{
  "best_global_step": 4097,
  "best_metric": 0.2517484128475189,
  "best_model_checkpoint": "saves_bts_preliminary/base/llama-3.2-1b-instruct/train_cola_42_1774791067/checkpoint-4097",
  "epoch": 5.0,
  "eval_steps": 241,
  "global_step": 4810,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005197505197505198,
      "grad_norm": 434.86285400390625,
      "learning_rate": 4.1580041580041583e-07,
      "loss": 1.2842,
      "num_input_tokens_seen": 2048,
      "step": 5
    },
    {
      "epoch": 0.010395010395010396,
      "grad_norm": 175.30227661132812,
      "learning_rate": 9.355509355509356e-07,
      "loss": 0.8545,
      "num_input_tokens_seen": 4224,
      "step": 10
    },
    {
      "epoch": 0.015592515592515593,
      "grad_norm": 26.521507263183594,
      "learning_rate": 1.4553014553014554e-06,
      "loss": 0.3278,
      "num_input_tokens_seen": 6272,
      "step": 15
    },
    {
      "epoch": 0.02079002079002079,
      "grad_norm": 86.457275390625,
      "learning_rate": 1.975051975051975e-06,
      "loss": 0.365,
      "num_input_tokens_seen": 8384,
      "step": 20
    },
    {
      "epoch": 0.02598752598752599,
      "grad_norm": 55.42351150512695,
      "learning_rate": 2.494802494802495e-06,
      "loss": 0.4433,
      "num_input_tokens_seen": 10496,
      "step": 25
    },
    {
      "epoch": 0.031185031185031187,
      "grad_norm": 22.866464614868164,
      "learning_rate": 3.014553014553015e-06,
      "loss": 0.2526,
      "num_input_tokens_seen": 12544,
      "step": 30
    },
    {
      "epoch": 0.036382536382536385,
      "grad_norm": 34.69972229003906,
      "learning_rate": 3.5343035343035348e-06,
      "loss": 0.3051,
      "num_input_tokens_seen": 14528,
      "step": 35
    },
    {
      "epoch": 0.04158004158004158,
      "grad_norm": 62.158714294433594,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 0.5129,
      "num_input_tokens_seen": 16576,
      "step": 40
    },
    {
      "epoch": 0.04677754677754678,
      "grad_norm": 24.736021041870117,
      "learning_rate": 4.573804573804574e-06,
      "loss": 0.283,
      "num_input_tokens_seen": 18560,
      "step": 45
    },
    {
      "epoch": 0.05197505197505198,
      "grad_norm": 103.87641906738281,
      "learning_rate": 5.093555093555094e-06,
      "loss": 0.454,
      "num_input_tokens_seen": 20608,
      "step": 50
    },
    {
      "epoch": 0.057172557172557176,
      "grad_norm": 85.98382568359375,
      "learning_rate": 5.613305613305614e-06,
      "loss": 0.2518,
      "num_input_tokens_seen": 22656,
      "step": 55
    },
    {
      "epoch": 0.062370062370062374,
      "grad_norm": 15.937103271484375,
      "learning_rate": 6.1330561330561335e-06,
      "loss": 0.2538,
      "num_input_tokens_seen": 24640,
      "step": 60
    },
    {
      "epoch": 0.06756756756756757,
      "grad_norm": 71.92253112792969,
      "learning_rate": 6.652806652806653e-06,
      "loss": 0.3571,
      "num_input_tokens_seen": 26752,
      "step": 65
    },
    {
      "epoch": 0.07276507276507277,
      "grad_norm": 41.08773422241211,
      "learning_rate": 7.172557172557173e-06,
      "loss": 0.3588,
      "num_input_tokens_seen": 28608,
      "step": 70
    },
    {
      "epoch": 0.07796257796257797,
      "grad_norm": 39.851627349853516,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.2553,
      "num_input_tokens_seen": 30912,
      "step": 75
    },
    {
      "epoch": 0.08316008316008316,
      "grad_norm": 39.69013595581055,
      "learning_rate": 8.212058212058212e-06,
      "loss": 0.3869,
      "num_input_tokens_seen": 32896,
      "step": 80
    },
    {
      "epoch": 0.08835758835758836,
      "grad_norm": 15.486207962036133,
      "learning_rate": 8.731808731808733e-06,
      "loss": 0.3812,
      "num_input_tokens_seen": 34816,
      "step": 85
    },
    {
      "epoch": 0.09355509355509356,
      "grad_norm": 54.213748931884766,
      "learning_rate": 9.251559251559252e-06,
      "loss": 0.735,
      "num_input_tokens_seen": 36736,
      "step": 90
    },
    {
      "epoch": 0.09875259875259876,
      "grad_norm": 33.433189392089844,
      "learning_rate": 9.771309771309773e-06,
      "loss": 0.4027,
      "num_input_tokens_seen": 38720,
      "step": 95
    },
    {
      "epoch": 0.10395010395010396,
      "grad_norm": 45.393463134765625,
      "learning_rate": 1.0291060291060291e-05,
      "loss": 0.4605,
      "num_input_tokens_seen": 40640,
      "step": 100
    },
    {
      "epoch": 0.10914760914760915,
      "grad_norm": 4.37235164642334,
      "learning_rate": 1.0810810810810812e-05,
      "loss": 0.2999,
      "num_input_tokens_seen": 42688,
      "step": 105
    },
    {
      "epoch": 0.11434511434511435,
      "grad_norm": 41.325050354003906,
      "learning_rate": 1.1330561330561331e-05,
      "loss": 0.2839,
      "num_input_tokens_seen": 44544,
      "step": 110
    },
    {
      "epoch": 0.11954261954261955,
      "grad_norm": 19.053668975830078,
      "learning_rate": 1.1850311850311852e-05,
      "loss": 0.3329,
      "num_input_tokens_seen": 46400,
      "step": 115
    },
    {
      "epoch": 0.12474012474012475,
      "grad_norm": 11.979599952697754,
      "learning_rate": 1.2370062370062372e-05,
      "loss": 0.2089,
      "num_input_tokens_seen": 48448,
      "step": 120
    },
    {
      "epoch": 0.12993762993762994,
      "grad_norm": 20.051830291748047,
      "learning_rate": 1.2889812889812891e-05,
      "loss": 0.2829,
      "num_input_tokens_seen": 50496,
      "step": 125
    },
    {
      "epoch": 0.13513513513513514,
      "grad_norm": 41.367645263671875,
      "learning_rate": 1.3409563409563412e-05,
      "loss": 0.2434,
      "num_input_tokens_seen": 52416,
      "step": 130
    },
    {
      "epoch": 0.14033264033264034,
      "grad_norm": 97.8139419555664,
      "learning_rate": 1.392931392931393e-05,
      "loss": 0.3621,
      "num_input_tokens_seen": 54464,
      "step": 135
    },
    {
      "epoch": 0.14553014553014554,
      "grad_norm": 25.77781867980957,
      "learning_rate": 1.4449064449064451e-05,
      "loss": 0.3338,
      "num_input_tokens_seen": 56448,
      "step": 140
    },
    {
      "epoch": 0.15072765072765074,
      "grad_norm": 22.036914825439453,
      "learning_rate": 1.496881496881497e-05,
      "loss": 0.3185,
      "num_input_tokens_seen": 58368,
      "step": 145
    },
    {
      "epoch": 0.15592515592515593,
      "grad_norm": 13.202033996582031,
      "learning_rate": 1.548856548856549e-05,
      "loss": 0.2379,
      "num_input_tokens_seen": 60544,
      "step": 150
    },
    {
      "epoch": 0.16112266112266113,
      "grad_norm": 27.529069900512695,
      "learning_rate": 1.600831600831601e-05,
      "loss": 0.3634,
      "num_input_tokens_seen": 62592,
      "step": 155
    },
    {
      "epoch": 0.16632016632016633,
      "grad_norm": 25.028240203857422,
      "learning_rate": 1.652806652806653e-05,
      "loss": 0.3093,
      "num_input_tokens_seen": 64576,
      "step": 160
    },
    {
      "epoch": 0.17151767151767153,
      "grad_norm": 10.42611026763916,
      "learning_rate": 1.704781704781705e-05,
      "loss": 0.3377,
      "num_input_tokens_seen": 66688,
      "step": 165
    },
    {
      "epoch": 0.17671517671517672,
      "grad_norm": 4.2057390213012695,
      "learning_rate": 1.756756756756757e-05,
      "loss": 0.3224,
      "num_input_tokens_seen": 68544,
      "step": 170
    },
    {
      "epoch": 0.18191268191268192,
      "grad_norm": 36.7962532043457,
      "learning_rate": 1.808731808731809e-05,
      "loss": 0.3814,
      "num_input_tokens_seen": 70592,
      "step": 175
    },
    {
      "epoch": 0.18711018711018712,
      "grad_norm": 9.922293663024902,
      "learning_rate": 1.8607068607068607e-05,
      "loss": 0.338,
      "num_input_tokens_seen": 72576,
      "step": 180
    },
    {
      "epoch": 0.19230769230769232,
      "grad_norm": 2.9066672325134277,
      "learning_rate": 1.9126819126819128e-05,
      "loss": 0.2853,
      "num_input_tokens_seen": 74624,
      "step": 185
    },
    {
      "epoch": 0.19750519750519752,
      "grad_norm": 30.27046775817871,
      "learning_rate": 1.964656964656965e-05,
      "loss": 0.3374,
      "num_input_tokens_seen": 76608,
      "step": 190
    },
    {
      "epoch": 0.20270270270270271,
      "grad_norm": 17.140790939331055,
      "learning_rate": 2.016632016632017e-05,
      "loss": 0.3221,
      "num_input_tokens_seen": 78720,
      "step": 195
    },
    {
      "epoch": 0.2079002079002079,
      "grad_norm": 75.4980239868164,
      "learning_rate": 2.068607068607069e-05,
      "loss": 0.3944,
      "num_input_tokens_seen": 81152,
      "step": 200
    },
    {
      "epoch": 0.2130977130977131,
      "grad_norm": 10.222508430480957,
      "learning_rate": 2.1205821205821207e-05,
      "loss": 0.3755,
      "num_input_tokens_seen": 83200,
      "step": 205
    },
    {
      "epoch": 0.2182952182952183,
      "grad_norm": 9.712939262390137,
      "learning_rate": 2.1725571725571728e-05,
      "loss": 0.2949,
      "num_input_tokens_seen": 85184,
      "step": 210
    },
    {
      "epoch": 0.2234927234927235,
      "grad_norm": 20.923433303833008,
      "learning_rate": 2.2245322245322248e-05,
      "loss": 0.2174,
      "num_input_tokens_seen": 87232,
      "step": 215
    },
    {
      "epoch": 0.2286902286902287,
      "grad_norm": 27.973743438720703,
      "learning_rate": 2.276507276507277e-05,
      "loss": 0.6706,
      "num_input_tokens_seen": 89152,
      "step": 220
    },
    {
      "epoch": 0.2338877338877339,
      "grad_norm": 23.218599319458008,
      "learning_rate": 2.3284823284823286e-05,
      "loss": 0.4387,
      "num_input_tokens_seen": 91328,
      "step": 225
    },
    {
      "epoch": 0.2390852390852391,
      "grad_norm": 3.9333198070526123,
      "learning_rate": 2.3804573804573807e-05,
      "loss": 0.3047,
      "num_input_tokens_seen": 93312,
      "step": 230
    },
    {
      "epoch": 0.2442827442827443,
      "grad_norm": 3.1809186935424805,
      "learning_rate": 2.4324324324324327e-05,
      "loss": 0.2717,
      "num_input_tokens_seen": 95296,
      "step": 235
    },
    {
      "epoch": 0.2494802494802495,
      "grad_norm": 15.368531227111816,
      "learning_rate": 2.4844074844074848e-05,
      "loss": 0.2682,
      "num_input_tokens_seen": 97216,
      "step": 240
    },
    {
      "epoch": 0.2505197505197505,
      "eval_loss": 0.38468512892723083,
      "eval_runtime": 1.6403,
      "eval_samples_per_second": 521.872,
      "eval_steps_per_second": 65.234,
      "num_input_tokens_seen": 97664,
      "step": 241
    },
    {
      "epoch": 0.25467775467775466,
      "grad_norm": 2.458040952682495,
      "learning_rate": 2.5363825363825365e-05,
      "loss": 0.344,
      "num_input_tokens_seen": 99264,
      "step": 245
    },
    {
      "epoch": 0.2598752598752599,
      "grad_norm": 26.132585525512695,
      "learning_rate": 2.5883575883575882e-05,
      "loss": 0.3993,
      "num_input_tokens_seen": 101184,
      "step": 250
    },
    {
      "epoch": 0.26507276507276506,
      "grad_norm": 28.206130981445312,
      "learning_rate": 2.6403326403326406e-05,
      "loss": 0.4414,
      "num_input_tokens_seen": 103296,
      "step": 255
    },
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 6.235988140106201,
      "learning_rate": 2.6923076923076923e-05,
      "loss": 0.3322,
      "num_input_tokens_seen": 105344,
      "step": 260
    },
    {
      "epoch": 0.27546777546777546,
      "grad_norm": 5.716452121734619,
      "learning_rate": 2.7442827442827447e-05,
      "loss": 0.3314,
      "num_input_tokens_seen": 107392,
      "step": 265
    },
    {
      "epoch": 0.2806652806652807,
      "grad_norm": 13.819499969482422,
      "learning_rate": 2.796257796257796e-05,
      "loss": 0.2951,
      "num_input_tokens_seen": 109440,
      "step": 270
    },
    {
      "epoch": 0.28586278586278585,
      "grad_norm": 27.50826644897461,
      "learning_rate": 2.8482328482328485e-05,
      "loss": 0.2986,
      "num_input_tokens_seen": 111424,
      "step": 275
    },
    {
      "epoch": 0.2910602910602911,
      "grad_norm": 24.54796600341797,
      "learning_rate": 2.9002079002079002e-05,
      "loss": 0.407,
      "num_input_tokens_seen": 113408,
      "step": 280
    },
    {
      "epoch": 0.29625779625779625,
      "grad_norm": 1.7183665037155151,
      "learning_rate": 2.9521829521829526e-05,
      "loss": 0.4506,
      "num_input_tokens_seen": 115392,
      "step": 285
    },
    {
      "epoch": 0.30145530145530147,
      "grad_norm": 38.75688552856445,
      "learning_rate": 3.0041580041580043e-05,
      "loss": 0.4738,
      "num_input_tokens_seen": 117440,
      "step": 290
    },
    {
      "epoch": 0.30665280665280664,
      "grad_norm": 20.901853561401367,
      "learning_rate": 3.056133056133057e-05,
      "loss": 0.5248,
      "num_input_tokens_seen": 119424,
      "step": 295
    },
    {
      "epoch": 0.31185031185031187,
      "grad_norm": 37.667694091796875,
      "learning_rate": 3.108108108108108e-05,
      "loss": 0.3321,
      "num_input_tokens_seen": 121344,
      "step": 300
    },
    {
      "epoch": 0.31704781704781704,
      "grad_norm": 21.214815139770508,
      "learning_rate": 3.16008316008316e-05,
      "loss": 0.3707,
      "num_input_tokens_seen": 123264,
      "step": 305
    },
    {
      "epoch": 0.32224532224532226,
      "grad_norm": 3.6351711750030518,
      "learning_rate": 3.212058212058212e-05,
      "loss": 0.3574,
      "num_input_tokens_seen": 125184,
      "step": 310
    },
    {
      "epoch": 0.32744282744282743,
      "grad_norm": 2.756866216659546,
      "learning_rate": 3.264033264033264e-05,
      "loss": 0.2851,
      "num_input_tokens_seen": 127296,
      "step": 315
    },
    {
      "epoch": 0.33264033264033266,
      "grad_norm": 8.542179107666016,
      "learning_rate": 3.3160083160083164e-05,
      "loss": 0.2794,
      "num_input_tokens_seen": 129408,
      "step": 320
    },
    {
      "epoch": 0.33783783783783783,
      "grad_norm": 5.599982738494873,
      "learning_rate": 3.3679833679833684e-05,
      "loss": 0.3015,
      "num_input_tokens_seen": 131520,
      "step": 325
    },
    {
      "epoch": 0.34303534303534305,
      "grad_norm": 0.6099509000778198,
      "learning_rate": 3.41995841995842e-05,
      "loss": 0.2826,
      "num_input_tokens_seen": 133568,
      "step": 330
    },
    {
      "epoch": 0.3482328482328482,
      "grad_norm": 10.96675968170166,
      "learning_rate": 3.4719334719334725e-05,
      "loss": 0.3704,
      "num_input_tokens_seen": 135616,
      "step": 335
    },
    {
      "epoch": 0.35343035343035345,
      "grad_norm": 36.376678466796875,
      "learning_rate": 3.523908523908524e-05,
      "loss": 0.4519,
      "num_input_tokens_seen": 137664,
      "step": 340
    },
    {
      "epoch": 0.3586278586278586,
      "grad_norm": 23.873687744140625,
      "learning_rate": 3.575883575883576e-05,
      "loss": 0.2863,
      "num_input_tokens_seen": 139584,
      "step": 345
    },
    {
      "epoch": 0.36382536382536385,
      "grad_norm": 3.04746675491333,
      "learning_rate": 3.627858627858628e-05,
      "loss": 0.4921,
      "num_input_tokens_seen": 141504,
      "step": 350
    },
    {
      "epoch": 0.369022869022869,
      "grad_norm": 22.159381866455078,
      "learning_rate": 3.67983367983368e-05,
      "loss": 0.3078,
      "num_input_tokens_seen": 143552,
      "step": 355
    },
    {
      "epoch": 0.37422037422037424,
      "grad_norm": 4.239355564117432,
      "learning_rate": 3.731808731808732e-05,
      "loss": 0.252,
      "num_input_tokens_seen": 145536,
      "step": 360
    },
    {
      "epoch": 0.3794178794178794,
      "grad_norm": 33.11611557006836,
      "learning_rate": 3.783783783783784e-05,
      "loss": 0.2894,
      "num_input_tokens_seen": 147456,
      "step": 365
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 3.5446720123291016,
      "learning_rate": 3.8357588357588356e-05,
      "loss": 0.1779,
      "num_input_tokens_seen": 149440,
      "step": 370
    },
    {
      "epoch": 0.3898128898128898,
      "grad_norm": 17.057636260986328,
      "learning_rate": 3.8877338877338883e-05,
      "loss": 0.4928,
      "num_input_tokens_seen": 151360,
      "step": 375
    },
    {
      "epoch": 0.39501039501039503,
      "grad_norm": 11.648699760437012,
      "learning_rate": 3.93970893970894e-05,
      "loss": 0.4363,
      "num_input_tokens_seen": 153344,
      "step": 380
    },
    {
      "epoch": 0.4002079002079002,
      "grad_norm": 6.347674369812012,
      "learning_rate": 3.991683991683992e-05,
      "loss": 0.2689,
      "num_input_tokens_seen": 155264,
      "step": 385
    },
    {
      "epoch": 0.40540540540540543,
      "grad_norm": 14.731024742126465,
      "learning_rate": 4.043659043659044e-05,
      "loss": 0.2539,
      "num_input_tokens_seen": 157248,
      "step": 390
    },
    {
      "epoch": 0.4106029106029106,
      "grad_norm": 11.619532585144043,
      "learning_rate": 4.095634095634096e-05,
      "loss": 0.332,
      "num_input_tokens_seen": 159296,
      "step": 395
    },
    {
      "epoch": 0.4158004158004158,
      "grad_norm": 20.94300079345703,
      "learning_rate": 4.147609147609148e-05,
      "loss": 0.3,
      "num_input_tokens_seen": 161344,
      "step": 400
    },
    {
      "epoch": 0.420997920997921,
      "grad_norm": 4.039135456085205,
      "learning_rate": 4.1995841995842e-05,
      "loss": 0.2376,
      "num_input_tokens_seen": 163328,
      "step": 405
    },
    {
      "epoch": 0.4261954261954262,
      "grad_norm": 9.615825653076172,
      "learning_rate": 4.2515592515592514e-05,
      "loss": 0.2709,
      "num_input_tokens_seen": 165312,
      "step": 410
    },
    {
      "epoch": 0.4313929313929314,
      "grad_norm": 4.961127758026123,
      "learning_rate": 4.303534303534304e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 167360,
      "step": 415
    },
    {
      "epoch": 0.4365904365904366,
      "grad_norm": 55.59805679321289,
      "learning_rate": 4.3555093555093555e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 169344,
      "step": 420
    },
    {
      "epoch": 0.4417879417879418,
      "grad_norm": 7.878658294677734,
      "learning_rate": 4.407484407484408e-05,
      "loss": 0.603,
      "num_input_tokens_seen": 171456,
      "step": 425
    },
    {
      "epoch": 0.446985446985447,
      "grad_norm": 5.158046722412109,
      "learning_rate": 4.4594594594594596e-05,
      "loss": 0.3054,
      "num_input_tokens_seen": 173568,
      "step": 430
    },
    {
      "epoch": 0.4521829521829522,
      "grad_norm": 1.331925630569458,
      "learning_rate": 4.511434511434512e-05,
      "loss": 0.3208,
      "num_input_tokens_seen": 175552,
      "step": 435
    },
    {
      "epoch": 0.4573804573804574,
      "grad_norm": 7.393728733062744,
      "learning_rate": 4.563409563409564e-05,
      "loss": 0.2942,
      "num_input_tokens_seen": 177536,
      "step": 440
    },
    {
      "epoch": 0.4625779625779626,
      "grad_norm": 15.99036693572998,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.3005,
      "num_input_tokens_seen": 179584,
      "step": 445
    },
    {
      "epoch": 0.4677754677754678,
      "grad_norm": 1.2690688371658325,
      "learning_rate": 4.667359667359668e-05,
      "loss": 1.7773,
      "num_input_tokens_seen": 181568,
      "step": 450
    },
    {
      "epoch": 0.47297297297297297,
      "grad_norm": 3.469167709350586,
      "learning_rate": 4.71933471933472e-05,
      "loss": 0.2889,
      "num_input_tokens_seen": 183552,
      "step": 455
    },
    {
      "epoch": 0.4781704781704782,
      "grad_norm": 90.938232421875,
      "learning_rate": 4.771309771309771e-05,
      "loss": 0.2556,
      "num_input_tokens_seen": 185600,
      "step": 460
    },
    {
      "epoch": 0.48336798336798337,
      "grad_norm": 2.4140217304229736,
      "learning_rate": 4.823284823284824e-05,
      "loss": 0.2378,
      "num_input_tokens_seen": 187584,
      "step": 465
    },
    {
      "epoch": 0.4885654885654886,
      "grad_norm": 4.565840721130371,
      "learning_rate": 4.8752598752598754e-05,
      "loss": 0.2859,
      "num_input_tokens_seen": 189568,
      "step": 470
    },
    {
      "epoch": 0.49376299376299376,
      "grad_norm": 12.585880279541016,
      "learning_rate": 4.9272349272349275e-05,
      "loss": 0.2438,
      "num_input_tokens_seen": 191680,
      "step": 475
    },
    {
      "epoch": 0.498960498960499,
      "grad_norm": 3.2450168132781982,
      "learning_rate": 4.9792099792099796e-05,
      "loss": 0.3383,
      "num_input_tokens_seen": 193728,
      "step": 480
    },
    {
      "epoch": 0.501039501039501,
      "eval_loss": 0.4080815315246582,
      "eval_runtime": 2.6441,
      "eval_samples_per_second": 323.736,
      "eval_steps_per_second": 40.467,
      "num_input_tokens_seen": 194560,
      "step": 482
    },
    {
      "epoch": 0.5041580041580042,
      "grad_norm": 29.770891189575195,
      "learning_rate": 4.999994075155936e-05,
      "loss": 0.381,
      "num_input_tokens_seen": 195776,
      "step": 485
    },
    {
      "epoch": 0.5093555093555093,
      "grad_norm": 19.471193313598633,
      "learning_rate": 4.999957867877242e-05,
      "loss": 0.2916,
      "num_input_tokens_seen": 197696,
      "step": 490
    },
    {
      "epoch": 0.5145530145530145,
      "grad_norm": 29.544788360595703,
      "learning_rate": 4.999888745376028e-05,
      "loss": 0.2766,
      "num_input_tokens_seen": 199680,
      "step": 495
    },
    {
      "epoch": 0.5197505197505198,
      "grad_norm": 11.05797290802002,
      "learning_rate": 4.9997867085623824e-05,
      "loss": 0.2596,
      "num_input_tokens_seen": 201792,
      "step": 500
    },
    {
      "epoch": 0.524948024948025,
      "grad_norm": 3.8948936462402344,
      "learning_rate": 4.999651758779754e-05,
      "loss": 0.2689,
      "num_input_tokens_seen": 203840,
      "step": 505
    },
    {
      "epoch": 0.5301455301455301,
      "grad_norm": 18.150039672851562,
      "learning_rate": 4.999483897804933e-05,
      "loss": 0.3363,
      "num_input_tokens_seen": 205824,
      "step": 510
    },
    {
      "epoch": 0.5353430353430353,
      "grad_norm": 1.8224319219589233,
      "learning_rate": 4.999283127848029e-05,
      "loss": 0.2583,
      "num_input_tokens_seen": 207936,
      "step": 515
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 9.910496711730957,
      "learning_rate": 4.999049451552443e-05,
      "loss": 0.3915,
      "num_input_tokens_seen": 209984,
      "step": 520
    },
    {
      "epoch": 0.5457380457380457,
      "grad_norm": 0.8137359619140625,
      "learning_rate": 4.9987828719948284e-05,
      "loss": 0.2461,
      "num_input_tokens_seen": 212096,
      "step": 525
    },
    {
      "epoch": 0.5509355509355509,
      "grad_norm": 2.0492968559265137,
      "learning_rate": 4.998483392685055e-05,
      "loss": 0.2863,
      "num_input_tokens_seen": 214080,
      "step": 530
    },
    {
      "epoch": 0.5561330561330561,
      "grad_norm": 24.322450637817383,
      "learning_rate": 4.9981510175661606e-05,
      "loss": 0.3052,
      "num_input_tokens_seen": 216128,
      "step": 535
    },
    {
      "epoch": 0.5613305613305614,
      "grad_norm": 28.344161987304688,
      "learning_rate": 4.9977857510143e-05,
      "loss": 0.3625,
      "num_input_tokens_seen": 218176,
      "step": 540
    },
    {
      "epoch": 0.5665280665280665,
      "grad_norm": 11.159996032714844,
      "learning_rate": 4.9973875978386843e-05,
      "loss": 0.2801,
      "num_input_tokens_seen": 220096,
      "step": 545
    },
    {
      "epoch": 0.5717255717255717,
      "grad_norm": 14.94636058807373,
      "learning_rate": 4.996956563281524e-05,
      "loss": 0.3041,
      "num_input_tokens_seen": 222080,
      "step": 550
    },
    {
      "epoch": 0.5769230769230769,
      "grad_norm": 8.506303787231445,
      "learning_rate": 4.996492653017952e-05,
      "loss": 0.2498,
      "num_input_tokens_seen": 224000,
      "step": 555
    },
    {
      "epoch": 0.5821205821205822,
      "grad_norm": 18.009078979492188,
      "learning_rate": 4.995995873155958e-05,
      "loss": 0.3224,
      "num_input_tokens_seen": 225984,
      "step": 560
    },
    {
      "epoch": 0.5873180873180873,
      "grad_norm": 19.32984733581543,
      "learning_rate": 4.9954662302362973e-05,
      "loss": 0.322,
      "num_input_tokens_seen": 227840,
      "step": 565
    },
    {
      "epoch": 0.5925155925155925,
      "grad_norm": 18.17301368713379,
      "learning_rate": 4.9949037312324155e-05,
      "loss": 0.302,
      "num_input_tokens_seen": 229824,
      "step": 570
    },
    {
      "epoch": 0.5977130977130977,
      "grad_norm": 14.943294525146484,
      "learning_rate": 4.9943083835503467e-05,
      "loss": 0.3875,
      "num_input_tokens_seen": 231872,
      "step": 575
    },
    {
      "epoch": 0.6029106029106029,
      "grad_norm": 12.56007194519043,
      "learning_rate": 4.993680195028626e-05,
      "loss": 0.314,
      "num_input_tokens_seen": 233920,
      "step": 580
    },
    {
      "epoch": 0.6081081081081081,
      "grad_norm": 12.372591972351074,
      "learning_rate": 4.9930191739381775e-05,
      "loss": 0.3189,
      "num_input_tokens_seen": 235840,
      "step": 585
    },
    {
      "epoch": 0.6133056133056133,
      "grad_norm": 11.213384628295898,
      "learning_rate": 4.9923253289822116e-05,
      "loss": 0.3418,
      "num_input_tokens_seen": 238016,
      "step": 590
    },
    {
      "epoch": 0.6185031185031185,
      "grad_norm": 12.969761848449707,
      "learning_rate": 4.9915986692961045e-05,
      "loss": 0.3307,
      "num_input_tokens_seen": 240064,
      "step": 595
    },
    {
      "epoch": 0.6237006237006237,
      "grad_norm": 1.1426069736480713,
      "learning_rate": 4.9908392044472865e-05,
      "loss": 0.2835,
      "num_input_tokens_seen": 242048,
      "step": 600
    },
    {
      "epoch": 0.6288981288981289,
      "grad_norm": 7.80474853515625,
      "learning_rate": 4.990046944435105e-05,
      "loss": 0.2584,
      "num_input_tokens_seen": 243968,
      "step": 605
    },
    {
      "epoch": 0.6340956340956341,
      "grad_norm": 1.6327171325683594,
      "learning_rate": 4.989221899690704e-05,
      "loss": 0.2953,
      "num_input_tokens_seen": 246016,
      "step": 610
    },
    {
      "epoch": 0.6392931392931392,
      "grad_norm": 16.204669952392578,
      "learning_rate": 4.9883640810768764e-05,
      "loss": 0.3214,
      "num_input_tokens_seen": 248000,
      "step": 615
    },
    {
      "epoch": 0.6444906444906445,
      "grad_norm": 0.29481250047683716,
      "learning_rate": 4.9874734998879316e-05,
      "loss": 0.312,
      "num_input_tokens_seen": 250048,
      "step": 620
    },
    {
      "epoch": 0.6496881496881497,
      "grad_norm": 18.51386833190918,
      "learning_rate": 4.9865501678495375e-05,
      "loss": 0.3215,
      "num_input_tokens_seen": 252096,
      "step": 625
    },
    {
      "epoch": 0.6548856548856549,
      "grad_norm": 5.452864646911621,
      "learning_rate": 4.98559409711857e-05,
      "loss": 0.2892,
      "num_input_tokens_seen": 254144,
      "step": 630
    },
    {
      "epoch": 0.66008316008316,
      "grad_norm": 11.4102783203125,
      "learning_rate": 4.984605300282954e-05,
      "loss": 0.3016,
      "num_input_tokens_seen": 256128,
      "step": 635
    },
    {
      "epoch": 0.6652806652806653,
      "grad_norm": 8.976471900939941,
      "learning_rate": 4.983583790361497e-05,
      "loss": 0.2596,
      "num_input_tokens_seen": 258048,
      "step": 640
    },
    {
      "epoch": 0.6704781704781705,
      "grad_norm": 27.530433654785156,
      "learning_rate": 4.982529580803714e-05,
      "loss": 0.363,
      "num_input_tokens_seen": 260352,
      "step": 645
    },
    {
      "epoch": 0.6756756756756757,
      "grad_norm": 12.061158180236816,
      "learning_rate": 4.981442685489659e-05,
      "loss": 0.2849,
      "num_input_tokens_seen": 262272,
      "step": 650
    },
    {
      "epoch": 0.6808731808731808,
      "grad_norm": 13.447040557861328,
      "learning_rate": 4.9803231187297304e-05,
      "loss": 0.2969,
      "num_input_tokens_seen": 264320,
      "step": 655
    },
    {
      "epoch": 0.6860706860706861,
      "grad_norm": 33.57612228393555,
      "learning_rate": 4.979170895264494e-05,
      "loss": 0.3962,
      "num_input_tokens_seen": 266240,
      "step": 660
    },
    {
      "epoch": 0.6912681912681913,
      "grad_norm": 4.9110002517700195,
      "learning_rate": 4.977986030264482e-05,
      "loss": 0.312,
      "num_input_tokens_seen": 268224,
      "step": 665
    },
    {
      "epoch": 0.6964656964656964,
      "grad_norm": 0.21739044785499573,
      "learning_rate": 4.976768539329994e-05,
      "loss": 0.2552,
      "num_input_tokens_seen": 270272,
      "step": 670
    },
    {
      "epoch": 0.7016632016632016,
      "grad_norm": 8.439529418945312,
      "learning_rate": 4.975518438490897e-05,
      "loss": 0.312,
      "num_input_tokens_seen": 272256,
      "step": 675
    },
    {
      "epoch": 0.7068607068607069,
      "grad_norm": 10.565657615661621,
      "learning_rate": 4.9742357442064045e-05,
      "loss": 0.2889,
      "num_input_tokens_seen": 274240,
      "step": 680
    },
    {
      "epoch": 0.7120582120582121,
      "grad_norm": 19.482200622558594,
      "learning_rate": 4.972920473364869e-05,
      "loss": 0.2841,
      "num_input_tokens_seen": 276288,
      "step": 685
    },
    {
      "epoch": 0.7172557172557172,
      "grad_norm": 21.85175132751465,
      "learning_rate": 4.971572643283557e-05,
      "loss": 0.3076,
      "num_input_tokens_seen": 278272,
      "step": 690
    },
    {
      "epoch": 0.7224532224532224,
      "grad_norm": 7.696235656738281,
      "learning_rate": 4.970192271708416e-05,
      "loss": 0.285,
      "num_input_tokens_seen": 280384,
      "step": 695
    },
    {
      "epoch": 0.7276507276507277,
      "grad_norm": 4.778509616851807,
      "learning_rate": 4.968779376813849e-05,
      "loss": 0.2376,
      "num_input_tokens_seen": 282368,
      "step": 700
    },
    {
      "epoch": 0.7328482328482329,
      "grad_norm": 9.415769577026367,
      "learning_rate": 4.967333977202469e-05,
      "loss": 0.2787,
      "num_input_tokens_seen": 284416,
      "step": 705
    },
    {
      "epoch": 0.738045738045738,
      "grad_norm": 4.963840484619141,
      "learning_rate": 4.965856091904855e-05,
      "loss": 0.214,
      "num_input_tokens_seen": 286464,
      "step": 710
    },
    {
      "epoch": 0.7432432432432432,
      "grad_norm": 12.690200805664062,
      "learning_rate": 4.964345740379307e-05,
      "loss": 0.2858,
      "num_input_tokens_seen": 288448,
      "step": 715
    },
    {
      "epoch": 0.7484407484407485,
      "grad_norm": 15.548615455627441,
      "learning_rate": 4.962802942511581e-05,
      "loss": 0.2962,
      "num_input_tokens_seen": 290496,
      "step": 720
    },
    {
      "epoch": 0.7515592515592515,
      "eval_loss": 0.29600390791893005,
      "eval_runtime": 3.3797,
      "eval_samples_per_second": 253.278,
      "eval_steps_per_second": 31.66,
      "num_input_tokens_seen": 291712,
      "step": 723
    },
    {
      "epoch": 0.7536382536382537,
      "grad_norm": 7.057698726654053,
      "learning_rate": 4.9612277186146335e-05,
      "loss": 0.3061,
      "num_input_tokens_seen": 292480,
      "step": 725
    },
    {
      "epoch": 0.7588357588357588,
      "grad_norm": 17.72517204284668,
      "learning_rate": 4.959620089428354e-05,
      "loss": 0.281,
      "num_input_tokens_seen": 294464,
      "step": 730
    },
    {
      "epoch": 0.764033264033264,
      "grad_norm": 1.6551860570907593,
      "learning_rate": 4.957980076119285e-05,
      "loss": 0.2702,
      "num_input_tokens_seen": 296448,
      "step": 735
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 9.719671249389648,
      "learning_rate": 4.956307700280354e-05,
      "loss": 0.3146,
      "num_input_tokens_seen": 298432,
      "step": 740
    },
    {
      "epoch": 0.7744282744282744,
      "grad_norm": 8.770389556884766,
      "learning_rate": 4.954602983930581e-05,
      "loss": 0.2567,
      "num_input_tokens_seen": 300480,
      "step": 745
    },
    {
      "epoch": 0.7796257796257796,
      "grad_norm": 7.721967697143555,
      "learning_rate": 4.95286594951479e-05,
      "loss": 0.2488,
      "num_input_tokens_seen": 302400,
      "step": 750
    },
    {
      "epoch": 0.7848232848232848,
      "grad_norm": 7.888345241546631,
      "learning_rate": 4.9510966199033174e-05,
      "loss": 0.2852,
      "num_input_tokens_seen": 304320,
      "step": 755
    },
    {
      "epoch": 0.7900207900207901,
      "grad_norm": 8.099255561828613,
      "learning_rate": 4.949295018391706e-05,
      "loss": 0.2968,
      "num_input_tokens_seen": 306240,
      "step": 760
    },
    {
      "epoch": 0.7952182952182952,
      "grad_norm": 7.542270183563232,
      "learning_rate": 4.947461168700402e-05,
      "loss": 0.2957,
      "num_input_tokens_seen": 308032,
      "step": 765
    },
    {
      "epoch": 0.8004158004158004,
      "grad_norm": 4.36430025100708,
      "learning_rate": 4.945595094974442e-05,
      "loss": 0.2531,
      "num_input_tokens_seen": 309952,
      "step": 770
    },
    {
      "epoch": 0.8056133056133056,
      "grad_norm": 11.39189624786377,
      "learning_rate": 4.94369682178313e-05,
      "loss": 0.2396,
      "num_input_tokens_seen": 311936,
      "step": 775
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 3.859365224838257,
      "learning_rate": 4.9417663741197236e-05,
      "loss": 0.3127,
      "num_input_tokens_seen": 313920,
      "step": 780
    },
    {
      "epoch": 0.816008316008316,
      "grad_norm": 10.639534950256348,
      "learning_rate": 4.939803777401095e-05,
      "loss": 0.2873,
      "num_input_tokens_seen": 315968,
      "step": 785
    },
    {
      "epoch": 0.8212058212058212,
      "grad_norm": 8.708645820617676,
      "learning_rate": 4.937809057467404e-05,
      "loss": 0.263,
      "num_input_tokens_seen": 317952,
      "step": 790
    },
    {
      "epoch": 0.8264033264033264,
      "grad_norm": 3.0771279335021973,
      "learning_rate": 4.935782240581752e-05,
      "loss": 0.2762,
      "num_input_tokens_seen": 319872,
      "step": 795
    },
    {
      "epoch": 0.8316008316008316,
      "grad_norm": 9.068258285522461,
      "learning_rate": 4.9337233534298425e-05,
      "loss": 0.2759,
      "num_input_tokens_seen": 321856,
      "step": 800
    },
    {
      "epoch": 0.8367983367983368,
      "grad_norm": 11.020606994628906,
      "learning_rate": 4.931632423119621e-05,
      "loss": 0.2849,
      "num_input_tokens_seen": 323968,
      "step": 805
    },
    {
      "epoch": 0.841995841995842,
      "grad_norm": 11.716889381408691,
      "learning_rate": 4.9295094771809285e-05,
      "loss": 0.2775,
      "num_input_tokens_seen": 325952,
      "step": 810
    },
    {
      "epoch": 0.8471933471933472,
      "grad_norm": 2.4100873470306396,
      "learning_rate": 4.92735454356513e-05,
      "loss": 0.1962,
      "num_input_tokens_seen": 328000,
      "step": 815
    },
    {
      "epoch": 0.8523908523908524,
      "grad_norm": 18.39780044555664,
      "learning_rate": 4.925167650644752e-05,
      "loss": 0.2237,
      "num_input_tokens_seen": 329984,
      "step": 820
    },
    {
      "epoch": 0.8575883575883576,
      "grad_norm": 31.814456939697266,
      "learning_rate": 4.9229488272131067e-05,
      "loss": 0.3432,
      "num_input_tokens_seen": 331904,
      "step": 825
    },
    {
      "epoch": 0.8627858627858628,
      "grad_norm": 13.310869216918945,
      "learning_rate": 4.920698102483912e-05,
      "loss": 0.3102,
      "num_input_tokens_seen": 333888,
      "step": 830
    },
    {
      "epoch": 0.867983367983368,
      "grad_norm": 1.8199642896652222,
      "learning_rate": 4.918415506090911e-05,
      "loss": 0.3035,
      "num_input_tokens_seen": 335872,
      "step": 835
    },
    {
      "epoch": 0.8731808731808732,
      "grad_norm": 14.503442764282227,
      "learning_rate": 4.916101068087476e-05,
      "loss": 0.2682,
      "num_input_tokens_seen": 337856,
      "step": 840
    },
    {
      "epoch": 0.8783783783783784,
      "grad_norm": 1.1979509592056274,
      "learning_rate": 4.913754818946219e-05,
      "loss": 0.2422,
      "num_input_tokens_seen": 339776,
      "step": 845
    },
    {
      "epoch": 0.8835758835758836,
      "grad_norm": 5.467031478881836,
      "learning_rate": 4.911376789558584e-05,
      "loss": 0.1949,
      "num_input_tokens_seen": 341760,
      "step": 850
    },
    {
      "epoch": 0.8887733887733887,
      "grad_norm": 26.362804412841797,
      "learning_rate": 4.9089670112344456e-05,
      "loss": 0.3477,
      "num_input_tokens_seen": 343680,
      "step": 855
    },
    {
      "epoch": 0.893970893970894,
      "grad_norm": 13.462319374084473,
      "learning_rate": 4.906525515701695e-05,
      "loss": 0.292,
      "num_input_tokens_seen": 345600,
      "step": 860
    },
    {
      "epoch": 0.8991683991683992,
      "grad_norm": 15.104516983032227,
      "learning_rate": 4.904052335105822e-05,
      "loss": 0.2896,
      "num_input_tokens_seen": 347520,
      "step": 865
    },
    {
      "epoch": 0.9043659043659044,
      "grad_norm": 2.1453566551208496,
      "learning_rate": 4.90154750200949e-05,
      "loss": 0.2644,
      "num_input_tokens_seen": 349568,
      "step": 870
    },
    {
      "epoch": 0.9095634095634095,
      "grad_norm": 30.04273223876953,
      "learning_rate": 4.8990110493921105e-05,
      "loss": 0.3212,
      "num_input_tokens_seen": 351552,
      "step": 875
    },
    {
      "epoch": 0.9147609147609148,
      "grad_norm": 9.758408546447754,
      "learning_rate": 4.8964430106494075e-05,
      "loss": 0.2787,
      "num_input_tokens_seen": 353472,
      "step": 880
    },
    {
      "epoch": 0.91995841995842,
      "grad_norm": 2.3094310760498047,
      "learning_rate": 4.893843419592977e-05,
      "loss": 0.2656,
      "num_input_tokens_seen": 355392,
      "step": 885
    },
    {
      "epoch": 0.9251559251559252,
      "grad_norm": 2.449030876159668,
      "learning_rate": 4.891212310449844e-05,
      "loss": 0.2593,
      "num_input_tokens_seen": 357440,
      "step": 890
    },
    {
      "epoch": 0.9303534303534303,
      "grad_norm": 5.086390972137451,
      "learning_rate": 4.8885497178620095e-05,
      "loss": 0.2785,
      "num_input_tokens_seen": 359488,
      "step": 895
    },
    {
      "epoch": 0.9355509355509356,
      "grad_norm": 12.254743576049805,
      "learning_rate": 4.8858556768859944e-05,
      "loss": 0.283,
      "num_input_tokens_seen": 361408,
      "step": 900
    },
    {
      "epoch": 0.9407484407484408,
      "grad_norm": 8.048048973083496,
      "learning_rate": 4.88313022299238e-05,
      "loss": 0.27,
      "num_input_tokens_seen": 363392,
      "step": 905
    },
    {
      "epoch": 0.9459459459459459,
      "grad_norm": 13.520707130432129,
      "learning_rate": 4.88037339206534e-05,
      "loss": 0.2987,
      "num_input_tokens_seen": 365440,
      "step": 910
    },
    {
      "epoch": 0.9511434511434511,
      "grad_norm": 18.515222549438477,
      "learning_rate": 4.8775852204021665e-05,
      "loss": 0.2998,
      "num_input_tokens_seen": 367616,
      "step": 915
    },
    {
      "epoch": 0.9563409563409564,
      "grad_norm": 25.785511016845703,
      "learning_rate": 4.874765744712796e-05,
      "loss": 0.358,
      "num_input_tokens_seen": 369600,
      "step": 920
    },
    {
      "epoch": 0.9615384615384616,
      "grad_norm": 8.675851821899414,
      "learning_rate": 4.871915002119321e-05,
      "loss": 0.2755,
      "num_input_tokens_seen": 371520,
      "step": 925
    },
    {
      "epoch": 0.9667359667359667,
      "grad_norm": 0.7464810013771057,
      "learning_rate": 4.8690330301555045e-05,
      "loss": 0.3085,
      "num_input_tokens_seen": 373568,
      "step": 930
    },
    {
      "epoch": 0.9719334719334719,
      "grad_norm": 21.288469314575195,
      "learning_rate": 4.8661198667662854e-05,
      "loss": 0.2548,
      "num_input_tokens_seen": 375488,
      "step": 935
    },
    {
      "epoch": 0.9771309771309772,
      "grad_norm": 15.424735069274902,
      "learning_rate": 4.86317555030728e-05,
      "loss": 0.3274,
      "num_input_tokens_seen": 377728,
      "step": 940
    },
    {
      "epoch": 0.9823284823284824,
      "grad_norm": 11.471841812133789,
      "learning_rate": 4.8602001195442725e-05,
      "loss": 0.2924,
      "num_input_tokens_seen": 379840,
      "step": 945
    },
    {
      "epoch": 0.9875259875259875,
      "grad_norm": 12.289169311523438,
      "learning_rate": 4.857193613652711e-05,
      "loss": 0.2685,
      "num_input_tokens_seen": 381760,
      "step": 950
    },
    {
      "epoch": 0.9927234927234927,
      "grad_norm": 2.2855138778686523,
      "learning_rate": 4.8541560722171855e-05,
      "loss": 0.2743,
      "num_input_tokens_seen": 383808,
      "step": 955
    },
    {
      "epoch": 0.997920997920998,
      "grad_norm": 14.171538352966309,
      "learning_rate": 4.8510875352309106e-05,
      "loss": 0.2807,
      "num_input_tokens_seen": 385856,
      "step": 960
    },
    {
      "epoch": 1.002079002079002,
      "eval_loss": 0.2738620638847351,
      "eval_runtime": 1.0463,
      "eval_samples_per_second": 818.137,
      "eval_steps_per_second": 102.267,
      "num_input_tokens_seen": 387464,
      "step": 964
    },
    {
      "epoch": 1.003118503118503,
      "grad_norm": 1.732373833656311,
      "learning_rate": 4.8479880430951995e-05,
      "loss": 0.2643,
      "num_input_tokens_seen": 387848,
      "step": 965
    },
    {
      "epoch": 1.0083160083160083,
      "grad_norm": 9.852456092834473,
      "learning_rate": 4.844857636618928e-05,
      "loss": 0.2613,
      "num_input_tokens_seen": 389640,
      "step": 970
    },
    {
      "epoch": 1.0135135135135136,
      "grad_norm": 13.70590877532959,
      "learning_rate": 4.8416963570180025e-05,
      "loss": 0.2824,
      "num_input_tokens_seen": 391624,
      "step": 975
    },
    {
      "epoch": 1.0187110187110187,
      "grad_norm": 10.222359657287598,
      "learning_rate": 4.838504245914812e-05,
      "loss": 0.3008,
      "num_input_tokens_seen": 393672,
      "step": 980
    },
    {
      "epoch": 1.023908523908524,
      "grad_norm": 24.265884399414062,
      "learning_rate": 4.8352813453376836e-05,
      "loss": 0.3048,
      "num_input_tokens_seen": 395784,
      "step": 985
    },
    {
      "epoch": 1.0291060291060292,
      "grad_norm": 20.11310577392578,
      "learning_rate": 4.83202769772033e-05,
      "loss": 0.3205,
      "num_input_tokens_seen": 397768,
      "step": 990
    },
    {
      "epoch": 1.0343035343035343,
      "grad_norm": 19.961780548095703,
      "learning_rate": 4.8287433459012844e-05,
      "loss": 0.2694,
      "num_input_tokens_seen": 399816,
      "step": 995
    },
    {
      "epoch": 1.0395010395010396,
      "grad_norm": 3.684941530227661,
      "learning_rate": 4.8254283331233464e-05,
      "loss": 0.3263,
      "num_input_tokens_seen": 401928,
      "step": 1000
    },
    {
      "epoch": 1.0446985446985446,
      "grad_norm": 1.1775870323181152,
      "learning_rate": 4.822082703033003e-05,
      "loss": 0.3028,
      "num_input_tokens_seen": 403912,
      "step": 1005
    },
    {
      "epoch": 1.04989604989605,
      "grad_norm": 7.7064595222473145,
      "learning_rate": 4.818706499679862e-05,
      "loss": 0.2501,
      "num_input_tokens_seen": 405832,
      "step": 1010
    },
    {
      "epoch": 1.0550935550935552,
      "grad_norm": 20.951908111572266,
      "learning_rate": 4.815299767516065e-05,
      "loss": 0.339,
      "num_input_tokens_seen": 407880,
      "step": 1015
    },
    {
      "epoch": 1.0602910602910602,
      "grad_norm": 0.5823318958282471,
      "learning_rate": 4.8118625513957074e-05,
      "loss": 0.2748,
      "num_input_tokens_seen": 410120,
      "step": 1020
    },
    {
      "epoch": 1.0654885654885655,
      "grad_norm": 5.90387487411499,
      "learning_rate": 4.808394896574245e-05,
      "loss": 0.3246,
      "num_input_tokens_seen": 412168,
      "step": 1025
    },
    {
      "epoch": 1.0706860706860706,
      "grad_norm": 13.779071807861328,
      "learning_rate": 4.8048968487079e-05,
      "loss": 0.2849,
      "num_input_tokens_seen": 414472,
      "step": 1030
    },
    {
      "epoch": 1.0758835758835759,
      "grad_norm": 15.818516731262207,
      "learning_rate": 4.8013684538530565e-05,
      "loss": 0.2908,
      "num_input_tokens_seen": 416520,
      "step": 1035
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 11.890518188476562,
      "learning_rate": 4.79780975846566e-05,
      "loss": 0.2774,
      "num_input_tokens_seen": 418568,
      "step": 1040
    },
    {
      "epoch": 1.0862785862785862,
      "grad_norm": 7.704615592956543,
      "learning_rate": 4.7942208094006e-05,
      "loss": 0.2417,
      "num_input_tokens_seen": 420488,
      "step": 1045
    },
    {
      "epoch": 1.0914760914760915,
      "grad_norm": 2.7734618186950684,
      "learning_rate": 4.790601653911094e-05,
      "loss": 0.2736,
      "num_input_tokens_seen": 422472,
      "step": 1050
    },
    {
      "epoch": 1.0966735966735968,
      "grad_norm": 12.339665412902832,
      "learning_rate": 4.786952339648071e-05,
      "loss": 0.3165,
      "num_input_tokens_seen": 424456,
      "step": 1055
    },
    {
      "epoch": 1.1018711018711018,
      "grad_norm": 3.863492250442505,
      "learning_rate": 4.783272914659535e-05,
      "loss": 0.305,
      "num_input_tokens_seen": 426568,
      "step": 1060
    },
    {
      "epoch": 1.107068607068607,
      "grad_norm": 5.703735828399658,
      "learning_rate": 4.77956342738994e-05,
      "loss": 0.2809,
      "num_input_tokens_seen": 428552,
      "step": 1065
    },
    {
      "epoch": 1.1122661122661124,
      "grad_norm": 5.661115646362305,
      "learning_rate": 4.775823926679548e-05,
      "loss": 0.2758,
      "num_input_tokens_seen": 430472,
      "step": 1070
    },
    {
      "epoch": 1.1174636174636174,
      "grad_norm": 5.488950729370117,
      "learning_rate": 4.77205446176379e-05,
      "loss": 0.2657,
      "num_input_tokens_seen": 432328,
      "step": 1075
    },
    {
      "epoch": 1.1226611226611227,
      "grad_norm": 10.569025993347168,
      "learning_rate": 4.768255082272611e-05,
      "loss": 0.2881,
      "num_input_tokens_seen": 434440,
      "step": 1080
    },
    {
      "epoch": 1.1278586278586278,
      "grad_norm": 3.0698752403259277,
      "learning_rate": 4.764425838229824e-05,
      "loss": 0.2938,
      "num_input_tokens_seen": 436488,
      "step": 1085
    },
    {
      "epoch": 1.133056133056133,
      "grad_norm": 23.730390548706055,
      "learning_rate": 4.760566780052445e-05,
      "loss": 0.4153,
      "num_input_tokens_seen": 438472,
      "step": 1090
    },
    {
      "epoch": 1.1382536382536383,
      "grad_norm": 0.6121450066566467,
      "learning_rate": 4.7566779585500347e-05,
      "loss": 0.3296,
      "num_input_tokens_seen": 440456,
      "step": 1095
    },
    {
      "epoch": 1.1434511434511434,
      "grad_norm": 4.771881103515625,
      "learning_rate": 4.7527594249240264e-05,
      "loss": 0.3082,
      "num_input_tokens_seen": 442440,
      "step": 1100
    },
    {
      "epoch": 1.1486486486486487,
      "grad_norm": 7.00406551361084,
      "learning_rate": 4.748811230767051e-05,
      "loss": 0.3109,
      "num_input_tokens_seen": 444424,
      "step": 1105
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 0.35836905241012573,
      "learning_rate": 4.744833428062262e-05,
      "loss": 0.2754,
      "num_input_tokens_seen": 446280,
      "step": 1110
    },
    {
      "epoch": 1.159043659043659,
      "grad_norm": 12.108142852783203,
      "learning_rate": 4.740826069182645e-05,
      "loss": 0.286,
      "num_input_tokens_seen": 448264,
      "step": 1115
    },
    {
      "epoch": 1.1642411642411643,
      "grad_norm": 2.960707902908325,
      "learning_rate": 4.736789206890332e-05,
      "loss": 0.2806,
      "num_input_tokens_seen": 450376,
      "step": 1120
    },
    {
      "epoch": 1.1694386694386694,
      "grad_norm": 1.6544653177261353,
      "learning_rate": 4.732722894335909e-05,
      "loss": 0.2575,
      "num_input_tokens_seen": 452552,
      "step": 1125
    },
    {
      "epoch": 1.1746361746361746,
      "grad_norm": 4.431951999664307,
      "learning_rate": 4.7286271850577105e-05,
      "loss": 0.2989,
      "num_input_tokens_seen": 454600,
      "step": 1130
    },
    {
      "epoch": 1.17983367983368,
      "grad_norm": 27.89082145690918,
      "learning_rate": 4.724502132981119e-05,
      "loss": 0.5214,
      "num_input_tokens_seen": 456648,
      "step": 1135
    },
    {
      "epoch": 1.185031185031185,
      "grad_norm": 12.9546480178833,
      "learning_rate": 4.7203477924178506e-05,
      "loss": 0.3325,
      "num_input_tokens_seen": 458632,
      "step": 1140
    },
    {
      "epoch": 1.1902286902286903,
      "grad_norm": 10.19326114654541,
      "learning_rate": 4.7161642180652464e-05,
      "loss": 0.2795,
      "num_input_tokens_seen": 460680,
      "step": 1145
    },
    {
      "epoch": 1.1954261954261955,
      "grad_norm": 11.509027481079102,
      "learning_rate": 4.7119514650055476e-05,
      "loss": 0.2697,
      "num_input_tokens_seen": 462728,
      "step": 1150
    },
    {
      "epoch": 1.2006237006237006,
      "grad_norm": 11.531643867492676,
      "learning_rate": 4.7077095887051686e-05,
      "loss": 0.2429,
      "num_input_tokens_seen": 464776,
      "step": 1155
    },
    {
      "epoch": 1.2058212058212059,
      "grad_norm": 2.4489336013793945,
      "learning_rate": 4.7034386450139735e-05,
      "loss": 0.2844,
      "num_input_tokens_seen": 466696,
      "step": 1160
    },
    {
      "epoch": 1.211018711018711,
      "grad_norm": 10.006536483764648,
      "learning_rate": 4.699138690164533e-05,
      "loss": 0.259,
      "num_input_tokens_seen": 468616,
      "step": 1165
    },
    {
      "epoch": 1.2162162162162162,
      "grad_norm": 5.630403995513916,
      "learning_rate": 4.694809780771391e-05,
      "loss": 0.2734,
      "num_input_tokens_seen": 470728,
      "step": 1170
    },
    {
      "epoch": 1.2214137214137215,
      "grad_norm": 9.379544258117676,
      "learning_rate": 4.690451973830313e-05,
      "loss": 0.2763,
      "num_input_tokens_seen": 472776,
      "step": 1175
    },
    {
      "epoch": 1.2266112266112266,
      "grad_norm": 4.9165167808532715,
      "learning_rate": 4.6860653267175416e-05,
      "loss": 0.2615,
      "num_input_tokens_seen": 474824,
      "step": 1180
    },
    {
      "epoch": 1.2318087318087318,
      "grad_norm": 7.163639068603516,
      "learning_rate": 4.681649897189036e-05,
      "loss": 0.2894,
      "num_input_tokens_seen": 476744,
      "step": 1185
    },
    {
      "epoch": 1.237006237006237,
      "grad_norm": 3.139596939086914,
      "learning_rate": 4.677205743379713e-05,
      "loss": 0.2065,
      "num_input_tokens_seen": 478856,
      "step": 1190
    },
    {
      "epoch": 1.2422037422037422,
      "grad_norm": 18.728031158447266,
      "learning_rate": 4.672732923802685e-05,
      "loss": 0.4129,
      "num_input_tokens_seen": 480776,
      "step": 1195
    },
    {
      "epoch": 1.2474012474012475,
      "grad_norm": 21.870121002197266,
      "learning_rate": 4.668231497348484e-05,
      "loss": 0.2716,
      "num_input_tokens_seen": 482952,
      "step": 1200
    },
    {
      "epoch": 1.2525987525987525,
      "grad_norm": 7.45676851272583,
      "learning_rate": 4.663701523284291e-05,
      "loss": 0.2836,
      "num_input_tokens_seen": 485192,
      "step": 1205
    },
    {
      "epoch": 1.2525987525987525,
      "eval_loss": 0.2581372559070587,
      "eval_runtime": 1.5453,
      "eval_samples_per_second": 553.946,
      "eval_steps_per_second": 69.243,
      "num_input_tokens_seen": 485192,
      "step": 1205
    },
    {
      "epoch": 1.2577962577962578,
      "grad_norm": 19.655874252319336,
      "learning_rate": 4.6591430612531515e-05,
      "loss": 0.2541,
      "num_input_tokens_seen": 487112,
      "step": 1210
    },
    {
      "epoch": 1.262993762993763,
      "grad_norm": 11.319998741149902,
      "learning_rate": 4.6545561712731954e-05,
      "loss": 0.3056,
      "num_input_tokens_seen": 489160,
      "step": 1215
    },
    {
      "epoch": 1.2681912681912682,
      "grad_norm": 6.425257205963135,
      "learning_rate": 4.649940913736841e-05,
      "loss": 0.2656,
      "num_input_tokens_seen": 491080,
      "step": 1220
    },
    {
      "epoch": 1.2733887733887734,
      "grad_norm": 7.79650354385376,
      "learning_rate": 4.645297349410005e-05,
      "loss": 0.2917,
      "num_input_tokens_seen": 493064,
      "step": 1225
    },
    {
      "epoch": 1.2785862785862787,
      "grad_norm": 8.849406242370605,
      "learning_rate": 4.640625539431298e-05,
      "loss": 0.2878,
      "num_input_tokens_seen": 494984,
      "step": 1230
    },
    {
      "epoch": 1.2837837837837838,
      "grad_norm": 10.014208793640137,
      "learning_rate": 4.635925545311224e-05,
      "loss": 0.2686,
      "num_input_tokens_seen": 496968,
      "step": 1235
    },
    {
      "epoch": 1.288981288981289,
      "grad_norm": 3.8225860595703125,
      "learning_rate": 4.6311974289313646e-05,
      "loss": 0.2747,
      "num_input_tokens_seen": 498824,
      "step": 1240
    },
    {
      "epoch": 1.2941787941787941,
      "grad_norm": 9.505216598510742,
      "learning_rate": 4.6264412525435716e-05,
      "loss": 0.2269,
      "num_input_tokens_seen": 500808,
      "step": 1245
    },
    {
      "epoch": 1.2993762993762994,
      "grad_norm": 12.160257339477539,
      "learning_rate": 4.6216570787691423e-05,
      "loss": 0.2595,
      "num_input_tokens_seen": 502856,
      "step": 1250
    },
    {
      "epoch": 1.3045738045738045,
      "grad_norm": 2.7872188091278076,
      "learning_rate": 4.6168449705979956e-05,
      "loss": 0.2367,
      "num_input_tokens_seen": 504712,
      "step": 1255
    },
    {
      "epoch": 1.3097713097713097,
      "grad_norm": 1.930995225906372,
      "learning_rate": 4.612004991387843e-05,
      "loss": 0.3177,
      "num_input_tokens_seen": 506696,
      "step": 1260
    },
    {
      "epoch": 1.314968814968815,
      "grad_norm": 0.9595280289649963,
      "learning_rate": 4.6071372048633566e-05,
      "loss": 0.2562,
      "num_input_tokens_seen": 508680,
      "step": 1265
    },
    {
      "epoch": 1.32016632016632,
      "grad_norm": 1.0570541620254517,
      "learning_rate": 4.6022416751153255e-05,
      "loss": 0.291,
      "num_input_tokens_seen": 510728,
      "step": 1270
    },
    {
      "epoch": 1.3253638253638254,
      "grad_norm": 3.702086925506592,
      "learning_rate": 4.5973184665998186e-05,
      "loss": 0.2441,
      "num_input_tokens_seen": 512712,
      "step": 1275
    },
    {
      "epoch": 1.3305613305613306,
      "grad_norm": 0.28220054507255554,
      "learning_rate": 4.5923676441373287e-05,
      "loss": 0.2674,
      "num_input_tokens_seen": 514696,
      "step": 1280
    },
    {
      "epoch": 1.3357588357588357,
      "grad_norm": 4.1724724769592285,
      "learning_rate": 4.5873892729119225e-05,
      "loss": 0.2628,
      "num_input_tokens_seen": 516808,
      "step": 1285
    },
    {
      "epoch": 1.340956340956341,
      "grad_norm": 5.697893142700195,
      "learning_rate": 4.582383418470386e-05,
      "loss": 0.208,
      "num_input_tokens_seen": 518792,
      "step": 1290
    },
    {
      "epoch": 1.3461538461538463,
      "grad_norm": 14.32170581817627,
      "learning_rate": 4.577350146721353e-05,
      "loss": 0.2791,
      "num_input_tokens_seen": 520840,
      "step": 1295
    },
    {
      "epoch": 1.3513513513513513,
      "grad_norm": 5.778555393218994,
      "learning_rate": 4.5722895239344435e-05,
      "loss": 0.2367,
      "num_input_tokens_seen": 522760,
      "step": 1300
    },
    {
      "epoch": 1.3565488565488566,
      "grad_norm": 10.224639892578125,
      "learning_rate": 4.567201616739393e-05,
      "loss": 0.2853,
      "num_input_tokens_seen": 524872,
      "step": 1305
    },
    {
      "epoch": 1.3617463617463619,
      "grad_norm": 10.455507278442383,
      "learning_rate": 4.562086492125167e-05,
      "loss": 0.2922,
      "num_input_tokens_seen": 526920,
      "step": 1310
    },
    {
      "epoch": 1.366943866943867,
      "grad_norm": 1.0684677362442017,
      "learning_rate": 4.556944217439088e-05,
      "loss": 0.2892,
      "num_input_tokens_seen": 528968,
      "step": 1315
    },
    {
      "epoch": 1.3721413721413722,
      "grad_norm": 8.995451927185059,
      "learning_rate": 4.5517748603859435e-05,
      "loss": 0.2689,
      "num_input_tokens_seen": 530888,
      "step": 1320
    },
    {
      "epoch": 1.3773388773388773,
      "grad_norm": 9.848129272460938,
      "learning_rate": 4.546578489027095e-05,
      "loss": 0.2348,
      "num_input_tokens_seen": 532872,
      "step": 1325
    },
    {
      "epoch": 1.3825363825363826,
      "grad_norm": 12.437982559204102,
      "learning_rate": 4.541355171779582e-05,
      "loss": 0.2971,
      "num_input_tokens_seen": 534920,
      "step": 1330
    },
    {
      "epoch": 1.3877338877338876,
      "grad_norm": 4.1738667488098145,
      "learning_rate": 4.5361049774152256e-05,
      "loss": 0.2582,
      "num_input_tokens_seen": 536840,
      "step": 1335
    },
    {
      "epoch": 1.392931392931393,
      "grad_norm": 11.655598640441895,
      "learning_rate": 4.530827975059715e-05,
      "loss": 0.2788,
      "num_input_tokens_seen": 538760,
      "step": 1340
    },
    {
      "epoch": 1.3981288981288982,
      "grad_norm": 16.621856689453125,
      "learning_rate": 4.5255242341917055e-05,
      "loss": 0.2367,
      "num_input_tokens_seen": 540680,
      "step": 1345
    },
    {
      "epoch": 1.4033264033264032,
      "grad_norm": 13.314364433288574,
      "learning_rate": 4.5201938246418976e-05,
      "loss": 0.2864,
      "num_input_tokens_seen": 542664,
      "step": 1350
    },
    {
      "epoch": 1.4085239085239085,
      "grad_norm": 0.3170285224914551,
      "learning_rate": 4.51483681659212e-05,
      "loss": 0.1771,
      "num_input_tokens_seen": 544712,
      "step": 1355
    },
    {
      "epoch": 1.4137214137214138,
      "grad_norm": 14.410113334655762,
      "learning_rate": 4.509453280574407e-05,
      "loss": 0.3864,
      "num_input_tokens_seen": 546824,
      "step": 1360
    },
    {
      "epoch": 1.4189189189189189,
      "grad_norm": 9.504006385803223,
      "learning_rate": 4.504043287470068e-05,
      "loss": 0.2952,
      "num_input_tokens_seen": 548936,
      "step": 1365
    },
    {
      "epoch": 1.4241164241164241,
      "grad_norm": 19.96672248840332,
      "learning_rate": 4.498606908508754e-05,
      "loss": 0.3433,
      "num_input_tokens_seen": 550920,
      "step": 1370
    },
    {
      "epoch": 1.4293139293139294,
      "grad_norm": 3.9341182708740234,
      "learning_rate": 4.4931442152675185e-05,
      "loss": 0.2757,
      "num_input_tokens_seen": 552904,
      "step": 1375
    },
    {
      "epoch": 1.4345114345114345,
      "grad_norm": 13.74646282196045,
      "learning_rate": 4.487655279669881e-05,
      "loss": 0.3025,
      "num_input_tokens_seen": 554824,
      "step": 1380
    },
    {
      "epoch": 1.4397089397089398,
      "grad_norm": 2.631537914276123,
      "learning_rate": 4.482140173984875e-05,
      "loss": 0.2663,
      "num_input_tokens_seen": 556872,
      "step": 1385
    },
    {
      "epoch": 1.444906444906445,
      "grad_norm": 6.440622329711914,
      "learning_rate": 4.476598970826094e-05,
      "loss": 0.2717,
      "num_input_tokens_seen": 558984,
      "step": 1390
    },
    {
      "epoch": 1.45010395010395,
      "grad_norm": 17.752002716064453,
      "learning_rate": 4.4710317431507434e-05,
      "loss": 0.2791,
      "num_input_tokens_seen": 560968,
      "step": 1395
    },
    {
      "epoch": 1.4553014553014554,
      "grad_norm": 5.554246425628662,
      "learning_rate": 4.465438564258673e-05,
      "loss": 0.2617,
      "num_input_tokens_seen": 562952,
      "step": 1400
    },
    {
      "epoch": 1.4604989604989604,
      "grad_norm": 5.866454601287842,
      "learning_rate": 4.4598195077914145e-05,
      "loss": 0.2452,
      "num_input_tokens_seen": 565064,
      "step": 1405
    },
    {
      "epoch": 1.4656964656964657,
      "grad_norm": 3.3153018951416016,
      "learning_rate": 4.454174647731213e-05,
      "loss": 0.2761,
      "num_input_tokens_seen": 567112,
      "step": 1410
    },
    {
      "epoch": 1.4708939708939708,
      "grad_norm": 20.00400733947754,
      "learning_rate": 4.4485040584000514e-05,
      "loss": 0.2875,
      "num_input_tokens_seen": 569160,
      "step": 1415
    },
    {
      "epoch": 1.476091476091476,
      "grad_norm": 12.827857971191406,
      "learning_rate": 4.442807814458672e-05,
      "loss": 0.2782,
      "num_input_tokens_seen": 571336,
      "step": 1420
    },
    {
      "epoch": 1.4812889812889813,
      "grad_norm": 1.1996235847473145,
      "learning_rate": 4.437085990905591e-05,
      "loss": 0.25,
      "num_input_tokens_seen": 573384,
      "step": 1425
    },
    {
      "epoch": 1.4864864864864864,
      "grad_norm": 1.4895740747451782,
      "learning_rate": 4.431338663076119e-05,
      "loss": 0.2596,
      "num_input_tokens_seen": 575304,
      "step": 1430
    },
    {
      "epoch": 1.4916839916839917,
      "grad_norm": 2.120258331298828,
      "learning_rate": 4.4255659066413595e-05,
      "loss": 0.2788,
      "num_input_tokens_seen": 577160,
      "step": 1435
    },
    {
      "epoch": 1.496881496881497,
      "grad_norm": 1.543442726135254,
      "learning_rate": 4.419767797607219e-05,
      "loss": 0.2892,
      "num_input_tokens_seen": 579208,
      "step": 1440
    },
    {
      "epoch": 1.502079002079002,
      "grad_norm": 7.000278472900391,
      "learning_rate": 4.413944412313405e-05,
      "loss": 0.2936,
      "num_input_tokens_seen": 581256,
      "step": 1445
    },
    {
      "epoch": 1.503118503118503,
      "eval_loss": 0.25699949264526367,
      "eval_runtime": 1.6561,
      "eval_samples_per_second": 516.892,
      "eval_steps_per_second": 64.612,
      "num_input_tokens_seen": 581704,
      "step": 1446
    },
    {
      "epoch": 1.5072765072765073,
      "grad_norm": 8.624284744262695,
      "learning_rate": 4.4080958274324155e-05,
      "loss": 0.2702,
      "num_input_tokens_seen": 583304,
      "step": 1450
    },
    {
      "epoch": 1.5124740124740126,
      "grad_norm": 1.7187201976776123,
      "learning_rate": 4.40222211996854e-05,
      "loss": 0.2252,
      "num_input_tokens_seen": 585224,
      "step": 1455
    },
    {
      "epoch": 1.5176715176715176,
      "grad_norm": 13.548686981201172,
      "learning_rate": 4.396323367256836e-05,
      "loss": 0.4066,
      "num_input_tokens_seen": 587272,
      "step": 1460
    },
    {
      "epoch": 1.5228690228690227,
      "grad_norm": 4.836349010467529,
      "learning_rate": 4.390399646962117e-05,
      "loss": 0.2413,
      "num_input_tokens_seen": 589320,
      "step": 1465
    },
    {
      "epoch": 1.5280665280665282,
      "grad_norm": 16.470077514648438,
      "learning_rate": 4.384451037077924e-05,
      "loss": 0.2593,
      "num_input_tokens_seen": 591304,
      "step": 1470
    },
    {
      "epoch": 1.5332640332640333,
      "grad_norm": 19.7336483001709,
      "learning_rate": 4.378477615925505e-05,
      "loss": 0.2499,
      "num_input_tokens_seen": 593224,
      "step": 1475
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 11.778541564941406,
      "learning_rate": 4.372479462152781e-05,
      "loss": 0.2672,
      "num_input_tokens_seen": 595336,
      "step": 1480
    },
    {
      "epoch": 1.5436590436590436,
      "grad_norm": 10.281831741333008,
      "learning_rate": 4.366456654733308e-05,
      "loss": 0.2898,
      "num_input_tokens_seen": 597256,
      "step": 1485
    },
    {
      "epoch": 1.5488565488565489,
      "grad_norm": 1.1301134824752808,
      "learning_rate": 4.360409272965242e-05,
      "loss": 0.2852,
      "num_input_tokens_seen": 599304,
      "step": 1490
    },
    {
      "epoch": 1.554054054054054,
      "grad_norm": 13.930249214172363,
      "learning_rate": 4.3543373964702907e-05,
      "loss": 0.2828,
      "num_input_tokens_seen": 601288,
      "step": 1495
    },
    {
      "epoch": 1.5592515592515592,
      "grad_norm": 6.923305988311768,
      "learning_rate": 4.348241105192668e-05,
      "loss": 0.2597,
      "num_input_tokens_seen": 603272,
      "step": 1500
    },
    {
      "epoch": 1.5644490644490645,
      "grad_norm": 6.9845356941223145,
      "learning_rate": 4.34212047939804e-05,
      "loss": 0.2584,
      "num_input_tokens_seen": 605256,
      "step": 1505
    },
    {
      "epoch": 1.5696465696465696,
      "grad_norm": 14.034126281738281,
      "learning_rate": 4.335975599672469e-05,
      "loss": 0.2713,
      "num_input_tokens_seen": 607304,
      "step": 1510
    },
    {
      "epoch": 1.5748440748440748,
      "grad_norm": 1.7358630895614624,
      "learning_rate": 4.329806546921353e-05,
      "loss": 0.2702,
      "num_input_tokens_seen": 609224,
      "step": 1515
    },
    {
      "epoch": 1.5800415800415801,
      "grad_norm": 7.067193031311035,
      "learning_rate": 4.323613402368357e-05,
      "loss": 0.2648,
      "num_input_tokens_seen": 611336,
      "step": 1520
    },
    {
      "epoch": 1.5852390852390852,
      "grad_norm": 2.2806382179260254,
      "learning_rate": 4.317396247554347e-05,
      "loss": 0.2879,
      "num_input_tokens_seen": 613320,
      "step": 1525
    },
    {
      "epoch": 1.5904365904365905,
      "grad_norm": 7.832094669342041,
      "learning_rate": 4.311155164336318e-05,
      "loss": 0.2953,
      "num_input_tokens_seen": 615176,
      "step": 1530
    },
    {
      "epoch": 1.5956340956340958,
      "grad_norm": 5.03207540512085,
      "learning_rate": 4.3048902348863116e-05,
      "loss": 0.2754,
      "num_input_tokens_seen": 617224,
      "step": 1535
    },
    {
      "epoch": 1.6008316008316008,
      "grad_norm": 9.225286483764648,
      "learning_rate": 4.298601541690336e-05,
      "loss": 0.2785,
      "num_input_tokens_seen": 619208,
      "step": 1540
    },
    {
      "epoch": 1.6060291060291059,
      "grad_norm": 8.74346923828125,
      "learning_rate": 4.292289167547281e-05,
      "loss": 0.278,
      "num_input_tokens_seen": 621192,
      "step": 1545
    },
    {
      "epoch": 1.6112266112266114,
      "grad_norm": 7.021914005279541,
      "learning_rate": 4.285953195567827e-05,
      "loss": 0.2618,
      "num_input_tokens_seen": 623176,
      "step": 1550
    },
    {
      "epoch": 1.6164241164241164,
      "grad_norm": 1.7642489671707153,
      "learning_rate": 4.2795937091733515e-05,
      "loss": 0.2506,
      "num_input_tokens_seen": 625160,
      "step": 1555
    },
    {
      "epoch": 1.6216216216216215,
      "grad_norm": 19.76347541809082,
      "learning_rate": 4.27321079209483e-05,
      "loss": 0.3095,
      "num_input_tokens_seen": 627144,
      "step": 1560
    },
    {
      "epoch": 1.6268191268191268,
      "grad_norm": 3.08026385307312,
      "learning_rate": 4.266804528371732e-05,
      "loss": 0.2951,
      "num_input_tokens_seen": 629192,
      "step": 1565
    },
    {
      "epoch": 1.632016632016632,
      "grad_norm": 6.732242584228516,
      "learning_rate": 4.260375002350917e-05,
      "loss": 0.2796,
      "num_input_tokens_seen": 631240,
      "step": 1570
    },
    {
      "epoch": 1.637214137214137,
      "grad_norm": 9.543598175048828,
      "learning_rate": 4.253922298685525e-05,
      "loss": 0.2407,
      "num_input_tokens_seen": 633224,
      "step": 1575
    },
    {
      "epoch": 1.6424116424116424,
      "grad_norm": 12.409037590026855,
      "learning_rate": 4.247446502333858e-05,
      "loss": 0.2386,
      "num_input_tokens_seen": 635208,
      "step": 1580
    },
    {
      "epoch": 1.6476091476091477,
      "grad_norm": 1.7029387950897217,
      "learning_rate": 4.2409476985582644e-05,
      "loss": 0.2872,
      "num_input_tokens_seen": 637256,
      "step": 1585
    },
    {
      "epoch": 1.6528066528066527,
      "grad_norm": 8.404769897460938,
      "learning_rate": 4.234425972924014e-05,
      "loss": 0.2806,
      "num_input_tokens_seen": 639176,
      "step": 1590
    },
    {
      "epoch": 1.658004158004158,
      "grad_norm": 3.5066511631011963,
      "learning_rate": 4.227881411298175e-05,
      "loss": 0.2715,
      "num_input_tokens_seen": 641224,
      "step": 1595
    },
    {
      "epoch": 1.6632016632016633,
      "grad_norm": 0.08306021988391876,
      "learning_rate": 4.221314099848481e-05,
      "loss": 0.2872,
      "num_input_tokens_seen": 643144,
      "step": 1600
    },
    {
      "epoch": 1.6683991683991684,
      "grad_norm": 3.17313551902771,
      "learning_rate": 4.2147241250421944e-05,
      "loss": 0.2204,
      "num_input_tokens_seen": 644936,
      "step": 1605
    },
    {
      "epoch": 1.6735966735966736,
      "grad_norm": 7.495476722717285,
      "learning_rate": 4.208111573644975e-05,
      "loss": 0.2557,
      "num_input_tokens_seen": 646984,
      "step": 1610
    },
    {
      "epoch": 1.678794178794179,
      "grad_norm": 11.988297462463379,
      "learning_rate": 4.201476532719728e-05,
      "loss": 0.2777,
      "num_input_tokens_seen": 649032,
      "step": 1615
    },
    {
      "epoch": 1.683991683991684,
      "grad_norm": 9.61047649383545,
      "learning_rate": 4.194819089625466e-05,
      "loss": 0.2778,
      "num_input_tokens_seen": 651080,
      "step": 1620
    },
    {
      "epoch": 1.689189189189189,
      "grad_norm": 12.550249099731445,
      "learning_rate": 4.188139332016154e-05,
      "loss": 0.2953,
      "num_input_tokens_seen": 653000,
      "step": 1625
    },
    {
      "epoch": 1.6943866943866945,
      "grad_norm": 9.836450576782227,
      "learning_rate": 4.1814373478395586e-05,
      "loss": 0.2955,
      "num_input_tokens_seen": 654920,
      "step": 1630
    },
    {
      "epoch": 1.6995841995841996,
      "grad_norm": 4.831801414489746,
      "learning_rate": 4.174713225336086e-05,
      "loss": 0.2599,
      "num_input_tokens_seen": 656904,
      "step": 1635
    },
    {
      "epoch": 1.7047817047817047,
      "grad_norm": 23.776752471923828,
      "learning_rate": 4.1679670530376244e-05,
      "loss": 0.2986,
      "num_input_tokens_seen": 658952,
      "step": 1640
    },
    {
      "epoch": 1.70997920997921,
      "grad_norm": 8.714804649353027,
      "learning_rate": 4.161198919766375e-05,
      "loss": 0.264,
      "num_input_tokens_seen": 660872,
      "step": 1645
    },
    {
      "epoch": 1.7151767151767152,
      "grad_norm": 7.620249271392822,
      "learning_rate": 4.154408914633685e-05,
      "loss": 0.2337,
      "num_input_tokens_seen": 662856,
      "step": 1650
    },
    {
      "epoch": 1.7203742203742203,
      "grad_norm": 12.147835731506348,
      "learning_rate": 4.147597127038873e-05,
      "loss": 0.2968,
      "num_input_tokens_seen": 664904,
      "step": 1655
    },
    {
      "epoch": 1.7255717255717256,
      "grad_norm": 4.487679481506348,
      "learning_rate": 4.140763646668052e-05,
      "loss": 0.2433,
      "num_input_tokens_seen": 666888,
      "step": 1660
    },
    {
      "epoch": 1.7307692307692308,
      "grad_norm": 4.300174236297607,
      "learning_rate": 4.1339085634929485e-05,
      "loss": 0.3127,
      "num_input_tokens_seen": 668936,
      "step": 1665
    },
    {
      "epoch": 1.735966735966736,
      "grad_norm": 5.227227687835693,
      "learning_rate": 4.12703196776972e-05,
      "loss": 0.2454,
      "num_input_tokens_seen": 670856,
      "step": 1670
    },
    {
      "epoch": 1.7411642411642412,
      "grad_norm": 3.5631868839263916,
      "learning_rate": 4.120133950037763e-05,
      "loss": 0.3639,
      "num_input_tokens_seen": 672840,
      "step": 1675
    },
    {
      "epoch": 1.7463617463617465,
      "grad_norm": 3.815650701522827,
      "learning_rate": 4.113214601118524e-05,
      "loss": 0.2468,
      "num_input_tokens_seen": 674824,
      "step": 1680
    },
    {
      "epoch": 1.7515592515592515,
      "grad_norm": 8.337157249450684,
      "learning_rate": 4.1062740121143016e-05,
      "loss": 0.2705,
      "num_input_tokens_seen": 676808,
      "step": 1685
    },
    {
      "epoch": 1.7536382536382535,
      "eval_loss": 0.25601524114608765,
      "eval_runtime": 1.0768,
      "eval_samples_per_second": 794.918,
      "eval_steps_per_second": 99.365,
      "num_input_tokens_seen": 677576,
      "step": 1687
    },
    {
      "epoch": 1.7567567567567568,
      "grad_norm": 5.897716999053955,
      "learning_rate": 4.099312274407048e-05,
      "loss": 0.2969,
      "num_input_tokens_seen": 678728,
      "step": 1690
    },
    {
      "epoch": 1.761954261954262,
      "grad_norm": 8.0389404296875,
      "learning_rate": 4.0923294796571676e-05,
      "loss": 0.2874,
      "num_input_tokens_seen": 680776,
      "step": 1695
    },
    {
      "epoch": 1.7671517671517671,
      "grad_norm": 8.596820831298828,
      "learning_rate": 4.085325719802307e-05,
      "loss": 0.2651,
      "num_input_tokens_seen": 683016,
      "step": 1700
    },
    {
      "epoch": 1.7723492723492722,
      "grad_norm": 9.364229202270508,
      "learning_rate": 4.078301087056144e-05,
      "loss": 0.2924,
      "num_input_tokens_seen": 685256,
      "step": 1705
    },
    {
      "epoch": 1.7775467775467777,
      "grad_norm": 17.756505966186523,
      "learning_rate": 4.0712556739071795e-05,
      "loss": 0.2762,
      "num_input_tokens_seen": 687304,
      "step": 1710
    },
    {
      "epoch": 1.7827442827442828,
      "grad_norm": 16.004009246826172,
      "learning_rate": 4.064189573117512e-05,
      "loss": 0.2888,
      "num_input_tokens_seen": 689224,
      "step": 1715
    },
    {
      "epoch": 1.7879417879417878,
      "grad_norm": 15.74494457244873,
      "learning_rate": 4.0571028777216214e-05,
      "loss": 0.2282,
      "num_input_tokens_seen": 691400,
      "step": 1720
    },
    {
      "epoch": 1.793139293139293,
      "grad_norm": 2.8662827014923096,
      "learning_rate": 4.049995681025143e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 693320,
      "step": 1725
    },
    {
      "epoch": 1.7983367983367984,
      "grad_norm": 8.626184463500977,
      "learning_rate": 4.0428680766036384e-05,
      "loss": 0.4406,
      "num_input_tokens_seen": 695432,
      "step": 1730
    },
    {
      "epoch": 1.8035343035343034,
      "grad_norm": 3.6228630542755127,
      "learning_rate": 4.035720158301363e-05,
      "loss": 0.3552,
      "num_input_tokens_seen": 697544,
      "step": 1735
    },
    {
      "epoch": 1.8087318087318087,
      "grad_norm": 34.01264953613281,
      "learning_rate": 4.028552020230031e-05,
      "loss": 0.3263,
      "num_input_tokens_seen": 699592,
      "step": 1740
    },
    {
      "epoch": 1.813929313929314,
      "grad_norm": 5.828328609466553,
      "learning_rate": 4.0213637567675774e-05,
      "loss": 0.2859,
      "num_input_tokens_seen": 701576,
      "step": 1745
    },
    {
      "epoch": 1.819126819126819,
      "grad_norm": 6.004729747772217,
      "learning_rate": 4.0141554625569125e-05,
      "loss": 0.2657,
      "num_input_tokens_seen": 703688,
      "step": 1750
    },
    {
      "epoch": 1.8243243243243243,
      "grad_norm": 4.201369285583496,
      "learning_rate": 4.0069272325046816e-05,
      "loss": 0.2842,
      "num_input_tokens_seen": 705736,
      "step": 1755
    },
    {
      "epoch": 1.8295218295218296,
      "grad_norm": 12.624285697937012,
      "learning_rate": 3.999679161780005e-05,
      "loss": 0.2479,
      "num_input_tokens_seen": 707720,
      "step": 1760
    },
    {
      "epoch": 1.8347193347193347,
      "grad_norm": 4.109714031219482,
      "learning_rate": 3.99241134581324e-05,
      "loss": 0.3132,
      "num_input_tokens_seen": 709896,
      "step": 1765
    },
    {
      "epoch": 1.83991683991684,
      "grad_norm": 5.0265302658081055,
      "learning_rate": 3.985123880294708e-05,
      "loss": 0.2661,
      "num_input_tokens_seen": 711944,
      "step": 1770
    },
    {
      "epoch": 1.8451143451143452,
      "grad_norm": 6.782411575317383,
      "learning_rate": 3.9778168611734456e-05,
      "loss": 0.2664,
      "num_input_tokens_seen": 713992,
      "step": 1775
    },
    {
      "epoch": 1.8503118503118503,
      "grad_norm": 1.0791548490524292,
      "learning_rate": 3.970490384655939e-05,
      "loss": 0.2443,
      "num_input_tokens_seen": 715976,
      "step": 1780
    },
    {
      "epoch": 1.8555093555093554,
      "grad_norm": 8.595144271850586,
      "learning_rate": 3.963144547204856e-05,
      "loss": 0.2659,
      "num_input_tokens_seen": 718024,
      "step": 1785
    },
    {
      "epoch": 1.8607068607068609,
      "grad_norm": 6.319913387298584,
      "learning_rate": 3.955779445537776e-05,
      "loss": 0.2441,
      "num_input_tokens_seen": 720072,
      "step": 1790
    },
    {
      "epoch": 1.865904365904366,
      "grad_norm": 16.376117706298828,
      "learning_rate": 3.948395176625918e-05,
      "loss": 0.279,
      "num_input_tokens_seen": 722120,
      "step": 1795
    },
    {
      "epoch": 1.871101871101871,
      "grad_norm": 12.614494323730469,
      "learning_rate": 3.9409918376928604e-05,
      "loss": 0.2851,
      "num_input_tokens_seen": 724168,
      "step": 1800
    },
    {
      "epoch": 1.8762993762993763,
      "grad_norm": 0.3612583577632904,
      "learning_rate": 3.933569526213268e-05,
      "loss": 0.2928,
      "num_input_tokens_seen": 726280,
      "step": 1805
    },
    {
      "epoch": 1.8814968814968815,
      "grad_norm": 5.5901103019714355,
      "learning_rate": 3.926128339911599e-05,
      "loss": 0.2677,
      "num_input_tokens_seen": 728264,
      "step": 1810
    },
    {
      "epoch": 1.8866943866943866,
      "grad_norm": 19.419448852539062,
      "learning_rate": 3.918668376760827e-05,
      "loss": 0.2924,
      "num_input_tokens_seen": 730312,
      "step": 1815
    },
    {
      "epoch": 1.8918918918918919,
      "grad_norm": 19.855844497680664,
      "learning_rate": 3.9111897349811454e-05,
      "loss": 0.2771,
      "num_input_tokens_seen": 732296,
      "step": 1820
    },
    {
      "epoch": 1.8970893970893972,
      "grad_norm": 0.2629048228263855,
      "learning_rate": 3.903692513038677e-05,
      "loss": 0.2412,
      "num_input_tokens_seen": 734088,
      "step": 1825
    },
    {
      "epoch": 1.9022869022869022,
      "grad_norm": 10.053337097167969,
      "learning_rate": 3.896176809644178e-05,
      "loss": 0.2897,
      "num_input_tokens_seen": 736072,
      "step": 1830
    },
    {
      "epoch": 1.9074844074844075,
      "grad_norm": 14.880932807922363,
      "learning_rate": 3.8886427237517344e-05,
      "loss": 0.3063,
      "num_input_tokens_seen": 738120,
      "step": 1835
    },
    {
      "epoch": 1.9126819126819128,
      "grad_norm": 7.023550510406494,
      "learning_rate": 3.881090354557463e-05,
      "loss": 0.3038,
      "num_input_tokens_seen": 740168,
      "step": 1840
    },
    {
      "epoch": 1.9178794178794178,
      "grad_norm": 14.33624267578125,
      "learning_rate": 3.8735198014982064e-05,
      "loss": 0.2716,
      "num_input_tokens_seen": 742280,
      "step": 1845
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 3.6817572116851807,
      "learning_rate": 3.865931164250219e-05,
      "loss": 0.2834,
      "num_input_tokens_seen": 744328,
      "step": 1850
    },
    {
      "epoch": 1.9282744282744284,
      "grad_norm": 6.452430248260498,
      "learning_rate": 3.8583245427278584e-05,
      "loss": 0.2845,
      "num_input_tokens_seen": 746440,
      "step": 1855
    },
    {
      "epoch": 1.9334719334719335,
      "grad_norm": 5.071720123291016,
      "learning_rate": 3.850700037082268e-05,
      "loss": 0.3004,
      "num_input_tokens_seen": 748488,
      "step": 1860
    },
    {
      "epoch": 1.9386694386694385,
      "grad_norm": 7.624428749084473,
      "learning_rate": 3.8430577477000595e-05,
      "loss": 0.2696,
      "num_input_tokens_seen": 750344,
      "step": 1865
    },
    {
      "epoch": 1.943866943866944,
      "grad_norm": 11.881917953491211,
      "learning_rate": 3.835397775201991e-05,
      "loss": 0.2567,
      "num_input_tokens_seen": 752328,
      "step": 1870
    },
    {
      "epoch": 1.949064449064449,
      "grad_norm": 3.7098724842071533,
      "learning_rate": 3.827720220441642e-05,
      "loss": 0.269,
      "num_input_tokens_seen": 754312,
      "step": 1875
    },
    {
      "epoch": 1.9542619542619541,
      "grad_norm": 8.87547492980957,
      "learning_rate": 3.8200251845040855e-05,
      "loss": 0.2816,
      "num_input_tokens_seen": 756232,
      "step": 1880
    },
    {
      "epoch": 1.9594594594594594,
      "grad_norm": 4.694116592407227,
      "learning_rate": 3.812312768704557e-05,
      "loss": 0.2706,
      "num_input_tokens_seen": 758280,
      "step": 1885
    },
    {
      "epoch": 1.9646569646569647,
      "grad_norm": 0.5913885235786438,
      "learning_rate": 3.8045830745871195e-05,
      "loss": 0.2412,
      "num_input_tokens_seen": 760328,
      "step": 1890
    },
    {
      "epoch": 1.9698544698544698,
      "grad_norm": 0.12654490768909454,
      "learning_rate": 3.7968362039233316e-05,
      "loss": 0.2593,
      "num_input_tokens_seen": 762248,
      "step": 1895
    },
    {
      "epoch": 1.975051975051975,
      "grad_norm": 25.26936149597168,
      "learning_rate": 3.789072258710898e-05,
      "loss": 0.2765,
      "num_input_tokens_seen": 764168,
      "step": 1900
    },
    {
      "epoch": 1.9802494802494803,
      "grad_norm": 13.081347465515137,
      "learning_rate": 3.781291341172338e-05,
      "loss": 0.2703,
      "num_input_tokens_seen": 766216,
      "step": 1905
    },
    {
      "epoch": 1.9854469854469854,
      "grad_norm": 5.589529037475586,
      "learning_rate": 3.7734935537536276e-05,
      "loss": 0.2418,
      "num_input_tokens_seen": 768264,
      "step": 1910
    },
    {
      "epoch": 1.9906444906444907,
      "grad_norm": 10.241774559020996,
      "learning_rate": 3.7656789991228636e-05,
      "loss": 0.2502,
      "num_input_tokens_seen": 770184,
      "step": 1915
    },
    {
      "epoch": 1.995841995841996,
      "grad_norm": 2.5858964920043945,
      "learning_rate": 3.7578477801689e-05,
      "loss": 0.2432,
      "num_input_tokens_seen": 772168,
      "step": 1920
    },
    {
      "epoch": 2.001039501039501,
      "grad_norm": 3.724403142929077,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.2243,
      "num_input_tokens_seen": 774160,
      "step": 1925
    },
    {
      "epoch": 2.004158004158004,
      "eval_loss": 0.2575376331806183,
      "eval_runtime": 1.0513,
      "eval_samples_per_second": 814.265,
      "eval_steps_per_second": 101.783,
      "num_input_tokens_seen": 775312,
      "step": 1928
    },
    {
      "epoch": 2.006237006237006,
      "grad_norm": 8.0248384475708,
      "learning_rate": 3.742135761942479e-05,
      "loss": 0.26,
      "num_input_tokens_seen": 776144,
      "step": 1930
    },
    {
      "epoch": 2.0114345114345116,
      "grad_norm": 8.558785438537598,
      "learning_rate": 3.734255169539337e-05,
      "loss": 0.2814,
      "num_input_tokens_seen": 778128,
      "step": 1935
    },
    {
      "epoch": 2.0166320166320166,
      "grad_norm": 29.213605880737305,
      "learning_rate": 3.7263583265489074e-05,
      "loss": 0.2911,
      "num_input_tokens_seen": 780176,
      "step": 1940
    },
    {
      "epoch": 2.0218295218295217,
      "grad_norm": 2.797147274017334,
      "learning_rate": 3.718445336943478e-05,
      "loss": 0.2723,
      "num_input_tokens_seen": 782160,
      "step": 1945
    },
    {
      "epoch": 2.027027027027027,
      "grad_norm": 11.302480697631836,
      "learning_rate": 3.710516304907931e-05,
      "loss": 0.3159,
      "num_input_tokens_seen": 784208,
      "step": 1950
    },
    {
      "epoch": 2.0322245322245323,
      "grad_norm": 10.434552192687988,
      "learning_rate": 3.702571334838365e-05,
      "loss": 0.2713,
      "num_input_tokens_seen": 786256,
      "step": 1955
    },
    {
      "epoch": 2.0374220374220373,
      "grad_norm": 18.884937286376953,
      "learning_rate": 3.694610531340729e-05,
      "loss": 0.2491,
      "num_input_tokens_seen": 788240,
      "step": 1960
    },
    {
      "epoch": 2.042619542619543,
      "grad_norm": 7.0856404304504395,
      "learning_rate": 3.6866339992294344e-05,
      "loss": 0.2663,
      "num_input_tokens_seen": 790288,
      "step": 1965
    },
    {
      "epoch": 2.047817047817048,
      "grad_norm": 8.749947547912598,
      "learning_rate": 3.6786418435259854e-05,
      "loss": 0.2681,
      "num_input_tokens_seen": 792272,
      "step": 1970
    },
    {
      "epoch": 2.053014553014553,
      "grad_norm": 7.0498785972595215,
      "learning_rate": 3.670634169457587e-05,
      "loss": 0.2757,
      "num_input_tokens_seen": 794384,
      "step": 1975
    },
    {
      "epoch": 2.0582120582120584,
      "grad_norm": 1.3708738088607788,
      "learning_rate": 3.662611082455766e-05,
      "loss": 0.2727,
      "num_input_tokens_seen": 796368,
      "step": 1980
    },
    {
      "epoch": 2.0634095634095635,
      "grad_norm": 17.901782989501953,
      "learning_rate": 3.654572688154979e-05,
      "loss": 0.2711,
      "num_input_tokens_seen": 798480,
      "step": 1985
    },
    {
      "epoch": 2.0686070686070686,
      "grad_norm": 3.3067057132720947,
      "learning_rate": 3.646519092391227e-05,
      "loss": 0.2843,
      "num_input_tokens_seen": 800528,
      "step": 1990
    },
    {
      "epoch": 2.0738045738045736,
      "grad_norm": 10.808554649353027,
      "learning_rate": 3.6384504012006544e-05,
      "loss": 0.2917,
      "num_input_tokens_seen": 802768,
      "step": 1995
    },
    {
      "epoch": 2.079002079002079,
      "grad_norm": 15.788969039916992,
      "learning_rate": 3.6303667208181575e-05,
      "loss": 0.2846,
      "num_input_tokens_seen": 804752,
      "step": 2000
    },
    {
      "epoch": 2.084199584199584,
      "grad_norm": 7.7113847732543945,
      "learning_rate": 3.622268157675986e-05,
      "loss": 0.2932,
      "num_input_tokens_seen": 806672,
      "step": 2005
    },
    {
      "epoch": 2.0893970893970892,
      "grad_norm": 1.0900391340255737,
      "learning_rate": 3.614154818402339e-05,
      "loss": 0.2602,
      "num_input_tokens_seen": 808656,
      "step": 2010
    },
    {
      "epoch": 2.0945945945945947,
      "grad_norm": 8.559063911437988,
      "learning_rate": 3.606026809819966e-05,
      "loss": 0.2402,
      "num_input_tokens_seen": 810640,
      "step": 2015
    },
    {
      "epoch": 2.0997920997921,
      "grad_norm": 13.94715690612793,
      "learning_rate": 3.597884238944752e-05,
      "loss": 0.2832,
      "num_input_tokens_seen": 812688,
      "step": 2020
    },
    {
      "epoch": 2.104989604989605,
      "grad_norm": 10.943618774414062,
      "learning_rate": 3.5897272129843194e-05,
      "loss": 0.262,
      "num_input_tokens_seen": 814800,
      "step": 2025
    },
    {
      "epoch": 2.1101871101871104,
      "grad_norm": 8.191487312316895,
      "learning_rate": 3.581555839336606e-05,
      "loss": 0.2348,
      "num_input_tokens_seen": 816912,
      "step": 2030
    },
    {
      "epoch": 2.1153846153846154,
      "grad_norm": 8.54643440246582,
      "learning_rate": 3.57337022558846e-05,
      "loss": 0.2981,
      "num_input_tokens_seen": 818896,
      "step": 2035
    },
    {
      "epoch": 2.1205821205821205,
      "grad_norm": 0.22179275751113892,
      "learning_rate": 3.565170479514214e-05,
      "loss": 0.2857,
      "num_input_tokens_seen": 820880,
      "step": 2040
    },
    {
      "epoch": 2.125779625779626,
      "grad_norm": 8.135652542114258,
      "learning_rate": 3.5569567090742764e-05,
      "loss": 0.2745,
      "num_input_tokens_seen": 822864,
      "step": 2045
    },
    {
      "epoch": 2.130977130977131,
      "grad_norm": 1.2629114389419556,
      "learning_rate": 3.548729022413701e-05,
      "loss": 0.2705,
      "num_input_tokens_seen": 825040,
      "step": 2050
    },
    {
      "epoch": 2.136174636174636,
      "grad_norm": 2.5352609157562256,
      "learning_rate": 3.540487527860769e-05,
      "loss": 0.2397,
      "num_input_tokens_seen": 827024,
      "step": 2055
    },
    {
      "epoch": 2.141372141372141,
      "grad_norm": 5.771927356719971,
      "learning_rate": 3.53223233392556e-05,
      "loss": 0.2921,
      "num_input_tokens_seen": 829136,
      "step": 2060
    },
    {
      "epoch": 2.1465696465696467,
      "grad_norm": 14.170849800109863,
      "learning_rate": 3.523963549298525e-05,
      "loss": 0.274,
      "num_input_tokens_seen": 831184,
      "step": 2065
    },
    {
      "epoch": 2.1517671517671517,
      "grad_norm": 35.1840705871582,
      "learning_rate": 3.51568128284905e-05,
      "loss": 0.3599,
      "num_input_tokens_seen": 833168,
      "step": 2070
    },
    {
      "epoch": 2.156964656964657,
      "grad_norm": 3.5509274005889893,
      "learning_rate": 3.5073856436240334e-05,
      "loss": 0.2991,
      "num_input_tokens_seen": 835216,
      "step": 2075
    },
    {
      "epoch": 2.1621621621621623,
      "grad_norm": 11.091768264770508,
      "learning_rate": 3.499076740846438e-05,
      "loss": 0.2711,
      "num_input_tokens_seen": 837136,
      "step": 2080
    },
    {
      "epoch": 2.1673596673596673,
      "grad_norm": 10.117902755737305,
      "learning_rate": 3.490754683913863e-05,
      "loss": 0.263,
      "num_input_tokens_seen": 839120,
      "step": 2085
    },
    {
      "epoch": 2.1725571725571724,
      "grad_norm": 8.886628150939941,
      "learning_rate": 3.482419582397095e-05,
      "loss": 0.2114,
      "num_input_tokens_seen": 841104,
      "step": 2090
    },
    {
      "epoch": 2.177754677754678,
      "grad_norm": 3.9707517623901367,
      "learning_rate": 3.474071546038673e-05,
      "loss": 0.3437,
      "num_input_tokens_seen": 843152,
      "step": 2095
    },
    {
      "epoch": 2.182952182952183,
      "grad_norm": 7.44064474105835,
      "learning_rate": 3.46571068475144e-05,
      "loss": 0.2665,
      "num_input_tokens_seen": 845136,
      "step": 2100
    },
    {
      "epoch": 2.188149688149688,
      "grad_norm": 15.758981704711914,
      "learning_rate": 3.4573371086170936e-05,
      "loss": 0.2736,
      "num_input_tokens_seen": 847120,
      "step": 2105
    },
    {
      "epoch": 2.1933471933471935,
      "grad_norm": 6.210805892944336,
      "learning_rate": 3.4489509278847414e-05,
      "loss": 0.2967,
      "num_input_tokens_seen": 849168,
      "step": 2110
    },
    {
      "epoch": 2.1985446985446986,
      "grad_norm": 5.870239734649658,
      "learning_rate": 3.4405522529694454e-05,
      "loss": 0.2804,
      "num_input_tokens_seen": 851152,
      "step": 2115
    },
    {
      "epoch": 2.2037422037422036,
      "grad_norm": 0.2958710789680481,
      "learning_rate": 3.432141194450772e-05,
      "loss": 0.2789,
      "num_input_tokens_seen": 853008,
      "step": 2120
    },
    {
      "epoch": 2.208939708939709,
      "grad_norm": 2.236769199371338,
      "learning_rate": 3.4237178630713314e-05,
      "loss": 0.2593,
      "num_input_tokens_seen": 855120,
      "step": 2125
    },
    {
      "epoch": 2.214137214137214,
      "grad_norm": 13.610929489135742,
      "learning_rate": 3.415282369735324e-05,
      "loss": 0.3184,
      "num_input_tokens_seen": 857232,
      "step": 2130
    },
    {
      "epoch": 2.2193347193347193,
      "grad_norm": 23.174781799316406,
      "learning_rate": 3.4068348255070763e-05,
      "loss": 0.2697,
      "num_input_tokens_seen": 859344,
      "step": 2135
    },
    {
      "epoch": 2.2245322245322248,
      "grad_norm": 5.864615440368652,
      "learning_rate": 3.3983753416095845e-05,
      "loss": 0.2666,
      "num_input_tokens_seen": 861328,
      "step": 2140
    },
    {
      "epoch": 2.22972972972973,
      "grad_norm": 4.22444486618042,
      "learning_rate": 3.389904029423041e-05,
      "loss": 0.2807,
      "num_input_tokens_seen": 863376,
      "step": 2145
    },
    {
      "epoch": 2.234927234927235,
      "grad_norm": 10.534299850463867,
      "learning_rate": 3.381421000483378e-05,
      "loss": 0.277,
      "num_input_tokens_seen": 865424,
      "step": 2150
    },
    {
      "epoch": 2.24012474012474,
      "grad_norm": 19.86652946472168,
      "learning_rate": 3.37292636648079e-05,
      "loss": 0.2884,
      "num_input_tokens_seen": 867472,
      "step": 2155
    },
    {
      "epoch": 2.2453222453222454,
      "grad_norm": 1.5927962064743042,
      "learning_rate": 3.36442023925827e-05,
      "loss": 0.259,
      "num_input_tokens_seen": 869584,
      "step": 2160
    },
    {
      "epoch": 2.2505197505197505,
      "grad_norm": 13.930359840393066,
      "learning_rate": 3.3559027308101345e-05,
      "loss": 0.2477,
      "num_input_tokens_seen": 871568,
      "step": 2165
    },
    {
      "epoch": 2.2546777546777546,
      "eval_loss": 0.2923731505870819,
      "eval_runtime": 2.2896,
      "eval_samples_per_second": 373.867,
      "eval_steps_per_second": 46.733,
      "num_input_tokens_seen": 873104,
      "step": 2169
    },
    {
      "epoch": 2.2557172557172556,
      "grad_norm": 13.714078903198242,
      "learning_rate": 3.3473739532805467e-05,
      "loss": 0.3482,
      "num_input_tokens_seen": 873488,
      "step": 2170
    },
    {
      "epoch": 2.260914760914761,
      "grad_norm": 6.342438697814941,
      "learning_rate": 3.3388340189620424e-05,
      "loss": 0.26,
      "num_input_tokens_seen": 875472,
      "step": 2175
    },
    {
      "epoch": 2.266112266112266,
      "grad_norm": 13.85311222076416,
      "learning_rate": 3.330283040294053e-05,
      "loss": 0.2554,
      "num_input_tokens_seen": 877392,
      "step": 2180
    },
    {
      "epoch": 2.271309771309771,
      "grad_norm": 4.5571370124816895,
      "learning_rate": 3.321721129861422e-05,
      "loss": 0.2621,
      "num_input_tokens_seen": 879504,
      "step": 2185
    },
    {
      "epoch": 2.2765072765072767,
      "grad_norm": 12.646343231201172,
      "learning_rate": 3.3131484003929246e-05,
      "loss": 0.29,
      "num_input_tokens_seen": 881360,
      "step": 2190
    },
    {
      "epoch": 2.2817047817047817,
      "grad_norm": 2.4754276275634766,
      "learning_rate": 3.3045649647597815e-05,
      "loss": 0.2916,
      "num_input_tokens_seen": 883280,
      "step": 2195
    },
    {
      "epoch": 2.286902286902287,
      "grad_norm": 9.653176307678223,
      "learning_rate": 3.2959709359741744e-05,
      "loss": 0.2572,
      "num_input_tokens_seen": 885328,
      "step": 2200
    },
    {
      "epoch": 2.2920997920997923,
      "grad_norm": 16.368059158325195,
      "learning_rate": 3.2873664271877584e-05,
      "loss": 0.3062,
      "num_input_tokens_seen": 887312,
      "step": 2205
    },
    {
      "epoch": 2.2972972972972974,
      "grad_norm": 11.021875381469727,
      "learning_rate": 3.278751551690172e-05,
      "loss": 0.2672,
      "num_input_tokens_seen": 889296,
      "step": 2210
    },
    {
      "epoch": 2.3024948024948024,
      "grad_norm": 8.907588958740234,
      "learning_rate": 3.270126422907544e-05,
      "loss": 0.2647,
      "num_input_tokens_seen": 891408,
      "step": 2215
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 9.290825843811035,
      "learning_rate": 3.261491154401001e-05,
      "loss": 0.2884,
      "num_input_tokens_seen": 893392,
      "step": 2220
    },
    {
      "epoch": 2.312889812889813,
      "grad_norm": 4.591054916381836,
      "learning_rate": 3.2528458598651734e-05,
      "loss": 0.2719,
      "num_input_tokens_seen": 895440,
      "step": 2225
    },
    {
      "epoch": 2.318087318087318,
      "grad_norm": 14.931044578552246,
      "learning_rate": 3.244190653126696e-05,
      "loss": 0.2718,
      "num_input_tokens_seen": 897616,
      "step": 2230
    },
    {
      "epoch": 2.323284823284823,
      "grad_norm": 5.752501487731934,
      "learning_rate": 3.2355256481427145e-05,
      "loss": 0.2636,
      "num_input_tokens_seen": 899536,
      "step": 2235
    },
    {
      "epoch": 2.3284823284823286,
      "grad_norm": 10.545312881469727,
      "learning_rate": 3.226850958999375e-05,
      "loss": 0.2544,
      "num_input_tokens_seen": 901648,
      "step": 2240
    },
    {
      "epoch": 2.3336798336798337,
      "grad_norm": 0.47291961312294006,
      "learning_rate": 3.2181666999103324e-05,
      "loss": 0.2258,
      "num_input_tokens_seen": 903696,
      "step": 2245
    },
    {
      "epoch": 2.3388773388773387,
      "grad_norm": 12.981254577636719,
      "learning_rate": 3.209472985215243e-05,
      "loss": 0.3054,
      "num_input_tokens_seen": 905552,
      "step": 2250
    },
    {
      "epoch": 2.3440748440748442,
      "grad_norm": 6.684492588043213,
      "learning_rate": 3.2007699293782555e-05,
      "loss": 0.3446,
      "num_input_tokens_seen": 907472,
      "step": 2255
    },
    {
      "epoch": 2.3492723492723493,
      "grad_norm": 4.928536891937256,
      "learning_rate": 3.1920576469865115e-05,
      "loss": 0.2495,
      "num_input_tokens_seen": 909584,
      "step": 2260
    },
    {
      "epoch": 2.3544698544698544,
      "grad_norm": 6.033250331878662,
      "learning_rate": 3.183336252748627e-05,
      "loss": 0.2754,
      "num_input_tokens_seen": 911632,
      "step": 2265
    },
    {
      "epoch": 2.35966735966736,
      "grad_norm": 2.3870551586151123,
      "learning_rate": 3.1746058614931916e-05,
      "loss": 0.2604,
      "num_input_tokens_seen": 913616,
      "step": 2270
    },
    {
      "epoch": 2.364864864864865,
      "grad_norm": 18.541162490844727,
      "learning_rate": 3.16586658816725e-05,
      "loss": 0.2711,
      "num_input_tokens_seen": 915728,
      "step": 2275
    },
    {
      "epoch": 2.37006237006237,
      "grad_norm": 13.122420310974121,
      "learning_rate": 3.157118547834793e-05,
      "loss": 0.2566,
      "num_input_tokens_seen": 917776,
      "step": 2280
    },
    {
      "epoch": 2.375259875259875,
      "grad_norm": 12.414369583129883,
      "learning_rate": 3.148361855675237e-05,
      "loss": 0.2684,
      "num_input_tokens_seen": 919952,
      "step": 2285
    },
    {
      "epoch": 2.3804573804573805,
      "grad_norm": 7.519947052001953,
      "learning_rate": 3.139596626981916e-05,
      "loss": 0.2294,
      "num_input_tokens_seen": 921872,
      "step": 2290
    },
    {
      "epoch": 2.3856548856548856,
      "grad_norm": 8.675261497497559,
      "learning_rate": 3.130822977160554e-05,
      "loss": 0.2603,
      "num_input_tokens_seen": 923856,
      "step": 2295
    },
    {
      "epoch": 2.390852390852391,
      "grad_norm": 5.88022518157959,
      "learning_rate": 3.122041021727755e-05,
      "loss": 0.3942,
      "num_input_tokens_seen": 925968,
      "step": 2300
    },
    {
      "epoch": 2.396049896049896,
      "grad_norm": 14.03941535949707,
      "learning_rate": 3.1132508763094715e-05,
      "loss": 0.3128,
      "num_input_tokens_seen": 927888,
      "step": 2305
    },
    {
      "epoch": 2.401247401247401,
      "grad_norm": 15.206921577453613,
      "learning_rate": 3.104452656639492e-05,
      "loss": 0.2467,
      "num_input_tokens_seen": 929808,
      "step": 2310
    },
    {
      "epoch": 2.4064449064449063,
      "grad_norm": 1.0389469861984253,
      "learning_rate": 3.0956464785579124e-05,
      "loss": 0.1963,
      "num_input_tokens_seen": 931728,
      "step": 2315
    },
    {
      "epoch": 2.4116424116424118,
      "grad_norm": 9.624686241149902,
      "learning_rate": 3.0868324580096114e-05,
      "loss": 0.3533,
      "num_input_tokens_seen": 933840,
      "step": 2320
    },
    {
      "epoch": 2.416839916839917,
      "grad_norm": 1.1855783462524414,
      "learning_rate": 3.078010711042723e-05,
      "loss": 0.2936,
      "num_input_tokens_seen": 935824,
      "step": 2325
    },
    {
      "epoch": 2.422037422037422,
      "grad_norm": 4.064991474151611,
      "learning_rate": 3.0691813538071105e-05,
      "loss": 0.274,
      "num_input_tokens_seen": 937872,
      "step": 2330
    },
    {
      "epoch": 2.4272349272349274,
      "grad_norm": 26.271160125732422,
      "learning_rate": 3.0603445025528376e-05,
      "loss": 0.3378,
      "num_input_tokens_seen": 939984,
      "step": 2335
    },
    {
      "epoch": 2.4324324324324325,
      "grad_norm": 12.682053565979004,
      "learning_rate": 3.051500273628633e-05,
      "loss": 0.2418,
      "num_input_tokens_seen": 941968,
      "step": 2340
    },
    {
      "epoch": 2.4376299376299375,
      "grad_norm": 3.6713008880615234,
      "learning_rate": 3.0426487834803657e-05,
      "loss": 0.2943,
      "num_input_tokens_seen": 943952,
      "step": 2345
    },
    {
      "epoch": 2.442827442827443,
      "grad_norm": 19.096328735351562,
      "learning_rate": 3.0337901486495073e-05,
      "loss": 0.2435,
      "num_input_tokens_seen": 945872,
      "step": 2350
    },
    {
      "epoch": 2.448024948024948,
      "grad_norm": 4.218629360198975,
      "learning_rate": 3.0249244857715976e-05,
      "loss": 0.267,
      "num_input_tokens_seen": 947856,
      "step": 2355
    },
    {
      "epoch": 2.453222453222453,
      "grad_norm": 0.270999550819397,
      "learning_rate": 3.01605191157471e-05,
      "loss": 0.2452,
      "num_input_tokens_seen": 949840,
      "step": 2360
    },
    {
      "epoch": 2.4584199584199586,
      "grad_norm": 1.739176630973816,
      "learning_rate": 3.007172542877915e-05,
      "loss": 0.2342,
      "num_input_tokens_seen": 951760,
      "step": 2365
    },
    {
      "epoch": 2.4636174636174637,
      "grad_norm": 13.98282241821289,
      "learning_rate": 2.998286496589742e-05,
      "loss": 0.3294,
      "num_input_tokens_seen": 953680,
      "step": 2370
    },
    {
      "epoch": 2.4688149688149688,
      "grad_norm": 9.587896347045898,
      "learning_rate": 2.9893938897066393e-05,
      "loss": 0.2417,
      "num_input_tokens_seen": 955600,
      "step": 2375
    },
    {
      "epoch": 2.474012474012474,
      "grad_norm": 6.101572036743164,
      "learning_rate": 2.9804948393114324e-05,
      "loss": 0.2781,
      "num_input_tokens_seen": 957456,
      "step": 2380
    },
    {
      "epoch": 2.4792099792099793,
      "grad_norm": 5.505262851715088,
      "learning_rate": 2.9715894625717866e-05,
      "loss": 0.2721,
      "num_input_tokens_seen": 959504,
      "step": 2385
    },
    {
      "epoch": 2.4844074844074844,
      "grad_norm": 0.05719423666596413,
      "learning_rate": 2.9626778767386604e-05,
      "loss": 0.277,
      "num_input_tokens_seen": 961488,
      "step": 2390
    },
    {
      "epoch": 2.4896049896049894,
      "grad_norm": 17.471576690673828,
      "learning_rate": 2.953760199144764e-05,
      "loss": 0.2759,
      "num_input_tokens_seen": 963408,
      "step": 2395
    },
    {
      "epoch": 2.494802494802495,
      "grad_norm": 12.931529998779297,
      "learning_rate": 2.9448365472030115e-05,
      "loss": 0.2633,
      "num_input_tokens_seen": 965392,
      "step": 2400
    },
    {
      "epoch": 2.5,
      "grad_norm": 8.689949989318848,
      "learning_rate": 2.935907038404981e-05,
      "loss": 0.2744,
      "num_input_tokens_seen": 967440,
      "step": 2405
    },
    {
      "epoch": 2.505197505197505,
      "grad_norm": 1.9627736806869507,
      "learning_rate": 2.92697179031936e-05,
      "loss": 0.2379,
      "num_input_tokens_seen": 969360,
      "step": 2410
    },
    {
      "epoch": 2.505197505197505,
      "eval_loss": 0.2576568126678467,
      "eval_runtime": 1.0822,
      "eval_samples_per_second": 790.985,
      "eval_steps_per_second": 98.873,
      "num_input_tokens_seen": 969360,
      "step": 2410
    },
    {
      "epoch": 2.51039501039501,
      "grad_norm": 0.891659140586853,
      "learning_rate": 2.9180309205904027e-05,
      "loss": 0.2923,
      "num_input_tokens_seen": 971472,
      "step": 2415
    },
    {
      "epoch": 2.5155925155925156,
      "grad_norm": 2.514449119567871,
      "learning_rate": 2.9090845469363805e-05,
      "loss": 0.296,
      "num_input_tokens_seen": 973456,
      "step": 2420
    },
    {
      "epoch": 2.5207900207900207,
      "grad_norm": 19.587871551513672,
      "learning_rate": 2.9001327871480294e-05,
      "loss": 0.2911,
      "num_input_tokens_seen": 975504,
      "step": 2425
    },
    {
      "epoch": 2.525987525987526,
      "grad_norm": 16.2788143157959,
      "learning_rate": 2.8911757590870027e-05,
      "loss": 0.285,
      "num_input_tokens_seen": 977552,
      "step": 2430
    },
    {
      "epoch": 2.5311850311850312,
      "grad_norm": 21.12313461303711,
      "learning_rate": 2.8822135806843154e-05,
      "loss": 0.2552,
      "num_input_tokens_seen": 979536,
      "step": 2435
    },
    {
      "epoch": 2.5363825363825363,
      "grad_norm": 1.4399539232254028,
      "learning_rate": 2.8732463699387968e-05,
      "loss": 0.2906,
      "num_input_tokens_seen": 981584,
      "step": 2440
    },
    {
      "epoch": 2.5415800415800414,
      "grad_norm": 13.594979286193848,
      "learning_rate": 2.8642742449155284e-05,
      "loss": 0.2795,
      "num_input_tokens_seen": 983632,
      "step": 2445
    },
    {
      "epoch": 2.546777546777547,
      "grad_norm": 16.668123245239258,
      "learning_rate": 2.855297323744301e-05,
      "loss": 0.228,
      "num_input_tokens_seen": 985680,
      "step": 2450
    },
    {
      "epoch": 2.551975051975052,
      "grad_norm": 1.3556967973709106,
      "learning_rate": 2.8463157246180468e-05,
      "loss": 0.2414,
      "num_input_tokens_seen": 987664,
      "step": 2455
    },
    {
      "epoch": 2.5571725571725574,
      "grad_norm": 8.519729614257812,
      "learning_rate": 2.8373295657912945e-05,
      "loss": 0.2636,
      "num_input_tokens_seen": 989648,
      "step": 2460
    },
    {
      "epoch": 2.5623700623700625,
      "grad_norm": 2.2857918739318848,
      "learning_rate": 2.828338965578603e-05,
      "loss": 0.2691,
      "num_input_tokens_seen": 991696,
      "step": 2465
    },
    {
      "epoch": 2.5675675675675675,
      "grad_norm": 5.966533184051514,
      "learning_rate": 2.8193440423530114e-05,
      "loss": 0.2598,
      "num_input_tokens_seen": 993616,
      "step": 2470
    },
    {
      "epoch": 2.5727650727650726,
      "grad_norm": 2.139225482940674,
      "learning_rate": 2.810344914544475e-05,
      "loss": 0.2688,
      "num_input_tokens_seen": 995664,
      "step": 2475
    },
    {
      "epoch": 2.577962577962578,
      "grad_norm": 1.0499998331069946,
      "learning_rate": 2.8013417006383076e-05,
      "loss": 0.295,
      "num_input_tokens_seen": 997648,
      "step": 2480
    },
    {
      "epoch": 2.583160083160083,
      "grad_norm": 8.98536205291748,
      "learning_rate": 2.792334519173624e-05,
      "loss": 0.2802,
      "num_input_tokens_seen": 999696,
      "step": 2485
    },
    {
      "epoch": 2.5883575883575882,
      "grad_norm": 9.025850296020508,
      "learning_rate": 2.7833234887417743e-05,
      "loss": 0.2897,
      "num_input_tokens_seen": 1001680,
      "step": 2490
    },
    {
      "epoch": 2.5935550935550937,
      "grad_norm": 4.961789131164551,
      "learning_rate": 2.7743087279847868e-05,
      "loss": 0.2723,
      "num_input_tokens_seen": 1003728,
      "step": 2495
    },
    {
      "epoch": 2.598752598752599,
      "grad_norm": 11.652027130126953,
      "learning_rate": 2.765290355593805e-05,
      "loss": 0.2874,
      "num_input_tokens_seen": 1005584,
      "step": 2500
    },
    {
      "epoch": 2.603950103950104,
      "grad_norm": 16.368242263793945,
      "learning_rate": 2.7562684903075238e-05,
      "loss": 0.2405,
      "num_input_tokens_seen": 1007696,
      "step": 2505
    },
    {
      "epoch": 2.609147609147609,
      "grad_norm": 12.281315803527832,
      "learning_rate": 2.7472432509106248e-05,
      "loss": 0.2737,
      "num_input_tokens_seen": 1009680,
      "step": 2510
    },
    {
      "epoch": 2.6143451143451144,
      "grad_norm": 16.369632720947266,
      "learning_rate": 2.7382147562322174e-05,
      "loss": 0.2753,
      "num_input_tokens_seen": 1011728,
      "step": 2515
    },
    {
      "epoch": 2.6195426195426195,
      "grad_norm": 3.395531177520752,
      "learning_rate": 2.729183125144269e-05,
      "loss": 0.2553,
      "num_input_tokens_seen": 1013840,
      "step": 2520
    },
    {
      "epoch": 2.624740124740125,
      "grad_norm": 7.065828323364258,
      "learning_rate": 2.7201484765600426e-05,
      "loss": 0.2564,
      "num_input_tokens_seen": 1015824,
      "step": 2525
    },
    {
      "epoch": 2.62993762993763,
      "grad_norm": 2.8372550010681152,
      "learning_rate": 2.7111109294325297e-05,
      "loss": 0.277,
      "num_input_tokens_seen": 1017744,
      "step": 2530
    },
    {
      "epoch": 2.635135135135135,
      "grad_norm": 12.833436012268066,
      "learning_rate": 2.702070602752887e-05,
      "loss": 0.2439,
      "num_input_tokens_seen": 1019728,
      "step": 2535
    },
    {
      "epoch": 2.64033264033264,
      "grad_norm": 0.17026355862617493,
      "learning_rate": 2.693027615548864e-05,
      "loss": 0.2958,
      "num_input_tokens_seen": 1021840,
      "step": 2540
    },
    {
      "epoch": 2.6455301455301456,
      "grad_norm": 10.469789505004883,
      "learning_rate": 2.6839820868832433e-05,
      "loss": 0.2809,
      "num_input_tokens_seen": 1023824,
      "step": 2545
    },
    {
      "epoch": 2.6507276507276507,
      "grad_norm": 10.419620513916016,
      "learning_rate": 2.6749341358522674e-05,
      "loss": 0.2583,
      "num_input_tokens_seen": 1025616,
      "step": 2550
    },
    {
      "epoch": 2.6559251559251558,
      "grad_norm": 3.5157597064971924,
      "learning_rate": 2.665883881584072e-05,
      "loss": 0.2494,
      "num_input_tokens_seen": 1027664,
      "step": 2555
    },
    {
      "epoch": 2.6611226611226613,
      "grad_norm": 1.5977754592895508,
      "learning_rate": 2.6568314432371183e-05,
      "loss": 0.2477,
      "num_input_tokens_seen": 1029648,
      "step": 2560
    },
    {
      "epoch": 2.6663201663201663,
      "grad_norm": 1.0980393886566162,
      "learning_rate": 2.6477769399986245e-05,
      "loss": 0.2402,
      "num_input_tokens_seen": 1031632,
      "step": 2565
    },
    {
      "epoch": 2.6715176715176714,
      "grad_norm": 8.981844902038574,
      "learning_rate": 2.6387204910829956e-05,
      "loss": 0.2583,
      "num_input_tokens_seen": 1033488,
      "step": 2570
    },
    {
      "epoch": 2.6767151767151764,
      "grad_norm": 3.0374419689178467,
      "learning_rate": 2.629662215730253e-05,
      "loss": 0.2162,
      "num_input_tokens_seen": 1035536,
      "step": 2575
    },
    {
      "epoch": 2.681912681912682,
      "grad_norm": 5.767524242401123,
      "learning_rate": 2.6206022332044667e-05,
      "loss": 0.2652,
      "num_input_tokens_seen": 1037584,
      "step": 2580
    },
    {
      "epoch": 2.687110187110187,
      "grad_norm": 0.40257641673088074,
      "learning_rate": 2.6115406627921825e-05,
      "loss": 0.2401,
      "num_input_tokens_seen": 1039568,
      "step": 2585
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 1.3719632625579834,
      "learning_rate": 2.6024776238008543e-05,
      "loss": 0.2416,
      "num_input_tokens_seen": 1041616,
      "step": 2590
    },
    {
      "epoch": 2.6975051975051976,
      "grad_norm": 0.23526664078235626,
      "learning_rate": 2.593413235557271e-05,
      "loss": 0.2479,
      "num_input_tokens_seen": 1043664,
      "step": 2595
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 5.976350784301758,
      "learning_rate": 2.5843476174059872e-05,
      "loss": 0.2624,
      "num_input_tokens_seen": 1045520,
      "step": 2600
    },
    {
      "epoch": 2.7079002079002077,
      "grad_norm": 1.9363540410995483,
      "learning_rate": 2.5752808887077477e-05,
      "loss": 0.2519,
      "num_input_tokens_seen": 1047376,
      "step": 2605
    },
    {
      "epoch": 2.713097713097713,
      "grad_norm": 0.9820289611816406,
      "learning_rate": 2.5662131688379242e-05,
      "loss": 0.2347,
      "num_input_tokens_seen": 1049360,
      "step": 2610
    },
    {
      "epoch": 2.7182952182952183,
      "grad_norm": 6.348036766052246,
      "learning_rate": 2.5571445771849327e-05,
      "loss": 0.2785,
      "num_input_tokens_seen": 1051344,
      "step": 2615
    },
    {
      "epoch": 2.7234927234927238,
      "grad_norm": 2.1040618419647217,
      "learning_rate": 2.548075233148674e-05,
      "loss": 0.2622,
      "num_input_tokens_seen": 1053264,
      "step": 2620
    },
    {
      "epoch": 2.728690228690229,
      "grad_norm": 3.914247989654541,
      "learning_rate": 2.5390052561389478e-05,
      "loss": 0.2798,
      "num_input_tokens_seen": 1055248,
      "step": 2625
    },
    {
      "epoch": 2.733887733887734,
      "grad_norm": 10.045319557189941,
      "learning_rate": 2.529934765573893e-05,
      "loss": 0.2568,
      "num_input_tokens_seen": 1057104,
      "step": 2630
    },
    {
      "epoch": 2.739085239085239,
      "grad_norm": 6.609117031097412,
      "learning_rate": 2.520863880878408e-05,
      "loss": 0.2622,
      "num_input_tokens_seen": 1059024,
      "step": 2635
    },
    {
      "epoch": 2.7442827442827444,
      "grad_norm": 8.553672790527344,
      "learning_rate": 2.511792721482581e-05,
      "loss": 0.2707,
      "num_input_tokens_seen": 1060944,
      "step": 2640
    },
    {
      "epoch": 2.7494802494802495,
      "grad_norm": 6.869534969329834,
      "learning_rate": 2.502721406820116e-05,
      "loss": 0.2525,
      "num_input_tokens_seen": 1062992,
      "step": 2645
    },
    {
      "epoch": 2.7546777546777546,
      "grad_norm": 8.741276741027832,
      "learning_rate": 2.4936500563267627e-05,
      "loss": 0.2934,
      "num_input_tokens_seen": 1064848,
      "step": 2650
    },
    {
      "epoch": 2.7557172557172556,
      "eval_loss": 0.2561495900154114,
      "eval_runtime": 1.071,
      "eval_samples_per_second": 799.235,
      "eval_steps_per_second": 99.904,
      "num_input_tokens_seen": 1065232,
      "step": 2651
    },
    {
      "epoch": 2.75987525987526,
      "grad_norm": 10.544535636901855,
      "learning_rate": 2.4845787894387425e-05,
      "loss": 0.2493,
      "num_input_tokens_seen": 1066832,
      "step": 2655
    },
    {
      "epoch": 2.765072765072765,
      "grad_norm": 7.622214317321777,
      "learning_rate": 2.4755077255911743e-05,
      "loss": 0.2661,
      "num_input_tokens_seen": 1068880,
      "step": 2660
    },
    {
      "epoch": 2.77027027027027,
      "grad_norm": 7.094521522521973,
      "learning_rate": 2.4664369842165068e-05,
      "loss": 0.2398,
      "num_input_tokens_seen": 1070864,
      "step": 2665
    },
    {
      "epoch": 2.7754677754677752,
      "grad_norm": 9.394824981689453,
      "learning_rate": 2.4573666847429384e-05,
      "loss": 0.2523,
      "num_input_tokens_seen": 1072848,
      "step": 2670
    },
    {
      "epoch": 2.7806652806652807,
      "grad_norm": 5.644352436065674,
      "learning_rate": 2.4482969465928543e-05,
      "loss": 0.3167,
      "num_input_tokens_seen": 1074832,
      "step": 2675
    },
    {
      "epoch": 2.785862785862786,
      "grad_norm": 11.34897232055664,
      "learning_rate": 2.4392278891812455e-05,
      "loss": 0.2951,
      "num_input_tokens_seen": 1076944,
      "step": 2680
    },
    {
      "epoch": 2.7910602910602913,
      "grad_norm": 4.7796196937561035,
      "learning_rate": 2.430159631914141e-05,
      "loss": 0.2844,
      "num_input_tokens_seen": 1078800,
      "step": 2685
    },
    {
      "epoch": 2.7962577962577964,
      "grad_norm": 6.175029277801514,
      "learning_rate": 2.4210922941870367e-05,
      "loss": 0.2706,
      "num_input_tokens_seen": 1080912,
      "step": 2690
    },
    {
      "epoch": 2.8014553014553014,
      "grad_norm": 2.2498137950897217,
      "learning_rate": 2.41202599538332e-05,
      "loss": 0.244,
      "num_input_tokens_seen": 1082960,
      "step": 2695
    },
    {
      "epoch": 2.8066528066528065,
      "grad_norm": 14.539408683776855,
      "learning_rate": 2.402960854872697e-05,
      "loss": 0.2877,
      "num_input_tokens_seen": 1085008,
      "step": 2700
    },
    {
      "epoch": 2.811850311850312,
      "grad_norm": 6.818408489227295,
      "learning_rate": 2.39389699200963e-05,
      "loss": 0.1818,
      "num_input_tokens_seen": 1087184,
      "step": 2705
    },
    {
      "epoch": 2.817047817047817,
      "grad_norm": 8.152698516845703,
      "learning_rate": 2.384834526131752e-05,
      "loss": 0.2384,
      "num_input_tokens_seen": 1089104,
      "step": 2710
    },
    {
      "epoch": 2.822245322245322,
      "grad_norm": 0.4875684976577759,
      "learning_rate": 2.3757735765583083e-05,
      "loss": 0.2707,
      "num_input_tokens_seen": 1091024,
      "step": 2715
    },
    {
      "epoch": 2.8274428274428276,
      "grad_norm": 0.5857343077659607,
      "learning_rate": 2.366714262588577e-05,
      "loss": 0.2399,
      "num_input_tokens_seen": 1093008,
      "step": 2720
    },
    {
      "epoch": 2.8326403326403327,
      "grad_norm": 0.10867218673229218,
      "learning_rate": 2.3576567035003027e-05,
      "loss": 0.2595,
      "num_input_tokens_seen": 1094992,
      "step": 2725
    },
    {
      "epoch": 2.8378378378378377,
      "grad_norm": 5.911508083343506,
      "learning_rate": 2.3486010185481248e-05,
      "loss": 0.2918,
      "num_input_tokens_seen": 1097040,
      "step": 2730
    },
    {
      "epoch": 2.8430353430353428,
      "grad_norm": 12.04796314239502,
      "learning_rate": 2.3395473269620056e-05,
      "loss": 0.274,
      "num_input_tokens_seen": 1098960,
      "step": 2735
    },
    {
      "epoch": 2.8482328482328483,
      "grad_norm": 15.308754920959473,
      "learning_rate": 2.330495747945665e-05,
      "loss": 0.2749,
      "num_input_tokens_seen": 1101200,
      "step": 2740
    },
    {
      "epoch": 2.8534303534303533,
      "grad_norm": 16.627357482910156,
      "learning_rate": 2.321446400675005e-05,
      "loss": 0.2766,
      "num_input_tokens_seen": 1103120,
      "step": 2745
    },
    {
      "epoch": 2.858627858627859,
      "grad_norm": 20.289390563964844,
      "learning_rate": 2.3123994042965453e-05,
      "loss": 0.255,
      "num_input_tokens_seen": 1105168,
      "step": 2750
    },
    {
      "epoch": 2.863825363825364,
      "grad_norm": 8.598325729370117,
      "learning_rate": 2.3033548779258535e-05,
      "loss": 0.2452,
      "num_input_tokens_seen": 1107152,
      "step": 2755
    },
    {
      "epoch": 2.869022869022869,
      "grad_norm": 13.070589065551758,
      "learning_rate": 2.294312940645975e-05,
      "loss": 0.3016,
      "num_input_tokens_seen": 1109200,
      "step": 2760
    },
    {
      "epoch": 2.874220374220374,
      "grad_norm": 2.4811654090881348,
      "learning_rate": 2.2852737115058682e-05,
      "loss": 0.282,
      "num_input_tokens_seen": 1111248,
      "step": 2765
    },
    {
      "epoch": 2.8794178794178795,
      "grad_norm": 9.034327507019043,
      "learning_rate": 2.276237309518834e-05,
      "loss": 0.2753,
      "num_input_tokens_seen": 1113232,
      "step": 2770
    },
    {
      "epoch": 2.8846153846153846,
      "grad_norm": 0.18150383234024048,
      "learning_rate": 2.2672038536609487e-05,
      "loss": 0.268,
      "num_input_tokens_seen": 1115216,
      "step": 2775
    },
    {
      "epoch": 2.88981288981289,
      "grad_norm": 4.699392318725586,
      "learning_rate": 2.2581734628695034e-05,
      "loss": 0.2816,
      "num_input_tokens_seen": 1117264,
      "step": 2780
    },
    {
      "epoch": 2.895010395010395,
      "grad_norm": 4.98245906829834,
      "learning_rate": 2.2491462560414287e-05,
      "loss": 0.2795,
      "num_input_tokens_seen": 1119376,
      "step": 2785
    },
    {
      "epoch": 2.9002079002079,
      "grad_norm": 7.173026084899902,
      "learning_rate": 2.2401223520317362e-05,
      "loss": 0.283,
      "num_input_tokens_seen": 1121424,
      "step": 2790
    },
    {
      "epoch": 2.9054054054054053,
      "grad_norm": 1.0886781215667725,
      "learning_rate": 2.2311018696519532e-05,
      "loss": 0.2663,
      "num_input_tokens_seen": 1123472,
      "step": 2795
    },
    {
      "epoch": 2.9106029106029108,
      "grad_norm": 3.959348440170288,
      "learning_rate": 2.222084927668553e-05,
      "loss": 0.218,
      "num_input_tokens_seen": 1125584,
      "step": 2800
    },
    {
      "epoch": 2.915800415800416,
      "grad_norm": 18.90048599243164,
      "learning_rate": 2.2130716448014e-05,
      "loss": 0.2474,
      "num_input_tokens_seen": 1127568,
      "step": 2805
    },
    {
      "epoch": 2.920997920997921,
      "grad_norm": 19.015106201171875,
      "learning_rate": 2.204062139722176e-05,
      "loss": 0.3308,
      "num_input_tokens_seen": 1129552,
      "step": 2810
    },
    {
      "epoch": 2.9261954261954264,
      "grad_norm": 3.917026996612549,
      "learning_rate": 2.1950565310528266e-05,
      "loss": 0.2914,
      "num_input_tokens_seen": 1131472,
      "step": 2815
    },
    {
      "epoch": 2.9313929313929314,
      "grad_norm": 11.03906536102295,
      "learning_rate": 2.186054937363996e-05,
      "loss": 0.3008,
      "num_input_tokens_seen": 1133392,
      "step": 2820
    },
    {
      "epoch": 2.9365904365904365,
      "grad_norm": 9.081315040588379,
      "learning_rate": 2.1770574771734642e-05,
      "loss": 0.2747,
      "num_input_tokens_seen": 1135440,
      "step": 2825
    },
    {
      "epoch": 2.9417879417879416,
      "grad_norm": 1.4751437902450562,
      "learning_rate": 2.168064268944591e-05,
      "loss": 0.2694,
      "num_input_tokens_seen": 1137424,
      "step": 2830
    },
    {
      "epoch": 2.946985446985447,
      "grad_norm": 0.2908379137516022,
      "learning_rate": 2.159075431084751e-05,
      "loss": 0.2574,
      "num_input_tokens_seen": 1139408,
      "step": 2835
    },
    {
      "epoch": 2.952182952182952,
      "grad_norm": 6.697641372680664,
      "learning_rate": 2.1500910819437766e-05,
      "loss": 0.2677,
      "num_input_tokens_seen": 1141456,
      "step": 2840
    },
    {
      "epoch": 2.9573804573804576,
      "grad_norm": 0.9215155243873596,
      "learning_rate": 2.141111339812405e-05,
      "loss": 0.2513,
      "num_input_tokens_seen": 1143440,
      "step": 2845
    },
    {
      "epoch": 2.9625779625779627,
      "grad_norm": 0.976637601852417,
      "learning_rate": 2.1321363229207096e-05,
      "loss": 0.256,
      "num_input_tokens_seen": 1145360,
      "step": 2850
    },
    {
      "epoch": 2.9677754677754677,
      "grad_norm": 13.551946640014648,
      "learning_rate": 2.123166149436556e-05,
      "loss": 0.2779,
      "num_input_tokens_seen": 1147280,
      "step": 2855
    },
    {
      "epoch": 2.972972972972973,
      "grad_norm": 11.918627738952637,
      "learning_rate": 2.114200937464035e-05,
      "loss": 0.2814,
      "num_input_tokens_seen": 1149200,
      "step": 2860
    },
    {
      "epoch": 2.9781704781704783,
      "grad_norm": 19.893056869506836,
      "learning_rate": 2.1052408050419152e-05,
      "loss": 0.2671,
      "num_input_tokens_seen": 1151184,
      "step": 2865
    },
    {
      "epoch": 2.9833679833679834,
      "grad_norm": 12.081989288330078,
      "learning_rate": 2.0962858701420866e-05,
      "loss": 0.2391,
      "num_input_tokens_seen": 1153232,
      "step": 2870
    },
    {
      "epoch": 2.9885654885654884,
      "grad_norm": 9.680088996887207,
      "learning_rate": 2.0873362506680057e-05,
      "loss": 0.251,
      "num_input_tokens_seen": 1155216,
      "step": 2875
    },
    {
      "epoch": 2.993762993762994,
      "grad_norm": 7.104877471923828,
      "learning_rate": 2.078392064453144e-05,
      "loss": 0.2661,
      "num_input_tokens_seen": 1157264,
      "step": 2880
    },
    {
      "epoch": 2.998960498960499,
      "grad_norm": 0.48920291662216187,
      "learning_rate": 2.0694534292594392e-05,
      "loss": 0.2609,
      "num_input_tokens_seen": 1159312,
      "step": 2885
    },
    {
      "epoch": 3.004158004158004,
      "grad_norm": 7.250494003295898,
      "learning_rate": 2.0605204627757403e-05,
      "loss": 0.2209,
      "num_input_tokens_seen": 1161248,
      "step": 2890
    },
    {
      "epoch": 3.006237006237006,
      "eval_loss": 0.257083535194397,
      "eval_runtime": 1.0821,
      "eval_samples_per_second": 791.084,
      "eval_steps_per_second": 98.885,
      "num_input_tokens_seen": 1162016,
      "step": 2892
    },
    {
      "epoch": 3.0093555093555096,
      "grad_norm": 14.574300765991211,
      "learning_rate": 2.051593282616262e-05,
      "loss": 0.2146,
      "num_input_tokens_seen": 1163168,
      "step": 2895
    },
    {
      "epoch": 3.0145530145530146,
      "grad_norm": 15.882224082946777,
      "learning_rate": 2.0426720063190335e-05,
      "loss": 0.2637,
      "num_input_tokens_seen": 1165088,
      "step": 2900
    },
    {
      "epoch": 3.0197505197505197,
      "grad_norm": 17.72052574157715,
      "learning_rate": 2.033756751344352e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 1167136,
      "step": 2905
    },
    {
      "epoch": 3.024948024948025,
      "grad_norm": 6.26023530960083,
      "learning_rate": 2.0248476350732368e-05,
      "loss": 0.2915,
      "num_input_tokens_seen": 1169120,
      "step": 2910
    },
    {
      "epoch": 3.0301455301455302,
      "grad_norm": 10.183695793151855,
      "learning_rate": 2.0159447748058805e-05,
      "loss": 0.3333,
      "num_input_tokens_seen": 1171040,
      "step": 2915
    },
    {
      "epoch": 3.0353430353430353,
      "grad_norm": 5.654576778411865,
      "learning_rate": 2.0070482877601127e-05,
      "loss": 0.2142,
      "num_input_tokens_seen": 1173024,
      "step": 2920
    },
    {
      "epoch": 3.0405405405405403,
      "grad_norm": 6.201435565948486,
      "learning_rate": 1.998158291069845e-05,
      "loss": 0.2912,
      "num_input_tokens_seen": 1174944,
      "step": 2925
    },
    {
      "epoch": 3.045738045738046,
      "grad_norm": 6.677492141723633,
      "learning_rate": 1.9892749017835384e-05,
      "loss": 0.2799,
      "num_input_tokens_seen": 1177056,
      "step": 2930
    },
    {
      "epoch": 3.050935550935551,
      "grad_norm": 22.7122859954834,
      "learning_rate": 1.9803982368626583e-05,
      "loss": 0.318,
      "num_input_tokens_seen": 1178976,
      "step": 2935
    },
    {
      "epoch": 3.056133056133056,
      "grad_norm": 8.921732902526855,
      "learning_rate": 1.9715284131801353e-05,
      "loss": 0.2836,
      "num_input_tokens_seen": 1181024,
      "step": 2940
    },
    {
      "epoch": 3.0613305613305615,
      "grad_norm": 1.3563765287399292,
      "learning_rate": 1.9626655475188238e-05,
      "loss": 0.2515,
      "num_input_tokens_seen": 1183008,
      "step": 2945
    },
    {
      "epoch": 3.0665280665280665,
      "grad_norm": 9.756816864013672,
      "learning_rate": 1.953809756569971e-05,
      "loss": 0.2949,
      "num_input_tokens_seen": 1185056,
      "step": 2950
    },
    {
      "epoch": 3.0717255717255716,
      "grad_norm": 10.324029922485352,
      "learning_rate": 1.9449611569316717e-05,
      "loss": 0.2646,
      "num_input_tokens_seen": 1186976,
      "step": 2955
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 1.5440117120742798,
      "learning_rate": 1.9361198651073408e-05,
      "loss": 0.3027,
      "num_input_tokens_seen": 1188960,
      "step": 2960
    },
    {
      "epoch": 3.082120582120582,
      "grad_norm": 11.017732620239258,
      "learning_rate": 1.9272859975041754e-05,
      "loss": 0.2629,
      "num_input_tokens_seen": 1190944,
      "step": 2965
    },
    {
      "epoch": 3.087318087318087,
      "grad_norm": 19.006444931030273,
      "learning_rate": 1.918459670431622e-05,
      "loss": 0.2261,
      "num_input_tokens_seen": 1192928,
      "step": 2970
    },
    {
      "epoch": 3.0925155925155927,
      "grad_norm": 7.811370849609375,
      "learning_rate": 1.9096410000998475e-05,
      "loss": 0.239,
      "num_input_tokens_seen": 1194848,
      "step": 2975
    },
    {
      "epoch": 3.0977130977130978,
      "grad_norm": 1.5552674531936646,
      "learning_rate": 1.900830102618206e-05,
      "loss": 0.2484,
      "num_input_tokens_seen": 1196768,
      "step": 2980
    },
    {
      "epoch": 3.102910602910603,
      "grad_norm": 9.481167793273926,
      "learning_rate": 1.892027093993716e-05,
      "loss": 0.2739,
      "num_input_tokens_seen": 1198688,
      "step": 2985
    },
    {
      "epoch": 3.108108108108108,
      "grad_norm": 8.231619834899902,
      "learning_rate": 1.8832320901295227e-05,
      "loss": 0.2701,
      "num_input_tokens_seen": 1200672,
      "step": 2990
    },
    {
      "epoch": 3.1133056133056134,
      "grad_norm": 9.055398941040039,
      "learning_rate": 1.8744452068233825e-05,
      "loss": 0.29,
      "num_input_tokens_seen": 1202720,
      "step": 2995
    },
    {
      "epoch": 3.1185031185031185,
      "grad_norm": 1.7919214963912964,
      "learning_rate": 1.8656665597661333e-05,
      "loss": 0.2816,
      "num_input_tokens_seen": 1204768,
      "step": 3000
    },
    {
      "epoch": 3.1237006237006235,
      "grad_norm": 9.331473350524902,
      "learning_rate": 1.85689626454017e-05,
      "loss": 0.2677,
      "num_input_tokens_seen": 1206944,
      "step": 3005
    },
    {
      "epoch": 3.128898128898129,
      "grad_norm": 9.339173316955566,
      "learning_rate": 1.8481344366179284e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 1209056,
      "step": 3010
    },
    {
      "epoch": 3.134095634095634,
      "grad_norm": 0.4089166224002838,
      "learning_rate": 1.839381191360358e-05,
      "loss": 0.2745,
      "num_input_tokens_seen": 1210976,
      "step": 3015
    },
    {
      "epoch": 3.139293139293139,
      "grad_norm": 13.044018745422363,
      "learning_rate": 1.8306366440154066e-05,
      "loss": 0.2645,
      "num_input_tokens_seen": 1213024,
      "step": 3020
    },
    {
      "epoch": 3.1444906444906446,
      "grad_norm": 3.4399261474609375,
      "learning_rate": 1.821900909716504e-05,
      "loss": 0.2854,
      "num_input_tokens_seen": 1215136,
      "step": 3025
    },
    {
      "epoch": 3.1496881496881497,
      "grad_norm": 16.901412963867188,
      "learning_rate": 1.8131741034810435e-05,
      "loss": 0.2667,
      "num_input_tokens_seen": 1217056,
      "step": 3030
    },
    {
      "epoch": 3.1548856548856548,
      "grad_norm": 4.719562530517578,
      "learning_rate": 1.8044563402088684e-05,
      "loss": 0.2799,
      "num_input_tokens_seen": 1219168,
      "step": 3035
    },
    {
      "epoch": 3.1600831600831603,
      "grad_norm": 0.4354883134365082,
      "learning_rate": 1.795747734680762e-05,
      "loss": 0.2724,
      "num_input_tokens_seen": 1221088,
      "step": 3040
    },
    {
      "epoch": 3.1652806652806653,
      "grad_norm": 4.838202953338623,
      "learning_rate": 1.7870484015569306e-05,
      "loss": 0.2666,
      "num_input_tokens_seen": 1223264,
      "step": 3045
    },
    {
      "epoch": 3.1704781704781704,
      "grad_norm": 0.40550848841667175,
      "learning_rate": 1.7783584553755006e-05,
      "loss": 0.2586,
      "num_input_tokens_seen": 1225440,
      "step": 3050
    },
    {
      "epoch": 3.175675675675676,
      "grad_norm": 8.755316734313965,
      "learning_rate": 1.769678010551003e-05,
      "loss": 0.2519,
      "num_input_tokens_seen": 1227424,
      "step": 3055
    },
    {
      "epoch": 3.180873180873181,
      "grad_norm": 8.232521057128906,
      "learning_rate": 1.761007181372874e-05,
      "loss": 0.241,
      "num_input_tokens_seen": 1229344,
      "step": 3060
    },
    {
      "epoch": 3.186070686070686,
      "grad_norm": 7.669497013092041,
      "learning_rate": 1.7523460820039464e-05,
      "loss": 0.2406,
      "num_input_tokens_seen": 1231456,
      "step": 3065
    },
    {
      "epoch": 3.1912681912681915,
      "grad_norm": 12.419413566589355,
      "learning_rate": 1.7436948264789466e-05,
      "loss": 0.3145,
      "num_input_tokens_seen": 1233440,
      "step": 3070
    },
    {
      "epoch": 3.1964656964656966,
      "grad_norm": 2.3379440307617188,
      "learning_rate": 1.7350535287029957e-05,
      "loss": 0.2568,
      "num_input_tokens_seen": 1235552,
      "step": 3075
    },
    {
      "epoch": 3.2016632016632016,
      "grad_norm": 10.220464706420898,
      "learning_rate": 1.7264223024501064e-05,
      "loss": 0.2696,
      "num_input_tokens_seen": 1237536,
      "step": 3080
    },
    {
      "epoch": 3.2068607068607067,
      "grad_norm": 17.75920867919922,
      "learning_rate": 1.717801261361685e-05,
      "loss": 0.2861,
      "num_input_tokens_seen": 1239584,
      "step": 3085
    },
    {
      "epoch": 3.212058212058212,
      "grad_norm": 7.69419002532959,
      "learning_rate": 1.7091905189450423e-05,
      "loss": 0.2501,
      "num_input_tokens_seen": 1241504,
      "step": 3090
    },
    {
      "epoch": 3.2172557172557172,
      "grad_norm": 6.530004501342773,
      "learning_rate": 1.700590188571887e-05,
      "loss": 0.2578,
      "num_input_tokens_seen": 1243552,
      "step": 3095
    },
    {
      "epoch": 3.2224532224532223,
      "grad_norm": 2.1304476261138916,
      "learning_rate": 1.6920003834768438e-05,
      "loss": 0.3075,
      "num_input_tokens_seen": 1245600,
      "step": 3100
    },
    {
      "epoch": 3.227650727650728,
      "grad_norm": 19.96734046936035,
      "learning_rate": 1.6834212167559575e-05,
      "loss": 0.2389,
      "num_input_tokens_seen": 1247712,
      "step": 3105
    },
    {
      "epoch": 3.232848232848233,
      "grad_norm": 12.024674415588379,
      "learning_rate": 1.674852801365203e-05,
      "loss": 0.26,
      "num_input_tokens_seen": 1249696,
      "step": 3110
    },
    {
      "epoch": 3.238045738045738,
      "grad_norm": 6.835387229919434,
      "learning_rate": 1.6662952501190033e-05,
      "loss": 0.3027,
      "num_input_tokens_seen": 1251808,
      "step": 3115
    },
    {
      "epoch": 3.2432432432432434,
      "grad_norm": 8.725172996520996,
      "learning_rate": 1.6577486756887374e-05,
      "loss": 0.2315,
      "num_input_tokens_seen": 1253728,
      "step": 3120
    },
    {
      "epoch": 3.2484407484407485,
      "grad_norm": 0.420453816652298,
      "learning_rate": 1.649213190601261e-05,
      "loss": 0.249,
      "num_input_tokens_seen": 1255840,
      "step": 3125
    },
    {
      "epoch": 3.2536382536382535,
      "grad_norm": 8.639700889587402,
      "learning_rate": 1.640688907237425e-05,
      "loss": 0.2647,
      "num_input_tokens_seen": 1257888,
      "step": 3130
    },
    {
      "epoch": 3.2567567567567566,
      "eval_loss": 0.2563324272632599,
      "eval_runtime": 1.0592,
      "eval_samples_per_second": 808.124,
      "eval_steps_per_second": 101.015,
      "num_input_tokens_seen": 1259168,
      "step": 3133
    },
    {
      "epoch": 3.258835758835759,
      "grad_norm": 13.672746658325195,
      "learning_rate": 1.632175937830594e-05,
      "loss": 0.2584,
      "num_input_tokens_seen": 1259936,
      "step": 3135
    },
    {
      "epoch": 3.264033264033264,
      "grad_norm": 14.988076210021973,
      "learning_rate": 1.6236743944651703e-05,
      "loss": 0.2355,
      "num_input_tokens_seen": 1262112,
      "step": 3140
    },
    {
      "epoch": 3.269230769230769,
      "grad_norm": 4.268918037414551,
      "learning_rate": 1.615184389075117e-05,
      "loss": 0.2481,
      "num_input_tokens_seen": 1263904,
      "step": 3145
    },
    {
      "epoch": 3.274428274428274,
      "grad_norm": 2.765204429626465,
      "learning_rate": 1.6067060334424835e-05,
      "loss": 0.2607,
      "num_input_tokens_seen": 1265952,
      "step": 3150
    },
    {
      "epoch": 3.2796257796257797,
      "grad_norm": 18.85066795349121,
      "learning_rate": 1.5982394391959382e-05,
      "loss": 0.3119,
      "num_input_tokens_seen": 1267872,
      "step": 3155
    },
    {
      "epoch": 3.284823284823285,
      "grad_norm": 2.8350369930267334,
      "learning_rate": 1.58978471780929e-05,
      "loss": 0.2925,
      "num_input_tokens_seen": 1269792,
      "step": 3160
    },
    {
      "epoch": 3.29002079002079,
      "grad_norm": 6.0720438957214355,
      "learning_rate": 1.581341980600033e-05,
      "loss": 0.2586,
      "num_input_tokens_seen": 1271776,
      "step": 3165
    },
    {
      "epoch": 3.2952182952182953,
      "grad_norm": 15.48050308227539,
      "learning_rate": 1.5729113387278673e-05,
      "loss": 0.2671,
      "num_input_tokens_seen": 1273760,
      "step": 3170
    },
    {
      "epoch": 3.3004158004158004,
      "grad_norm": 4.232337951660156,
      "learning_rate": 1.5644929031932454e-05,
      "loss": 0.2864,
      "num_input_tokens_seen": 1275808,
      "step": 3175
    },
    {
      "epoch": 3.3056133056133055,
      "grad_norm": 4.032689094543457,
      "learning_rate": 1.5560867848359077e-05,
      "loss": 0.2666,
      "num_input_tokens_seen": 1277792,
      "step": 3180
    },
    {
      "epoch": 3.310810810810811,
      "grad_norm": 3.819479465484619,
      "learning_rate": 1.547693094333421e-05,
      "loss": 0.2648,
      "num_input_tokens_seen": 1279776,
      "step": 3185
    },
    {
      "epoch": 3.316008316008316,
      "grad_norm": 5.751096725463867,
      "learning_rate": 1.539311942199725e-05,
      "loss": 0.2586,
      "num_input_tokens_seen": 1281760,
      "step": 3190
    },
    {
      "epoch": 3.321205821205821,
      "grad_norm": 5.748921871185303,
      "learning_rate": 1.5309434387836735e-05,
      "loss": 0.2391,
      "num_input_tokens_seen": 1283744,
      "step": 3195
    },
    {
      "epoch": 3.3264033264033266,
      "grad_norm": 4.8532915115356445,
      "learning_rate": 1.5225876942675842e-05,
      "loss": 0.2907,
      "num_input_tokens_seen": 1285792,
      "step": 3200
    },
    {
      "epoch": 3.3316008316008316,
      "grad_norm": 2.2807741165161133,
      "learning_rate": 1.5142448186657878e-05,
      "loss": 0.2942,
      "num_input_tokens_seen": 1287776,
      "step": 3205
    },
    {
      "epoch": 3.3367983367983367,
      "grad_norm": 0.5820121169090271,
      "learning_rate": 1.505914921823178e-05,
      "loss": 0.3321,
      "num_input_tokens_seen": 1289696,
      "step": 3210
    },
    {
      "epoch": 3.3419958419958418,
      "grad_norm": 2.5034902095794678,
      "learning_rate": 1.4975981134137659e-05,
      "loss": 0.2486,
      "num_input_tokens_seen": 1291680,
      "step": 3215
    },
    {
      "epoch": 3.3471933471933473,
      "grad_norm": 1.1145634651184082,
      "learning_rate": 1.489294502939238e-05,
      "loss": 0.2502,
      "num_input_tokens_seen": 1293536,
      "step": 3220
    },
    {
      "epoch": 3.3523908523908523,
      "grad_norm": 7.527857780456543,
      "learning_rate": 1.4810041997275092e-05,
      "loss": 0.2878,
      "num_input_tokens_seen": 1295712,
      "step": 3225
    },
    {
      "epoch": 3.357588357588358,
      "grad_norm": 12.356915473937988,
      "learning_rate": 1.4727273129312918e-05,
      "loss": 0.2824,
      "num_input_tokens_seen": 1297760,
      "step": 3230
    },
    {
      "epoch": 3.362785862785863,
      "grad_norm": 4.496915817260742,
      "learning_rate": 1.4644639515266483e-05,
      "loss": 0.2772,
      "num_input_tokens_seen": 1299808,
      "step": 3235
    },
    {
      "epoch": 3.367983367983368,
      "grad_norm": 2.9005799293518066,
      "learning_rate": 1.4562142243115644e-05,
      "loss": 0.2602,
      "num_input_tokens_seen": 1301920,
      "step": 3240
    },
    {
      "epoch": 3.373180873180873,
      "grad_norm": 1.413136601448059,
      "learning_rate": 1.4479782399045152e-05,
      "loss": 0.2737,
      "num_input_tokens_seen": 1303904,
      "step": 3245
    },
    {
      "epoch": 3.3783783783783785,
      "grad_norm": 15.236632347106934,
      "learning_rate": 1.4397561067430298e-05,
      "loss": 0.2683,
      "num_input_tokens_seen": 1305888,
      "step": 3250
    },
    {
      "epoch": 3.3835758835758836,
      "grad_norm": 13.490982055664062,
      "learning_rate": 1.4315479330822712e-05,
      "loss": 0.261,
      "num_input_tokens_seen": 1308064,
      "step": 3255
    },
    {
      "epoch": 3.3887733887733886,
      "grad_norm": 4.421963691711426,
      "learning_rate": 1.4233538269936042e-05,
      "loss": 0.2702,
      "num_input_tokens_seen": 1310048,
      "step": 3260
    },
    {
      "epoch": 3.393970893970894,
      "grad_norm": 10.610108375549316,
      "learning_rate": 1.415173896363178e-05,
      "loss": 0.2761,
      "num_input_tokens_seen": 1311968,
      "step": 3265
    },
    {
      "epoch": 3.399168399168399,
      "grad_norm": 4.008954048156738,
      "learning_rate": 1.4070082488905034e-05,
      "loss": 0.2631,
      "num_input_tokens_seen": 1313888,
      "step": 3270
    },
    {
      "epoch": 3.4043659043659042,
      "grad_norm": 2.906602144241333,
      "learning_rate": 1.3988569920870314e-05,
      "loss": 0.2702,
      "num_input_tokens_seen": 1316064,
      "step": 3275
    },
    {
      "epoch": 3.4095634095634098,
      "grad_norm": 12.885231018066406,
      "learning_rate": 1.3907202332747454e-05,
      "loss": 0.2643,
      "num_input_tokens_seen": 1318112,
      "step": 3280
    },
    {
      "epoch": 3.414760914760915,
      "grad_norm": 0.1723124235868454,
      "learning_rate": 1.3825980795847402e-05,
      "loss": 0.2877,
      "num_input_tokens_seen": 1319968,
      "step": 3285
    },
    {
      "epoch": 3.41995841995842,
      "grad_norm": 7.7092390060424805,
      "learning_rate": 1.3744906379558165e-05,
      "loss": 0.271,
      "num_input_tokens_seen": 1322016,
      "step": 3290
    },
    {
      "epoch": 3.4251559251559254,
      "grad_norm": 1.4101628065109253,
      "learning_rate": 1.3663980151330732e-05,
      "loss": 0.2729,
      "num_input_tokens_seen": 1323936,
      "step": 3295
    },
    {
      "epoch": 3.4303534303534304,
      "grad_norm": 1.401780128479004,
      "learning_rate": 1.3583203176664961e-05,
      "loss": 0.261,
      "num_input_tokens_seen": 1325920,
      "step": 3300
    },
    {
      "epoch": 3.4355509355509355,
      "grad_norm": 21.787748336791992,
      "learning_rate": 1.350257651909562e-05,
      "loss": 0.2498,
      "num_input_tokens_seen": 1327840,
      "step": 3305
    },
    {
      "epoch": 3.4407484407484406,
      "grad_norm": 8.913102149963379,
      "learning_rate": 1.3422101240178365e-05,
      "loss": 0.2384,
      "num_input_tokens_seen": 1329760,
      "step": 3310
    },
    {
      "epoch": 3.445945945945946,
      "grad_norm": 2.156329870223999,
      "learning_rate": 1.3341778399475713e-05,
      "loss": 0.2789,
      "num_input_tokens_seen": 1331744,
      "step": 3315
    },
    {
      "epoch": 3.451143451143451,
      "grad_norm": 6.306075096130371,
      "learning_rate": 1.3261609054543179e-05,
      "loss": 0.26,
      "num_input_tokens_seen": 1333792,
      "step": 3320
    },
    {
      "epoch": 3.456340956340956,
      "grad_norm": 8.915762901306152,
      "learning_rate": 1.3181594260915262e-05,
      "loss": 0.2975,
      "num_input_tokens_seen": 1335776,
      "step": 3325
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 2.6008245944976807,
      "learning_rate": 1.3101735072091622e-05,
      "loss": 0.2479,
      "num_input_tokens_seen": 1337824,
      "step": 3330
    },
    {
      "epoch": 3.4667359667359667,
      "grad_norm": 10.535400390625,
      "learning_rate": 1.3022032539523176e-05,
      "loss": 0.223,
      "num_input_tokens_seen": 1339872,
      "step": 3335
    },
    {
      "epoch": 3.471933471933472,
      "grad_norm": 7.183012008666992,
      "learning_rate": 1.2942487712598234e-05,
      "loss": 0.2543,
      "num_input_tokens_seen": 1341920,
      "step": 3340
    },
    {
      "epoch": 3.4771309771309773,
      "grad_norm": 8.969158172607422,
      "learning_rate": 1.2863101638628717e-05,
      "loss": 0.2449,
      "num_input_tokens_seen": 1343904,
      "step": 3345
    },
    {
      "epoch": 3.4823284823284824,
      "grad_norm": 14.45447826385498,
      "learning_rate": 1.2783875362836373e-05,
      "loss": 0.2881,
      "num_input_tokens_seen": 1345952,
      "step": 3350
    },
    {
      "epoch": 3.4875259875259874,
      "grad_norm": 5.68330717086792,
      "learning_rate": 1.2704809928338956e-05,
      "loss": 0.2574,
      "num_input_tokens_seen": 1348128,
      "step": 3355
    },
    {
      "epoch": 3.492723492723493,
      "grad_norm": 4.9651079177856445,
      "learning_rate": 1.2625906376136581e-05,
      "loss": 0.2915,
      "num_input_tokens_seen": 1350048,
      "step": 3360
    },
    {
      "epoch": 3.497920997920998,
      "grad_norm": 0.1253027617931366,
      "learning_rate": 1.2547165745097928e-05,
      "loss": 0.2653,
      "num_input_tokens_seen": 1351968,
      "step": 3365
    },
    {
      "epoch": 3.503118503118503,
      "grad_norm": 7.796263217926025,
      "learning_rate": 1.2468589071946632e-05,
      "loss": 0.2795,
      "num_input_tokens_seen": 1353952,
      "step": 3370
    },
    {
      "epoch": 3.507276507276507,
      "eval_loss": 0.2641850709915161,
      "eval_runtime": 1.0564,
      "eval_samples_per_second": 810.335,
      "eval_steps_per_second": 101.292,
      "num_input_tokens_seen": 1355552,
      "step": 3374
    },
    {
      "epoch": 3.508316008316008,
      "grad_norm": 0.7158005237579346,
      "learning_rate": 1.2390177391247614e-05,
      "loss": 0.2673,
      "num_input_tokens_seen": 1356000,
      "step": 3375
    },
    {
      "epoch": 3.5135135135135136,
      "grad_norm": 2.2780776023864746,
      "learning_rate": 1.2311931735393417e-05,
      "loss": 0.2527,
      "num_input_tokens_seen": 1357984,
      "step": 3380
    },
    {
      "epoch": 3.5187110187110187,
      "grad_norm": 2.900038957595825,
      "learning_rate": 1.2233853134590697e-05,
      "loss": 0.2635,
      "num_input_tokens_seen": 1359904,
      "step": 3385
    },
    {
      "epoch": 3.523908523908524,
      "grad_norm": 9.76952075958252,
      "learning_rate": 1.215594261684656e-05,
      "loss": 0.2737,
      "num_input_tokens_seen": 1361952,
      "step": 3390
    },
    {
      "epoch": 3.529106029106029,
      "grad_norm": 0.9151262044906616,
      "learning_rate": 1.2078201207955123e-05,
      "loss": 0.2521,
      "num_input_tokens_seen": 1364000,
      "step": 3395
    },
    {
      "epoch": 3.5343035343035343,
      "grad_norm": 7.289126873016357,
      "learning_rate": 1.2000629931483947e-05,
      "loss": 0.246,
      "num_input_tokens_seen": 1366112,
      "step": 3400
    },
    {
      "epoch": 3.5395010395010393,
      "grad_norm": 7.83226203918457,
      "learning_rate": 1.1923229808760564e-05,
      "loss": 0.3136,
      "num_input_tokens_seen": 1368096,
      "step": 3405
    },
    {
      "epoch": 3.544698544698545,
      "grad_norm": 0.7650713324546814,
      "learning_rate": 1.1846001858859054e-05,
      "loss": 0.2365,
      "num_input_tokens_seen": 1370208,
      "step": 3410
    },
    {
      "epoch": 3.54989604989605,
      "grad_norm": 8.69223403930664,
      "learning_rate": 1.1768947098586628e-05,
      "loss": 0.2545,
      "num_input_tokens_seen": 1372192,
      "step": 3415
    },
    {
      "epoch": 3.555093555093555,
      "grad_norm": 8.523820877075195,
      "learning_rate": 1.1692066542470201e-05,
      "loss": 0.2743,
      "num_input_tokens_seen": 1374240,
      "step": 3420
    },
    {
      "epoch": 3.5602910602910605,
      "grad_norm": 13.287330627441406,
      "learning_rate": 1.1615361202743088e-05,
      "loss": 0.2821,
      "num_input_tokens_seen": 1376160,
      "step": 3425
    },
    {
      "epoch": 3.5654885654885655,
      "grad_norm": 6.943297386169434,
      "learning_rate": 1.1538832089331628e-05,
      "loss": 0.2579,
      "num_input_tokens_seen": 1378208,
      "step": 3430
    },
    {
      "epoch": 3.5706860706860706,
      "grad_norm": 9.258183479309082,
      "learning_rate": 1.1462480209841928e-05,
      "loss": 0.2526,
      "num_input_tokens_seen": 1380192,
      "step": 3435
    },
    {
      "epoch": 3.5758835758835756,
      "grad_norm": 0.5721746683120728,
      "learning_rate": 1.138630656954658e-05,
      "loss": 0.2369,
      "num_input_tokens_seen": 1382368,
      "step": 3440
    },
    {
      "epoch": 3.581081081081081,
      "grad_norm": 7.4510297775268555,
      "learning_rate": 1.1310312171371393e-05,
      "loss": 0.3155,
      "num_input_tokens_seen": 1384608,
      "step": 3445
    },
    {
      "epoch": 3.586278586278586,
      "grad_norm": 17.350297927856445,
      "learning_rate": 1.1234498015882261e-05,
      "loss": 0.2615,
      "num_input_tokens_seen": 1386592,
      "step": 3450
    },
    {
      "epoch": 3.5914760914760917,
      "grad_norm": 5.5357513427734375,
      "learning_rate": 1.1158865101271906e-05,
      "loss": 0.2614,
      "num_input_tokens_seen": 1388448,
      "step": 3455
    },
    {
      "epoch": 3.5966735966735968,
      "grad_norm": 7.910511016845703,
      "learning_rate": 1.1083414423346807e-05,
      "loss": 0.2222,
      "num_input_tokens_seen": 1390560,
      "step": 3460
    },
    {
      "epoch": 3.601871101871102,
      "grad_norm": 13.7831392288208,
      "learning_rate": 1.1008146975514059e-05,
      "loss": 0.3213,
      "num_input_tokens_seen": 1392736,
      "step": 3465
    },
    {
      "epoch": 3.607068607068607,
      "grad_norm": 0.9229243397712708,
      "learning_rate": 1.0933063748768254e-05,
      "loss": 0.2762,
      "num_input_tokens_seen": 1394720,
      "step": 3470
    },
    {
      "epoch": 3.6122661122661124,
      "grad_norm": 8.663399696350098,
      "learning_rate": 1.0858165731678513e-05,
      "loss": 0.23,
      "num_input_tokens_seen": 1396640,
      "step": 3475
    },
    {
      "epoch": 3.6174636174636174,
      "grad_norm": 13.28499984741211,
      "learning_rate": 1.0783453910375424e-05,
      "loss": 0.3322,
      "num_input_tokens_seen": 1398752,
      "step": 3480
    },
    {
      "epoch": 3.6226611226611225,
      "grad_norm": 8.6603364944458,
      "learning_rate": 1.0708929268538034e-05,
      "loss": 0.2494,
      "num_input_tokens_seen": 1400800,
      "step": 3485
    },
    {
      "epoch": 3.627858627858628,
      "grad_norm": 16.12564468383789,
      "learning_rate": 1.0634592787380965e-05,
      "loss": 0.2596,
      "num_input_tokens_seen": 1402720,
      "step": 3490
    },
    {
      "epoch": 3.633056133056133,
      "grad_norm": 7.862550735473633,
      "learning_rate": 1.0560445445641423e-05,
      "loss": 0.2563,
      "num_input_tokens_seen": 1404704,
      "step": 3495
    },
    {
      "epoch": 3.638253638253638,
      "grad_norm": 0.07106157392263412,
      "learning_rate": 1.048648821956637e-05,
      "loss": 0.2589,
      "num_input_tokens_seen": 1406560,
      "step": 3500
    },
    {
      "epoch": 3.643451143451143,
      "grad_norm": 0.42181286215782166,
      "learning_rate": 1.0412722082899644e-05,
      "loss": 0.2386,
      "num_input_tokens_seen": 1408544,
      "step": 3505
    },
    {
      "epoch": 3.6486486486486487,
      "grad_norm": 0.11339768022298813,
      "learning_rate": 1.033914800686912e-05,
      "loss": 0.2512,
      "num_input_tokens_seen": 1410464,
      "step": 3510
    },
    {
      "epoch": 3.6538461538461537,
      "grad_norm": 16.115554809570312,
      "learning_rate": 1.0265766960173965e-05,
      "loss": 0.2277,
      "num_input_tokens_seen": 1412448,
      "step": 3515
    },
    {
      "epoch": 3.6590436590436592,
      "grad_norm": 18.590940475463867,
      "learning_rate": 1.019257990897185e-05,
      "loss": 0.2869,
      "num_input_tokens_seen": 1414688,
      "step": 3520
    },
    {
      "epoch": 3.6642411642411643,
      "grad_norm": 5.8544158935546875,
      "learning_rate": 1.0119587816866258e-05,
      "loss": 0.2914,
      "num_input_tokens_seen": 1416672,
      "step": 3525
    },
    {
      "epoch": 3.6694386694386694,
      "grad_norm": 8.909049034118652,
      "learning_rate": 1.0046791644893758e-05,
      "loss": 0.2836,
      "num_input_tokens_seen": 1418592,
      "step": 3530
    },
    {
      "epoch": 3.6746361746361744,
      "grad_norm": 5.8211212158203125,
      "learning_rate": 9.974192351511368e-06,
      "loss": 0.2675,
      "num_input_tokens_seen": 1420576,
      "step": 3535
    },
    {
      "epoch": 3.67983367983368,
      "grad_norm": 7.190529823303223,
      "learning_rate": 9.901790892583974e-06,
      "loss": 0.2679,
      "num_input_tokens_seen": 1422560,
      "step": 3540
    },
    {
      "epoch": 3.685031185031185,
      "grad_norm": 6.408370018005371,
      "learning_rate": 9.829588221371694e-06,
      "loss": 0.2697,
      "num_input_tokens_seen": 1424608,
      "step": 3545
    },
    {
      "epoch": 3.6902286902286905,
      "grad_norm": 3.801816940307617,
      "learning_rate": 9.757585288517328e-06,
      "loss": 0.2612,
      "num_input_tokens_seen": 1426784,
      "step": 3550
    },
    {
      "epoch": 3.6954261954261955,
      "grad_norm": 2.5229806900024414,
      "learning_rate": 9.6857830420339e-06,
      "loss": 0.2501,
      "num_input_tokens_seen": 1428896,
      "step": 3555
    },
    {
      "epoch": 3.7006237006237006,
      "grad_norm": 7.06850004196167,
      "learning_rate": 9.614182427292077e-06,
      "loss": 0.2586,
      "num_input_tokens_seen": 1430880,
      "step": 3560
    },
    {
      "epoch": 3.7058212058212057,
      "grad_norm": 6.542532920837402,
      "learning_rate": 9.54278438700785e-06,
      "loss": 0.2484,
      "num_input_tokens_seen": 1432864,
      "step": 3565
    },
    {
      "epoch": 3.711018711018711,
      "grad_norm": 15.00790023803711,
      "learning_rate": 9.471589861229998e-06,
      "loss": 0.287,
      "num_input_tokens_seen": 1434912,
      "step": 3570
    },
    {
      "epoch": 3.7162162162162162,
      "grad_norm": 3.730722188949585,
      "learning_rate": 9.400599787327773e-06,
      "loss": 0.3025,
      "num_input_tokens_seen": 1436832,
      "step": 3575
    },
    {
      "epoch": 3.7214137214137213,
      "grad_norm": 0.24840876460075378,
      "learning_rate": 9.329815099978568e-06,
      "loss": 0.2433,
      "num_input_tokens_seen": 1438752,
      "step": 3580
    },
    {
      "epoch": 3.726611226611227,
      "grad_norm": 12.727954864501953,
      "learning_rate": 9.259236731155582e-06,
      "loss": 0.3809,
      "num_input_tokens_seen": 1440672,
      "step": 3585
    },
    {
      "epoch": 3.731808731808732,
      "grad_norm": 3.5419039726257324,
      "learning_rate": 9.18886561011557e-06,
      "loss": 0.2221,
      "num_input_tokens_seen": 1442784,
      "step": 3590
    },
    {
      "epoch": 3.737006237006237,
      "grad_norm": 4.471946716308594,
      "learning_rate": 9.118702663386584e-06,
      "loss": 0.2622,
      "num_input_tokens_seen": 1444960,
      "step": 3595
    },
    {
      "epoch": 3.742203742203742,
      "grad_norm": 7.390607833862305,
      "learning_rate": 9.048748814755784e-06,
      "loss": 0.3094,
      "num_input_tokens_seen": 1446880,
      "step": 3600
    },
    {
      "epoch": 3.7474012474012475,
      "grad_norm": 0.6072008013725281,
      "learning_rate": 8.979004985257294e-06,
      "loss": 0.2723,
      "num_input_tokens_seen": 1448992,
      "step": 3605
    },
    {
      "epoch": 3.7525987525987525,
      "grad_norm": 5.596619606018066,
      "learning_rate": 8.909472093160065e-06,
      "loss": 0.2755,
      "num_input_tokens_seen": 1450976,
      "step": 3610
    },
    {
      "epoch": 3.757796257796258,
      "grad_norm": 4.79258394241333,
      "learning_rate": 8.840151053955773e-06,
      "loss": 0.2751,
      "num_input_tokens_seen": 1453088,
      "step": 3615
    },
    {
      "epoch": 3.757796257796258,
      "eval_loss": 0.25871533155441284,
      "eval_runtime": 1.0494,
      "eval_samples_per_second": 815.737,
      "eval_steps_per_second": 101.967,
      "num_input_tokens_seen": 1453088,
      "step": 3615
    },
    {
      "epoch": 3.762993762993763,
      "grad_norm": 9.075246810913086,
      "learning_rate": 8.771042780346766e-06,
      "loss": 0.2834,
      "num_input_tokens_seen": 1455136,
      "step": 3620
    },
    {
      "epoch": 3.768191268191268,
      "grad_norm": 3.0494143962860107,
      "learning_rate": 8.702148182234043e-06,
      "loss": 0.2472,
      "num_input_tokens_seen": 1457120,
      "step": 3625
    },
    {
      "epoch": 3.773388773388773,
      "grad_norm": 3.5605342388153076,
      "learning_rate": 8.633468166705336e-06,
      "loss": 0.2772,
      "num_input_tokens_seen": 1459168,
      "step": 3630
    },
    {
      "epoch": 3.7785862785862787,
      "grad_norm": 3.971622943878174,
      "learning_rate": 8.565003638023065e-06,
      "loss": 0.2651,
      "num_input_tokens_seen": 1461152,
      "step": 3635
    },
    {
      "epoch": 3.7837837837837838,
      "grad_norm": 2.8847694396972656,
      "learning_rate": 8.496755497612492e-06,
      "loss": 0.2756,
      "num_input_tokens_seen": 1463136,
      "step": 3640
    },
    {
      "epoch": 3.788981288981289,
      "grad_norm": 9.094311714172363,
      "learning_rate": 8.42872464404986e-06,
      "loss": 0.2693,
      "num_input_tokens_seen": 1465120,
      "step": 3645
    },
    {
      "epoch": 3.7941787941787943,
      "grad_norm": 3.9456887245178223,
      "learning_rate": 8.360911973050537e-06,
      "loss": 0.2816,
      "num_input_tokens_seen": 1467104,
      "step": 3650
    },
    {
      "epoch": 3.7993762993762994,
      "grad_norm": 9.080158233642578,
      "learning_rate": 8.293318377457241e-06,
      "loss": 0.2571,
      "num_input_tokens_seen": 1469152,
      "step": 3655
    },
    {
      "epoch": 3.8045738045738045,
      "grad_norm": 3.958771228790283,
      "learning_rate": 8.225944747228257e-06,
      "loss": 0.268,
      "num_input_tokens_seen": 1471264,
      "step": 3660
    },
    {
      "epoch": 3.8097713097713095,
      "grad_norm": 8.621827125549316,
      "learning_rate": 8.158791969425738e-06,
      "loss": 0.2128,
      "num_input_tokens_seen": 1473248,
      "step": 3665
    },
    {
      "epoch": 3.814968814968815,
      "grad_norm": 14.013006210327148,
      "learning_rate": 8.091860928204049e-06,
      "loss": 0.3101,
      "num_input_tokens_seen": 1475360,
      "step": 3670
    },
    {
      "epoch": 3.82016632016632,
      "grad_norm": 11.601419448852539,
      "learning_rate": 8.025152504798078e-06,
      "loss": 0.3044,
      "num_input_tokens_seen": 1477472,
      "step": 3675
    },
    {
      "epoch": 3.8253638253638256,
      "grad_norm": 3.5920257568359375,
      "learning_rate": 7.958667577511683e-06,
      "loss": 0.2471,
      "num_input_tokens_seen": 1479328,
      "step": 3680
    },
    {
      "epoch": 3.8305613305613306,
      "grad_norm": 11.748164176940918,
      "learning_rate": 7.892407021706063e-06,
      "loss": 0.2552,
      "num_input_tokens_seen": 1481248,
      "step": 3685
    },
    {
      "epoch": 3.8357588357588357,
      "grad_norm": 11.761496543884277,
      "learning_rate": 7.826371709788313e-06,
      "loss": 0.3112,
      "num_input_tokens_seen": 1483168,
      "step": 3690
    },
    {
      "epoch": 3.8409563409563408,
      "grad_norm": 2.8758435249328613,
      "learning_rate": 7.760562511199882e-06,
      "loss": 0.2585,
      "num_input_tokens_seen": 1485152,
      "step": 3695
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 3.748758554458618,
      "learning_rate": 7.694980292405122e-06,
      "loss": 0.2673,
      "num_input_tokens_seen": 1487200,
      "step": 3700
    },
    {
      "epoch": 3.8513513513513513,
      "grad_norm": 5.409223556518555,
      "learning_rate": 7.629625916879932e-06,
      "loss": 0.2763,
      "num_input_tokens_seen": 1489184,
      "step": 3705
    },
    {
      "epoch": 3.856548856548857,
      "grad_norm": 13.337824821472168,
      "learning_rate": 7.564500245100325e-06,
      "loss": 0.278,
      "num_input_tokens_seen": 1491168,
      "step": 3710
    },
    {
      "epoch": 3.861746361746362,
      "grad_norm": 1.357743740081787,
      "learning_rate": 7.499604134531149e-06,
      "loss": 0.2727,
      "num_input_tokens_seen": 1493216,
      "step": 3715
    },
    {
      "epoch": 3.866943866943867,
      "grad_norm": 0.9974175095558167,
      "learning_rate": 7.434938439614781e-06,
      "loss": 0.2667,
      "num_input_tokens_seen": 1495200,
      "step": 3720
    },
    {
      "epoch": 3.872141372141372,
      "grad_norm": 1.3982412815093994,
      "learning_rate": 7.370504011759855e-06,
      "loss": 0.2707,
      "num_input_tokens_seen": 1497184,
      "step": 3725
    },
    {
      "epoch": 3.8773388773388775,
      "grad_norm": 2.715106248855591,
      "learning_rate": 7.306301699330065e-06,
      "loss": 0.2656,
      "num_input_tokens_seen": 1499040,
      "step": 3730
    },
    {
      "epoch": 3.8825363825363826,
      "grad_norm": 10.939849853515625,
      "learning_rate": 7.242332347633052e-06,
      "loss": 0.2423,
      "num_input_tokens_seen": 1501024,
      "step": 3735
    },
    {
      "epoch": 3.8877338877338876,
      "grad_norm": 8.844660758972168,
      "learning_rate": 7.178596798909159e-06,
      "loss": 0.2487,
      "num_input_tokens_seen": 1503072,
      "step": 3740
    },
    {
      "epoch": 3.892931392931393,
      "grad_norm": 0.9195927381515503,
      "learning_rate": 7.115095892320456e-06,
      "loss": 0.2847,
      "num_input_tokens_seen": 1505248,
      "step": 3745
    },
    {
      "epoch": 3.898128898128898,
      "grad_norm": 15.075143814086914,
      "learning_rate": 7.051830463939604e-06,
      "loss": 0.2334,
      "num_input_tokens_seen": 1507296,
      "step": 3750
    },
    {
      "epoch": 3.9033264033264032,
      "grad_norm": 5.580709934234619,
      "learning_rate": 6.98880134673891e-06,
      "loss": 0.2595,
      "num_input_tokens_seen": 1509344,
      "step": 3755
    },
    {
      "epoch": 3.9085239085239083,
      "grad_norm": 5.476256370544434,
      "learning_rate": 6.926009370579334e-06,
      "loss": 0.2098,
      "num_input_tokens_seen": 1511456,
      "step": 3760
    },
    {
      "epoch": 3.913721413721414,
      "grad_norm": 3.832021713256836,
      "learning_rate": 6.8634553621995416e-06,
      "loss": 0.2277,
      "num_input_tokens_seen": 1513440,
      "step": 3765
    },
    {
      "epoch": 3.918918918918919,
      "grad_norm": 15.132575035095215,
      "learning_rate": 6.80114014520507e-06,
      "loss": 0.3251,
      "num_input_tokens_seen": 1515488,
      "step": 3770
    },
    {
      "epoch": 3.9241164241164244,
      "grad_norm": 5.868591785430908,
      "learning_rate": 6.739064540057424e-06,
      "loss": 0.2604,
      "num_input_tokens_seen": 1517408,
      "step": 3775
    },
    {
      "epoch": 3.9293139293139294,
      "grad_norm": 11.784462928771973,
      "learning_rate": 6.677229364063328e-06,
      "loss": 0.2458,
      "num_input_tokens_seen": 1519392,
      "step": 3780
    },
    {
      "epoch": 3.9345114345114345,
      "grad_norm": 3.2488772869110107,
      "learning_rate": 6.615635431363942e-06,
      "loss": 0.2596,
      "num_input_tokens_seen": 1521440,
      "step": 3785
    },
    {
      "epoch": 3.9397089397089395,
      "grad_norm": 3.991219997406006,
      "learning_rate": 6.554283552924118e-06,
      "loss": 0.2766,
      "num_input_tokens_seen": 1523488,
      "step": 3790
    },
    {
      "epoch": 3.944906444906445,
      "grad_norm": 2.877652168273926,
      "learning_rate": 6.493174536521768e-06,
      "loss": 0.2551,
      "num_input_tokens_seen": 1525600,
      "step": 3795
    },
    {
      "epoch": 3.95010395010395,
      "grad_norm": 10.219852447509766,
      "learning_rate": 6.4323091867372095e-06,
      "loss": 0.2612,
      "num_input_tokens_seen": 1527584,
      "step": 3800
    },
    {
      "epoch": 3.955301455301455,
      "grad_norm": 9.425738334655762,
      "learning_rate": 6.371688304942544e-06,
      "loss": 0.2575,
      "num_input_tokens_seen": 1529504,
      "step": 3805
    },
    {
      "epoch": 3.9604989604989607,
      "grad_norm": 13.004129409790039,
      "learning_rate": 6.311312689291166e-06,
      "loss": 0.2897,
      "num_input_tokens_seen": 1531424,
      "step": 3810
    },
    {
      "epoch": 3.9656964656964657,
      "grad_norm": 7.159772872924805,
      "learning_rate": 6.251183134707184e-06,
      "loss": 0.2366,
      "num_input_tokens_seen": 1533408,
      "step": 3815
    },
    {
      "epoch": 3.970893970893971,
      "grad_norm": 6.52066707611084,
      "learning_rate": 6.191300432875017e-06,
      "loss": 0.2668,
      "num_input_tokens_seen": 1535392,
      "step": 3820
    },
    {
      "epoch": 3.976091476091476,
      "grad_norm": 13.160318374633789,
      "learning_rate": 6.13166537222894e-06,
      "loss": 0.1661,
      "num_input_tokens_seen": 1537312,
      "step": 3825
    },
    {
      "epoch": 3.9812889812889813,
      "grad_norm": 16.36733055114746,
      "learning_rate": 6.072278737942691e-06,
      "loss": 0.28,
      "num_input_tokens_seen": 1539360,
      "step": 3830
    },
    {
      "epoch": 3.9864864864864864,
      "grad_norm": 4.4265875816345215,
      "learning_rate": 6.0131413119191685e-06,
      "loss": 0.3038,
      "num_input_tokens_seen": 1541280,
      "step": 3835
    },
    {
      "epoch": 3.991683991683992,
      "grad_norm": 4.7540717124938965,
      "learning_rate": 5.954253872780102e-06,
      "loss": 0.2598,
      "num_input_tokens_seen": 1543136,
      "step": 3840
    },
    {
      "epoch": 3.996881496881497,
      "grad_norm": 1.5352482795715332,
      "learning_rate": 5.8956171958558266e-06,
      "loss": 0.2387,
      "num_input_tokens_seen": 1545120,
      "step": 3845
    },
    {
      "epoch": 4.002079002079002,
      "grad_norm": 0.05950174108147621,
      "learning_rate": 5.8372320531750655e-06,
      "loss": 0.2258,
      "num_input_tokens_seen": 1547056,
      "step": 3850
    },
    {
      "epoch": 4.007276507276507,
      "grad_norm": 9.390141487121582,
      "learning_rate": 5.77909921345475e-06,
      "loss": 0.279,
      "num_input_tokens_seen": 1548976,
      "step": 3855
    },
    {
      "epoch": 4.008316008316008,
      "eval_loss": 0.25588732957839966,
      "eval_runtime": 1.0892,
      "eval_samples_per_second": 785.884,
      "eval_steps_per_second": 98.235,
      "num_input_tokens_seen": 1549360,
      "step": 3856
    },
    {
      "epoch": 4.012474012474012,
      "grad_norm": 4.292252063751221,
      "learning_rate": 5.721219442089926e-06,
      "loss": 0.2636,
      "num_input_tokens_seen": 1550960,
      "step": 3860
    },
    {
      "epoch": 4.017671517671518,
      "grad_norm": 3.9697649478912354,
      "learning_rate": 5.663593501143663e-06,
      "loss": 0.2772,
      "num_input_tokens_seen": 1552944,
      "step": 3865
    },
    {
      "epoch": 4.022869022869023,
      "grad_norm": 1.579316258430481,
      "learning_rate": 5.6062221493370035e-06,
      "loss": 0.2654,
      "num_input_tokens_seen": 1554992,
      "step": 3870
    },
    {
      "epoch": 4.028066528066528,
      "grad_norm": 14.188766479492188,
      "learning_rate": 5.549106142039018e-06,
      "loss": 0.2682,
      "num_input_tokens_seen": 1557104,
      "step": 3875
    },
    {
      "epoch": 4.033264033264033,
      "grad_norm": 7.652184963226318,
      "learning_rate": 5.492246231256798e-06,
      "loss": 0.2818,
      "num_input_tokens_seen": 1559088,
      "step": 3880
    },
    {
      "epoch": 4.038461538461538,
      "grad_norm": 7.617486953735352,
      "learning_rate": 5.435643165625614e-06,
      "loss": 0.2739,
      "num_input_tokens_seen": 1561008,
      "step": 3885
    },
    {
      "epoch": 4.043659043659043,
      "grad_norm": 14.909598350524902,
      "learning_rate": 5.379297690399035e-06,
      "loss": 0.273,
      "num_input_tokens_seen": 1563056,
      "step": 3890
    },
    {
      "epoch": 4.048856548856548,
      "grad_norm": 13.364591598510742,
      "learning_rate": 5.3232105474390895e-06,
      "loss": 0.2571,
      "num_input_tokens_seen": 1565040,
      "step": 3895
    },
    {
      "epoch": 4.054054054054054,
      "grad_norm": 4.7986159324646,
      "learning_rate": 5.267382475206548e-06,
      "loss": 0.2529,
      "num_input_tokens_seen": 1567024,
      "step": 3900
    },
    {
      "epoch": 4.0592515592515594,
      "grad_norm": 4.830848693847656,
      "learning_rate": 5.2118142087511705e-06,
      "loss": 0.264,
      "num_input_tokens_seen": 1569136,
      "step": 3905
    },
    {
      "epoch": 4.0644490644490645,
      "grad_norm": 11.776159286499023,
      "learning_rate": 5.156506479702019e-06,
      "loss": 0.2638,
      "num_input_tokens_seen": 1571120,
      "step": 3910
    },
    {
      "epoch": 4.06964656964657,
      "grad_norm": 0.2953391969203949,
      "learning_rate": 5.101460016257859e-06,
      "loss": 0.2652,
      "num_input_tokens_seen": 1573040,
      "step": 3915
    },
    {
      "epoch": 4.074844074844075,
      "grad_norm": 0.2401997148990631,
      "learning_rate": 5.0466755431775316e-06,
      "loss": 0.2582,
      "num_input_tokens_seen": 1574896,
      "step": 3920
    },
    {
      "epoch": 4.08004158004158,
      "grad_norm": 0.6011865139007568,
      "learning_rate": 4.992153781770448e-06,
      "loss": 0.2618,
      "num_input_tokens_seen": 1576880,
      "step": 3925
    },
    {
      "epoch": 4.085239085239086,
      "grad_norm": 12.98713493347168,
      "learning_rate": 4.937895449887075e-06,
      "loss": 0.2455,
      "num_input_tokens_seen": 1578864,
      "step": 3930
    },
    {
      "epoch": 4.090436590436591,
      "grad_norm": 13.129234313964844,
      "learning_rate": 4.883901261909465e-06,
      "loss": 0.2813,
      "num_input_tokens_seen": 1580848,
      "step": 3935
    },
    {
      "epoch": 4.095634095634096,
      "grad_norm": 11.796737670898438,
      "learning_rate": 4.8301719287419e-06,
      "loss": 0.2815,
      "num_input_tokens_seen": 1582704,
      "step": 3940
    },
    {
      "epoch": 4.100831600831601,
      "grad_norm": 9.959646224975586,
      "learning_rate": 4.776708157801463e-06,
      "loss": 0.2796,
      "num_input_tokens_seen": 1584816,
      "step": 3945
    },
    {
      "epoch": 4.106029106029106,
      "grad_norm": 11.752809524536133,
      "learning_rate": 4.7235106530088085e-06,
      "loss": 0.2491,
      "num_input_tokens_seen": 1586800,
      "step": 3950
    },
    {
      "epoch": 4.111226611226611,
      "grad_norm": 5.360927104949951,
      "learning_rate": 4.670580114778813e-06,
      "loss": 0.2528,
      "num_input_tokens_seen": 1588720,
      "step": 3955
    },
    {
      "epoch": 4.116424116424117,
      "grad_norm": 5.705671310424805,
      "learning_rate": 4.617917240011394e-06,
      "loss": 0.2732,
      "num_input_tokens_seen": 1590576,
      "step": 3960
    },
    {
      "epoch": 4.121621621621622,
      "grad_norm": 1.100892186164856,
      "learning_rate": 4.565522722082336e-06,
      "loss": 0.2573,
      "num_input_tokens_seen": 1592496,
      "step": 3965
    },
    {
      "epoch": 4.126819126819127,
      "grad_norm": 1.0973014831542969,
      "learning_rate": 4.513397250834159e-06,
      "loss": 0.2638,
      "num_input_tokens_seen": 1594544,
      "step": 3970
    },
    {
      "epoch": 4.132016632016632,
      "grad_norm": 8.093790054321289,
      "learning_rate": 4.461541512567011e-06,
      "loss": 0.2925,
      "num_input_tokens_seen": 1596400,
      "step": 3975
    },
    {
      "epoch": 4.137214137214137,
      "grad_norm": 0.3267320990562439,
      "learning_rate": 4.409956190029674e-06,
      "loss": 0.2786,
      "num_input_tokens_seen": 1598320,
      "step": 3980
    },
    {
      "epoch": 4.142411642411642,
      "grad_norm": 0.30618995428085327,
      "learning_rate": 4.358641962410537e-06,
      "loss": 0.2286,
      "num_input_tokens_seen": 1600368,
      "step": 3985
    },
    {
      "epoch": 4.147609147609147,
      "grad_norm": 12.98468017578125,
      "learning_rate": 4.307599505328672e-06,
      "loss": 0.2986,
      "num_input_tokens_seen": 1602352,
      "step": 3990
    },
    {
      "epoch": 4.152806652806653,
      "grad_norm": 0.18394550681114197,
      "learning_rate": 4.256829490824949e-06,
      "loss": 0.2363,
      "num_input_tokens_seen": 1604336,
      "step": 3995
    },
    {
      "epoch": 4.158004158004158,
      "grad_norm": 6.980307102203369,
      "learning_rate": 4.206332587353149e-06,
      "loss": 0.2652,
      "num_input_tokens_seen": 1606256,
      "step": 4000
    },
    {
      "epoch": 4.163201663201663,
      "grad_norm": 0.6148359775543213,
      "learning_rate": 4.1561094597712155e-06,
      "loss": 0.2641,
      "num_input_tokens_seen": 1608304,
      "step": 4005
    },
    {
      "epoch": 4.168399168399168,
      "grad_norm": 9.185464859008789,
      "learning_rate": 4.106160769332443e-06,
      "loss": 0.2415,
      "num_input_tokens_seen": 1610480,
      "step": 4010
    },
    {
      "epoch": 4.173596673596673,
      "grad_norm": 0.5142713189125061,
      "learning_rate": 4.056487173676843e-06,
      "loss": 0.2449,
      "num_input_tokens_seen": 1612528,
      "step": 4015
    },
    {
      "epoch": 4.1787941787941785,
      "grad_norm": 15.734733581542969,
      "learning_rate": 4.007089326822405e-06,
      "loss": 0.2742,
      "num_input_tokens_seen": 1614576,
      "step": 4020
    },
    {
      "epoch": 4.183991683991684,
      "grad_norm": 1.1118292808532715,
      "learning_rate": 3.957967879156533e-06,
      "loss": 0.264,
      "num_input_tokens_seen": 1616624,
      "step": 4025
    },
    {
      "epoch": 4.1891891891891895,
      "grad_norm": 12.48095989227295,
      "learning_rate": 3.909123477427487e-06,
      "loss": 0.2548,
      "num_input_tokens_seen": 1618672,
      "step": 4030
    },
    {
      "epoch": 4.1943866943866945,
      "grad_norm": 5.244612693786621,
      "learning_rate": 3.860556764735842e-06,
      "loss": 0.2113,
      "num_input_tokens_seen": 1620784,
      "step": 4035
    },
    {
      "epoch": 4.1995841995842,
      "grad_norm": 21.30678367614746,
      "learning_rate": 3.812268380526046e-06,
      "loss": 0.3261,
      "num_input_tokens_seen": 1622768,
      "step": 4040
    },
    {
      "epoch": 4.204781704781705,
      "grad_norm": 8.002152442932129,
      "learning_rate": 3.764258960577971e-06,
      "loss": 0.2481,
      "num_input_tokens_seen": 1624688,
      "step": 4045
    },
    {
      "epoch": 4.20997920997921,
      "grad_norm": 2.3525662422180176,
      "learning_rate": 3.7165291369985618e-06,
      "loss": 0.2599,
      "num_input_tokens_seen": 1626672,
      "step": 4050
    },
    {
      "epoch": 4.215176715176715,
      "grad_norm": 4.8225202560424805,
      "learning_rate": 3.6690795382135186e-06,
      "loss": 0.258,
      "num_input_tokens_seen": 1628848,
      "step": 4055
    },
    {
      "epoch": 4.220374220374221,
      "grad_norm": 1.9619829654693604,
      "learning_rate": 3.6219107889590155e-06,
      "loss": 0.2809,
      "num_input_tokens_seen": 1630832,
      "step": 4060
    },
    {
      "epoch": 4.225571725571726,
      "grad_norm": 20.82301139831543,
      "learning_rate": 3.575023510273462e-06,
      "loss": 0.2292,
      "num_input_tokens_seen": 1632880,
      "step": 4065
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 1.0340510606765747,
      "learning_rate": 3.5284183194893488e-06,
      "loss": 0.2871,
      "num_input_tokens_seen": 1634992,
      "step": 4070
    },
    {
      "epoch": 4.235966735966736,
      "grad_norm": 12.200738906860352,
      "learning_rate": 3.48209583022511e-06,
      "loss": 0.2634,
      "num_input_tokens_seen": 1636912,
      "step": 4075
    },
    {
      "epoch": 4.241164241164241,
      "grad_norm": 14.348763465881348,
      "learning_rate": 3.4360566523770426e-06,
      "loss": 0.2477,
      "num_input_tokens_seen": 1638832,
      "step": 4080
    },
    {
      "epoch": 4.246361746361746,
      "grad_norm": 3.283851385116577,
      "learning_rate": 3.3903013921112755e-06,
      "loss": 0.2351,
      "num_input_tokens_seen": 1641072,
      "step": 4085
    },
    {
      "epoch": 4.251559251559252,
      "grad_norm": 1.4203660488128662,
      "learning_rate": 3.3448306518557795e-06,
      "loss": 0.2808,
      "num_input_tokens_seen": 1642992,
      "step": 4090
    },
    {
      "epoch": 4.256756756756757,
      "grad_norm": 5.4867753982543945,
      "learning_rate": 3.299645030292467e-06,
      "loss": 0.2511,
      "num_input_tokens_seen": 1645040,
      "step": 4095
    },
    {
      "epoch": 4.258835758835759,
      "eval_loss": 0.2517484128475189,
      "eval_runtime": 1.069,
      "eval_samples_per_second": 800.752,
      "eval_steps_per_second": 100.094,
      "num_input_tokens_seen": 1645808,
      "step": 4097
    },
    {
      "epoch": 4.261954261954262,
      "grad_norm": 15.803162574768066,
      "learning_rate": 3.2547451223492786e-06,
      "loss": 0.2495,
      "num_input_tokens_seen": 1647024,
      "step": 4100
    },
    {
      "epoch": 4.267151767151767,
      "grad_norm": 6.709695816040039,
      "learning_rate": 3.2101315191923663e-06,
      "loss": 0.2774,
      "num_input_tokens_seen": 1649008,
      "step": 4105
    },
    {
      "epoch": 4.272349272349272,
      "grad_norm": 12.675971984863281,
      "learning_rate": 3.165804808218292e-06,
      "loss": 0.2208,
      "num_input_tokens_seen": 1651056,
      "step": 4110
    },
    {
      "epoch": 4.277546777546777,
      "grad_norm": 20.12381935119629,
      "learning_rate": 3.1217655730463093e-06,
      "loss": 0.2601,
      "num_input_tokens_seen": 1653104,
      "step": 4115
    },
    {
      "epoch": 4.282744282744282,
      "grad_norm": 5.935137748718262,
      "learning_rate": 3.078014393510695e-06,
      "loss": 0.2427,
      "num_input_tokens_seen": 1655344,
      "step": 4120
    },
    {
      "epoch": 4.287941787941788,
      "grad_norm": 3.3360893726348877,
      "learning_rate": 3.0345518456530665e-06,
      "loss": 0.2959,
      "num_input_tokens_seen": 1657392,
      "step": 4125
    },
    {
      "epoch": 4.293139293139293,
      "grad_norm": 8.30186653137207,
      "learning_rate": 2.991378501714856e-06,
      "loss": 0.2807,
      "num_input_tokens_seen": 1659312,
      "step": 4130
    },
    {
      "epoch": 4.298336798336798,
      "grad_norm": 7.54670524597168,
      "learning_rate": 2.9484949301297166e-06,
      "loss": 0.2579,
      "num_input_tokens_seen": 1661424,
      "step": 4135
    },
    {
      "epoch": 4.303534303534303,
      "grad_norm": 14.763995170593262,
      "learning_rate": 2.9059016955160916e-06,
      "loss": 0.2498,
      "num_input_tokens_seen": 1663408,
      "step": 4140
    },
    {
      "epoch": 4.3087318087318085,
      "grad_norm": 1.220461368560791,
      "learning_rate": 2.8635993586697553e-06,
      "loss": 0.226,
      "num_input_tokens_seen": 1665328,
      "step": 4145
    },
    {
      "epoch": 4.313929313929314,
      "grad_norm": 23.395973205566406,
      "learning_rate": 2.8215884765564193e-06,
      "loss": 0.2687,
      "num_input_tokens_seen": 1667312,
      "step": 4150
    },
    {
      "epoch": 4.3191268191268195,
      "grad_norm": 4.078779697418213,
      "learning_rate": 2.7798696023044163e-06,
      "loss": 0.1693,
      "num_input_tokens_seen": 1669296,
      "step": 4155
    },
    {
      "epoch": 4.324324324324325,
      "grad_norm": 5.839596748352051,
      "learning_rate": 2.73844328519742e-06,
      "loss": 0.2217,
      "num_input_tokens_seen": 1671280,
      "step": 4160
    },
    {
      "epoch": 4.32952182952183,
      "grad_norm": 21.18253517150879,
      "learning_rate": 2.6973100706672e-06,
      "loss": 0.2814,
      "num_input_tokens_seen": 1673456,
      "step": 4165
    },
    {
      "epoch": 4.334719334719335,
      "grad_norm": 6.5861101150512695,
      "learning_rate": 2.656470500286451e-06,
      "loss": 0.2523,
      "num_input_tokens_seen": 1675504,
      "step": 4170
    },
    {
      "epoch": 4.33991683991684,
      "grad_norm": 6.993749618530273,
      "learning_rate": 2.615925111761647e-06,
      "loss": 0.228,
      "num_input_tokens_seen": 1677488,
      "step": 4175
    },
    {
      "epoch": 4.345114345114345,
      "grad_norm": 11.91810417175293,
      "learning_rate": 2.5756744389259734e-06,
      "loss": 0.2399,
      "num_input_tokens_seen": 1679536,
      "step": 4180
    },
    {
      "epoch": 4.350311850311851,
      "grad_norm": 4.944723606109619,
      "learning_rate": 2.535719011732321e-06,
      "loss": 0.2345,
      "num_input_tokens_seen": 1681520,
      "step": 4185
    },
    {
      "epoch": 4.355509355509356,
      "grad_norm": 18.20621681213379,
      "learning_rate": 2.49605935624625e-06,
      "loss": 0.2726,
      "num_input_tokens_seen": 1683568,
      "step": 4190
    },
    {
      "epoch": 4.360706860706861,
      "grad_norm": 8.823187828063965,
      "learning_rate": 2.4566959946391243e-06,
      "loss": 0.236,
      "num_input_tokens_seen": 1685488,
      "step": 4195
    },
    {
      "epoch": 4.365904365904366,
      "grad_norm": 8.236385345458984,
      "learning_rate": 2.417629445181194e-06,
      "loss": 0.2476,
      "num_input_tokens_seen": 1687408,
      "step": 4200
    },
    {
      "epoch": 4.371101871101871,
      "grad_norm": 3.388185739517212,
      "learning_rate": 2.378860222234794e-06,
      "loss": 0.2325,
      "num_input_tokens_seen": 1689520,
      "step": 4205
    },
    {
      "epoch": 4.376299376299376,
      "grad_norm": 2.7209372520446777,
      "learning_rate": 2.3403888362475782e-06,
      "loss": 0.2749,
      "num_input_tokens_seen": 1691568,
      "step": 4210
    },
    {
      "epoch": 4.381496881496881,
      "grad_norm": 7.729001045227051,
      "learning_rate": 2.3022157937457627e-06,
      "loss": 0.2235,
      "num_input_tokens_seen": 1693616,
      "step": 4215
    },
    {
      "epoch": 4.386694386694387,
      "grad_norm": 9.053695678710938,
      "learning_rate": 2.2643415973275016e-06,
      "loss": 0.2369,
      "num_input_tokens_seen": 1695600,
      "step": 4220
    },
    {
      "epoch": 4.391891891891892,
      "grad_norm": 4.420502662658691,
      "learning_rate": 2.2267667456562307e-06,
      "loss": 0.285,
      "num_input_tokens_seen": 1697584,
      "step": 4225
    },
    {
      "epoch": 4.397089397089397,
      "grad_norm": 3.5097815990448,
      "learning_rate": 2.1894917334541354e-06,
      "loss": 0.2273,
      "num_input_tokens_seen": 1699568,
      "step": 4230
    },
    {
      "epoch": 4.402286902286902,
      "grad_norm": 23.638469696044922,
      "learning_rate": 2.15251705149562e-06,
      "loss": 0.2953,
      "num_input_tokens_seen": 1701744,
      "step": 4235
    },
    {
      "epoch": 4.407484407484407,
      "grad_norm": 4.72571325302124,
      "learning_rate": 2.11584318660083e-06,
      "loss": 0.2404,
      "num_input_tokens_seen": 1703600,
      "step": 4240
    },
    {
      "epoch": 4.412681912681912,
      "grad_norm": 6.068181991577148,
      "learning_rate": 2.0794706216292813e-06,
      "loss": 0.3067,
      "num_input_tokens_seen": 1705712,
      "step": 4245
    },
    {
      "epoch": 4.417879417879418,
      "grad_norm": 10.372715950012207,
      "learning_rate": 2.043399835473475e-06,
      "loss": 0.2088,
      "num_input_tokens_seen": 1707696,
      "step": 4250
    },
    {
      "epoch": 4.423076923076923,
      "grad_norm": 4.278080463409424,
      "learning_rate": 2.0076313030525844e-06,
      "loss": 0.2195,
      "num_input_tokens_seen": 1709744,
      "step": 4255
    },
    {
      "epoch": 4.428274428274428,
      "grad_norm": 9.075408935546875,
      "learning_rate": 1.972165495306241e-06,
      "loss": 0.2399,
      "num_input_tokens_seen": 1711792,
      "step": 4260
    },
    {
      "epoch": 4.4334719334719335,
      "grad_norm": 4.037149429321289,
      "learning_rate": 1.937002879188285e-06,
      "loss": 0.2491,
      "num_input_tokens_seen": 1713904,
      "step": 4265
    },
    {
      "epoch": 4.4386694386694385,
      "grad_norm": 7.768908977508545,
      "learning_rate": 1.9021439176606564e-06,
      "loss": 0.2257,
      "num_input_tokens_seen": 1715824,
      "step": 4270
    },
    {
      "epoch": 4.443866943866944,
      "grad_norm": 4.410028457641602,
      "learning_rate": 1.8675890696872838e-06,
      "loss": 0.2438,
      "num_input_tokens_seen": 1717808,
      "step": 4275
    },
    {
      "epoch": 4.4490644490644495,
      "grad_norm": 4.74402379989624,
      "learning_rate": 1.8333387902280314e-06,
      "loss": 0.2773,
      "num_input_tokens_seen": 1719856,
      "step": 4280
    },
    {
      "epoch": 4.454261954261955,
      "grad_norm": 6.120370864868164,
      "learning_rate": 1.7993935302327292e-06,
      "loss": 0.2193,
      "num_input_tokens_seen": 1721776,
      "step": 4285
    },
    {
      "epoch": 4.45945945945946,
      "grad_norm": 10.760017395019531,
      "learning_rate": 1.7657537366352338e-06,
      "loss": 0.238,
      "num_input_tokens_seen": 1723632,
      "step": 4290
    },
    {
      "epoch": 4.464656964656965,
      "grad_norm": 6.990230083465576,
      "learning_rate": 1.732419852347511e-06,
      "loss": 0.1772,
      "num_input_tokens_seen": 1725488,
      "step": 4295
    },
    {
      "epoch": 4.46985446985447,
      "grad_norm": 15.208341598510742,
      "learning_rate": 1.699392316253856e-06,
      "loss": 0.2837,
      "num_input_tokens_seen": 1727600,
      "step": 4300
    },
    {
      "epoch": 4.475051975051975,
      "grad_norm": 16.286996841430664,
      "learning_rate": 1.666671563205069e-06,
      "loss": 0.2494,
      "num_input_tokens_seen": 1729712,
      "step": 4305
    },
    {
      "epoch": 4.48024948024948,
      "grad_norm": 2.84039306640625,
      "learning_rate": 1.6342580240127582e-06,
      "loss": 0.269,
      "num_input_tokens_seen": 1731696,
      "step": 4310
    },
    {
      "epoch": 4.485446985446986,
      "grad_norm": 15.725839614868164,
      "learning_rate": 1.6021521254436678e-06,
      "loss": 0.2551,
      "num_input_tokens_seen": 1733744,
      "step": 4315
    },
    {
      "epoch": 4.490644490644491,
      "grad_norm": 4.358521461486816,
      "learning_rate": 1.5703542902140294e-06,
      "loss": 0.2408,
      "num_input_tokens_seen": 1735728,
      "step": 4320
    },
    {
      "epoch": 4.495841995841996,
      "grad_norm": 14.235426902770996,
      "learning_rate": 1.5388649369840357e-06,
      "loss": 0.1891,
      "num_input_tokens_seen": 1737776,
      "step": 4325
    },
    {
      "epoch": 4.501039501039501,
      "grad_norm": 7.304314136505127,
      "learning_rate": 1.5076844803522922e-06,
      "loss": 0.2684,
      "num_input_tokens_seen": 1739824,
      "step": 4330
    },
    {
      "epoch": 4.506237006237006,
      "grad_norm": 18.58903694152832,
      "learning_rate": 1.476813330850388e-06,
      "loss": 0.2709,
      "num_input_tokens_seen": 1741744,
      "step": 4335
    },
    {
      "epoch": 4.509355509355509,
      "eval_loss": 0.2577267587184906,
      "eval_runtime": 1.0636,
      "eval_samples_per_second": 804.851,
      "eval_steps_per_second": 100.606,
      "num_input_tokens_seen": 1742960,
      "step": 4338
    },
    {
      "epoch": 4.511434511434511,
      "grad_norm": 18.499958038330078,
      "learning_rate": 1.4462518949374838e-06,
      "loss": 0.2731,
      "num_input_tokens_seen": 1743728,
      "step": 4340
    },
    {
      "epoch": 4.516632016632016,
      "grad_norm": 4.451554298400879,
      "learning_rate": 1.4160005749949328e-06,
      "loss": 0.2431,
      "num_input_tokens_seen": 1745904,
      "step": 4345
    },
    {
      "epoch": 4.521829521829522,
      "grad_norm": 5.922986030578613,
      "learning_rate": 1.386059769321027e-06,
      "loss": 0.2649,
      "num_input_tokens_seen": 1747824,
      "step": 4350
    },
    {
      "epoch": 4.527027027027027,
      "grad_norm": 11.740480422973633,
      "learning_rate": 1.3564298721257223e-06,
      "loss": 0.2569,
      "num_input_tokens_seen": 1749872,
      "step": 4355
    },
    {
      "epoch": 4.532224532224532,
      "grad_norm": 10.447465896606445,
      "learning_rate": 1.3271112735254498e-06,
      "loss": 0.1998,
      "num_input_tokens_seen": 1751792,
      "step": 4360
    },
    {
      "epoch": 4.537422037422037,
      "grad_norm": 5.754942893981934,
      "learning_rate": 1.298104359538005e-06,
      "loss": 0.251,
      "num_input_tokens_seen": 1753776,
      "step": 4365
    },
    {
      "epoch": 4.542619542619542,
      "grad_norm": 8.049335479736328,
      "learning_rate": 1.269409512077427e-06,
      "loss": 0.2705,
      "num_input_tokens_seen": 1755824,
      "step": 4370
    },
    {
      "epoch": 4.547817047817047,
      "grad_norm": 4.818149089813232,
      "learning_rate": 1.241027108949e-06,
      "loss": 0.202,
      "num_input_tokens_seen": 1758000,
      "step": 4375
    },
    {
      "epoch": 4.553014553014553,
      "grad_norm": 10.757091522216797,
      "learning_rate": 1.2129575238442715e-06,
      "loss": 0.2565,
      "num_input_tokens_seen": 1759984,
      "step": 4380
    },
    {
      "epoch": 4.558212058212058,
      "grad_norm": 23.19622230529785,
      "learning_rate": 1.185201126336122e-06,
      "loss": 0.2607,
      "num_input_tokens_seen": 1761968,
      "step": 4385
    },
    {
      "epoch": 4.5634095634095635,
      "grad_norm": 6.780999660491943,
      "learning_rate": 1.1577582818739135e-06,
      "loss": 0.2392,
      "num_input_tokens_seen": 1764016,
      "step": 4390
    },
    {
      "epoch": 4.5686070686070686,
      "grad_norm": 6.052558422088623,
      "learning_rate": 1.1306293517786614e-06,
      "loss": 0.2808,
      "num_input_tokens_seen": 1765936,
      "step": 4395
    },
    {
      "epoch": 4.573804573804574,
      "grad_norm": 11.568082809448242,
      "learning_rate": 1.1038146932383004e-06,
      "loss": 0.1891,
      "num_input_tokens_seen": 1767984,
      "step": 4400
    },
    {
      "epoch": 4.579002079002079,
      "grad_norm": 17.461746215820312,
      "learning_rate": 1.0773146593029637e-06,
      "loss": 0.2029,
      "num_input_tokens_seen": 1769904,
      "step": 4405
    },
    {
      "epoch": 4.584199584199585,
      "grad_norm": 9.160344123840332,
      "learning_rate": 1.0511295988803294e-06,
      "loss": 0.2743,
      "num_input_tokens_seen": 1771888,
      "step": 4410
    },
    {
      "epoch": 4.58939708939709,
      "grad_norm": 8.415828704833984,
      "learning_rate": 1.0252598567310451e-06,
      "loss": 0.2115,
      "num_input_tokens_seen": 1773936,
      "step": 4415
    },
    {
      "epoch": 4.594594594594595,
      "grad_norm": 17.489177703857422,
      "learning_rate": 9.99705773464185e-07,
      "loss": 0.2778,
      "num_input_tokens_seen": 1775984,
      "step": 4420
    },
    {
      "epoch": 4.5997920997921,
      "grad_norm": 11.403511047363281,
      "learning_rate": 9.744676855327483e-07,
      "loss": 0.2428,
      "num_input_tokens_seen": 1777840,
      "step": 4425
    },
    {
      "epoch": 4.604989604989605,
      "grad_norm": 23.2257022857666,
      "learning_rate": 9.495459252292504e-07,
      "loss": 0.2124,
      "num_input_tokens_seen": 1779824,
      "step": 4430
    },
    {
      "epoch": 4.61018711018711,
      "grad_norm": 4.646495342254639,
      "learning_rate": 9.249408206813332e-07,
      "loss": 0.1939,
      "num_input_tokens_seen": 1781872,
      "step": 4435
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 3.444089412689209,
      "learning_rate": 9.006526958474509e-07,
      "loss": 0.2364,
      "num_input_tokens_seen": 1783984,
      "step": 4440
    },
    {
      "epoch": 4.620582120582121,
      "grad_norm": 18.236093521118164,
      "learning_rate": 8.766818705126134e-07,
      "loss": 0.2023,
      "num_input_tokens_seen": 1786032,
      "step": 4445
    },
    {
      "epoch": 4.625779625779626,
      "grad_norm": 7.719527244567871,
      "learning_rate": 8.530286602841525e-07,
      "loss": 0.2455,
      "num_input_tokens_seen": 1788016,
      "step": 4450
    },
    {
      "epoch": 4.630977130977131,
      "grad_norm": 6.857132911682129,
      "learning_rate": 8.296933765875897e-07,
      "loss": 0.2154,
      "num_input_tokens_seen": 1790064,
      "step": 4455
    },
    {
      "epoch": 4.636174636174636,
      "grad_norm": 10.924470901489258,
      "learning_rate": 8.066763266625282e-07,
      "loss": 0.2046,
      "num_input_tokens_seen": 1791984,
      "step": 4460
    },
    {
      "epoch": 4.641372141372141,
      "grad_norm": 3.512413740158081,
      "learning_rate": 7.839778135586007e-07,
      "loss": 0.1884,
      "num_input_tokens_seen": 1793904,
      "step": 4465
    },
    {
      "epoch": 4.646569646569646,
      "grad_norm": 20.026166915893555,
      "learning_rate": 7.615981361314889e-07,
      "loss": 0.223,
      "num_input_tokens_seen": 1795888,
      "step": 4470
    },
    {
      "epoch": 4.651767151767151,
      "grad_norm": 13.404367446899414,
      "learning_rate": 7.3953758903898e-07,
      "loss": 0.3424,
      "num_input_tokens_seen": 1797872,
      "step": 4475
    },
    {
      "epoch": 4.656964656964657,
      "grad_norm": 8.044845581054688,
      "learning_rate": 7.177964627370997e-07,
      "loss": 0.2152,
      "num_input_tokens_seen": 1799920,
      "step": 4480
    },
    {
      "epoch": 4.662162162162162,
      "grad_norm": 6.080995082855225,
      "learning_rate": 6.963750434762745e-07,
      "loss": 0.2628,
      "num_input_tokens_seen": 1801776,
      "step": 4485
    },
    {
      "epoch": 4.667359667359667,
      "grad_norm": 7.769125461578369,
      "learning_rate": 6.752736132975696e-07,
      "loss": 0.2974,
      "num_input_tokens_seen": 1803824,
      "step": 4490
    },
    {
      "epoch": 4.672557172557172,
      "grad_norm": 9.061914443969727,
      "learning_rate": 6.54492450028979e-07,
      "loss": 0.1697,
      "num_input_tokens_seen": 1805744,
      "step": 4495
    },
    {
      "epoch": 4.6777546777546775,
      "grad_norm": 8.759089469909668,
      "learning_rate": 6.340318272817474e-07,
      "loss": 0.2489,
      "num_input_tokens_seen": 1807728,
      "step": 4500
    },
    {
      "epoch": 4.682952182952183,
      "grad_norm": 14.477424621582031,
      "learning_rate": 6.138920144468124e-07,
      "loss": 0.255,
      "num_input_tokens_seen": 1809712,
      "step": 4505
    },
    {
      "epoch": 4.6881496881496885,
      "grad_norm": 16.341243743896484,
      "learning_rate": 5.94073276691201e-07,
      "loss": 0.2946,
      "num_input_tokens_seen": 1811632,
      "step": 4510
    },
    {
      "epoch": 4.6933471933471935,
      "grad_norm": 4.825214385986328,
      "learning_rate": 5.745758749545749e-07,
      "loss": 0.2011,
      "num_input_tokens_seen": 1813552,
      "step": 4515
    },
    {
      "epoch": 4.698544698544699,
      "grad_norm": 6.129444122314453,
      "learning_rate": 5.554000659457881e-07,
      "loss": 0.2354,
      "num_input_tokens_seen": 1815664,
      "step": 4520
    },
    {
      "epoch": 4.703742203742204,
      "grad_norm": 4.319976806640625,
      "learning_rate": 5.365461021395096e-07,
      "loss": 0.2284,
      "num_input_tokens_seen": 1817648,
      "step": 4525
    },
    {
      "epoch": 4.708939708939709,
      "grad_norm": 4.276791572570801,
      "learning_rate": 5.180142317728815e-07,
      "loss": 0.2259,
      "num_input_tokens_seen": 1819696,
      "step": 4530
    },
    {
      "epoch": 4.714137214137214,
      "grad_norm": 10.404797554016113,
      "learning_rate": 4.998046988422766e-07,
      "loss": 0.269,
      "num_input_tokens_seen": 1821680,
      "step": 4535
    },
    {
      "epoch": 4.71933471933472,
      "grad_norm": 11.949972152709961,
      "learning_rate": 4.819177431000604e-07,
      "loss": 0.2786,
      "num_input_tokens_seen": 1823728,
      "step": 4540
    },
    {
      "epoch": 4.724532224532225,
      "grad_norm": 9.46826171875,
      "learning_rate": 4.6435360005145644e-07,
      "loss": 0.3228,
      "num_input_tokens_seen": 1825712,
      "step": 4545
    },
    {
      "epoch": 4.72972972972973,
      "grad_norm": 13.263773918151855,
      "learning_rate": 4.4711250095143267e-07,
      "loss": 0.2664,
      "num_input_tokens_seen": 1827760,
      "step": 4550
    },
    {
      "epoch": 4.734927234927235,
      "grad_norm": 5.771205902099609,
      "learning_rate": 4.30194672801662e-07,
      "loss": 0.2014,
      "num_input_tokens_seen": 1829680,
      "step": 4555
    },
    {
      "epoch": 4.74012474012474,
      "grad_norm": 23.836517333984375,
      "learning_rate": 4.136003383475251e-07,
      "loss": 0.2992,
      "num_input_tokens_seen": 1831728,
      "step": 4560
    },
    {
      "epoch": 4.745322245322245,
      "grad_norm": 12.346782684326172,
      "learning_rate": 3.9732971607519265e-07,
      "loss": 0.2033,
      "num_input_tokens_seen": 1833648,
      "step": 4565
    },
    {
      "epoch": 4.75051975051975,
      "grad_norm": 5.421136856079102,
      "learning_rate": 3.8138302020873373e-07,
      "loss": 0.2388,
      "num_input_tokens_seen": 1835696,
      "step": 4570
    },
    {
      "epoch": 4.755717255717256,
      "grad_norm": 12.613269805908203,
      "learning_rate": 3.6576046070730675e-07,
      "loss": 0.2582,
      "num_input_tokens_seen": 1837808,
      "step": 4575
    },
    {
      "epoch": 4.75987525987526,
      "eval_loss": 0.2604904770851135,
      "eval_runtime": 1.0828,
      "eval_samples_per_second": 790.52,
      "eval_steps_per_second": 98.815,
      "num_input_tokens_seen": 1839344,
      "step": 4579
    },
    {
      "epoch": 4.760914760914761,
      "grad_norm": 10.182611465454102,
      "learning_rate": 3.5046224326238107e-07,
      "loss": 0.2365,
      "num_input_tokens_seen": 1839728,
      "step": 4580
    },
    {
      "epoch": 4.766112266112266,
      "grad_norm": 5.537731647491455,
      "learning_rate": 3.3548856929505047e-07,
      "loss": 0.2569,
      "num_input_tokens_seen": 1841776,
      "step": 4585
    },
    {
      "epoch": 4.771309771309771,
      "grad_norm": 4.337884902954102,
      "learning_rate": 3.208396359533572e-07,
      "loss": 0.2548,
      "num_input_tokens_seen": 1843696,
      "step": 4590
    },
    {
      "epoch": 4.776507276507276,
      "grad_norm": 15.579423904418945,
      "learning_rate": 3.065156361097138e-07,
      "loss": 0.2391,
      "num_input_tokens_seen": 1845744,
      "step": 4595
    },
    {
      "epoch": 4.781704781704782,
      "grad_norm": 3.1390204429626465,
      "learning_rate": 2.925167583583577e-07,
      "loss": 0.2364,
      "num_input_tokens_seen": 1847792,
      "step": 4600
    },
    {
      "epoch": 4.786902286902287,
      "grad_norm": 12.236397743225098,
      "learning_rate": 2.7884318701285885e-07,
      "loss": 0.2843,
      "num_input_tokens_seen": 1849776,
      "step": 4605
    },
    {
      "epoch": 4.792099792099792,
      "grad_norm": 6.024417877197266,
      "learning_rate": 2.6549510210371607e-07,
      "loss": 0.2187,
      "num_input_tokens_seen": 1852016,
      "step": 4610
    },
    {
      "epoch": 4.797297297297297,
      "grad_norm": 13.414298057556152,
      "learning_rate": 2.524726793759591e-07,
      "loss": 0.246,
      "num_input_tokens_seen": 1854064,
      "step": 4615
    },
    {
      "epoch": 4.802494802494802,
      "grad_norm": 2.9425196647644043,
      "learning_rate": 2.397760902868612e-07,
      "loss": 0.2382,
      "num_input_tokens_seen": 1856112,
      "step": 4620
    },
    {
      "epoch": 4.8076923076923075,
      "grad_norm": 18.276071548461914,
      "learning_rate": 2.274055020036553e-07,
      "loss": 0.2361,
      "num_input_tokens_seen": 1858096,
      "step": 4625
    },
    {
      "epoch": 4.8128898128898125,
      "grad_norm": 3.6439883708953857,
      "learning_rate": 2.1536107740135482e-07,
      "loss": 0.2671,
      "num_input_tokens_seen": 1860272,
      "step": 4630
    },
    {
      "epoch": 4.8180873180873185,
      "grad_norm": 10.452295303344727,
      "learning_rate": 2.0364297506060003e-07,
      "loss": 0.2501,
      "num_input_tokens_seen": 1862256,
      "step": 4635
    },
    {
      "epoch": 4.8232848232848236,
      "grad_norm": 7.328881740570068,
      "learning_rate": 1.922513492655653e-07,
      "loss": 0.1984,
      "num_input_tokens_seen": 1864304,
      "step": 4640
    },
    {
      "epoch": 4.828482328482329,
      "grad_norm": 4.626097202301025,
      "learning_rate": 1.8118635000194396e-07,
      "loss": 0.2677,
      "num_input_tokens_seen": 1866224,
      "step": 4645
    },
    {
      "epoch": 4.833679833679834,
      "grad_norm": 6.644436836242676,
      "learning_rate": 1.704481229549526e-07,
      "loss": 0.2349,
      "num_input_tokens_seen": 1868336,
      "step": 4650
    },
    {
      "epoch": 4.838877338877339,
      "grad_norm": 6.079159259796143,
      "learning_rate": 1.6003680950742728e-07,
      "loss": 0.2927,
      "num_input_tokens_seen": 1870448,
      "step": 4655
    },
    {
      "epoch": 4.844074844074844,
      "grad_norm": 11.934816360473633,
      "learning_rate": 1.4995254673795812e-07,
      "loss": 0.2206,
      "num_input_tokens_seen": 1872368,
      "step": 4660
    },
    {
      "epoch": 4.849272349272349,
      "grad_norm": 2.1284427642822266,
      "learning_rate": 1.4019546741908251e-07,
      "loss": 0.2374,
      "num_input_tokens_seen": 1874480,
      "step": 4665
    },
    {
      "epoch": 4.854469854469855,
      "grad_norm": 7.95443868637085,
      "learning_rate": 1.3076570001553934e-07,
      "loss": 0.2712,
      "num_input_tokens_seen": 1876464,
      "step": 4670
    },
    {
      "epoch": 4.85966735966736,
      "grad_norm": 10.991232872009277,
      "learning_rate": 1.216633686825841e-07,
      "loss": 0.2406,
      "num_input_tokens_seen": 1878448,
      "step": 4675
    },
    {
      "epoch": 4.864864864864865,
      "grad_norm": 16.078187942504883,
      "learning_rate": 1.1288859326433477e-07,
      "loss": 0.2743,
      "num_input_tokens_seen": 1880432,
      "step": 4680
    },
    {
      "epoch": 4.87006237006237,
      "grad_norm": 19.031558990478516,
      "learning_rate": 1.0444148929221464e-07,
      "loss": 0.2828,
      "num_input_tokens_seen": 1882544,
      "step": 4685
    },
    {
      "epoch": 4.875259875259875,
      "grad_norm": 16.82112693786621,
      "learning_rate": 9.63221679834203e-08,
      "loss": 0.274,
      "num_input_tokens_seen": 1884528,
      "step": 4690
    },
    {
      "epoch": 4.88045738045738,
      "grad_norm": 6.512580394744873,
      "learning_rate": 8.853073623946162e-08,
      "loss": 0.2457,
      "num_input_tokens_seen": 1886640,
      "step": 4695
    },
    {
      "epoch": 4.885654885654886,
      "grad_norm": 9.699633598327637,
      "learning_rate": 8.106729664475176e-08,
      "loss": 0.2663,
      "num_input_tokens_seen": 1888688,
      "step": 4700
    },
    {
      "epoch": 4.890852390852391,
      "grad_norm": 14.048091888427734,
      "learning_rate": 7.393194746525279e-08,
      "loss": 0.2572,
      "num_input_tokens_seen": 1890736,
      "step": 4705
    },
    {
      "epoch": 4.896049896049896,
      "grad_norm": 7.062888145446777,
      "learning_rate": 6.712478264719601e-08,
      "loss": 0.2399,
      "num_input_tokens_seen": 1892720,
      "step": 4710
    },
    {
      "epoch": 4.901247401247401,
      "grad_norm": 11.242563247680664,
      "learning_rate": 6.064589181582481e-08,
      "loss": 0.2267,
      "num_input_tokens_seen": 1894704,
      "step": 4715
    },
    {
      "epoch": 4.906444906444906,
      "grad_norm": 1.6842583417892456,
      "learning_rate": 5.4495360274231524e-08,
      "loss": 0.2652,
      "num_input_tokens_seen": 1896624,
      "step": 4720
    },
    {
      "epoch": 4.911642411642411,
      "grad_norm": 4.220650672912598,
      "learning_rate": 4.867326900223068e-08,
      "loss": 0.2254,
      "num_input_tokens_seen": 1898544,
      "step": 4725
    },
    {
      "epoch": 4.916839916839917,
      "grad_norm": 3.8804283142089844,
      "learning_rate": 4.317969465527927e-08,
      "loss": 0.2342,
      "num_input_tokens_seen": 1900592,
      "step": 4730
    },
    {
      "epoch": 4.922037422037422,
      "grad_norm": 6.306464195251465,
      "learning_rate": 3.8014709563488625e-08,
      "loss": 0.2151,
      "num_input_tokens_seen": 1902576,
      "step": 4735
    },
    {
      "epoch": 4.927234927234927,
      "grad_norm": 5.872474193572998,
      "learning_rate": 3.317838173066135e-08,
      "loss": 0.2368,
      "num_input_tokens_seen": 1904624,
      "step": 4740
    },
    {
      "epoch": 4.9324324324324325,
      "grad_norm": 23.829078674316406,
      "learning_rate": 2.8670774833386426e-08,
      "loss": 0.2662,
      "num_input_tokens_seen": 1906736,
      "step": 4745
    },
    {
      "epoch": 4.9376299376299375,
      "grad_norm": 2.1569466590881348,
      "learning_rate": 2.449194822022327e-08,
      "loss": 0.221,
      "num_input_tokens_seen": 1908592,
      "step": 4750
    },
    {
      "epoch": 4.942827442827443,
      "grad_norm": 5.917527675628662,
      "learning_rate": 2.064195691089954e-08,
      "loss": 0.2286,
      "num_input_tokens_seen": 1910576,
      "step": 4755
    },
    {
      "epoch": 4.948024948024948,
      "grad_norm": 8.833841323852539,
      "learning_rate": 1.712085159559784e-08,
      "loss": 0.2104,
      "num_input_tokens_seen": 1912624,
      "step": 4760
    },
    {
      "epoch": 4.953222453222454,
      "grad_norm": 17.08448600769043,
      "learning_rate": 1.3928678634289593e-08,
      "loss": 0.2802,
      "num_input_tokens_seen": 1914608,
      "step": 4765
    },
    {
      "epoch": 4.958419958419959,
      "grad_norm": 2.2440571784973145,
      "learning_rate": 1.1065480056110522e-08,
      "loss": 0.2125,
      "num_input_tokens_seen": 1916592,
      "step": 4770
    },
    {
      "epoch": 4.963617463617464,
      "grad_norm": 6.129010200500488,
      "learning_rate": 8.531293558824982e-09,
      "loss": 0.2162,
      "num_input_tokens_seen": 1918704,
      "step": 4775
    },
    {
      "epoch": 4.968814968814969,
      "grad_norm": 2.9528563022613525,
      "learning_rate": 6.326152508320804e-09,
      "loss": 0.2508,
      "num_input_tokens_seen": 1920624,
      "step": 4780
    },
    {
      "epoch": 4.974012474012474,
      "grad_norm": 7.140697479248047,
      "learning_rate": 4.450085938170756e-09,
      "loss": 0.2453,
      "num_input_tokens_seen": 1922480,
      "step": 4785
    },
    {
      "epoch": 4.979209979209979,
      "grad_norm": 13.262871742248535,
      "learning_rate": 2.9031185492522926e-09,
      "loss": 0.2758,
      "num_input_tokens_seen": 1924464,
      "step": 4790
    },
    {
      "epoch": 4.984407484407484,
      "grad_norm": 8.832655906677246,
      "learning_rate": 1.6852707094172636e-09,
      "loss": 0.2275,
      "num_input_tokens_seen": 1926448,
      "step": 4795
    },
    {
      "epoch": 4.98960498960499,
      "grad_norm": 11.241654396057129,
      "learning_rate": 7.965584532282355e-10,
      "loss": 0.2635,
      "num_input_tokens_seen": 1928560,
      "step": 4800
    },
    {
      "epoch": 4.994802494802495,
      "grad_norm": 5.555950164794922,
      "learning_rate": 2.3699348174754945e-10,
      "loss": 0.2384,
      "num_input_tokens_seen": 1930544,
      "step": 4805
    },
    {
      "epoch": 5.0,
      "grad_norm": 12.928886413574219,
      "learning_rate": 6.583162381890162e-12,
      "loss": 0.2943,
      "num_input_tokens_seen": 1932608,
      "step": 4810
    },
    {
      "epoch": 5.0,
      "num_input_tokens_seen": 1932608,
      "step": 4810,
      "total_flos": 1.1284259767320576e+16,
      "train_loss": 0.28020706261022177,
      "train_runtime": 1219.1473,
      "train_samples_per_second": 31.559,
      "train_steps_per_second": 3.945
    }
  ],
  "logging_steps": 5,
  "max_steps": 4810,
  "num_input_tokens_seen": 1932608,
  "num_train_epochs": 5,
  "save_steps": 241,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1284259767320576e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}