tokens_seen": 2515072,
      "step": 5120
    },
    {
      "epoch": 0.6763890721921605,
      "grad_norm": 0.26867377758026123,
      "learning_rate": 5.734538311027819e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 2517376,
      "step": 5125
    },
    {
      "epoch": 0.677048963969909,
      "grad_norm": 0.730602502822876,
      "learning_rate": 5.713714750781533e-07,
      "loss": 0.003,
      "num_input_tokens_seen": 2520064,
      "step": 5130
    },
    {
      "epoch": 0.6777088557476574,
      "grad_norm": 0.5069525241851807,
      "learning_rate": 5.692913935184862e-07,
      "loss": 0.0716,
      "num_input_tokens_seen": 2522688,
      "step": 5135
    },
    {
      "epoch": 0.6783687475254059,
      "grad_norm": 0.17478005588054657,
      "learning_rate": 5.672135974614794e-07,
      "loss": 0.0165,
      "num_input_tokens_seen": 2525184,
      "step": 5140
    },
    {
      "epoch": 0.6790286393031543,
      "grad_norm": 0.08164742588996887,
      "learning_rate": 5.651380979327034e-07,
      "loss": 0.0008,
      "num_input_tokens_seen": 2527552,
      "step": 5145
    },
    {
      "epoch": 0.6796885310809028,
      "grad_norm": 2.0063724517822266,
      "learning_rate": 5.630649059455444e-07,
      "loss": 0.0435,
      "num_input_tokens_seen": 2530240,
      "step": 5150
    },
    {
      "epoch": 0.6803484228586512,
      "grad_norm": 1.745144248008728,
      "learning_rate": 5.609940325011413e-07,
      "loss": 0.0116,
      "num_input_tokens_seen": 2532480,
      "step": 5155
    },
    {
      "epoch": 0.6810083146363997,
      "grad_norm": 0.13738574087619781,
      "learning_rate": 5.589254885883325e-07,
      "loss": 0.0006,
      "num_input_tokens_seen": 2534912,
      "step": 5160
    },
    {
      "epoch": 0.681668206414148,
      "grad_norm": 0.19333359599113464,
      "learning_rate": 5.568592851835936e-07,
      "loss": 0.0753,
      "num_input_tokens_seen": 2537408,
      "step": 5165
    },
    {
      "epoch": 0.6823280981918965,
      "grad_norm": 0.1132221594452858,
      "learning_rate": 5.547954332509805e-07,
      "loss": 0.2943,
      "num_input_tokens_seen": 2539776,
      "step": 5170
    },
    {
      "epoch": 0.6829879899696449,
      "grad_norm": 0.03335703909397125,
      "learning_rate": 5.527339437420717e-07,
      "loss": 0.001,
      "num_input_tokens_seen": 2542208,
      "step": 5175
    },
    {
      "epoch": 0.6836478817473934,
      "grad_norm": 0.27844464778900146,
      "learning_rate": 5.506748275959094e-07,
      "loss": 0.0691,
      "num_input_tokens_seen": 2544704,
      "step": 5180
    },
    {
      "epoch": 0.6843077735251418,
      "grad_norm": 0.07211031764745712,
      "learning_rate": 5.48618095738943e-07,
      "loss": 0.0527,
      "num_input_tokens_seen": 2547072,
      "step": 5185
    },
    {
      "epoch": 0.6849676653028903,
      "grad_norm": 0.14520291984081268,
      "learning_rate": 5.465637590849681e-07,
      "loss": 0.121,
      "num_input_tokens_seen": 2549440,
      "step": 5190
    },
    {
      "epoch": 0.6856275570806388,
      "grad_norm": 43.26601028442383,
      "learning_rate": 5.445118285350723e-07,
      "loss": 0.1132,
      "num_input_tokens_seen": 2552128,
      "step": 5195
    },
    {
      "epoch": 0.6862874488583872,
      "grad_norm": 0.13170602917671204,
      "learning_rate": 5.424623149775745e-07,
      "loss": 0.0694,
      "num_input_tokens_seen": 2554368,
      "step": 5200
    },
    {
      "epoch": 0.6869473406361357,
      "grad_norm": 0.018987668678164482,
      "learning_rate": 5.404152292879676e-07,
      "loss": 0.1189,
      "num_input_tokens_seen": 2556928,
      "step": 5205
    },
    {
      "epoch": 0.6876072324138841,
      "grad_norm": 20.207496643066406,
      "learning_rate": 5.38370582328863e-07,
      "loss": 0.1717,
      "num_input_tokens_seen": 2559360,
      "step": 5210
    },
    {
      "epoch": 0.6882671241916326,
      "grad_norm": 0.22772113978862762,
      "learning_rate": 5.363283849499293e-07,
      "loss": 0.1758,
      "num_input_tokens_seen": 2561856,
      "step": 5215
    },
    {
      "epoch": 0.688927015969381,
      "grad_norm": 17.795719146728516,
      "learning_rate": 5.342886479878387e-07,
      "loss": 0.1671,
      "num_input_tokens_seen": 2564352,
      "step": 5220
    },
    {
      "epoch": 0.6895869077471295,
      "grad_norm": 0.4601105749607086,
      "learning_rate": 5.32251382266206e-07,
      "loss": 0.0526,
      "num_input_tokens_seen": 2566784,
      "step": 5225
    },
    {
      "epoch": 0.6902467995248779,
      "grad_norm": 0.38812679052352905,
      "learning_rate": 5.302165985955327e-07,
      "loss": 0.1115,
      "num_input_tokens_seen": 2569152,
      "step": 5230
    },
    {
      "epoch": 0.6909066913026264,
      "grad_norm": 0.11644639819860458,
      "learning_rate": 5.281843077731511e-07,
      "loss": 0.0649,
      "num_input_tokens_seen": 2571520,
      "step": 5235
    },
    {
      "epoch": 0.6915665830803748,
      "grad_norm": 79.44305419921875,
      "learning_rate": 5.26154520583163e-07,
      "loss": 0.0934,
      "num_input_tokens_seen": 2574080,
      "step": 5240
    },
    {
      "epoch": 0.6922264748581233,
      "grad_norm": 0.16522075235843658,
      "learning_rate": 5.241272477963877e-07,
      "loss": 0.1196,
      "num_input_tokens_seen": 2576320,
      "step": 5245
    },
    {
      "epoch": 0.6928863666358717,
      "grad_norm": 0.037316884845495224,
      "learning_rate": 5.221025001703e-07,
      "loss": 0.0494,
      "num_input_tokens_seen": 2578752,
      "step": 5250
    },
    {
      "epoch": 0.6935462584136202,
      "grad_norm": 19.88916778564453,
      "learning_rate": 5.200802884489768e-07,
      "loss": 0.1279,
      "num_input_tokens_seen": 2581184,
      "step": 5255
    },
    {
      "epoch": 0.6942061501913687,
      "grad_norm": 17.280460357666016,
      "learning_rate": 5.180606233630374e-07,
      "loss": 0.2016,
      "num_input_tokens_seen": 2583872,
      "step": 5260
    },
    {
      "epoch": 0.694866041969117,
      "grad_norm": 0.17049993574619293,
      "learning_rate": 5.160435156295879e-07,
      "loss": 0.1809,
      "num_input_tokens_seen": 2586304,
      "step": 5265
    },
    {
      "epoch": 0.6955259337468656,
      "grad_norm": 0.04596101865172386,
      "learning_rate": 5.14028975952165e-07,
      "loss": 0.0593,
      "num_input_tokens_seen": 2589056,
      "step": 5270
    },
    {
      "epoch": 0.6961858255246139,
      "grad_norm": 13.068889617919922,
      "learning_rate": 5.120170150206768e-07,
      "loss": 0.0839,
      "num_input_tokens_seen": 2591488,
      "step": 5275
    },
    {
      "epoch": 0.6968457173023624,
      "grad_norm": 25.771705627441406,
      "learning_rate": 5.100076435113496e-07,
      "loss": 0.0112,
      "num_input_tokens_seen": 2593792,
      "step": 5280
    },
    {
      "epoch": 0.6975056090801108,
      "grad_norm": 176.9816436767578,
      "learning_rate": 5.080008720866673e-07,
      "loss": 0.1564,
      "num_input_tokens_seen": 2595968,
      "step": 5285
    },
    {
      "epoch": 0.6981655008578593,
      "grad_norm": 14.442154884338379,
      "learning_rate": 5.059967113953173e-07,
      "loss": 0.2333,
      "num_input_tokens_seen": 2598144,
      "step": 5290
    },
    {
      "epoch": 0.6988253926356077,
      "grad_norm": 0.13442066311836243,
      "learning_rate": 5.039951720721349e-07,
      "loss": 0.0942,
      "num_input_tokens_seen": 2600448,
      "step": 5295
    },
    {
      "epoch": 0.6994852844133562,
      "grad_norm": 0.41721171140670776,
      "learning_rate": 5.019962647380429e-07,
      "loss": 0.0922,
      "num_input_tokens_seen": 2602944,
      "step": 5300
    },
    {
      "epoch": 0.7001451761911046,
      "grad_norm": 78.33120727539062,
      "learning_rate": 5.000000000000002e-07,
      "loss": 0.1829,
      "num_input_tokens_seen": 2605120,
      "step": 5305
    },
    {
      "epoch": 0.7002771545466544,
      "eval_loss": 0.09362584352493286,
      "eval_runtime": 8.0352,
      "eval_samples_per_second": 838.191,
      "eval_steps_per_second": 104.789,
      "num_input_tokens_seen": 2605504,
      "step": 5306
    },
    {
      "epoch": 0.7008050679688531,
      "grad_norm": 0.36335432529449463,
      "learning_rate": 4.980063884509414e-07,
      "loss": 0.0412,
      "num_input_tokens_seen": 2607296,
      "step": 5310
    },
    {
      "epoch": 0.7014649597466015,
      "grad_norm": 0.11268818378448486,
      "learning_rate": 4.960154406697229e-07,
      "loss": 0.0804,
      "num_input_tokens_seen": 2609728,
      "step": 5315
    },
    {
      "epoch": 0.70212485152435,
      "grad_norm": 14.370553016662598,
      "learning_rate": 4.940271672210667e-07,
      "loss": 0.2263,
      "num_input_tokens_seen": 2612224,
      "step": 5320
    },
    {
      "epoch": 0.7027847433020985,
      "grad_norm": 0.1433858722448349,
      "learning_rate": 4.920415786555025e-07,
      "loss": 0.0293,
      "num_input_tokens_seen": 2614720,
      "step": 5325
    },
    {
      "epoch": 0.7034446350798469,
      "grad_norm": 0.18172575533390045,
      "learning_rate": 4.900586855093144e-07,
      "loss": 0.2921,
      "num_input_tokens_seen": 2617344,
      "step": 5330
    },
    {
      "epoch": 0.7041045268575954,
      "grad_norm": 20.320568084716797,
      "learning_rate": 4.880784983044827e-07,
      "loss": 0.1245,
      "num_input_tokens_seen": 2619584,
      "step": 5335
    },
    {
      "epoch": 0.7047644186353438,
      "grad_norm": 0.19204528629779816,
      "learning_rate": 4.861010275486284e-07,
      "loss": 0.0299,
      "num_input_tokens_seen": 2621888,
      "step": 5340
    },
    {
      "epoch": 0.7054243104130923,
      "grad_norm": 0.053933948278427124,
      "learning_rate": 4.8412628373496e-07,
      "loss": 0.0976,
      "num_input_tokens_seen": 2624512,
      "step": 5345
    },
    {
      "epoch": 0.7060842021908407,
      "grad_norm": 0.2728992700576782,
      "learning_rate": 4.821542773422136e-07,
      "loss": 0.0023,
      "num_input_tokens_seen": 2627008,
      "step": 5350
    },
    {
      "epoch": 0.7067440939685892,
      "grad_norm": 0.5393227934837341,
      "learning_rate": 4.801850188346012e-07,
      "loss": 0.0019,
      "num_input_tokens_seen": 2629440,
      "step": 5355
    },
    {
      "epoch": 0.7074039857463376,
      "grad_norm": 3.791325092315674,
      "learning_rate": 4.782185186617523e-07,
      "loss": 0.1053,
      "num_input_tokens_seen": 2631872,
      "step": 5360
    },
    {
      "epoch": 0.7080638775240861,
      "grad_norm": 0.12838907539844513,
      "learning_rate": 4.762547872586603e-07,
      "loss": 0.0618,
      "num_input_tokens_seen": 2634560,
      "step": 5365
    },
    {
      "epoch": 0.7087237693018344,
      "grad_norm": 14.399864196777344,
      "learning_rate": 4.7429383504562605e-07,
      "loss": 0.1344,
      "num_input_tokens_seen": 2637120,
      "step": 5370
    },
    {
      "epoch": 0.709383661079583,
      "grad_norm": 0.2652495503425598,
      "learning_rate": 4.723356724282029e-07,
      "loss": 0.0227,
      "num_input_tokens_seen": 2639552,
      "step": 5375
    },
    {
      "epoch": 0.7100435528573315,
      "grad_norm": 20.664852142333984,
      "learning_rate": 4.703803097971426e-07,
      "loss": 0.128,
      "num_input_tokens_seen": 2641984,
      "step": 5380
    },
    {
      "epoch": 0.7107034446350798,
      "grad_norm": 11.795884132385254,
      "learning_rate": 4.6842775752833763e-07,
      "loss": 0.0719,
      "num_input_tokens_seen": 2644352,
      "step": 5385
    },
    {
      "epoch": 0.7113633364128283,
      "grad_norm": 0.1572529375553131,
      "learning_rate": 4.664780259827689e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 2647040,
      "step": 5390
    },
    {
      "epoch": 0.7120232281905767,
      "grad_norm": 0.04447837546467781,
      "learning_rate": 4.6453112550644857e-07,
      "loss": 0.001,
      "num_input_tokens_seen": 2649472,
      "step": 5395
    },
    {
      "epoch": 0.7126831199683252,
      "grad_norm": 0.3557925522327423,
      "learning_rate": 4.625870664303663e-07,
      "loss": 0.0643,
      "num_input_tokens_seen": 2651840,
      "step": 5400
    },
    {
      "epoch": 0.7133430117460736,
      "grad_norm": 0.15744389593601227,
      "learning_rate": 4.6064585907043486e-07,
      "loss": 0.0024,
      "num_input_tokens_seen": 2654464,
      "step": 5405
    },
    {
      "epoch": 0.7140029035238221,
      "grad_norm": 0.013112170621752739,
      "learning_rate": 4.587075137274334e-07,
      "loss": 0.0598,
      "num_input_tokens_seen": 2656576,
      "step": 5410
    },
    {
      "epoch": 0.7146627953015705,
      "grad_norm": 0.7360963821411133,
      "learning_rate": 4.5677204068695597e-07,
      "loss": 0.0527,
      "num_input_tokens_seen": 2659008,
      "step": 5415
    },
    {
      "epoch": 0.715322687079319,
      "grad_norm": 0.024803217500448227,
      "learning_rate": 4.5483945021935356e-07,
      "loss": 0.037,
      "num_input_tokens_seen": 2661632,
      "step": 5420
    },
    {
      "epoch": 0.7159825788570674,
      "grad_norm": 0.0293336883187294,
      "learning_rate": 4.5290975257968155e-07,
      "loss": 0.1053,
      "num_input_tokens_seen": 2664192,
      "step": 5425
    },
    {
      "epoch": 0.7166424706348159,
      "grad_norm": 37.842960357666016,
      "learning_rate": 4.509829580076452e-07,
      "loss": 0.185,
      "num_input_tokens_seen": 2666624,
      "step": 5430
    },
    {
      "epoch": 0.7173023624125643,
      "grad_norm": 0.14816348254680634,
      "learning_rate": 4.490590767275442e-07,
      "loss": 0.1415,
      "num_input_tokens_seen": 2669120,
      "step": 5435
    },
    {
      "epoch": 0.7179622541903128,
      "grad_norm": 0.13981057703495026,
      "learning_rate": 4.4713811894822064e-07,
      "loss": 0.1221,
      "num_input_tokens_seen": 2671552,
      "step": 5440
    },
    {
      "epoch": 0.7186221459680613,
      "grad_norm": 0.21771204471588135,
      "learning_rate": 4.4522009486300204e-07,
      "loss": 0.0782,
      "num_input_tokens_seen": 2674240,
      "step": 5445
    },
    {
      "epoch": 0.7192820377458097,
      "grad_norm": 0.08881372213363647,
      "learning_rate": 4.43305014649649e-07,
      "loss": 0.0865,
      "num_input_tokens_seen": 2676544,
      "step": 5450
    },
    {
      "epoch": 0.7199419295235582,
      "grad_norm": 0.0907829999923706,
      "learning_rate": 4.4139288847030155e-07,
      "loss": 0.0004,
      "num_input_tokens_seen": 2678912,
      "step": 5455
    },
    {
      "epoch": 0.7206018213013066,
      "grad_norm": 0.06906948238611221,
      "learning_rate": 4.394837264714233e-07,
      "loss": 0.0538,
      "num_input_tokens_seen": 2681344,
      "step": 5460
    },
    {
      "epoch": 0.7212617130790551,
      "grad_norm": 0.9202603697776794,
      "learning_rate": 4.3757753878375005e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 2683776,
      "step": 5465
    },
    {
      "epoch": 0.7219216048568035,
      "grad_norm": 0.03742950037121773,
      "learning_rate": 4.3567433552223375e-07,
      "loss": 0.0553,
      "num_input_tokens_seen": 2686016,
      "step": 5470
    },
    {
      "epoch": 0.722581496634552,
      "grad_norm": 0.49609553813934326,
      "learning_rate": 4.3377412678599e-07,
      "loss": 0.1592,
      "num_input_tokens_seen": 2688128,
      "step": 5475
    },
    {
      "epoch": 0.7232413884123003,
      "grad_norm": 0.0728292167186737,
      "learning_rate": 4.318769226582454e-07,
      "loss": 0.1678,
      "num_input_tokens_seen": 2690368,
      "step": 5480
    },
    {
      "epoch": 0.7239012801900488,
      "grad_norm": 0.5910032391548157,
      "learning_rate": 4.299827332062811e-07,
      "loss": 0.0807,
      "num_input_tokens_seen": 2692992,
      "step": 5485
    },
    {
      "epoch": 0.7245611719677972,
      "grad_norm": 51.44681930541992,
      "learning_rate": 4.2809156848138363e-07,
      "loss": 0.0951,
      "num_input_tokens_seen": 2695424,
      "step": 5490
    },
    {
      "epoch": 0.7252210637455457,
      "grad_norm": 76.25809478759766,
      "learning_rate": 4.2620343851878616e-07,
      "loss": 0.1498,
      "num_input_tokens_seen": 2697856,
      "step": 5495
    },
    {
      "epoch": 0.7258809555232941,
      "grad_norm": 0.06275495141744614,
      "learning_rate": 4.2431835333762123e-07,
      "loss": 0.0717,
      "num_input_tokens_seen": 2700608,
      "step": 5500
    },
    {
      "epoch": 0.7265408473010426,
      "grad_norm": 0.07024704664945602,
      "learning_rate": 4.224363229408628e-07,
      "loss": 0.0003,
      "num_input_tokens_seen": 2703104,
      "step": 5505
    },
    {
      "epoch": 0.7272007390787911,
      "grad_norm": 78.2071533203125,
      "learning_rate": 4.205573573152753e-07,
      "loss": 0.2012,
      "num_input_tokens_seen": 2705344,
      "step": 5510
    },
    {
      "epoch": 0.7278606308565395,
      "grad_norm": 0.2134246826171875,
      "learning_rate": 4.18681466431361e-07,
      "loss": 0.0556,
      "num_input_tokens_seen": 2707520,
      "step": 5515
    },
    {
      "epoch": 0.728520522634288,
      "grad_norm": 0.714931070804596,
      "learning_rate": 4.168086602433055e-07,
      "loss": 0.0709,
      "num_input_tokens_seen": 2709888,
      "step": 5520
    },
    {
      "epoch": 0.7291804144120364,
      "grad_norm": 0.3997064232826233,
      "learning_rate": 4.1493894868892676e-07,
      "loss": 0.2017,
      "num_input_tokens_seen": 2712192,
      "step": 5525
    },
    {
      "epoch": 0.7298403061897849,
      "grad_norm": 0.4337630271911621,
      "learning_rate": 4.1307234168962093e-07,
      "loss": 0.0165,
      "num_input_tokens_seen": 2714368,
      "step": 5530
    },
    {
      "epoch": 0.7305001979675333,
      "grad_norm": 0.09288299828767776,
      "learning_rate": 4.112088491503095e-07,
      "loss": 0.0494,
      "num_input_tokens_seen": 2716608,
      "step": 5535
    },
    {
      "epoch": 0.7311600897452818,
      "grad_norm": 0.06428027153015137,
      "learning_rate": 4.0934848095938937e-07,
      "loss": 0.0011,
      "num_input_tokens_seen": 2718656,
      "step": 5540
    },
    {
      "epoch": 0.7318199815230302,
      "grad_norm": 0.03982330486178398,
      "learning_rate": 4.074912469886763e-07,
      "loss": 0.097,
      "num_input_tokens_seen": 2721152,
      "step": 5545
    },
    {
      "epoch": 0.7324798733007787,
      "grad_norm": 5.011870861053467,
      "learning_rate": 4.0563715709335657e-07,
      "loss": 0.0014,
      "num_input_tokens_seen": 2723264,
      "step": 5550
    },
    {
      "epoch": 0.7331397650785271,
      "grad_norm": 143.84207153320312,
      "learning_rate": 4.037862211119315e-07,
      "loss": 0.198,
      "num_input_tokens_seen": 2725568,
      "step": 5555
    },
    {
      "epoch": 0.7337996568562756,
      "grad_norm": 0.1886509656906128,
      "learning_rate": 4.0193844886616715e-07,
      "loss": 0.0634,
      "num_input_tokens_seen": 2728192,
      "step": 5560
    },
    {
      "epoch": 0.7344595486340241,
      "grad_norm": 46.56074142456055,
      "learning_rate": 4.0009385016104137e-07,
      "loss": 0.2335,
      "num_input_tokens_seen": 2731072,
      "step": 5565
    },
    {
      "epoch": 0.7351194404117725,
      "grad_norm": 0.030098050832748413,
      "learning_rate": 3.9825243478469164e-07,
      "loss": 0.0761,
      "num_input_tokens_seen": 2733440,
      "step": 5570
    },
    {
      "epoch": 0.735779332189521,
      "grad_norm": 11.673151016235352,
      "learning_rate": 3.9641421250836484e-07,
      "loss": 0.0718,
      "num_input_tokens_seen": 2736064,
      "step": 5575
    },
    {
      "epoch": 0.7364392239672694,
      "grad_norm": 70.8289794921875,
      "learning_rate": 3.945791930863622e-07,
      "loss": 0.0312,
      "num_input_tokens_seen": 2738496,
      "step": 5580
    },
    {
      "epoch": 0.7370991157450179,
      "grad_norm": 0.15004687011241913,
      "learning_rate": 3.9274738625599137e-07,
      "loss": 0.0019,
      "num_input_tokens_seen": 2740800,
      "step": 5585
    },
    {
      "epoch": 0.7377590075227662,
      "grad_norm": 3.4134578704833984,
      "learning_rate": 3.909188017375112e-07,
      "loss": 0.0637,
      "num_input_tokens_seen": 2743104,
      "step": 5590
    },
    {
      "epoch": 0.7384188993005147,
      "grad_norm": 0.05561397597193718,
      "learning_rate": 3.890934492340819e-07,
      "loss": 0.137,
      "num_input_tokens_seen": 2745344,
      "step": 5595
    },
    {
      "epoch": 0.7390787910782631,
      "grad_norm": 19.8775577545166,
      "learning_rate": 3.872713384317147e-07,
      "loss": 0.0649,
      "num_input_tokens_seen": 2747520,
      "step": 5600
    },
    {
      "epoch": 0.7397386828560116,
      "grad_norm": 0.08291494101285934,
      "learning_rate": 3.8545247899921776e-07,
      "loss": 0.1226,
      "num_input_tokens_seen": 2750016,
      "step": 5605
    },
    {
      "epoch": 0.74039857463376,
      "grad_norm": 0.08636556565761566,
      "learning_rate": 3.8363688058814614e-07,
      "loss": 0.123,
      "num_input_tokens_seen": 2752704,
      "step": 5610
    },
    {
      "epoch": 0.7410584664115085,
      "grad_norm": 15.197744369506836,
      "learning_rate": 3.818245528327526e-07,
      "loss": 0.1293,
      "num_input_tokens_seen": 2755328,
      "step": 5615
    },
    {
      "epoch": 0.7417183581892569,
      "grad_norm": 14.93408203125,
      "learning_rate": 3.8001550534993164e-07,
      "loss": 0.0808,
      "num_input_tokens_seen": 2757632,
      "step": 5620
    },
    {
      "epoch": 0.7423782499670054,
      "grad_norm": 0.16387708485126495,
      "learning_rate": 3.7820974773917413e-07,
      "loss": 0.0508,
      "num_input_tokens_seen": 2760192,
      "step": 5625
    },
    {
      "epoch": 0.7430381417447539,
      "grad_norm": 0.04745183512568474,
      "learning_rate": 3.764072895825117e-07,
      "loss": 0.0011,
      "num_input_tokens_seen": 2762816,
      "step": 5630
    },
    {
      "epoch": 0.7436980335225023,
      "grad_norm": 13.764304161071777,
      "learning_rate": 3.7460814044446934e-07,
      "loss": 0.0735,
      "num_input_tokens_seen": 2765120,
      "step": 5635
    },
    {
      "epoch": 0.7443579253002508,
      "grad_norm": 0.09650138765573502,
      "learning_rate": 3.72812309872012e-07,
      "loss": 0.1235,
      "num_input_tokens_seen": 2767808,
      "step": 5640
    },
    {
      "epoch": 0.7450178170779992,
      "grad_norm": 0.39747825264930725,
      "learning_rate": 3.71019807394495e-07,
      "loss": 0.0109,
      "num_input_tokens_seen": 2770176,
      "step": 5645
    },
    {
      "epoch": 0.7456777088557477,
      "grad_norm": 0.43890663981437683,
      "learning_rate": 3.6923064252361505e-07,
      "loss": 0.1873,
      "num_input_tokens_seen": 2772672,
      "step": 5650
    },
    {
      "epoch": 0.7463376006334961,
      "grad_norm": 0.17724739015102386,
      "learning_rate": 3.674448247533561e-07,
      "loss": 0.1134,
      "num_input_tokens_seen": 2775104,
      "step": 5655
    },
    {
      "epoch": 0.7469974924112446,
      "grad_norm": 1.322895884513855,
      "learning_rate": 3.656623635599432e-07,
      "loss": 0.2252,
      "num_input_tokens_seen": 2777792,
      "step": 5660
    },
    {
      "epoch": 0.747657384188993,
      "grad_norm": 64.87480163574219,
      "learning_rate": 3.6388326840178865e-07,
      "loss": 0.1396,
      "num_input_tokens_seen": 2780416,
      "step": 5665
    },
    {
      "epoch": 0.7483172759667415,
      "grad_norm": 0.14808109402656555,
      "learning_rate": 3.621075487194435e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 2783232,
      "step": 5670
    },
    {
      "epoch": 0.7489771677444899,
      "grad_norm": 7.88364839553833,
      "learning_rate": 3.603352139355483e-07,
      "loss": 0.1226,
      "num_input_tokens_seen": 2785664,
      "step": 5675
    },
    {
      "epoch": 0.7496370595222384,
      "grad_norm": 0.24036580324172974,
      "learning_rate": 3.58566273454781e-07,
      "loss": 0.0986,
      "num_input_tokens_seen": 2788224,
      "step": 5680
    },
    {
      "epoch": 0.7502969512999867,
      "grad_norm": 0.09827781468629837,
      "learning_rate": 3.5680073666380817e-07,
      "loss": 0.0314,
      "num_input_tokens_seen": 2790656,
      "step": 5685
    },
    {
      "epoch": 0.7502969512999867,
      "eval_loss": 0.10055803507566452,
      "eval_runtime": 7.8216,
      "eval_samples_per_second": 861.081,
      "eval_steps_per_second": 107.651,
      "num_input_tokens_seen": 2790656,
      "step": 5685
    },
    {
      "epoch": 0.7509568430777352,
      "grad_norm": 0.024036020040512085,
      "learning_rate": 3.5503861293123514e-07,
      "loss": 0.1421,
      "num_input_tokens_seen": 2792960,
      "step": 5690
    },
    {
      "epoch": 0.7516167348554837,
      "grad_norm": 0.17825555801391602,
      "learning_rate": 3.532799116075571e-07,
      "loss": 0.092,
      "num_input_tokens_seen": 2795648,
      "step": 5695
    },
    {
      "epoch": 0.7522766266332321,
      "grad_norm": 19.003950119018555,
      "learning_rate": 3.5152464202510777e-07,
      "loss": 0.104,
      "num_input_tokens_seen": 2797696,
      "step": 5700
    },
    {
      "epoch": 0.7529365184109806,
      "grad_norm": 17.081161499023438,
      "learning_rate": 3.4977281349801056e-07,
      "loss": 0.1443,
      "num_input_tokens_seen": 2800192,
      "step": 5705
    },
    {
      "epoch": 0.753596410188729,
      "grad_norm": 0.1098131462931633,
      "learning_rate": 3.4802443532213056e-07,
      "loss": 0.0584,
      "num_input_tokens_seen": 2802560,
      "step": 5710
    },
    {
      "epoch": 0.7542563019664775,
      "grad_norm": 0.023672033101320267,
      "learning_rate": 3.4627951677502233e-07,
      "loss": 0.2492,
      "num_input_tokens_seen": 2804992,
      "step": 5715
    },
    {
      "epoch": 0.7549161937442259,
      "grad_norm": 0.9849115610122681,
      "learning_rate": 3.4453806711588397e-07,
      "loss": 0.0061,
      "num_input_tokens_seen": 2807296,
      "step": 5720
    },
    {
      "epoch": 0.7555760855219744,
      "grad_norm": 0.09603046625852585,
      "learning_rate": 3.428000955855054e-07,
      "loss": 0.0309,
      "num_input_tokens_seen": 2809984,
      "step": 5725
    },
    {
      "epoch": 0.7562359772997228,
      "grad_norm": 0.11034424602985382,
      "learning_rate": 3.4106561140621983e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 2812736,
      "step": 5730
    },
    {
      "epoch": 0.7568958690774713,
      "grad_norm": 0.09973105043172836,
      "learning_rate": 3.393346237818567e-07,
      "loss": 0.173,
      "num_input_tokens_seen": 2815040,
      "step": 5735
    },
    {
      "epoch": 0.7575557608552197,
      "grad_norm": 47.743797302246094,
      "learning_rate": 3.3760714189769015e-07,
      "loss": 0.1591,
      "num_input_tokens_seen": 2817344,
      "step": 5740
    },
    {
      "epoch": 0.7582156526329682,
      "grad_norm": 0.07995446026325226,
      "learning_rate": 3.3588317492039266e-07,
      "loss": 0.024,
      "num_input_tokens_seen": 2819648,
      "step": 5745
    },
    {
      "epoch": 0.7588755444107167,
      "grad_norm": 40.636417388916016,
      "learning_rate": 3.341627319979834e-07,
      "loss": 0.1302,
      "num_input_tokens_seen": 2822464,
      "step": 5750
    },
    {
      "epoch": 0.7595354361884651,
      "grad_norm": 84.80926513671875,
      "learning_rate": 3.324458222597839e-07,
      "loss": 0.2033,
      "num_input_tokens_seen": 2824896,
      "step": 5755
    },
    {
      "epoch": 0.7601953279662136,
      "grad_norm": 0.16351526975631714,
      "learning_rate": 3.307324548163657e-07,
      "loss": 0.0718,
      "num_input_tokens_seen": 2827648,
      "step": 5760
    },
    {
      "epoch": 0.760855219743962,
      "grad_norm": 1.872399926185608,
      "learning_rate": 3.2902263875950374e-07,
      "loss": 0.0575,
      "num_input_tokens_seen": 2830336,
      "step": 5765
    },
    {
      "epoch": 0.7615151115217105,
      "grad_norm": 14.924793243408203,
      "learning_rate": 3.2731638316212894e-07,
      "loss": 0.0498,
      "num_input_tokens_seen": 2832640,
      "step": 5770
    },
    {
      "epoch": 0.7621750032994589,
      "grad_norm": 0.5178006291389465,
      "learning_rate": 3.256136970782782e-07,
      "loss": 0.086,
      "num_input_tokens_seen": 2834880,
      "step": 5775
    },
    {
      "epoch": 0.7628348950772074,
      "grad_norm": 0.028780991211533546,
      "learning_rate": 3.23914589543047e-07,
      "loss": 0.1474,
      "num_input_tokens_seen": 2837440,
      "step": 5780
    },
    {
      "epoch": 0.7634947868549558,
      "grad_norm": 29.812068939208984,
      "learning_rate": 3.2221906957254276e-07,
      "loss": 0.0621,
      "num_input_tokens_seen": 2839808,
      "step": 5785
    },
    {
      "epoch": 0.7641546786327043,
      "grad_norm": 0.09762804210186005,
      "learning_rate": 3.205271461638346e-07,
      "loss": 0.1094,
      "num_input_tokens_seen": 2842432,
      "step": 5790
    },
    {
      "epoch": 0.7648145704104526,
      "grad_norm": 0.17772820591926575,
      "learning_rate": 3.188388282949085e-07,
      "loss": 0.1338,
      "num_input_tokens_seen": 2845120,
      "step": 5795
    },
    {
      "epoch": 0.7654744621882011,
      "grad_norm": 46.48423385620117,
      "learning_rate": 3.171541249246166e-07,
      "loss": 0.1247,
      "num_input_tokens_seen": 2848000,
      "step": 5800
    },
    {
      "epoch": 0.7661343539659495,
      "grad_norm": 20.354982376098633,
      "learning_rate": 3.154730449926316e-07,
      "loss": 0.1788,
      "num_input_tokens_seen": 2850624,
      "step": 5805
    },
    {
      "epoch": 0.766794245743698,
      "grad_norm": 0.2434767633676529,
      "learning_rate": 3.137955974194e-07,
      "loss": 0.1394,
      "num_input_tokens_seen": 2852992,
      "step": 5810
    },
    {
      "epoch": 0.7674541375214465,
      "grad_norm": 14.338163375854492,
      "learning_rate": 3.1212179110609125e-07,
      "loss": 0.0413,
      "num_input_tokens_seen": 2855424,
      "step": 5815
    },
    {
      "epoch": 0.7681140292991949,
      "grad_norm": 0.05543454736471176,
      "learning_rate": 3.104516349345553e-07,
      "loss": 0.1604,
      "num_input_tokens_seen": 2857984,
      "step": 5820
    },
    {
      "epoch": 0.7687739210769434,
      "grad_norm": 0.143641859292984,
      "learning_rate": 3.0878513776727144e-07,
      "loss": 0.0702,
      "num_input_tokens_seen": 2860672,
      "step": 5825
    },
    {
      "epoch": 0.7694338128546918,
      "grad_norm": 15.186430931091309,
      "learning_rate": 3.0712230844730414e-07,
      "loss": 0.1228,
      "num_input_tokens_seen": 2863040,
      "step": 5830
    },
    {
      "epoch": 0.7700937046324403,
      "grad_norm": 35.79595947265625,
      "learning_rate": 3.054631557982539e-07,
      "loss": 0.1412,
      "num_input_tokens_seen": 2865856,
      "step": 5835
    },
    {
      "epoch": 0.7707535964101887,
      "grad_norm": 0.11609415709972382,
      "learning_rate": 3.0380768862421156e-07,
      "loss": 0.108,
      "num_input_tokens_seen": 2868096,
      "step": 5840
    },
    {
      "epoch": 0.7714134881879372,
      "grad_norm": 0.14167936146259308,
      "learning_rate": 3.0215591570971234e-07,
      "loss": 0.001,
      "num_input_tokens_seen": 2870784,
      "step": 5845
    },
    {
      "epoch": 0.7720733799656856,
      "grad_norm": 0.7140024304389954,
      "learning_rate": 3.005078458196868e-07,
      "loss": 0.0313,
      "num_input_tokens_seen": 2873216,
      "step": 5850
    },
    {
      "epoch": 0.7727332717434341,
      "grad_norm": 0.13905233144760132,
      "learning_rate": 2.988634876994175e-07,
      "loss": 0.0011,
      "num_input_tokens_seen": 2875776,
      "step": 5855
    },
    {
      "epoch": 0.7733931635211825,
      "grad_norm": 0.11488426476716995,
      "learning_rate": 2.972228500744898e-07,
      "loss": 0.0322,
      "num_input_tokens_seen": 2878336,
      "step": 5860
    },
    {
      "epoch": 0.774053055298931,
      "grad_norm": 18.05517578125,
      "learning_rate": 2.955859416507467e-07,
      "loss": 0.1583,
      "num_input_tokens_seen": 2880896,
      "step": 5865
    },
    {
      "epoch": 0.7747129470766794,
      "grad_norm": 0.03384196758270264,
      "learning_rate": 2.9395277111424357e-07,
      "loss": 0.0666,
      "num_input_tokens_seen": 2883648,
      "step": 5870
    },
    {
      "epoch": 0.7753728388544279,
      "grad_norm": 0.047510821372270584,
      "learning_rate": 2.9232334713120035e-07,
      "loss": 0.0659,
      "num_input_tokens_seen": 2885952,
      "step": 5875
    },
    {
      "epoch": 0.7760327306321764,
      "grad_norm": 0.21850033104419708,
      "learning_rate": 2.9069767834795655e-07,
      "loss": 0.0666,
      "num_input_tokens_seen": 2888576,
      "step": 5880
    },
    {
      "epoch": 0.7766926224099248,
      "grad_norm": 1.7232064008712769,
      "learning_rate": 2.8907577339092483e-07,
      "loss": 0.139,
      "num_input_tokens_seen": 2891136,
      "step": 5885
    },
    {
      "epoch": 0.7773525141876733,
      "grad_norm": 0.034715685993433,
      "learning_rate": 2.8745764086654654e-07,
      "loss": 0.0791,
      "num_input_tokens_seen": 2893696,
      "step": 5890
    },
    {
      "epoch": 0.7780124059654216,
      "grad_norm": 0.3917783796787262,
      "learning_rate": 2.8584328936124424e-07,
      "loss": 0.0014,
      "num_input_tokens_seen": 2896512,
      "step": 5895
    },
    {
      "epoch": 0.7786722977431701,
      "grad_norm": 15.660504341125488,
      "learning_rate": 2.8423272744137674e-07,
      "loss": 0.1734,
      "num_input_tokens_seen": 2899008,
      "step": 5900
    },
    {
      "epoch": 0.7793321895209185,
      "grad_norm": 0.46314364671707153,
      "learning_rate": 2.82625963653195e-07,
      "loss": 0.0576,
      "num_input_tokens_seen": 2901376,
      "step": 5905
    },
    {
      "epoch": 0.779992081298667,
      "grad_norm": 12.847580909729004,
      "learning_rate": 2.810230065227944e-07,
      "loss": 0.192,
      "num_input_tokens_seen": 2903872,
      "step": 5910
    },
    {
      "epoch": 0.7806519730764154,
      "grad_norm": 0.09266316145658493,
      "learning_rate": 2.7942386455607203e-07,
      "loss": 0.0012,
      "num_input_tokens_seen": 2906240,
      "step": 5915
    },
    {
      "epoch": 0.7813118648541639,
      "grad_norm": 0.7352420687675476,
      "learning_rate": 2.77828546238679e-07,
      "loss": 0.068,
      "num_input_tokens_seen": 2908736,
      "step": 5920
    },
    {
      "epoch": 0.7819717566319123,
      "grad_norm": 0.6023374199867249,
      "learning_rate": 2.762370600359774e-07,
      "loss": 0.1235,
      "num_input_tokens_seen": 2911104,
      "step": 5925
    },
    {
      "epoch": 0.7826316484096608,
      "grad_norm": 0.05204444006085396,
      "learning_rate": 2.7464941439299484e-07,
      "loss": 0.0383,
      "num_input_tokens_seen": 2913472,
      "step": 5930
    },
    {
      "epoch": 0.7832915401874093,
      "grad_norm": 13.93787670135498,
      "learning_rate": 2.7306561773437887e-07,
      "loss": 0.1809,
      "num_input_tokens_seen": 2915840,
      "step": 5935
    },
    {
      "epoch": 0.7839514319651577,
      "grad_norm": 0.12649060785770416,
      "learning_rate": 2.714856784643533e-07,
      "loss": 0.0033,
      "num_input_tokens_seen": 2918144,
      "step": 5940
    },
    {
      "epoch": 0.7846113237429062,
      "grad_norm": 0.04809973016381264,
      "learning_rate": 2.6990960496667313e-07,
      "loss": 0.1415,
      "num_input_tokens_seen": 2920768,
      "step": 5945
    },
    {
      "epoch": 0.7852712155206546,
      "grad_norm": 0.12313824892044067,
      "learning_rate": 2.6833740560457976e-07,
      "loss": 0.0698,
      "num_input_tokens_seen": 2923136,
      "step": 5950
    },
    {
      "epoch": 0.7859311072984031,
      "grad_norm": 0.06649130582809448,
      "learning_rate": 2.6676908872075757e-07,
      "loss": 0.0759,
      "num_input_tokens_seen": 2925568,
      "step": 5955
    },
    {
      "epoch": 0.7865909990761515,
      "grad_norm": 0.062394797801971436,
      "learning_rate": 2.6520466263728836e-07,
      "loss": 0.0742,
      "num_input_tokens_seen": 2928064,
      "step": 5960
    },
    {
      "epoch": 0.7872508908539,
      "grad_norm": 19.138282775878906,
      "learning_rate": 2.636441356556087e-07,
      "loss": 0.2564,
      "num_input_tokens_seen": 2930368,
      "step": 5965
    },
    {
      "epoch": 0.7879107826316484,
      "grad_norm": 15.425389289855957,
      "learning_rate": 2.620875160564645e-07,
      "loss": 0.1123,
      "num_input_tokens_seen": 2932928,
      "step": 5970
    },
    {
      "epoch": 0.7885706744093969,
      "grad_norm": 35.93429946899414,
      "learning_rate": 2.6053481209986715e-07,
      "loss": 0.3593,
      "num_input_tokens_seen": 2935360,
      "step": 5975
    },
    {
      "epoch": 0.7892305661871453,
      "grad_norm": 0.08447807282209396,
      "learning_rate": 2.5898603202505155e-07,
      "loss": 0.059,
      "num_input_tokens_seen": 2937920,
      "step": 5980
    },
    {
      "epoch": 0.7898904579648938,
      "grad_norm": 0.1833449751138687,
      "learning_rate": 2.5744118405042923e-07,
      "loss": 0.0823,
      "num_input_tokens_seen": 2940224,
      "step": 5985
    },
    {
      "epoch": 0.7905503497426422,
      "grad_norm": 0.16191020607948303,
      "learning_rate": 2.559002763735485e-07,
      "loss": 0.0012,
      "num_input_tokens_seen": 2942848,
      "step": 5990
    },
    {
      "epoch": 0.7912102415203907,
      "grad_norm": 0.1544707715511322,
      "learning_rate": 2.543633171710472e-07,
      "loss": 0.0604,
      "num_input_tokens_seen": 2945344,
      "step": 5995
    },
    {
      "epoch": 0.7918701332981392,
      "grad_norm": 12.624445915222168,
      "learning_rate": 2.5283031459861205e-07,
      "loss": 0.008,
      "num_input_tokens_seen": 2947840,
      "step": 6000
    },
    {
      "epoch": 0.7925300250758875,
      "grad_norm": 0.09669650346040726,
      "learning_rate": 2.5130127679093396e-07,
      "loss": 0.0374,
      "num_input_tokens_seen": 2950144,
      "step": 6005
    },
    {
      "epoch": 0.793189916853636,
      "grad_norm": 110.95389556884766,
      "learning_rate": 2.497762118616652e-07,
      "loss": 0.0336,
      "num_input_tokens_seen": 2952384,
      "step": 6010
    },
    {
      "epoch": 0.7938498086313844,
      "grad_norm": 0.06578873097896576,
      "learning_rate": 2.4825512790337745e-07,
      "loss": 0.071,
      "num_input_tokens_seen": 2955136,
      "step": 6015
    },
    {
      "epoch": 0.7945097004091329,
      "grad_norm": 24.038101196289062,
      "learning_rate": 2.467380329875163e-07,
      "loss": 0.0369,
      "num_input_tokens_seen": 2957824,
      "step": 6020
    },
    {
      "epoch": 0.7951695921868813,
      "grad_norm": 0.2718788683414459,
      "learning_rate": 2.452249351643615e-07,
      "loss": 0.0014,
      "num_input_tokens_seen": 2960256,
      "step": 6025
    },
    {
      "epoch": 0.7958294839646298,
      "grad_norm": 0.3302108943462372,
      "learning_rate": 2.437158424629817e-07,
      "loss": 0.1182,
      "num_input_tokens_seen": 2962944,
      "step": 6030
    },
    {
      "epoch": 0.7964893757423782,
      "grad_norm": 66.44668579101562,
      "learning_rate": 2.422107628911929e-07,
      "loss": 0.2256,
      "num_input_tokens_seen": 2965504,
      "step": 6035
    },
    {
      "epoch": 0.7971492675201267,
      "grad_norm": 0.32538238167762756,
      "learning_rate": 2.4070970443551673e-07,
      "loss": 0.274,
      "num_input_tokens_seen": 2967744,
      "step": 6040
    },
    {
      "epoch": 0.7978091592978751,
      "grad_norm": 0.16842018067836761,
      "learning_rate": 2.392126750611362e-07,
      "loss": 0.0032,
      "num_input_tokens_seen": 2970240,
      "step": 6045
    },
    {
      "epoch": 0.7984690510756236,
      "grad_norm": 14.741007804870605,
      "learning_rate": 2.3771968271185538e-07,
      "loss": 0.1761,
      "num_input_tokens_seen": 2972928,
      "step": 6050
    },
    {
      "epoch": 0.799128942853372,
      "grad_norm": 0.018400076776742935,
      "learning_rate": 2.3623073531005579e-07,
      "loss": 0.1485,
      "num_input_tokens_seen": 2975168,
      "step": 6055
    },
    {
      "epoch": 0.7997888346311205,
      "grad_norm": 0.2021035999059677,
      "learning_rate": 2.3474584075665493e-07,
      "loss": 0.1396,
      "num_input_tokens_seen": 2977408,
      "step": 6060
    },
    {
      "epoch": 0.8003167480533192,
      "eval_loss": 0.09634685516357422,
      "eval_runtime": 8.0558,
      "eval_samples_per_second": 836.039,
      "eval_steps_per_second": 104.52,
      "num_input_tokens_seen": 2979456,
      "step": 6064
    },
    {
      "epoch": 0.800448726408869,
      "grad_norm": 0.04960273206233978,
      "learning_rate": 2.3326500693106533e-07,
      "loss": 0.0051,
      "num_input_tokens_seen": 2979968,
      "step": 6065
    },
    {
      "epoch": 0.8011086181866174,
      "grad_norm": 59.42499542236328,
      "learning_rate": 2.3178824169114975e-07,
      "loss": 0.2232,
      "num_input_tokens_seen": 2982528,
      "step": 6070
    },
    {
      "epoch": 0.8017685099643659,
      "grad_norm": 0.30131280422210693,
      "learning_rate": 2.303155528731837e-07,
      "loss": 0.0557,
      "num_input_tokens_seen": 2984832,
      "step": 6075
    },
    {
      "epoch": 0.8024284017421143,
      "grad_norm": 0.18146446347236633,
      "learning_rate": 2.2884694829181016e-07,
      "loss": 0.0009,
      "num_input_tokens_seen": 2987328,
      "step": 6080
    },
    {
      "epoch": 0.8030882935198628,
      "grad_norm": 0.08123955875635147,
      "learning_rate": 2.273824357400005e-07,
      "loss": 0.0703,
      "num_input_tokens_seen": 2989760,
      "step": 6085
    },
    {
      "epoch": 0.8037481852976112,
      "grad_norm": 0.2558766305446625,
      "learning_rate": 2.2592202298901174e-07,
      "loss": 0.0457,
      "num_input_tokens_seen": 2992320,
      "step": 6090
    },
    {
      "epoch": 0.8044080770753597,
      "grad_norm": 0.09454154968261719,
      "learning_rate": 2.2446571778834555e-07,
      "loss": 0.0022,
      "num_input_tokens_seen": 2995136,
      "step": 6095
    },
    {
      "epoch": 0.805067968853108,
      "grad_norm": 0.08113214373588562,
      "learning_rate": 2.2301352786570827e-07,
      "loss": 0.0008,
      "num_input_tokens_seen": 2998016,
      "step": 6100
    },
    {
      "epoch": 0.8057278606308566,
      "grad_norm": 0.04773727431893349,
      "learning_rate": 2.215654609269685e-07,
      "loss": 0.1533,
      "num_input_tokens_seen": 3000832,
      "step": 6105
    },
    {
      "epoch": 0.8063877524086049,
      "grad_norm": 19.62008285522461,
      "learning_rate": 2.201215246561161e-07,
      "loss": 0.1716,
      "num_input_tokens_seen": 3003584,
      "step": 6110
    },
    {
      "epoch": 0.8070476441863534,
      "grad_norm": 36.953800201416016,
      "learning_rate": 2.1868172671522357e-07,
      "loss": 0.0731,
      "num_input_tokens_seen": 3006464,
      "step": 6115
    },
    {
      "epoch": 0.8077075359641019,
      "grad_norm": 0.20668062567710876,
      "learning_rate": 2.1724607474440216e-07,
      "loss": 0.107,
      "num_input_tokens_seen": 3008896,
      "step": 6120
    },
    {
      "epoch": 0.8083674277418503,
      "grad_norm": 25.05734634399414,
      "learning_rate": 2.158145763617646e-07,
      "loss": 0.1277,
      "num_input_tokens_seen": 3011392,
      "step": 6125
    },
    {
      "epoch": 0.8090273195195988,
      "grad_norm": 0.5729929208755493,
      "learning_rate": 2.1438723916338198e-07,
      "loss": 0.2135,
      "num_input_tokens_seen": 3014016,
      "step": 6130
    },
    {
      "epoch": 0.8096872112973472,
      "grad_norm": 0.34563905000686646,
      "learning_rate": 2.1296407072324495e-07,
      "loss": 0.1745,
      "num_input_tokens_seen": 3016576,
      "step": 6135
    },
    {
      "epoch": 0.8103471030750957,
      "grad_norm": 24.23431396484375,
      "learning_rate": 2.1154507859322336e-07,
      "loss": 0.0661,
      "num_input_tokens_seen": 3019008,
      "step": 6140
    },
    {
      "epoch": 0.8110069948528441,
      "grad_norm": 56.11611557006836,
      "learning_rate": 2.101302703030252e-07,
      "loss": 0.1033,
      "num_input_tokens_seen": 3021504,
      "step": 6145
    },
    {
      "epoch": 0.8116668866305926,
      "grad_norm": 0.6069444417953491,
      "learning_rate": 2.0871965336015885e-07,
      "loss": 0.0688,
      "num_input_tokens_seen": 3023552,
      "step": 6150
    },
    {
      "epoch": 0.812326778408341,
      "grad_norm": 319.5021057128906,
      "learning_rate": 2.0731323524989031e-07,
      "loss": 0.0896,
      "num_input_tokens_seen": 3025856,
      "step": 6155
    },
    {
      "epoch": 0.8129866701860895,
      "grad_norm": 10.510120391845703,
      "learning_rate": 2.0591102343520616e-07,
      "loss": 0.1915,
      "num_input_tokens_seen": 3028096,
      "step": 6160
    },
    {
      "epoch": 0.8136465619638379,
      "grad_norm": 25.165172576904297,
      "learning_rate": 2.0451302535677206e-07,
      "loss": 0.1774,
      "num_input_tokens_seen": 3030528,
      "step": 6165
    },
    {
      "epoch": 0.8143064537415864,
      "grad_norm": 142.08734130859375,
      "learning_rate": 2.0311924843289396e-07,
      "loss": 0.129,
      "num_input_tokens_seen": 3033088,
      "step": 6170
    },
    {
      "epoch": 0.8149663455193348,
      "grad_norm": 0.14164146780967712,
      "learning_rate": 2.017297000594794e-07,
      "loss": 0.0907,
      "num_input_tokens_seen": 3035200,
      "step": 6175
    },
    {
      "epoch": 0.8156262372970833,
      "grad_norm": 0.1182747557759285,
      "learning_rate": 2.0034438760999696e-07,
      "loss": 0.048,
      "num_input_tokens_seen": 3037696,
      "step": 6180
    },
    {
      "epoch": 0.8162861290748318,
      "grad_norm": 16.164520263671875,
      "learning_rate": 1.9896331843543856e-07,
      "loss": 0.1593,
      "num_input_tokens_seen": 3040128,
      "step": 6185
    },
    {
      "epoch": 0.8169460208525802,
      "grad_norm": 0.241964653134346,
      "learning_rate": 1.975864998642789e-07,
      "loss": 0.1511,
      "num_input_tokens_seen": 3042560,
      "step": 6190
    },
    {
      "epoch": 0.8176059126303287,
      "grad_norm": 39.79319381713867,
      "learning_rate": 1.9621393920243767e-07,
      "loss": 0.2786,
      "num_input_tokens_seen": 3044800,
      "step": 6195
    },
    {
      "epoch": 0.8182658044080771,
      "grad_norm": 1.3789013624191284,
      "learning_rate": 1.9484564373324074e-07,
      "loss": 0.062,
      "num_input_tokens_seen": 3047040,
      "step": 6200
    },
    {
      "epoch": 0.8189256961858256,
      "grad_norm": 0.04190517216920853,
      "learning_rate": 1.934816207173805e-07,
      "loss": 0.0633,
      "num_input_tokens_seen": 3049600,
      "step": 6205
    },
    {
      "epoch": 0.819585587963574,
      "grad_norm": 0.25988835096359253,
      "learning_rate": 1.9212187739287943e-07,
      "loss": 0.1436,
      "num_input_tokens_seen": 3052416,
      "step": 6210
    },
    {
      "epoch": 0.8202454797413224,
      "grad_norm": 0.2508937418460846,
      "learning_rate": 1.907664209750488e-07,
      "loss": 0.0028,
      "num_input_tokens_seen": 3055040,
      "step": 6215
    },
    {
      "epoch": 0.8209053715190708,
      "grad_norm": 0.35706770420074463,
      "learning_rate": 1.8941525865645336e-07,
      "loss": 0.0553,
      "num_input_tokens_seen": 3057856,
      "step": 6220
    },
    {
      "epoch": 0.8215652632968193,
      "grad_norm": 83.89398193359375,
      "learning_rate": 1.8806839760687076e-07,
      "loss": 0.209,
      "num_input_tokens_seen": 3060160,
      "step": 6225
    },
    {
      "epoch": 0.8222251550745677,
      "grad_norm": 0.15399783849716187,
      "learning_rate": 1.867258449732545e-07,
      "loss": 0.1147,
      "num_input_tokens_seen": 3062592,
      "step": 6230
    },
    {
      "epoch": 0.8228850468523162,
      "grad_norm": 27.60859489440918,
      "learning_rate": 1.8538760787969676e-07,
      "loss": 0.0562,
      "num_input_tokens_seen": 3065088,
      "step": 6235
    },
    {
      "epoch": 0.8235449386300646,
      "grad_norm": 0.4883517026901245,
      "learning_rate": 1.8405369342738907e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 3067712,
      "step": 6240
    },
    {
      "epoch": 0.8242048304078131,
      "grad_norm": 0.11648905277252197,
      "learning_rate": 1.8272410869458598e-07,
      "loss": 0.0602,
      "num_input_tokens_seen": 3070144,
      "step": 6245
    },
    {
      "epoch": 0.8248647221855616,
      "grad_norm": 0.22444108128547668,
      "learning_rate": 1.8139886073656653e-07,
      "loss": 0.2637,
      "num_input_tokens_seen": 3072448,
      "step": 6250
    },
    {
      "epoch": 0.82552461396331,
      "grad_norm": 0.2944018840789795,
      "learning_rate": 1.800779565855971e-07,
      "loss": 0.2488,
      "num_input_tokens_seen": 3075072,
      "step": 6255
    },
    {
      "epoch": 0.8261845057410585,
      "grad_norm": 0.07640720903873444,
      "learning_rate": 1.7876140325089463e-07,
      "loss": 0.0066,
      "num_input_tokens_seen": 3077376,
      "step": 6260
    },
    {
      "epoch": 0.8268443975188069,
      "grad_norm": 50.32258605957031,
      "learning_rate": 1.774492077185883e-07,
      "loss": 0.1657,
      "num_input_tokens_seen": 3079808,
      "step": 6265
    },
    {
      "epoch": 0.8275042892965554,
      "grad_norm": 0.03833077847957611,
      "learning_rate": 1.7614137695168408e-07,
      "loss": 0.0009,
      "num_input_tokens_seen": 3082560,
      "step": 6270
    },
    {
      "epoch": 0.8281641810743038,
      "grad_norm": 2.642076015472412,
      "learning_rate": 1.748379178900261e-07,
      "loss": 0.0715,
      "num_input_tokens_seen": 3084608,
      "step": 6275
    },
    {
      "epoch": 0.8288240728520523,
      "grad_norm": 98.66949462890625,
      "learning_rate": 1.7353883745026055e-07,
      "loss": 0.229,
      "num_input_tokens_seen": 3087104,
      "step": 6280
    },
    {
      "epoch": 0.8294839646298007,
      "grad_norm": 4.167269706726074,
      "learning_rate": 1.722441425257999e-07,
      "loss": 0.1264,
      "num_input_tokens_seen": 3089408,
      "step": 6285
    },
    {
      "epoch": 0.8301438564075492,
      "grad_norm": 0.2311401516199112,
      "learning_rate": 1.7095383998678402e-07,
      "loss": 0.0197,
      "num_input_tokens_seen": 3091776,
      "step": 6290
    },
    {
      "epoch": 0.8308037481852976,
      "grad_norm": 0.08988851308822632,
      "learning_rate": 1.6966793668004653e-07,
      "loss": 0.113,
      "num_input_tokens_seen": 3094208,
      "step": 6295
    },
    {
      "epoch": 0.8314636399630461,
      "grad_norm": 0.2825041711330414,
      "learning_rate": 1.6838643942907625e-07,
      "loss": 0.006,
      "num_input_tokens_seen": 3096768,
      "step": 6300
    },
    {
      "epoch": 0.8321235317407946,
      "grad_norm": 39.45750427246094,
      "learning_rate": 1.671093550339815e-07,
      "loss": 0.0423,
      "num_input_tokens_seen": 3099456,
      "step": 6305
    },
    {
      "epoch": 0.832783423518543,
      "grad_norm": 414.50775146484375,
      "learning_rate": 1.6583669027145542e-07,
      "loss": 0.0422,
      "num_input_tokens_seen": 3102208,
      "step": 6310
    },
    {
      "epoch": 0.8334433152962915,
      "grad_norm": 0.2666016221046448,
      "learning_rate": 1.6456845189473767e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 3104896,
      "step": 6315
    },
    {
      "epoch": 0.8341032070740398,
      "grad_norm": 0.1291654258966446,
      "learning_rate": 1.6330464663358123e-07,
      "loss": 0.0676,
      "num_input_tokens_seen": 3107520,
      "step": 6320
    },
    {
      "epoch": 0.8347630988517883,
      "grad_norm": 0.08973629027605057,
      "learning_rate": 1.6204528119421346e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 3110144,
      "step": 6325
    },
    {
      "epoch": 0.8354229906295367,
      "grad_norm": 0.12144739925861359,
      "learning_rate": 1.607903622593042e-07,
      "loss": 0.0483,
      "num_input_tokens_seen": 3112768,
      "step": 6330
    },
    {
      "epoch": 0.8360828824072852,
      "grad_norm": 0.026064425706863403,
      "learning_rate": 1.5953989648792743e-07,
      "loss": 0.0005,
      "num_input_tokens_seen": 3115328,
      "step": 6335
    },
    {
      "epoch": 0.8367427741850336,
      "grad_norm": 0.3229873478412628,
      "learning_rate": 1.5829389051552678e-07,
      "loss": 0.0166,
      "num_input_tokens_seen": 3117888,
      "step": 6340
    },
    {
      "epoch": 0.8374026659627821,
      "grad_norm": 107.5579833984375,
      "learning_rate": 1.5705235095388136e-07,
      "loss": 0.0081,
      "num_input_tokens_seen": 3120384,
      "step": 6345
    },
    {
      "epoch": 0.8380625577405305,
      "grad_norm": 0.10149043798446655,
      "learning_rate": 1.5581528439106907e-07,
      "loss": 0.0288,
      "num_input_tokens_seen": 3123008,
      "step": 6350
    },
    {
      "epoch": 0.838722449518279,
      "grad_norm": 0.22135783731937408,
      "learning_rate": 1.5458269739143292e-07,
      "loss": 0.0373,
      "num_input_tokens_seen": 3125504,
      "step": 6355
    },
    {
      "epoch": 0.8393823412960274,
      "grad_norm": 1.0634685754776,
      "learning_rate": 1.5335459649554538e-07,
      "loss": 0.0028,
      "num_input_tokens_seen": 3127744,
      "step": 6360
    },
    {
      "epoch": 0.8400422330737759,
      "grad_norm": 0.06553785502910614,
      "learning_rate": 1.5213098822017357e-07,
      "loss": 0.1164,
      "num_input_tokens_seen": 3130048,
      "step": 6365
    },
    {
      "epoch": 0.8407021248515244,
      "grad_norm": 0.2539460361003876,
      "learning_rate": 1.50911879058246e-07,
      "loss": 0.0148,
      "num_input_tokens_seen": 3132480,
      "step": 6370
    },
    {
      "epoch": 0.8413620166292728,
      "grad_norm": 1.5171750783920288,
      "learning_rate": 1.4969727547881628e-07,
      "loss": 0.0826,
      "num_input_tokens_seen": 3135104,
      "step": 6375
    },
    {
      "epoch": 0.8420219084070213,
      "grad_norm": 16.333852767944336,
      "learning_rate": 1.4848718392703052e-07,
      "loss": 0.1787,
      "num_input_tokens_seen": 3137344,
      "step": 6380
    },
    {
      "epoch": 0.8426818001847697,
      "grad_norm": 19.633960723876953,
      "learning_rate": 1.472816108240915e-07,
      "loss": 0.1727,
      "num_input_tokens_seen": 3140096,
      "step": 6385
    },
    {
      "epoch": 0.8433416919625182,
      "grad_norm": 0.48766201734542847,
      "learning_rate": 1.46080562567226e-07,
      "loss": 0.0868,
      "num_input_tokens_seen": 3142400,
      "step": 6390
    },
    {
      "epoch": 0.8440015837402666,
      "grad_norm": 0.19222283363342285,
      "learning_rate": 1.4488404552964993e-07,
      "loss": 0.0302,
      "num_input_tokens_seen": 3144512,
      "step": 6395
    },
    {
      "epoch": 0.8446614755180151,
      "grad_norm": 35.827850341796875,
      "learning_rate": 1.4369206606053463e-07,
      "loss": 0.0434,
      "num_input_tokens_seen": 3146944,
      "step": 6400
    },
    {
      "epoch": 0.8453213672957635,
      "grad_norm": 0.27888330817222595,
      "learning_rate": 1.425046304849742e-07,
      "loss": 0.0951,
      "num_input_tokens_seen": 3149376,
      "step": 6405
    },
    {
      "epoch": 0.845981259073512,
      "grad_norm": 8.114583015441895,
      "learning_rate": 1.4132174510395024e-07,
      "loss": 0.1082,
      "num_input_tokens_seen": 3151744,
      "step": 6410
    },
    {
      "epoch": 0.8466411508512603,
      "grad_norm": 0.21486525237560272,
      "learning_rate": 1.4014341619430003e-07,
      "loss": 0.0054,
      "num_input_tokens_seen": 3154112,
      "step": 6415
    },
    {
      "epoch": 0.8473010426290088,
      "grad_norm": 0.10914035141468048,
      "learning_rate": 1.3896965000868188e-07,
      "loss": 0.0044,
      "num_input_tokens_seen": 3156480,
      "step": 6420
    },
    {
      "epoch": 0.8479609344067572,
      "grad_norm": 135.10137939453125,
      "learning_rate": 1.3780045277554276e-07,
      "loss": 0.1166,
      "num_input_tokens_seen": 3158784,
      "step": 6425
    },
    {
      "epoch": 0.8486208261845057,
      "grad_norm": 0.17047731578350067,
      "learning_rate": 1.3663583069908535e-07,
      "loss": 0.1708,
      "num_input_tokens_seen": 3161152,
      "step": 6430
    },
    {
      "epoch": 0.8492807179622542,
      "grad_norm": 0.020379789173603058,
      "learning_rate": 1.3547578995923447e-07,
      "loss": 0.0101,
      "num_input_tokens_seen": 3163776,
      "step": 6435
    },
    {
      "epoch": 0.8499406097400026,
      "grad_norm": 41.39136505126953,
      "learning_rate": 1.3432033671160458e-07,
      "loss": 0.171,
      "num_input_tokens_seen": 3166272,
      "step": 6440
    },
    {
      "epoch": 0.8503365448066517,
      "eval_loss": 0.10069070011377335,
      "eval_runtime": 7.9053,
      "eval_samples_per_second": 851.955,
      "eval_steps_per_second": 106.51,
      "num_input_tokens_seen": 3167488,
      "step": 6443
    },
    {
      "epoch": 0.8506005015177511,
      "grad_norm": 0.44015660881996155,
      "learning_rate": 1.3316947708746762e-07,
      "loss": 0.0712,
      "num_input_tokens_seen": 3168640,
      "step": 6445
    },
    {
      "epoch": 0.8512603932954995,
      "grad_norm": 0.051551755517721176,
      "learning_rate": 1.3202321719371967e-07,
      "loss": 0.0922,
      "num_input_tokens_seen": 3171008,
      "step": 6450
    },
    {
      "epoch": 0.851920285073248,
      "grad_norm": 0.023698126897215843,
      "learning_rate": 1.3088156311284893e-07,
      "loss": 0.0878,
      "num_input_tokens_seen": 3173312,
      "step": 6455
    },
    {
      "epoch": 0.8525801768509964,
      "grad_norm": 0.08452281355857849,
      "learning_rate": 1.2974452090290322e-07,
      "loss": 0.2677,
      "num_input_tokens_seen": 3175808,
      "step": 6460
    },
    {
      "epoch": 0.8532400686287449,
      "grad_norm": 0.07100846618413925,
      "learning_rate": 1.2861209659745865e-07,
      "loss": 0.0748,
      "num_input_tokens_seen": 3178048,
      "step": 6465
    },
    {
      "epoch": 0.8538999604064933,
      "grad_norm": 0.09944210201501846,
      "learning_rate": 1.2748429620558654e-07,
      "loss": 0.1044,
      "num_input_tokens_seen": 3180544,
      "step": 6470
    },
    {
      "epoch": 0.8545598521842418,
      "grad_norm": 13.341742515563965,
      "learning_rate": 1.2636112571182167e-07,
      "loss": 0.1888,
      "num_input_tokens_seen": 3183040,
      "step": 6475
    },
    {
      "epoch": 0.8552197439619902,
      "grad_norm": 0.08588089793920517,
      "learning_rate": 1.2524259107613178e-07,
      "loss": 0.1613,
      "num_input_tokens_seen": 3185664,
      "step": 6480
    },
    {
      "epoch": 0.8558796357397387,
      "grad_norm": 15.565762519836426,
      "learning_rate": 1.2412869823388382e-07,
      "loss": 0.1552,
      "num_input_tokens_seen": 3188672,
      "step": 6485
    },
    {
      "epoch": 0.8565395275174872,
      "grad_norm": 0.09630011022090912,
      "learning_rate": 1.2301945309581486e-07,
      "loss": 0.0598,
      "num_input_tokens_seen": 3191168,
      "step": 6490
    },
    {
      "epoch": 0.8571994192952356,
      "grad_norm": 139.6576690673828,
      "learning_rate": 1.2191486154799846e-07,
      "loss": 0.116,
      "num_input_tokens_seen": 3193664,
      "step": 6495
    },
    {
      "epoch": 0.8578593110729841,
      "grad_norm": 0.07933595776557922,
      "learning_rate": 1.208149294518147e-07,
      "loss": 0.001,
      "num_input_tokens_seen": 3196224,
      "step": 6500
    },
    {
      "epoch": 0.8585192028507325,
      "grad_norm": 0.09007618576288223,
      "learning_rate": 1.1971966264391954e-07,
      "loss": 0.1906,
      "num_input_tokens_seen": 3198784,
      "step": 6505
    },
    {
      "epoch": 0.859179094628481,
      "grad_norm": 0.08821459859609604,
      "learning_rate": 1.1862906693621233e-07,
      "loss": 0.1289,
      "num_input_tokens_seen": 3201472,
      "step": 6510
    },
    {
      "epoch": 0.8598389864062294,
      "grad_norm": 0.09566283971071243,
      "learning_rate": 1.1754314811580623e-07,
      "loss": 0.1718,
      "num_input_tokens_seen": 3203584,
      "step": 6515
    },
    {
      "epoch": 0.8604988781839779,
      "grad_norm": 0.14483189582824707,
      "learning_rate": 1.1646191194499655e-07,
      "loss": 0.0728,
      "num_input_tokens_seen": 3205888,
      "step": 6520
    },
    {
      "epoch": 0.8611587699617262,
      "grad_norm": 5.497840881347656,
      "learning_rate": 1.1538536416123168e-07,
      "loss": 0.1805,
      "num_input_tokens_seen": 3208000,
      "step": 6525
    },
    {
      "epoch": 0.8618186617394747,
      "grad_norm": 0.27670353651046753,
      "learning_rate": 1.1431351047708072e-07,
      "loss": 0.0119,
      "num_input_tokens_seen": 3210240,
      "step": 6530
    },
    {
      "epoch": 0.8624785535172231,
      "grad_norm": 85.52823638916016,
      "learning_rate": 1.1324635658020432e-07,
      "loss": 0.1218,
      "num_input_tokens_seen": 3212672,
      "step": 6535
    },
    {
      "epoch": 0.8631384452949716,
      "grad_norm": 51.2424430847168,
      "learning_rate": 1.1218390813332479e-07,
      "loss": 0.1695,
      "num_input_tokens_seen": 3215360,
      "step": 6540
    },
    {
      "epoch": 0.86379833707272,
      "grad_norm": 0.06896913796663284,
      "learning_rate": 1.1112617077419472e-07,
      "loss": 0.1279,
      "num_input_tokens_seen": 3218112,
      "step": 6545
    },
    {
      "epoch": 0.8644582288504685,
      "grad_norm": 0.21461394429206848,
      "learning_rate": 1.1007315011556884e-07,
      "loss": 0.0386,
      "num_input_tokens_seen": 3220288,
      "step": 6550
    },
    {
      "epoch": 0.865118120628217,
      "grad_norm": 0.17472003400325775,
      "learning_rate": 1.0902485174517251e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 3222976,
      "step": 6555
    },
    {
      "epoch": 0.8657780124059654,
      "grad_norm": 0.14662866294384003,
      "learning_rate": 1.0798128122567285e-07,
      "loss": 0.0532,
      "num_input_tokens_seen": 3225472,
      "step": 6560
    },
    {
      "epoch": 0.8664379041837139,
      "grad_norm": 115.46258544921875,
      "learning_rate": 1.0694244409464992e-07,
      "loss": 0.1791,
      "num_input_tokens_seen": 3228096,
      "step": 6565
    },
    {
      "epoch": 0.8670977959614623,
      "grad_norm": 0.0434236116707325,
      "learning_rate": 1.0590834586456577e-07,
      "loss": 0.1216,
      "num_input_tokens_seen": 3230720,
      "step": 6570
    },
    {
      "epoch": 0.8677576877392108,
      "grad_norm": 28.532846450805664,
      "learning_rate": 1.0487899202273708e-07,
      "loss": 0.1303,
      "num_input_tokens_seen": 3233088,
      "step": 6575
    },
    {
      "epoch": 0.8684175795169592,
      "grad_norm": 32.79645919799805,
      "learning_rate": 1.0385438803130364e-07,
      "loss": 0.1208,
      "num_input_tokens_seen": 3235712,
      "step": 6580
    },
    {
      "epoch": 0.8690774712947077,
      "grad_norm": 0.07639691233634949,
      "learning_rate": 1.0283453932720199e-07,
      "loss": 0.1107,
      "num_input_tokens_seen": 3238528,
      "step": 6585
    },
    {
      "epoch": 0.8697373630724561,
      "grad_norm": 0.7503368258476257,
      "learning_rate": 1.0181945132213476e-07,
      "loss": 0.0808,
      "num_input_tokens_seen": 3240896,
      "step": 6590
    },
    {
      "epoch": 0.8703972548502046,
      "grad_norm": 0.1267632097005844,
      "learning_rate": 1.0080912940254227e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 3243392,
      "step": 6595
    },
    {
      "epoch": 0.871057146627953,
      "grad_norm": 0.19074136018753052,
      "learning_rate": 9.980357892957492e-08,
      "loss": 0.0143,
      "num_input_tokens_seen": 3245824,
      "step": 6600
    },
    {
      "epoch": 0.8717170384057015,
      "grad_norm": 2.3936076164245605,
      "learning_rate": 9.880280523906337e-08,
      "loss": 0.0015,
      "num_input_tokens_seen": 3248128,
      "step": 6605
    },
    {
      "epoch": 0.8723769301834499,
      "grad_norm": 17.094982147216797,
      "learning_rate": 9.780681364149091e-08,
      "loss": 0.1882,
      "num_input_tokens_seen": 3250624,
      "step": 6610
    },
    {
      "epoch": 0.8730368219611984,
      "grad_norm": 0.6837690472602844,
      "learning_rate": 9.681560942196587e-08,
      "loss": 0.108,
      "num_input_tokens_seen": 3253312,
      "step": 6615
    },
    {
      "epoch": 0.8736967137389469,
      "grad_norm": 0.05720958858728409,
      "learning_rate": 9.582919784019194e-08,
      "loss": 0.1163,
      "num_input_tokens_seen": 3255488,
      "step": 6620
    },
    {
      "epoch": 0.8743566055166953,
      "grad_norm": 0.04769499972462654,
      "learning_rate": 9.484758413044236e-08,
      "loss": 0.0734,
      "num_input_tokens_seen": 3257664,
      "step": 6625
    },
    {
      "epoch": 0.8750164972944438,
      "grad_norm": 0.1479378342628479,
      "learning_rate": 9.387077350153017e-08,
      "loss": 0.099,
      "num_input_tokens_seen": 3260160,
      "step": 6630
    },
    {
      "epoch": 0.8756763890721921,
      "grad_norm": 18.90147590637207,
      "learning_rate": 9.289877113678168e-08,
      "loss": 0.0644,
      "num_input_tokens_seen": 3262528,
      "step": 6635
    },
    {
      "epoch": 0.8763362808499406,
      "grad_norm": 0.020846739411354065,
      "learning_rate": 9.19315821940092e-08,
      "loss": 0.0668,
      "num_input_tokens_seen": 3265024,
      "step": 6640
    },
    {
      "epoch": 0.876996172627689,
      "grad_norm": 0.25527259707450867,
      "learning_rate": 9.096921180548234e-08,
      "loss": 0.157,
      "num_input_tokens_seen": 3267456,
      "step": 6645
    },
    {
      "epoch": 0.8776560644054375,
      "grad_norm": 0.35352909564971924,
      "learning_rate": 9.001166507790259e-08,
      "loss": 0.1047,
      "num_input_tokens_seen": 3270208,
      "step": 6650
    },
    {
      "epoch": 0.8783159561831859,
      "grad_norm": 21.19246482849121,
      "learning_rate": 8.905894709237427e-08,
      "loss": 0.1139,
      "num_input_tokens_seen": 3272960,
      "step": 6655
    },
    {
      "epoch": 0.8789758479609344,
      "grad_norm": 0.26658570766448975,
      "learning_rate": 8.811106290437975e-08,
      "loss": 0.083,
      "num_input_tokens_seen": 3275136,
      "step": 6660
    },
    {
      "epoch": 0.8796357397386828,
      "grad_norm": 0.2020629346370697,
      "learning_rate": 8.716801754375036e-08,
      "loss": 0.0826,
      "num_input_tokens_seen": 3277696,
      "step": 6665
    },
    {
      "epoch": 0.8802956315164313,
      "grad_norm": 14.342360496520996,
      "learning_rate": 8.62298160146413e-08,
      "loss": 0.1297,
      "num_input_tokens_seen": 3280064,
      "step": 6670
    },
    {
      "epoch": 0.8809555232941798,
      "grad_norm": 0.19681094586849213,
      "learning_rate": 8.529646329550466e-08,
      "loss": 0.0166,
      "num_input_tokens_seen": 3282304,
      "step": 6675
    },
    {
      "epoch": 0.8816154150719282,
      "grad_norm": 0.16262973845005035,
      "learning_rate": 8.436796433906235e-08,
      "loss": 0.03,
      "num_input_tokens_seen": 3284736,
      "step": 6680
    },
    {
      "epoch": 0.8822753068496767,
      "grad_norm": 40.94983673095703,
      "learning_rate": 8.344432407228141e-08,
      "loss": 0.0714,
      "num_input_tokens_seen": 3287168,
      "step": 6685
    },
    {
      "epoch": 0.8829351986274251,
      "grad_norm": 69.14717102050781,
      "learning_rate": 8.252554739634577e-08,
      "loss": 0.0623,
      "num_input_tokens_seen": 3289600,
      "step": 6690
    },
    {
      "epoch": 0.8835950904051736,
      "grad_norm": 0.04172307625412941,
      "learning_rate": 8.16116391866316e-08,
      "loss": 0.0589,
      "num_input_tokens_seen": 3292160,
      "step": 6695
    },
    {
      "epoch": 0.884254982182922,
      "grad_norm": 14.343789100646973,
      "learning_rate": 8.070260429268172e-08,
      "loss": 0.1393,
      "num_input_tokens_seen": 3294592,
      "step": 6700
    },
    {
      "epoch": 0.8849148739606705,
      "grad_norm": 21.479198455810547,
      "learning_rate": 7.979844753817855e-08,
      "loss": 0.0539,
      "num_input_tokens_seen": 3296960,
      "step": 6705
    },
    {
      "epoch": 0.8855747657384189,
      "grad_norm": 22.292438507080078,
      "learning_rate": 7.889917372091982e-08,
      "loss": 0.0709,
      "num_input_tokens_seen": 3299200,
      "step": 6710
    },
    {
      "epoch": 0.8862346575161674,
      "grad_norm": 13.006369590759277,
      "learning_rate": 7.800478761279183e-08,
      "loss": 0.2047,
      "num_input_tokens_seen": 3301568,
      "step": 6715
    },
    {
      "epoch": 0.8868945492939158,
      "grad_norm": 14.817943572998047,
      "learning_rate": 7.711529395974592e-08,
      "loss": 0.1478,
      "num_input_tokens_seen": 3304064,
      "step": 6720
    },
    {
      "epoch": 0.8875544410716643,
      "grad_norm": 0.05923435837030411,
      "learning_rate": 7.623069748177135e-08,
      "loss": 0.1855,
      "num_input_tokens_seen": 3306432,
      "step": 6725
    },
    {
      "epoch": 0.8882143328494126,
      "grad_norm": 0.2706301510334015,
      "learning_rate": 7.535100287287111e-08,
      "loss": 0.0866,
      "num_input_tokens_seen": 3308736,
      "step": 6730
    },
    {
      "epoch": 0.8888742246271611,
      "grad_norm": 0.10623639822006226,
      "learning_rate": 7.447621480103783e-08,
      "loss": 0.0024,
      "num_input_tokens_seen": 3311168,
      "step": 6735
    },
    {
      "epoch": 0.8895341164049096,
      "grad_norm": 31.206645965576172,
      "learning_rate": 7.360633790822713e-08,
      "loss": 0.2758,
      "num_input_tokens_seen": 3313664,
      "step": 6740
    },
    {
      "epoch": 0.890194008182658,
      "grad_norm": 0.22586752474308014,
      "learning_rate": 7.274137681033498e-08,
      "loss": 0.044,
      "num_input_tokens_seen": 3316224,
      "step": 6745
    },
    {
      "epoch": 0.8908538999604065,
      "grad_norm": 0.25102493166923523,
      "learning_rate": 7.188133609717184e-08,
      "loss": 0.0791,
      "num_input_tokens_seen": 3318464,
      "step": 6750
    },
    {
      "epoch": 0.8915137917381549,
      "grad_norm": 0.05701738968491554,
      "learning_rate": 7.102622033243843e-08,
      "loss": 0.0008,
      "num_input_tokens_seen": 3320896,
      "step": 6755
    },
    {
      "epoch": 0.8921736835159034,
      "grad_norm": 0.2995738983154297,
      "learning_rate": 7.017603405370276e-08,
      "loss": 0.1241,
      "num_input_tokens_seen": 3323648,
      "step": 6760
    },
    {
      "epoch": 0.8928335752936518,
      "grad_norm": 0.19232375919818878,
      "learning_rate": 6.933078177237429e-08,
      "loss": 0.1568,
      "num_input_tokens_seen": 3326208,
      "step": 6765
    },
    {
      "epoch": 0.8934934670714003,
      "grad_norm": 0.24058905243873596,
      "learning_rate": 6.849046797368108e-08,
      "loss": 0.0287,
      "num_input_tokens_seen": 3328576,
      "step": 6770
    },
    {
      "epoch": 0.8941533588491487,
      "grad_norm": 4.3670430183410645,
      "learning_rate": 6.765509711664574e-08,
      "loss": 0.0018,
      "num_input_tokens_seen": 3331520,
      "step": 6775
    },
    {
      "epoch": 0.8948132506268972,
      "grad_norm": 0.28316637873649597,
      "learning_rate": 6.682467363406174e-08,
      "loss": 0.0488,
      "num_input_tokens_seen": 3334336,
      "step": 6780
    },
    {
      "epoch": 0.8954731424046456,
      "grad_norm": 0.2006041258573532,
      "learning_rate": 6.59992019324701e-08,
      "loss": 0.0703,
      "num_input_tokens_seen": 3336896,
      "step": 6785
    },
    {
      "epoch": 0.8961330341823941,
      "grad_norm": 12.302181243896484,
      "learning_rate": 6.517868639213553e-08,
      "loss": 0.0708,
      "num_input_tokens_seen": 3339328,
      "step": 6790
    },
    {
      "epoch": 0.8967929259601425,
      "grad_norm": 0.21891701221466064,
      "learning_rate": 6.436313136702387e-08,
      "loss": 0.0182,
      "num_input_tokens_seen": 3341760,
      "step": 6795
    },
    {
      "epoch": 0.897452817737891,
      "grad_norm": 18.251375198364258,
      "learning_rate": 6.355254118477815e-08,
      "loss": 0.0659,
      "num_input_tokens_seen": 3344448,
      "step": 6800
    },
    {
      "epoch": 0.8981127095156395,
      "grad_norm": 0.09570349752902985,
      "learning_rate": 6.274692014669602e-08,
      "loss": 0.001,
      "num_input_tokens_seen": 3347008,
      "step": 6805
    },
    {
      "epoch": 0.8987726012933879,
      "grad_norm": 0.09717115759849548,
      "learning_rate": 6.194627252770768e-08,
      "loss": 0.0006,
      "num_input_tokens_seen": 3349824,
      "step": 6810
    },
    {
      "epoch": 0.8994324930711364,
      "grad_norm": 0.06764009594917297,
      "learning_rate": 6.115060257635174e-08,
      "loss": 0.0903,
      "num_input_tokens_seen": 3352320,
      "step": 6815
    },
    {
      "epoch": 0.9000923848488848,
      "grad_norm": 0.19751790165901184,
      "learning_rate": 6.035991451475375e-08,
      "loss": 0.0013,
      "num_input_tokens_seen": 3354688,
      "step": 6820
    },
    {
      "epoch": 0.9003563415599841,
      "eval_loss": 0.09756959229707718,
      "eval_runtime": 8.0458,
      "eval_samples_per_second": 837.078,
      "eval_steps_per_second": 104.65,
      "num_input_tokens_seen": 3355520,
      "step": 6822
    },
    {
      "epoch": 0.9007522766266333,
      "grad_norm": 0.013898173347115517,
      "learning_rate": 5.9574212538603505e-08,
      "loss": 0.0663,
      "num_input_tokens_seen": 3357056,
      "step": 6825
    },
    {
      "epoch": 0.9014121684043817,
      "grad_norm": 26.278072357177734,
      "learning_rate": 5.879350081713252e-08,
      "loss": 0.066,
      "num_input_tokens_seen": 3359488,
      "step": 6830
    },
    {
      "epoch": 0.9020720601821302,
      "grad_norm": 0.15908479690551758,
      "learning_rate": 5.8017783493092386e-08,
      "loss": 0.2437,
      "num_input_tokens_seen": 3361920,
      "step": 6835
    },
    {
      "epoch": 0.9027319519598785,
      "grad_norm": 0.18942204117774963,
      "learning_rate": 5.7247064682732104e-08,
      "loss": 0.0012,
      "num_input_tokens_seen": 3364416,
      "step": 6840
    },
    {
      "epoch": 0.903391843737627,
      "grad_norm": 0.23464049398899078,
      "learning_rate": 5.6481348475777566e-08,
      "loss": 0.0617,
      "num_input_tokens_seen": 3366912,
      "step": 6845
    },
    {
      "epoch": 0.9040517355153754,
      "grad_norm": 0.01882772147655487,
      "learning_rate": 5.5720638935407796e-08,
      "loss": 0.0012,
      "num_input_tokens_seen": 3369088,
      "step": 6850
    },
    {
      "epoch": 0.9047116272931239,
      "grad_norm": 0.01236939337104559,
      "learning_rate": 5.49649400982356e-08,
      "loss": 0.1105,
      "num_input_tokens_seen": 3371520,
      "step": 6855
    },
    {
      "epoch": 0.9053715190708723,
      "grad_norm": 0.031979262828826904,
      "learning_rate": 5.421425597428442e-08,
      "loss": 0.0005,
      "num_input_tokens_seen": 3374080,
      "step": 6860
    },
    {
      "epoch": 0.9060314108486208,
      "grad_norm": 104.13761138916016,
      "learning_rate": 5.346859054696784e-08,
      "loss": 0.1294,
      "num_input_tokens_seen": 3376640,
      "step": 6865
    },
    {
      "epoch": 0.9066913026263693,
      "grad_norm": 0.019581180065870285,
      "learning_rate": 5.2727947773068773e-08,
      "loss": 0.0709,
      "num_input_tokens_seen": 3379072,
      "step": 6870
    },
    {
      "epoch": 0.9073511944041177,
      "grad_norm": 0.8969870805740356,
      "learning_rate": 5.199233158271732e-08,
      "loss": 0.069,
      "num_input_tokens_seen": 3381696,
      "step": 6875
    },
    {
      "epoch": 0.9080110861818662,
      "grad_norm": 15.742879867553711,
      "learning_rate": 5.126174587937149e-08,
      "loss": 0.2041,
      "num_input_tokens_seen": 3384064,
      "step": 6880
    },
    {
      "epoch": 0.9086709779596146,
      "grad_norm": 0.0772380605340004,
      "learning_rate": 5.053619453979485e-08,
      "loss": 0.0024,
      "num_input_tokens_seen": 3386304,
      "step": 6885
    },
    {
      "epoch": 0.9093308697373631,
      "grad_norm": 0.030766058713197708,
      "learning_rate": 4.9815681414037025e-08,
      "loss": 0.1257,
      "num_input_tokens_seen": 3388800,
      "step": 6890
    },
    {
      "epoch": 0.9099907615151115,
      "grad_norm": 0.5838080048561096,
      "learning_rate": 4.910021032541334e-08,
      "loss": 0.097,
      "num_input_tokens_seen": 3391232,
      "step": 6895
    },
    {
      "epoch": 0.91065065329286,
      "grad_norm": 25.251256942749023,
      "learning_rate": 4.838978507048319e-08,
      "loss": 0.1039,
      "num_input_tokens_seen": 3393664,
      "step": 6900
    },
    {
      "epoch": 0.9113105450706084,
      "grad_norm": 0.1599751114845276,
      "learning_rate": 4.768440941903207e-08,
      "loss": 0.0018,
      "num_input_tokens_seen": 3395968,
      "step": 6905
    },
    {
      "epoch": 0.9119704368483569,
      "grad_norm": 0.09924237430095673,
      "learning_rate": 4.698408711404944e-08,
      "loss": 0.0407,
      "num_input_tokens_seen": 3398272,
      "step": 6910
    },
    {
      "epoch": 0.9126303286261053,
      "grad_norm": 22.53904914855957,
      "learning_rate": 4.628882187171046e-08,
      "loss": 0.0894,
      "num_input_tokens_seen": 3400960,
      "step": 6915
    },
    {
      "epoch": 0.9132902204038538,
      "grad_norm": 23.106115341186523,
      "learning_rate": 4.559861738135506e-08,
      "loss": 0.1429,
      "num_input_tokens_seen": 3403520,
      "step": 6920
    },
    {
      "epoch": 0.9139501121816023,
      "grad_norm": 0.8851416110992432,
      "learning_rate": 4.491347730546913e-08,
      "loss": 0.2526,
      "num_input_tokens_seen": 3405952,
      "step": 6925
    },
    {
      "epoch": 0.9146100039593507,
      "grad_norm": 18.905851364135742,
      "learning_rate": 4.423340527966512e-08,
      "loss": 0.1429,
      "num_input_tokens_seen": 3408320,
      "step": 6930
    },
    {
      "epoch": 0.9152698957370992,
      "grad_norm": 0.1432974636554718,
      "learning_rate": 4.355840491266205e-08,
      "loss": 0.0675,
      "num_input_tokens_seen": 3410880,
      "step": 6935
    },
    {
      "epoch": 0.9159297875148475,
      "grad_norm": 0.04884126037359238,
      "learning_rate": 4.288847978626686e-08,
      "loss": 0.0589,
      "num_input_tokens_seen": 3413440,
      "step": 6940
    },
    {
      "epoch": 0.916589679292596,
      "grad_norm": 490.322265625,
      "learning_rate": 4.222363345535585e-08,
      "loss": 0.0656,
      "num_input_tokens_seen": 3416000,
      "step": 6945
    },
    {
      "epoch": 0.9172495710703444,
      "grad_norm": 17.165851593017578,
      "learning_rate": 4.1563869447854505e-08,
      "loss": 0.1151,
      "num_input_tokens_seen": 3418240,
      "step": 6950
    },
    {
      "epoch": 0.9179094628480929,
      "grad_norm": 0.17861710488796234,
      "learning_rate": 4.090919126472048e-08,
      "loss": 0.1535,
      "num_input_tokens_seen": 3420672,
      "step": 6955
    },
    {
      "epoch": 0.9185693546258413,
      "grad_norm": 0.6773964166641235,
      "learning_rate": 4.025960237992332e-08,
      "loss": 0.0099,
      "num_input_tokens_seen": 3422912,
      "step": 6960
    },
    {
      "epoch": 0.9192292464035898,
      "grad_norm": 79.64431762695312,
      "learning_rate": 3.961510624042741e-08,
      "loss": 0.0129,
      "num_input_tokens_seen": 3425408,
      "step": 6965
    },
    {
      "epoch": 0.9198891381813382,
      "grad_norm": 22.402997970581055,
      "learning_rate": 3.8975706266172636e-08,
      "loss": 0.1038,
      "num_input_tokens_seen": 3427776,
      "step": 6970
    },
    {
      "epoch": 0.9205490299590867,
      "grad_norm": 17.22422218322754,
      "learning_rate": 3.834140585005696e-08,
      "loss": 0.1456,
      "num_input_tokens_seen": 3430336,
      "step": 6975
    },
    {
      "epoch": 0.9212089217368351,
      "grad_norm": 35.29188919067383,
      "learning_rate": 3.771220835791844e-08,
      "loss": 0.192,
      "num_input_tokens_seen": 3432896,
      "step": 6980
    },
    {
      "epoch": 0.9218688135145836,
      "grad_norm": 0.24460290372371674,
      "learning_rate": 3.708811712851634e-08,
      "loss": 0.023,
      "num_input_tokens_seen": 3435136,
      "step": 6985
    },
    {
      "epoch": 0.9225287052923321,
      "grad_norm": 0.08124759048223495,
      "learning_rate": 3.6469135473514936e-08,
      "loss": 0.0816,
      "num_input_tokens_seen": 3437824,
      "step": 6990
    },
    {
      "epoch": 0.9231885970700805,
      "grad_norm": 0.0361192487180233,
      "learning_rate": 3.5855266677464744e-08,
      "loss": 0.0381,
      "num_input_tokens_seen": 3440320,
      "step": 6995
    },
    {
      "epoch": 0.923848488847829,
      "grad_norm": 0.23591069877147675,
      "learning_rate": 3.524651399778555e-08,
      "loss": 0.0565,
      "num_input_tokens_seen": 3442880,
      "step": 7000
    },
    {
      "epoch": 0.9245083806255774,
      "grad_norm": 0.06923260539770126,
      "learning_rate": 3.4642880664749296e-08,
      "loss": 0.0016,
      "num_input_tokens_seen": 3445120,
      "step": 7005
    },
    {
      "epoch": 0.9251682724033259,
      "grad_norm": 0.1492045372724533,
      "learning_rate": 3.404436988146242e-08,
      "loss": 0.1966,
      "num_input_tokens_seen": 3447424,
      "step": 7010
    },
    {
      "epoch": 0.9258281641810743,
      "grad_norm": 0.01087052933871746,
      "learning_rate": 3.345098482384956e-08,
      "loss": 0.0537,
      "num_input_tokens_seen": 3449920,
      "step": 7015
    },
    {
      "epoch": 0.9264880559588228,
      "grad_norm": 0.2497768998146057,
      "learning_rate": 3.2862728640636105e-08,
      "loss": 0.0494,
      "num_input_tokens_seen": 3452416,
      "step": 7020
    },
    {
      "epoch": 0.9271479477365712,
      "grad_norm": 18.70515251159668,
      "learning_rate": 3.227960445333155e-08,
      "loss": 0.1016,
      "num_input_tokens_seen": 3454912,
      "step": 7025
    },
    {
      "epoch": 0.9278078395143197,
      "grad_norm": 0.039576705545186996,
      "learning_rate": 3.1701615356213295e-08,
      "loss": 0.0633,
      "num_input_tokens_seen": 3457472,
      "step": 7030
    },
    {
      "epoch": 0.928467731292068,
      "grad_norm": 0.12152862548828125,
      "learning_rate": 3.112876441630985e-08,
      "loss": 0.0683,
      "num_input_tokens_seen": 3459712,
      "step": 7035
    },
    {
      "epoch": 0.9291276230698166,
      "grad_norm": 0.22948475182056427,
      "learning_rate": 3.05610546733851e-08,
      "loss": 0.0577,
      "num_input_tokens_seen": 3462144,
      "step": 7040
    },
    {
      "epoch": 0.9297875148475649,
      "grad_norm": 18.062097549438477,
      "learning_rate": 2.99984891399212e-08,
      "loss": 0.2195,
      "num_input_tokens_seen": 3464512,
      "step": 7045
    },
    {
      "epoch": 0.9304474066253134,
      "grad_norm": 0.03376483917236328,
      "learning_rate": 2.9441070801103808e-08,
      "loss": 0.0013,
      "num_input_tokens_seen": 3466880,
      "step": 7050
    },
    {
      "epoch": 0.931107298403062,
      "grad_norm": 0.17363989353179932,
      "learning_rate": 2.8888802614805085e-08,
      "loss": 0.1014,
      "num_input_tokens_seen": 3469248,
      "step": 7055
    },
    {
      "epoch": 0.9317671901808103,
      "grad_norm": 75.9137191772461,
      "learning_rate": 2.8341687511568734e-08,
      "loss": 0.2786,
      "num_input_tokens_seen": 3471616,
      "step": 7060
    },
    {
      "epoch": 0.9324270819585588,
      "grad_norm": 0.2475104182958603,
      "learning_rate": 2.7799728394594547e-08,
      "loss": 0.082,
      "num_input_tokens_seen": 3474240,
      "step": 7065
    },
    {
      "epoch": 0.9330869737363072,
      "grad_norm": 0.09209995716810226,
      "learning_rate": 2.7262928139722198e-08,
      "loss": 0.0789,
      "num_input_tokens_seen": 3476800,
      "step": 7070
    },
    {
      "epoch": 0.9337468655140557,
      "grad_norm": 0.04198440536856651,
      "learning_rate": 2.673128959541693e-08,
      "loss": 0.0807,
      "num_input_tokens_seen": 3479488,
      "step": 7075
    },
    {
      "epoch": 0.9344067572918041,
      "grad_norm": 0.019190721213817596,
      "learning_rate": 2.620481558275367e-08,
      "loss": 0.0005,
      "num_input_tokens_seen": 3482176,
      "step": 7080
    },
    {
      "epoch": 0.9350666490695526,
      "grad_norm": 16.029815673828125,
      "learning_rate": 2.5683508895402382e-08,
      "loss": 0.0278,
      "num_input_tokens_seen": 3484800,
      "step": 7085
    },
    {
      "epoch": 0.935726540847301,
      "grad_norm": 14.575031280517578,
      "learning_rate": 2.5167372299613853e-08,
      "loss": 0.1436,
      "num_input_tokens_seen": 3487488,
      "step": 7090
    },
    {
      "epoch": 0.9363864326250495,
      "grad_norm": 14.435125350952148,
      "learning_rate": 2.4656408534203365e-08,
      "loss": 0.2508,
      "num_input_tokens_seen": 3489728,
      "step": 7095
    },
    {
      "epoch": 0.9370463244027979,
      "grad_norm": 0.028600651770830154,
      "learning_rate": 2.4150620310538273e-08,
      "loss": 0.2469,
      "num_input_tokens_seen": 3491904,
      "step": 7100
    },
    {
      "epoch": 0.9377062161805464,
      "grad_norm": 15.751771926879883,
      "learning_rate": 2.3650010312521673e-08,
      "loss": 0.0832,
      "num_input_tokens_seen": 3494592,
      "step": 7105
    },
    {
      "epoch": 0.9383661079582949,
      "grad_norm": 0.0997326597571373,
      "learning_rate": 2.3154581196579648e-08,
      "loss": 0.1798,
      "num_input_tokens_seen": 3497088,
      "step": 7110
    },
    {
      "epoch": 0.9390259997360433,
      "grad_norm": 0.07149897515773773,
      "learning_rate": 2.2664335591646377e-08,
      "loss": 0.0559,
      "num_input_tokens_seen": 3499520,
      "step": 7115
    },
    {
      "epoch": 0.9396858915137918,
      "grad_norm": 0.024605661630630493,
      "learning_rate": 2.2179276099150158e-08,
      "loss": 0.2113,
      "num_input_tokens_seen": 3502208,
      "step": 7120
    },
    {
      "epoch": 0.9403457832915402,
      "grad_norm": 81.53313446044922,
      "learning_rate": 2.1699405293000182e-08,
      "loss": 0.1584,
      "num_input_tokens_seen": 3504640,
      "step": 7125
    },
    {
      "epoch": 0.9410056750692887,
      "grad_norm": 0.09356902539730072,
      "learning_rate": 2.1224725719572235e-08,
      "loss": 0.0448,
      "num_input_tokens_seen": 3506944,
      "step": 7130
    },
    {
      "epoch": 0.9416655668470371,
      "grad_norm": 0.28894391655921936,
      "learning_rate": 2.0755239897695453e-08,
      "loss": 0.0454,
      "num_input_tokens_seen": 3509376,
      "step": 7135
    },
    {
      "epoch": 0.9423254586247856,
      "grad_norm": 0.5616478323936462,
      "learning_rate": 2.0290950318639256e-08,
      "loss": 0.1705,
      "num_input_tokens_seen": 3511680,
      "step": 7140
    },
    {
      "epoch": 0.942985350402534,
      "grad_norm": 2.1705074310302734,
      "learning_rate": 1.983185944609944e-08,
      "loss": 0.0576,
      "num_input_tokens_seen": 3514112,
      "step": 7145
    },
    {
      "epoch": 0.9436452421802825,
      "grad_norm": 0.15767821669578552,
      "learning_rate": 1.9377969716185994e-08,
      "loss": 0.0709,
      "num_input_tokens_seen": 3516480,
      "step": 7150
    },
    {
      "epoch": 0.9443051339580308,
      "grad_norm": 0.03708624094724655,
      "learning_rate": 1.8929283537408968e-08,
      "loss": 0.1222,
      "num_input_tokens_seen": 3518720,
      "step": 7155
    },
    {
      "epoch": 0.9449650257357793,
      "grad_norm": 0.16703498363494873,
      "learning_rate": 1.848580329066718e-08,
      "loss": 0.0012,
      "num_input_tokens_seen": 3521216,
      "step": 7160
    },
    {
      "epoch": 0.9456249175135277,
      "grad_norm": 17.98069953918457,
      "learning_rate": 1.804753132923431e-08,
      "loss": 0.4089,
      "num_input_tokens_seen": 3523776,
      "step": 7165
    },
    {
      "epoch": 0.9462848092912762,
      "grad_norm": 0.2615845799446106,
      "learning_rate": 1.7614469978746827e-08,
      "loss": 0.0013,
      "num_input_tokens_seen": 3526272,
      "step": 7170
    },
    {
      "epoch": 0.9469447010690247,
      "grad_norm": 61.300594329833984,
      "learning_rate": 1.7186621537192304e-08,
      "loss": 0.0095,
      "num_input_tokens_seen": 3528576,
      "step": 7175
    },
    {
      "epoch": 0.9476045928467731,
      "grad_norm": 0.127578467130661,
      "learning_rate": 1.6763988274896003e-08,
      "loss": 0.0009,
      "num_input_tokens_seen": 3531136,
      "step": 7180
    },
    {
      "epoch": 0.9482644846245216,
      "grad_norm": 15.952384948730469,
      "learning_rate": 1.6346572434509876e-08,
      "loss": 0.159,
      "num_input_tokens_seen": 3533696,
      "step": 7185
    },
    {
      "epoch": 0.94892437640227,
      "grad_norm": 1.98032808303833,
      "learning_rate": 1.5934376231000248e-08,
      "loss": 0.1343,
      "num_input_tokens_seen": 3536064,
      "step": 7190
    },
    {
      "epoch": 0.9495842681800185,
      "grad_norm": 0.20912101864814758,
      "learning_rate": 1.55274018516357e-08,
      "loss": 0.0016,
      "num_input_tokens_seen": 3538432,
      "step": 7195
    },
    {
      "epoch": 0.9502441599577669,
      "grad_norm": 0.0482788160443306,
      "learning_rate": 1.512565145597633e-08,
      "loss": 0.0683,
      "num_input_tokens_seen": 3541120,
      "step": 7200
    },
    {
      "epoch": 0.9503761383133166,
      "eval_loss": 0.09811879694461823,
      "eval_runtime": 7.9135,
      "eval_samples_per_second": 851.075,
      "eval_steps_per_second": 106.4,
      "num_input_tokens_seen": 3541632,
      "step": 7201
    },
    {
      "epoch": 0.9509040517355154,
      "grad_norm": 32.42230987548828,
      "learning_rate": 1.47291271758615e-08,
      "loss": 0.0802,
      "num_input_tokens_seen": 3543680,
      "step": 7205
    },
    {
      "epoch": 0.9515639435132638,
      "grad_norm": 22.951187133789062,
      "learning_rate": 1.4337831115398991e-08,
      "loss": 0.1379,
      "num_input_tokens_seen": 3545984,
      "step": 7210
    },
    {
      "epoch": 0.9522238352910123,
      "grad_norm": 21.046493530273438,
      "learning_rate": 1.3951765350953548e-08,
      "loss": 0.1338,
      "num_input_tokens_seen": 3548544,
      "step": 7215
    },
    {
      "epoch": 0.9528837270687607,
      "grad_norm": 0.031218115240335464,
      "learning_rate": 1.3570931931136009e-08,
      "loss": 0.0956,
      "num_input_tokens_seen": 3551040,
      "step": 7220
    },
    {
      "epoch": 0.9535436188465092,
      "grad_norm": 0.17024770379066467,
      "learning_rate": 1.3195332876792532e-08,
      "loss": 0.0867,
      "num_input_tokens_seen": 3553536,
      "step": 7225
    },
    {
      "epoch": 0.9542035106242576,
      "grad_norm": 0.06077006831765175,
      "learning_rate": 1.2824970180993488e-08,
      "loss": 0.1628,
      "num_input_tokens_seen": 3555712,
      "step": 7230
    },
    {
      "epoch": 0.9548634024020061,
      "grad_norm": 0.11309027671813965,
      "learning_rate": 1.2459845809023484e-08,
      "loss": 0.1277,
      "num_input_tokens_seen": 3558080,
      "step": 7235
    },
    {
      "epoch": 0.9555232941797546,
      "grad_norm": 27.433135986328125,
      "learning_rate": 1.2099961698370353e-08,
      "loss": 0.3486,
      "num_input_tokens_seen": 3560640,
      "step": 7240
    },
    {
      "epoch": 0.956183185957503,
      "grad_norm": 14.181526184082031,
      "learning_rate": 1.1745319758715288e-08,
      "loss": 0.0935,
      "num_input_tokens_seen": 3563392,
      "step": 7245
    },
    {
      "epoch": 0.9568430777352515,
      "grad_norm": 17.118762969970703,
      "learning_rate": 1.1395921871922509e-08,
      "loss": 0.1487,
      "num_input_tokens_seen": 3565824,
      "step": 7250
    },
    {
      "epoch": 0.9575029695129998,
      "grad_norm": 0.10188383609056473,
      "learning_rate": 1.105176989202905e-08,
      "loss": 0.0008,
      "num_input_tokens_seen": 3568256,
      "step": 7255
    },
    {
      "epoch": 0.9581628612907483,
      "grad_norm": 52.97846603393555,
      "learning_rate": 1.0712865645235659e-08,
      "loss": 0.0075,
      "num_input_tokens_seen": 3570752,
      "step": 7260
    },
    {
      "epoch": 0.9588227530684967,
      "grad_norm": 0.1891770213842392,
      "learning_rate": 1.0379210929896131e-08,
      "loss": 0.0681,
      "num_input_tokens_seen": 3572928,
      "step": 7265
    },
    {
      "epoch": 0.9594826448462452,
      "grad_norm": 63.22892761230469,
      "learning_rate": 1.0050807516508553e-08,
      "loss": 0.1822,
      "num_input_tokens_seen": 3575296,
      "step": 7270
    },
    {
      "epoch": 0.9601425366239936,
      "grad_norm": 0.0791056752204895,
      "learning_rate": 9.727657147705737e-09,
      "loss": 0.0006,
      "num_input_tokens_seen": 3577664,
      "step": 7275
    },
    {
      "epoch": 0.9608024284017421,
      "grad_norm": 0.34656962752342224,
      "learning_rate": 9.409761538245575e-09,
      "loss": 0.2031,
      "num_input_tokens_seen": 3580160,
      "step": 7280
    },
    {
      "epoch": 0.9614623201794905,
      "grad_norm": 0.3039737045764923,
      "learning_rate": 9.097122375002264e-09,
      "loss": 0.1012,
      "num_input_tokens_seen": 3582464,
      "step": 7285
    },
    {
      "epoch": 0.962122211957239,
      "grad_norm": 0.30660831928253174,
      "learning_rate": 8.789741316957312e-09,
      "loss": 0.1369,
      "num_input_tokens_seen": 3584896,
      "step": 7290
    },
    {
      "epoch": 0.9627821037349875,
      "grad_norm": 0.5878282189369202,
      "learning_rate": 8.487619995190986e-09,
      "loss": 0.0152,
      "num_input_tokens_seen": 3587584,
      "step": 7295
    },
    {
      "epoch": 0.9634419955127359,
      "grad_norm": 18.56205177307129,
      "learning_rate": 8.19076001287311e-09,
      "loss": 0.1341,
      "num_input_tokens_seen": 3590144,
      "step": 7300
    },
    {
      "epoch": 0.9641018872904844,
      "grad_norm": 0.07710211724042892,
      "learning_rate": 7.899162945254945e-09,
      "loss": 0.001,
      "num_input_tokens_seen": 3592832,
      "step": 7305
    },
    {
      "epoch": 0.9647617790682328,
      "grad_norm": 23.307418823242188,
      "learning_rate": 7.612830339660758e-09,
      "loss": 0.0999,
      "num_input_tokens_seen": 3595456,
      "step": 7310
    },
    {
      "epoch": 0.9654216708459813,
      "grad_norm": 12.40549087524414,
      "learning_rate": 7.3317637154796105e-09,
      "loss": 0.1011,
      "num_input_tokens_seen": 3597888,
      "step": 7315
    },
    {
      "epoch": 0.9660815626237297,
      "grad_norm": 18.231792449951172,
      "learning_rate": 7.0559645641572465e-09,
      "loss": 0.0699,
      "num_input_tokens_seen": 3600384,
      "step": 7320
    },
    {
      "epoch": 0.9667414544014782,
      "grad_norm": 93.2337646484375,
      "learning_rate": 6.785434349188102e-09,
      "loss": 0.1521,
      "num_input_tokens_seen": 3602880,
      "step": 7325
    },
    {
      "epoch": 0.9674013461792266,
      "grad_norm": 0.19403959810733795,
      "learning_rate": 6.520174506107867e-09,
      "loss": 0.045,
      "num_input_tokens_seen": 3605248,
      "step": 7330
    },
    {
      "epoch": 0.9680612379569751,
      "grad_norm": 0.17541255056858063,
      "learning_rate": 6.260186442485494e-09,
      "loss": 0.001,
      "num_input_tokens_seen": 3607808,
      "step": 7335
    },
    {
      "epoch": 0.9687211297347235,
      "grad_norm": 1.3419514894485474,
      "learning_rate": 6.005471537915863e-09,
      "loss": 0.1172,
      "num_input_tokens_seen": 3610112,
      "step": 7340
    },
    {
      "epoch": 0.969381021512472,
      "grad_norm": 35.56879806518555,
      "learning_rate": 5.756031144012685e-09,
      "loss": 0.0319,
      "num_input_tokens_seen": 3612352,
      "step": 7345
    },
    {
      "epoch": 0.9700409132902204,
      "grad_norm": 21.5054874420166,
      "learning_rate": 5.511866584400837e-09,
      "loss": 0.1853,
      "num_input_tokens_seen": 3614848,
      "step": 7350
    },
    {
      "epoch": 0.9707008050679689,
      "grad_norm": 0.22669829428195953,
      "learning_rate": 5.2729791547097e-09,
      "loss": 0.0038,
      "num_input_tokens_seen": 3617408,
      "step": 7355
    },
    {
      "epoch": 0.9713606968457174,
      "grad_norm": 26.995397567749023,
      "learning_rate": 5.039370122566389e-09,
      "loss": 0.1072,
      "num_input_tokens_seen": 3619968,
      "step": 7360
    },
    {
      "epoch": 0.9720205886234657,
      "grad_norm": 5.098914623260498,
      "learning_rate": 4.811040727588755e-09,
      "loss": 0.099,
      "num_input_tokens_seen": 3622016,
      "step": 7365
    },
    {
      "epoch": 0.9726804804012142,
      "grad_norm": 18.19532012939453,
      "learning_rate": 4.58799218137873e-09,
      "loss": 0.1695,
      "num_input_tokens_seen": 3624192,
      "step": 7370
    },
    {
      "epoch": 0.9733403721789626,
      "grad_norm": 0.12403739243745804,
      "learning_rate": 4.370225667516325e-09,
      "loss": 0.0006,
      "num_input_tokens_seen": 3626624,
      "step": 7375
    },
    {
      "epoch": 0.9740002639567111,
      "grad_norm": 14.581175804138184,
      "learning_rate": 4.157742341552861e-09,
      "loss": 0.1853,
      "num_input_tokens_seen": 3628928,
      "step": 7380
    },
    {
      "epoch": 0.9746601557344595,
      "grad_norm": 0.035725370049476624,
      "learning_rate": 3.950543331005307e-09,
      "loss": 0.1044,
      "num_input_tokens_seen": 3631552,
      "step": 7385
    },
    {
      "epoch": 0.975320047512208,
      "grad_norm": 13.293424606323242,
      "learning_rate": 3.748629735349839e-09,
      "loss": 0.0016,
      "num_input_tokens_seen": 3633984,
      "step": 7390
    },
    {
      "epoch": 0.9759799392899564,
      "grad_norm": 0.3222048878669739,
      "learning_rate": 3.552002626016293e-09,
      "loss": 0.2102,
      "num_input_tokens_seen": 3636224,
      "step": 7395
    },
    {
      "epoch": 0.9766398310677049,
      "grad_norm": 0.12731555104255676,
      "learning_rate": 3.3606630463824947e-09,
      "loss": 0.2375,
      "num_input_tokens_seen": 3638656,
      "step": 7400
    },
    {
      "epoch": 0.9772997228454533,
      "grad_norm": 0.33530130982398987,
      "learning_rate": 3.174612011768607e-09,
      "loss": 0.0009,
      "num_input_tokens_seen": 3641408,
      "step": 7405
    },
    {
      "epoch": 0.9779596146232018,
      "grad_norm": 15.725085258483887,
      "learning_rate": 2.9938505094316834e-09,
      "loss": 0.1826,
      "num_input_tokens_seen": 3643840,
      "step": 7410
    },
    {
      "epoch": 0.9786195064009502,
      "grad_norm": 0.08356328308582306,
      "learning_rate": 2.8183794985605637e-09,
      "loss": 0.0801,
      "num_input_tokens_seen": 3646336,
      "step": 7415
    },
    {
      "epoch": 0.9792793981786987,
      "grad_norm": 0.0940588191151619,
      "learning_rate": 2.6481999102707654e-09,
      "loss": 0.0731,
      "num_input_tokens_seen": 3648960,
      "step": 7420
    },
    {
      "epoch": 0.9799392899564472,
      "grad_norm": 0.3441082239151001,
      "learning_rate": 2.4833126475994894e-09,
      "loss": 0.0011,
      "num_input_tokens_seen": 3651200,
      "step": 7425
    },
    {
      "epoch": 0.9805991817341956,
      "grad_norm": 0.3757539689540863,
      "learning_rate": 2.3237185855008443e-09,
      "loss": 0.0073,
      "num_input_tokens_seen": 3653504,
      "step": 7430
    },
    {
      "epoch": 0.9812590735119441,
      "grad_norm": 150.4886474609375,
      "learning_rate": 2.1694185708414083e-09,
      "loss": 0.2482,
      "num_input_tokens_seen": 3656064,
      "step": 7435
    },
    {
      "epoch": 0.9819189652896925,
      "grad_norm": 92.5436019897461,
      "learning_rate": 2.0204134223952284e-09,
      "loss": 0.2875,
      "num_input_tokens_seen": 3658112,
      "step": 7440
    },
    {
      "epoch": 0.982578857067441,
      "grad_norm": 19.585914611816406,
      "learning_rate": 1.87670393083994e-09,
      "loss": 0.1509,
      "num_input_tokens_seen": 3660928,
      "step": 7445
    },
    {
      "epoch": 0.9832387488451894,
      "grad_norm": 39.90876388549805,
      "learning_rate": 1.7382908587525447e-09,
      "loss": 0.0779,
      "num_input_tokens_seen": 3663232,
      "step": 7450
    },
    {
      "epoch": 0.9838986406229379,
      "grad_norm": 0.06113344430923462,
      "learning_rate": 1.6051749406049697e-09,
      "loss": 0.0016,
      "num_input_tokens_seen": 3665600,
      "step": 7455
    },
    {
      "epoch": 0.9845585324006862,
      "grad_norm": 0.033529337495565414,
      "learning_rate": 1.4773568827607386e-09,
      "loss": 0.0008,
      "num_input_tokens_seen": 3668096,
      "step": 7460
    },
    {
      "epoch": 0.9852184241784347,
      "grad_norm": 0.1556025892496109,
      "learning_rate": 1.354837363470529e-09,
      "loss": 0.0016,
      "num_input_tokens_seen": 3670656,
      "step": 7465
    },
    {
      "epoch": 0.9858783159561831,
      "grad_norm": 52.730857849121094,
      "learning_rate": 1.23761703286962e-09,
      "loss": 0.0971,
      "num_input_tokens_seen": 3673024,
      "step": 7470
    },
    {
      "epoch": 0.9865382077339316,
      "grad_norm": 0.017705397680401802,
      "learning_rate": 1.1256965129730068e-09,
      "loss": 0.0012,
      "num_input_tokens_seen": 3675712,
      "step": 7475
    },
    {
      "epoch": 0.9871980995116801,
      "grad_norm": 0.13253773748874664,
      "learning_rate": 1.0190763976734018e-09,
      "loss": 0.0803,
      "num_input_tokens_seen": 3678080,
      "step": 7480
    },
    {
      "epoch": 0.9878579912894285,
      "grad_norm": 19.60429573059082,
      "learning_rate": 9.177572527375721e-10,
      "loss": 0.114,
      "num_input_tokens_seen": 3680448,
      "step": 7485
    },
    {
      "epoch": 0.988517883067177,
      "grad_norm": 19.607324600219727,
      "learning_rate": 8.217396158030076e-10,
      "loss": 0.0938,
      "num_input_tokens_seen": 3682752,
      "step": 7490
    },
    {
      "epoch": 0.9891777748449254,
      "grad_norm": 31.56887435913086,
      "learning_rate": 7.310239963755904e-10,
      "loss": 0.108,
      "num_input_tokens_seen": 3685376,
      "step": 7495
    },
    {
      "epoch": 0.9898376666226739,
      "grad_norm": 0.1998118758201599,
      "learning_rate": 6.456108758268186e-10,
      "loss": 0.0012,
      "num_input_tokens_seen": 3687744,
      "step": 7500
    },
    {
      "epoch": 0.9904975584004223,
      "grad_norm": 113.81692504882812,
      "learning_rate": 5.655007073909202e-10,
      "loss": 0.1473,
      "num_input_tokens_seen": 3690240,
      "step": 7505
    },
    {
      "epoch": 0.9911574501781708,
      "grad_norm": 0.039434801787137985,
      "learning_rate": 4.906939161627432e-10,
      "loss": 0.0775,
      "num_input_tokens_seen": 3692736,
      "step": 7510
    },
    {
      "epoch": 0.9918173419559192,
      "grad_norm": 18.382701873779297,
      "learning_rate": 4.2119089909542495e-10,
      "loss": 0.2057,
      "num_input_tokens_seen": 3695360,
      "step": 7515
    },
    {
      "epoch": 0.9924772337336677,
      "grad_norm": 0.3242112398147583,
      "learning_rate": 3.569920249981706e-10,
      "loss": 0.0823,
      "num_input_tokens_seen": 3697856,
      "step": 7520
    },
    {
      "epoch": 0.9931371255114161,
      "grad_norm": 1.300443172454834,
      "learning_rate": 2.980976345344777e-10,
      "loss": 0.0409,
      "num_input_tokens_seen": 3700224,
      "step": 7525
    },
    {
      "epoch": 0.9937970172891646,
      "grad_norm": 0.07868517935276031,
      "learning_rate": 2.445080402202482e-10,
      "loss": 0.0481,
      "num_input_tokens_seen": 3702592,
      "step": 7530
    },
    {
      "epoch": 0.994456909066913,
      "grad_norm": 0.6571361422538757,
      "learning_rate": 1.962235264222345e-10,
      "loss": 0.1701,
      "num_input_tokens_seen": 3704896,
      "step": 7535
    },
    {
      "epoch": 0.9951168008446615,
      "grad_norm": 0.6455036401748657,
      "learning_rate": 1.5324434935615195e-10,
      "loss": 0.0378,
      "num_input_tokens_seen": 3707264,
      "step": 7540
    },
    {
      "epoch": 0.99577669262241,
      "grad_norm": 15.328804016113281,
      "learning_rate": 1.1557073708579057e-10,
      "loss": 0.1023,
      "num_input_tokens_seen": 3709824,
      "step": 7545
    },
    {
      "epoch": 0.9964365844001584,
      "grad_norm": 0.0523066483438015,
      "learning_rate": 8.320288952168297e-11,
      "loss": 0.0658,
      "num_input_tokens_seen": 3712192,
      "step": 7550
    },
    {
      "epoch": 0.9970964761779069,
      "grad_norm": 49.20591354370117,
      "learning_rate": 5.614097841988297e-11,
      "loss": 0.1099,
      "num_input_tokens_seen": 3714880,
      "step": 7555
    },
    {
      "epoch": 0.9977563679556553,
      "grad_norm": 0.39536726474761963,
      "learning_rate": 3.43851473808554e-11,
      "loss": 0.0216,
      "num_input_tokens_seen": 3717184,
      "step": 7560
    },
    {
      "epoch": 0.9984162597334038,
      "grad_norm": 0.10730185359716415,
      "learning_rate": 1.7935511849587192e-11,
      "loss": 0.0521,
      "num_input_tokens_seen": 3719424,
      "step": 7565
    },
    {
      "epoch": 0.9990761515111521,
      "grad_norm": 0.08098907023668289,
      "learning_rate": 6.792159113921947e-12,
      "loss": 0.1418,
      "num_input_tokens_seen": 3721920,
      "step": 7570
    },
    {
      "epoch": 0.9997360432889006,
      "grad_norm": 14.478204727172852,
      "learning_rate": 9.55148304560005e-13,
      "loss": 0.0767,
      "num_input_tokens_seen": 3724288,
      "step": 7575
    },
    {
      "epoch": 1.0,
      "num_input_tokens_seen": 3725120,
      "step": 7577,
      "total_flos": 2.175051626840064e+16,
      "train_loss": 0.12628782090838944,
      "train_runtime": 1516.0655,
      "train_samples_per_second": 39.981,
      "train_steps_per_second": 4.998
    }
  ],
  "logging_steps": 5,
  "max_steps": 7577,
  "num_input_tokens_seen": 3725120,
  "num_train_epochs": 1,
  "save_steps": 379,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.175051626840064e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}