{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9999074502545118,
  "eval_steps": 100,
  "global_step": 2701,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0018509949097639982,
      "grad_norm": 2.749824922426202,
      "learning_rate": 3.690036900369004e-07,
      "loss": 1.1027,
      "step": 5
    },
    {
      "epoch": 0.0037019898195279964,
      "grad_norm": 2.7133127995482353,
      "learning_rate": 7.380073800738008e-07,
      "loss": 1.107,
      "step": 10
    },
    {
      "epoch": 0.005552984729291994,
      "grad_norm": 2.5060043757547894,
      "learning_rate": 1.1070110701107011e-06,
      "loss": 1.0411,
      "step": 15
    },
    {
      "epoch": 0.007403979639055993,
      "grad_norm": 2.580301155440442,
      "learning_rate": 1.4760147601476015e-06,
      "loss": 1.0394,
      "step": 20
    },
    {
      "epoch": 0.00925497454881999,
      "grad_norm": 2.101103627946831,
      "learning_rate": 1.845018450184502e-06,
      "loss": 1.025,
      "step": 25
    },
    {
      "epoch": 0.011105969458583989,
      "grad_norm": 1.7798274009053514,
      "learning_rate": 2.2140221402214023e-06,
      "loss": 1.0642,
      "step": 30
    },
    {
      "epoch": 0.012956964368347987,
      "grad_norm": 1.4788395927397677,
      "learning_rate": 2.5830258302583027e-06,
      "loss": 0.9636,
      "step": 35
    },
    {
      "epoch": 0.014807959278111986,
      "grad_norm": 1.4694159866777265,
      "learning_rate": 2.952029520295203e-06,
      "loss": 1.0495,
      "step": 40
    },
    {
      "epoch": 0.016658954187875982,
      "grad_norm": 1.529975198759594,
      "learning_rate": 3.3210332103321034e-06,
      "loss": 0.9308,
      "step": 45
    },
    {
      "epoch": 0.01850994909763998,
      "grad_norm": 1.4286077278093599,
      "learning_rate": 3.690036900369004e-06,
      "loss": 0.9909,
      "step": 50
    },
    {
      "epoch": 0.02036094400740398,
      "grad_norm": 1.336405308330729,
      "learning_rate": 4.059040590405905e-06,
      "loss": 0.948,
      "step": 55
    },
    {
      "epoch": 0.022211938917167977,
      "grad_norm": 1.411657382339046,
      "learning_rate": 4.428044280442805e-06,
      "loss": 0.9434,
      "step": 60
    },
    {
      "epoch": 0.024062933826931976,
      "grad_norm": 1.3370491817510695,
      "learning_rate": 4.797047970479705e-06,
      "loss": 0.9112,
      "step": 65
    },
    {
      "epoch": 0.025913928736695974,
      "grad_norm": 1.238804041054237,
      "learning_rate": 5.166051660516605e-06,
      "loss": 0.864,
      "step": 70
    },
    {
      "epoch": 0.027764923646459973,
      "grad_norm": 1.5683504507074115,
      "learning_rate": 5.535055350553506e-06,
      "loss": 0.927,
      "step": 75
    },
    {
      "epoch": 0.02961591855622397,
      "grad_norm": 1.3407079562724307,
      "learning_rate": 5.904059040590406e-06,
      "loss": 0.8787,
      "step": 80
    },
    {
      "epoch": 0.03146691346598797,
      "grad_norm": 1.1838954340614056,
      "learning_rate": 6.273062730627307e-06,
      "loss": 0.9208,
      "step": 85
    },
    {
      "epoch": 0.033317908375751965,
      "grad_norm": 1.5881922549502974,
      "learning_rate": 6.642066420664207e-06,
      "loss": 0.8529,
      "step": 90
    },
    {
      "epoch": 0.035168903285515966,
      "grad_norm": 1.4451601927412367,
      "learning_rate": 7.011070110701108e-06,
      "loss": 0.9031,
      "step": 95
    },
    {
      "epoch": 0.03701989819527996,
      "grad_norm": 1.2272450676249622,
      "learning_rate": 7.380073800738008e-06,
      "loss": 0.8384,
      "step": 100
    },
    {
      "epoch": 0.03701989819527996,
      "eval_loss": 0.8794825673103333,
      "eval_runtime": 125.7829,
      "eval_samples_per_second": 1.018,
      "eval_steps_per_second": 0.509,
      "step": 100
    },
    {
      "epoch": 0.03887089310504396,
      "grad_norm": 1.1893073354157337,
      "learning_rate": 7.749077490774908e-06,
      "loss": 0.889,
      "step": 105
    },
    {
      "epoch": 0.04072188801480796,
      "grad_norm": 1.3052090560256295,
      "learning_rate": 8.11808118081181e-06,
      "loss": 0.8133,
      "step": 110
    },
    {
      "epoch": 0.04257288292457196,
      "grad_norm": 1.37698008425289,
      "learning_rate": 8.48708487084871e-06,
      "loss": 0.865,
      "step": 115
    },
    {
      "epoch": 0.044423877834335955,
      "grad_norm": 1.303834447656376,
      "learning_rate": 8.85608856088561e-06,
      "loss": 0.846,
      "step": 120
    },
    {
      "epoch": 0.04627487274409996,
      "grad_norm": 1.2327474858734326,
      "learning_rate": 9.22509225092251e-06,
      "loss": 0.7942,
      "step": 125
    },
    {
      "epoch": 0.04812586765386395,
      "grad_norm": 1.2058003679999099,
      "learning_rate": 9.59409594095941e-06,
      "loss": 0.8417,
      "step": 130
    },
    {
      "epoch": 0.04997686256362795,
      "grad_norm": 1.2290254882715699,
      "learning_rate": 9.963099630996312e-06,
      "loss": 0.8701,
      "step": 135
    },
    {
      "epoch": 0.05182785747339195,
      "grad_norm": 1.1760267952662564,
      "learning_rate": 1.033210332103321e-05,
      "loss": 0.8252,
      "step": 140
    },
    {
      "epoch": 0.053678852383155944,
      "grad_norm": 1.269131376133505,
      "learning_rate": 1.0701107011070112e-05,
      "loss": 0.8302,
      "step": 145
    },
    {
      "epoch": 0.055529847292919945,
      "grad_norm": 1.1597954970515953,
      "learning_rate": 1.1070110701107012e-05,
      "loss": 0.8479,
      "step": 150
    },
    {
      "epoch": 0.05738084220268394,
      "grad_norm": 1.2448895450141289,
      "learning_rate": 1.1439114391143913e-05,
      "loss": 0.8521,
      "step": 155
    },
    {
      "epoch": 0.05923183711244794,
      "grad_norm": 1.1897405950550726,
      "learning_rate": 1.1808118081180812e-05,
      "loss": 0.7592,
      "step": 160
    },
    {
      "epoch": 0.06108283202221194,
      "grad_norm": 1.314365342475427,
      "learning_rate": 1.2177121771217713e-05,
      "loss": 0.8289,
      "step": 165
    },
    {
      "epoch": 0.06293382693197594,
      "grad_norm": 1.3600021570595577,
      "learning_rate": 1.2546125461254614e-05,
      "loss": 0.8119,
      "step": 170
    },
    {
      "epoch": 0.06478482184173993,
      "grad_norm": 1.2381592999546391,
      "learning_rate": 1.2915129151291515e-05,
      "loss": 0.8081,
      "step": 175
    },
    {
      "epoch": 0.06663581675150393,
      "grad_norm": 1.2650464435907114,
      "learning_rate": 1.3284132841328414e-05,
      "loss": 0.8173,
      "step": 180
    },
    {
      "epoch": 0.06848681166126794,
      "grad_norm": 1.2523135217657222,
      "learning_rate": 1.3653136531365315e-05,
      "loss": 0.8105,
      "step": 185
    },
    {
      "epoch": 0.07033780657103193,
      "grad_norm": 1.1760807838355647,
      "learning_rate": 1.4022140221402215e-05,
      "loss": 0.8644,
      "step": 190
    },
    {
      "epoch": 0.07218880148079593,
      "grad_norm": 1.2557569983889052,
      "learning_rate": 1.4391143911439116e-05,
      "loss": 0.8031,
      "step": 195
    },
    {
      "epoch": 0.07403979639055992,
      "grad_norm": 1.2032960225457723,
      "learning_rate": 1.4760147601476015e-05,
      "loss": 0.8639,
      "step": 200
    },
    {
      "epoch": 0.07403979639055992,
      "eval_loss": 0.8400982022285461,
      "eval_runtime": 125.9532,
      "eval_samples_per_second": 1.016,
      "eval_steps_per_second": 0.508,
      "step": 200
    },
    {
      "epoch": 0.07589079130032392,
      "grad_norm": 1.394672996108549,
      "learning_rate": 1.5129151291512916e-05,
      "loss": 0.8164,
      "step": 205
    },
    {
      "epoch": 0.07774178621008793,
      "grad_norm": 1.3016173569055736,
      "learning_rate": 1.5498154981549817e-05,
      "loss": 0.8478,
      "step": 210
    },
    {
      "epoch": 0.07959278111985192,
      "grad_norm": 1.406780438265723,
      "learning_rate": 1.5867158671586716e-05,
      "loss": 0.841,
      "step": 215
    },
    {
      "epoch": 0.08144377602961592,
      "grad_norm": 1.2377690323264836,
      "learning_rate": 1.623616236162362e-05,
      "loss": 0.8287,
      "step": 220
    },
    {
      "epoch": 0.08329477093937991,
      "grad_norm": 1.39607937433599,
      "learning_rate": 1.6605166051660518e-05,
      "loss": 0.8176,
      "step": 225
    },
    {
      "epoch": 0.08514576584914392,
      "grad_norm": 1.3980920756979587,
      "learning_rate": 1.697416974169742e-05,
      "loss": 0.7867,
      "step": 230
    },
    {
      "epoch": 0.08699676075890792,
      "grad_norm": 1.2563235907029433,
      "learning_rate": 1.734317343173432e-05,
      "loss": 0.8447,
      "step": 235
    },
    {
      "epoch": 0.08884775566867191,
      "grad_norm": 1.4806853438242649,
      "learning_rate": 1.771217712177122e-05,
      "loss": 0.807,
      "step": 240
    },
    {
      "epoch": 0.0906987505784359,
      "grad_norm": 1.2447141754207756,
      "learning_rate": 1.8081180811808117e-05,
      "loss": 0.7962,
      "step": 245
    },
    {
      "epoch": 0.09254974548819991,
      "grad_norm": 1.3080325661809735,
      "learning_rate": 1.845018450184502e-05,
      "loss": 0.7745,
      "step": 250
    },
    {
      "epoch": 0.09440074039796391,
      "grad_norm": 1.179323447687895,
      "learning_rate": 1.8819188191881922e-05,
      "loss": 0.8574,
      "step": 255
    },
    {
      "epoch": 0.0962517353077279,
      "grad_norm": 1.2896705705328326,
      "learning_rate": 1.918819188191882e-05,
      "loss": 0.8296,
      "step": 260
    },
    {
      "epoch": 0.0981027302174919,
      "grad_norm": 1.3644542064844416,
      "learning_rate": 1.955719557195572e-05,
      "loss": 0.8251,
      "step": 265
    },
    {
      "epoch": 0.0999537251272559,
      "grad_norm": 1.286282268357473,
      "learning_rate": 1.9926199261992623e-05,
      "loss": 0.7754,
      "step": 270
    },
    {
      "epoch": 0.1018047200370199,
      "grad_norm": 1.2329112299042364,
      "learning_rate": 1.999986628620426e-05,
      "loss": 0.7915,
      "step": 275
    },
    {
      "epoch": 0.1036557149467839,
      "grad_norm": 1.3173883195813032,
      "learning_rate": 1.9999323080037623e-05,
      "loss": 0.866,
      "step": 280
    },
    {
      "epoch": 0.10550670985654789,
      "grad_norm": 1.2611947477770145,
      "learning_rate": 1.9998362047068548e-05,
      "loss": 0.83,
      "step": 285
    },
    {
      "epoch": 0.10735770476631189,
      "grad_norm": 1.1796656758558173,
      "learning_rate": 1.9996983227454284e-05,
      "loss": 0.8357,
      "step": 290
    },
    {
      "epoch": 0.1092086996760759,
      "grad_norm": 1.1353613990508302,
      "learning_rate": 1.9995186678809513e-05,
      "loss": 0.7181,
      "step": 295
    },
    {
      "epoch": 0.11105969458583989,
      "grad_norm": 1.2567062023606574,
      "learning_rate": 1.999297247620393e-05,
      "loss": 0.8598,
      "step": 300
    },
    {
      "epoch": 0.11105969458583989,
      "eval_loss": 0.8292834758758545,
      "eval_runtime": 125.837,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.509,
      "step": 300
    },
    {
      "epoch": 0.11291068949560389,
      "grad_norm": 1.3854824655270053,
      "learning_rate": 1.999034071215912e-05,
      "loss": 0.869,
      "step": 305
    },
    {
      "epoch": 0.11476168440536788,
      "grad_norm": 1.4180966128082728,
      "learning_rate": 1.998729149664468e-05,
      "loss": 0.7802,
      "step": 310
    },
    {
      "epoch": 0.11661267931513189,
      "grad_norm": 1.2687340435719516,
      "learning_rate": 1.9983824957073632e-05,
      "loss": 0.7751,
      "step": 315
    },
    {
      "epoch": 0.11846367422489588,
      "grad_norm": 1.2012101992327227,
      "learning_rate": 1.99799412382971e-05,
      "loss": 0.7831,
      "step": 320
    },
    {
      "epoch": 0.12031466913465988,
      "grad_norm": 1.2085415925635907,
      "learning_rate": 1.9975640502598243e-05,
      "loss": 0.788,
      "step": 325
    },
    {
      "epoch": 0.12216566404442387,
      "grad_norm": 1.2183184035114982,
      "learning_rate": 1.9970922929685496e-05,
      "loss": 0.7998,
      "step": 330
    },
    {
      "epoch": 0.12401665895418787,
      "grad_norm": 1.2657715995907186,
      "learning_rate": 1.996578871668504e-05,
      "loss": 0.7978,
      "step": 335
    },
    {
      "epoch": 0.12586765386395188,
      "grad_norm": 1.1898542144979154,
      "learning_rate": 1.996023807813258e-05,
      "loss": 0.7861,
      "step": 340
    },
    {
      "epoch": 0.12771864877371586,
      "grad_norm": 1.2193829841446657,
      "learning_rate": 1.995427124596437e-05,
      "loss": 0.8067,
      "step": 345
    },
    {
      "epoch": 0.12956964368347987,
      "grad_norm": 1.1563840168462576,
      "learning_rate": 1.9947888469507527e-05,
      "loss": 0.7666,
      "step": 350
    },
    {
      "epoch": 0.13142063859324388,
      "grad_norm": 1.3445764875998873,
      "learning_rate": 1.9941090015469614e-05,
      "loss": 0.7855,
      "step": 355
    },
    {
      "epoch": 0.13327163350300786,
      "grad_norm": 1.2515125913797396,
      "learning_rate": 1.9933876167927494e-05,
      "loss": 0.8211,
      "step": 360
    },
    {
      "epoch": 0.13512262841277187,
      "grad_norm": 1.2363244217826992,
      "learning_rate": 1.9926247228315455e-05,
      "loss": 0.7726,
      "step": 365
    },
    {
      "epoch": 0.13697362332253588,
      "grad_norm": 1.1398366050989162,
      "learning_rate": 1.9918203515412616e-05,
      "loss": 0.7809,
      "step": 370
    },
    {
      "epoch": 0.13882461823229986,
      "grad_norm": 1.1609837477400817,
      "learning_rate": 1.9909745365329623e-05,
      "loss": 0.7981,
      "step": 375
    },
    {
      "epoch": 0.14067561314206387,
      "grad_norm": 1.1428050730123682,
      "learning_rate": 1.990087313149457e-05,
      "loss": 0.8055,
      "step": 380
    },
    {
      "epoch": 0.14252660805182785,
      "grad_norm": 1.1811407177495057,
      "learning_rate": 1.9891587184638274e-05,
      "loss": 0.7898,
      "step": 385
    },
    {
      "epoch": 0.14437760296159186,
      "grad_norm": 1.1089125886531135,
      "learning_rate": 1.9881887912778738e-05,
      "loss": 0.8435,
      "step": 390
    },
    {
      "epoch": 0.14622859787135586,
      "grad_norm": 1.1264567214650973,
      "learning_rate": 1.9871775721204973e-05,
      "loss": 0.8027,
      "step": 395
    },
    {
      "epoch": 0.14807959278111985,
      "grad_norm": 1.2984452085681775,
      "learning_rate": 1.9861251032460053e-05,
      "loss": 0.8238,
      "step": 400
    },
    {
      "epoch": 0.14807959278111985,
      "eval_loss": 0.8187944889068604,
      "eval_runtime": 125.8941,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 400
    },
    {
      "epoch": 0.14993058769088385,
      "grad_norm": 1.262457467293623,
      "learning_rate": 1.985031428632345e-05,
      "loss": 0.7656,
      "step": 405
    },
    {
      "epoch": 0.15178158260064784,
      "grad_norm": 1.222235652953353,
      "learning_rate": 1.9838965939792666e-05,
      "loss": 0.8033,
      "step": 410
    },
    {
      "epoch": 0.15363257751041184,
      "grad_norm": 1.0872174045726444,
      "learning_rate": 1.9827206467064133e-05,
      "loss": 0.8079,
      "step": 415
    },
    {
      "epoch": 0.15548357242017585,
      "grad_norm": 1.188894944380026,
      "learning_rate": 1.9815036359513408e-05,
      "loss": 0.7931,
      "step": 420
    },
    {
      "epoch": 0.15733456732993983,
      "grad_norm": 1.0474799030998936,
      "learning_rate": 1.980245612567462e-05,
      "loss": 0.7752,
      "step": 425
    },
    {
      "epoch": 0.15918556223970384,
      "grad_norm": 1.0714711020851349,
      "learning_rate": 1.9789466291219246e-05,
      "loss": 0.7787,
      "step": 430
    },
    {
      "epoch": 0.16103655714946785,
      "grad_norm": 1.2916090137573928,
      "learning_rate": 1.9776067398934122e-05,
      "loss": 0.8464,
      "step": 435
    },
    {
      "epoch": 0.16288755205923183,
      "grad_norm": 1.0837760464706834,
      "learning_rate": 1.9762260008698787e-05,
      "loss": 0.742,
      "step": 440
    },
    {
      "epoch": 0.16473854696899584,
      "grad_norm": 1.0638766335715402,
      "learning_rate": 1.974804469746206e-05,
      "loss": 0.7909,
      "step": 445
    },
    {
      "epoch": 0.16658954187875982,
      "grad_norm": 1.1650137243413088,
      "learning_rate": 1.9733422059217954e-05,
      "loss": 0.8142,
      "step": 450
    },
    {
      "epoch": 0.16844053678852383,
      "grad_norm": 1.1494193680521951,
      "learning_rate": 1.9718392704980852e-05,
      "loss": 0.7187,
      "step": 455
    },
    {
      "epoch": 0.17029153169828784,
      "grad_norm": 1.141831785166932,
      "learning_rate": 1.9702957262759964e-05,
      "loss": 0.7338,
      "step": 460
    },
    {
      "epoch": 0.17214252660805182,
      "grad_norm": 1.1597583259361357,
      "learning_rate": 1.9687116377533102e-05,
      "loss": 0.802,
      "step": 465
    },
    {
      "epoch": 0.17399352151781583,
      "grad_norm": 1.1829435674368602,
      "learning_rate": 1.9670870711219708e-05,
      "loss": 0.7982,
      "step": 470
    },
    {
      "epoch": 0.1758445164275798,
      "grad_norm": 1.2456510178235929,
      "learning_rate": 1.9654220942653223e-05,
      "loss": 0.796,
      "step": 475
    },
    {
      "epoch": 0.17769551133734382,
      "grad_norm": 1.1776301109970682,
      "learning_rate": 1.9637167767552687e-05,
      "loss": 0.7754,
      "step": 480
    },
    {
      "epoch": 0.17954650624710783,
      "grad_norm": 1.116849386127364,
      "learning_rate": 1.9619711898493707e-05,
      "loss": 0.7919,
      "step": 485
    },
    {
      "epoch": 0.1813975011568718,
      "grad_norm": 1.090344008956795,
      "learning_rate": 1.9601854064878645e-05,
      "loss": 0.84,
      "step": 490
    },
    {
      "epoch": 0.18324849606663582,
      "grad_norm": 1.0862481361440726,
      "learning_rate": 1.9583595012906173e-05,
      "loss": 0.7697,
      "step": 495
    },
    {
      "epoch": 0.18509949097639983,
      "grad_norm": 1.142567519924341,
      "learning_rate": 1.956493550554006e-05,
      "loss": 0.7635,
      "step": 500
    },
    {
      "epoch": 0.18509949097639983,
      "eval_loss": 0.8105577230453491,
      "eval_runtime": 125.9442,
      "eval_samples_per_second": 1.016,
      "eval_steps_per_second": 0.508,
      "step": 500
    },
    {
      "epoch": 0.1869504858861638,
      "grad_norm": 1.023479803471901,
      "learning_rate": 1.954587632247732e-05,
      "loss": 0.7747,
      "step": 505
    },
    {
      "epoch": 0.18880148079592782,
      "grad_norm": 1.063385408309198,
      "learning_rate": 1.9526418260115615e-05,
      "loss": 0.75,
      "step": 510
    },
    {
      "epoch": 0.1906524757056918,
      "grad_norm": 1.1699474105718337,
      "learning_rate": 1.9506562131519978e-05,
      "loss": 0.7914,
      "step": 515
    },
    {
      "epoch": 0.1925034706154558,
      "grad_norm": 1.2205261625534392,
      "learning_rate": 1.9486308766388843e-05,
      "loss": 0.8228,
      "step": 520
    },
    {
      "epoch": 0.19435446552521982,
      "grad_norm": 1.2493323461591765,
      "learning_rate": 1.9465659011019384e-05,
      "loss": 0.7494,
      "step": 525
    },
    {
      "epoch": 0.1962054604349838,
      "grad_norm": 1.083050859566808,
      "learning_rate": 1.9444613728272127e-05,
      "loss": 0.7478,
      "step": 530
    },
    {
      "epoch": 0.1980564553447478,
      "grad_norm": 1.122573251625821,
      "learning_rate": 1.9423173797534924e-05,
      "loss": 0.8088,
      "step": 535
    },
    {
      "epoch": 0.1999074502545118,
      "grad_norm": 1.2528116581295516,
      "learning_rate": 1.9401340114686187e-05,
      "loss": 0.7854,
      "step": 540
    },
    {
      "epoch": 0.2017584451642758,
      "grad_norm": 1.2245250113750425,
      "learning_rate": 1.9379113592057468e-05,
      "loss": 0.7745,
      "step": 545
    },
    {
      "epoch": 0.2036094400740398,
      "grad_norm": 1.3603783721746314,
      "learning_rate": 1.9356495158395317e-05,
      "loss": 0.8214,
      "step": 550
    },
    {
      "epoch": 0.20546043498380379,
      "grad_norm": 1.1676680192314415,
      "learning_rate": 1.9333485758822495e-05,
      "loss": 0.7731,
      "step": 555
    },
    {
      "epoch": 0.2073114298935678,
      "grad_norm": 1.2142076942764124,
      "learning_rate": 1.931008635479847e-05,
      "loss": 0.8128,
      "step": 560
    },
    {
      "epoch": 0.2091624248033318,
      "grad_norm": 1.1799986315552842,
      "learning_rate": 1.9286297924079244e-05,
      "loss": 0.7536,
      "step": 565
    },
    {
      "epoch": 0.21101341971309578,
      "grad_norm": 1.0965045706780743,
      "learning_rate": 1.92621214606765e-05,
      "loss": 0.7744,
      "step": 570
    },
    {
      "epoch": 0.2128644146228598,
      "grad_norm": 1.2434542202635104,
      "learning_rate": 1.9237557974816063e-05,
      "loss": 0.7965,
      "step": 575
    },
    {
      "epoch": 0.21471540953262377,
      "grad_norm": 1.2091983359476026,
      "learning_rate": 1.921260849289568e-05,
      "loss": 0.7837,
      "step": 580
    },
    {
      "epoch": 0.21656640444238778,
      "grad_norm": 1.1940311597580997,
      "learning_rate": 1.9187274057442153e-05,
      "loss": 0.774,
      "step": 585
    },
    {
      "epoch": 0.2184173993521518,
      "grad_norm": 1.0743701754676238,
      "learning_rate": 1.916155572706776e-05,
      "loss": 0.8083,
      "step": 590
    },
    {
      "epoch": 0.22026839426191577,
      "grad_norm": 1.119330031173641,
      "learning_rate": 1.913545457642601e-05,
      "loss": 0.7692,
      "step": 595
    },
    {
      "epoch": 0.22211938917167978,
      "grad_norm": 1.1838482419139507,
      "learning_rate": 1.9108971696166778e-05,
      "loss": 0.7388,
      "step": 600
    },
    {
      "epoch": 0.22211938917167978,
      "eval_loss": 0.8023950457572937,
      "eval_runtime": 125.9122,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 600
    },
    {
      "epoch": 0.22397038408144376,
      "grad_norm": 1.1918846048877687,
      "learning_rate": 1.9082108192890673e-05,
      "loss": 0.806,
      "step": 605
    },
    {
      "epoch": 0.22582137899120777,
      "grad_norm": 1.1474696064483147,
      "learning_rate": 1.905486518910286e-05,
      "loss": 0.7756,
      "step": 610
    },
    {
      "epoch": 0.22767237390097178,
      "grad_norm": 1.1873077682787359,
      "learning_rate": 1.9027243823166107e-05,
      "loss": 0.7807,
      "step": 615
    },
    {
      "epoch": 0.22952336881073576,
      "grad_norm": 1.1370410152260748,
      "learning_rate": 1.8999245249253247e-05,
      "loss": 0.8518,
      "step": 620
    },
    {
      "epoch": 0.23137436372049977,
      "grad_norm": 1.1592714123770727,
      "learning_rate": 1.8970870637298936e-05,
      "loss": 0.7947,
      "step": 625
    },
    {
      "epoch": 0.23322535863026378,
      "grad_norm": 1.167379979369555,
      "learning_rate": 1.8942121172950765e-05,
      "loss": 0.8175,
      "step": 630
    },
    {
      "epoch": 0.23507635354002776,
      "grad_norm": 1.0148428174894528,
      "learning_rate": 1.8912998057519735e-05,
      "loss": 0.7813,
      "step": 635
    },
    {
      "epoch": 0.23692734844979177,
      "grad_norm": 1.2294496173862697,
      "learning_rate": 1.8883502507930044e-05,
      "loss": 0.8282,
      "step": 640
    },
    {
      "epoch": 0.23877834335955575,
      "grad_norm": 1.153834698104633,
      "learning_rate": 1.885363575666823e-05,
      "loss": 0.819,
      "step": 645
    },
    {
      "epoch": 0.24062933826931976,
      "grad_norm": 1.0699702689958812,
      "learning_rate": 1.8823399051731698e-05,
      "loss": 0.7333,
      "step": 650
    },
    {
      "epoch": 0.24248033317908377,
      "grad_norm": 1.0339682532907033,
      "learning_rate": 1.8792793656576544e-05,
      "loss": 0.7723,
      "step": 655
    },
    {
      "epoch": 0.24433132808884775,
      "grad_norm": 1.083476975643199,
      "learning_rate": 1.876182085006478e-05,
      "loss": 0.7932,
      "step": 660
    },
    {
      "epoch": 0.24618232299861176,
      "grad_norm": 0.9842008335663365,
      "learning_rate": 1.873048192641088e-05,
      "loss": 0.7978,
      "step": 665
    },
    {
      "epoch": 0.24803331790837574,
      "grad_norm": 1.0859880941267326,
      "learning_rate": 1.8698778195127715e-05,
      "loss": 0.7903,
      "step": 670
    },
    {
      "epoch": 0.24988431281813975,
      "grad_norm": 1.0920881647401042,
      "learning_rate": 1.866671098097183e-05,
      "loss": 0.7805,
      "step": 675
    },
    {
      "epoch": 0.25173530772790376,
      "grad_norm": 1.0447814725747497,
      "learning_rate": 1.863428162388808e-05,
      "loss": 0.8156,
      "step": 680
    },
    {
      "epoch": 0.25358630263766774,
      "grad_norm": 1.122277511505867,
      "learning_rate": 1.860149147895366e-05,
      "loss": 0.7803,
      "step": 685
    },
    {
      "epoch": 0.2554372975474317,
      "grad_norm": 1.1085629206709198,
      "learning_rate": 1.856834191632144e-05,
      "loss": 0.8045,
      "step": 690
    },
    {
      "epoch": 0.25728829245719576,
      "grad_norm": 1.087490803470686,
      "learning_rate": 1.8534834321162778e-05,
      "loss": 0.822,
      "step": 695
    },
    {
      "epoch": 0.25913928736695974,
      "grad_norm": 1.0114699174712363,
      "learning_rate": 1.850097009360958e-05,
      "loss": 0.7234,
      "step": 700
    },
    {
      "epoch": 0.25913928736695974,
      "eval_loss": 0.7961795926094055,
      "eval_runtime": 125.8984,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 700
    },
    {
      "epoch": 0.2609902822767237,
      "grad_norm": 1.1666771129689475,
      "learning_rate": 1.8466750648695826e-05,
      "loss": 0.8284,
      "step": 705
    },
    {
      "epoch": 0.26284127718648775,
      "grad_norm": 1.0304043488373609,
      "learning_rate": 1.843217741629843e-05,
      "loss": 0.7025,
      "step": 710
    },
    {
      "epoch": 0.26469227209625174,
      "grad_norm": 1.0509818064952419,
      "learning_rate": 1.83972518410775e-05,
      "loss": 0.739,
      "step": 715
    },
    {
      "epoch": 0.2665432670060157,
      "grad_norm": 0.9979057999152209,
      "learning_rate": 1.836197538241596e-05,
      "loss": 0.7452,
      "step": 720
    },
    {
      "epoch": 0.26839426191577975,
      "grad_norm": 1.0805505088606497,
      "learning_rate": 1.8326349514358594e-05,
      "loss": 0.7787,
      "step": 725
    },
    {
      "epoch": 0.27024525682554373,
      "grad_norm": 1.091857948129845,
      "learning_rate": 1.8290375725550417e-05,
      "loss": 0.8327,
      "step": 730
    },
    {
      "epoch": 0.2720962517353077,
      "grad_norm": 1.0581119370365102,
      "learning_rate": 1.8254055519174502e-05,
      "loss": 0.7103,
      "step": 735
    },
    {
      "epoch": 0.27394724664507175,
      "grad_norm": 1.1460921794947827,
      "learning_rate": 1.821739041288915e-05,
      "loss": 0.7927,
      "step": 740
    },
    {
      "epoch": 0.27579824155483573,
      "grad_norm": 0.9651613777166085,
      "learning_rate": 1.818038193876448e-05,
      "loss": 0.7406,
      "step": 745
    },
    {
      "epoch": 0.2776492364645997,
      "grad_norm": 0.9502996970486759,
      "learning_rate": 1.8143031643218413e-05,
      "loss": 0.7027,
      "step": 750
    },
    {
      "epoch": 0.2795002313743637,
      "grad_norm": 1.1463924114726602,
      "learning_rate": 1.8105341086952052e-05,
      "loss": 0.8102,
      "step": 755
    },
    {
      "epoch": 0.28135122628412773,
      "grad_norm": 1.1234130601114334,
      "learning_rate": 1.806731184488447e-05,
      "loss": 0.7802,
      "step": 760
    },
    {
      "epoch": 0.2832022211938917,
      "grad_norm": 1.0971581054230324,
      "learning_rate": 1.8028945506086898e-05,
      "loss": 0.7516,
      "step": 765
    },
    {
      "epoch": 0.2850532161036557,
      "grad_norm": 1.107825346177917,
      "learning_rate": 1.799024367371631e-05,
      "loss": 0.7435,
      "step": 770
    },
    {
      "epoch": 0.28690421101341973,
      "grad_norm": 1.1319764102322498,
      "learning_rate": 1.795120796494848e-05,
      "loss": 0.8346,
      "step": 775
    },
    {
      "epoch": 0.2887552059231837,
      "grad_norm": 1.054146904567116,
      "learning_rate": 1.791184001091035e-05,
      "loss": 0.7668,
      "step": 780
    },
    {
      "epoch": 0.2906062008329477,
      "grad_norm": 0.9546703867947438,
      "learning_rate": 1.7872141456611904e-05,
      "loss": 0.7729,
      "step": 785
    },
    {
      "epoch": 0.29245719574271173,
      "grad_norm": 1.0520354517409871,
      "learning_rate": 1.7832113960877445e-05,
      "loss": 0.7666,
      "step": 790
    },
    {
      "epoch": 0.2943081906524757,
      "grad_norm": 1.0339039958969705,
      "learning_rate": 1.779175919627624e-05,
      "loss": 0.7879,
      "step": 795
    },
    {
      "epoch": 0.2961591855622397,
      "grad_norm": 1.0374069302809197,
      "learning_rate": 1.775107884905266e-05,
      "loss": 0.7693,
      "step": 800
    },
    {
      "epoch": 0.2961591855622397,
      "eval_loss": 0.7911367416381836,
      "eval_runtime": 125.8736,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 800
    },
    {
      "epoch": 0.2980101804720037,
      "grad_norm": 1.023031947891097,
      "learning_rate": 1.7710074619055707e-05,
      "loss": 0.7701,
      "step": 805
    },
    {
      "epoch": 0.2998611753817677,
      "grad_norm": 1.02308459176092,
      "learning_rate": 1.7668748219668007e-05,
      "loss": 0.7079,
      "step": 810
    },
    {
      "epoch": 0.3017121702915317,
      "grad_norm": 1.0193146042109342,
      "learning_rate": 1.7627101377734176e-05,
      "loss": 0.7814,
      "step": 815
    },
    {
      "epoch": 0.30356316520129567,
      "grad_norm": 1.1175824255386086,
      "learning_rate": 1.7585135833488692e-05,
      "loss": 0.8352,
      "step": 820
    },
    {
      "epoch": 0.3054141601110597,
      "grad_norm": 0.9758167149410084,
      "learning_rate": 1.7542853340483175e-05,
      "loss": 0.7525,
      "step": 825
    },
    {
      "epoch": 0.3072651550208237,
      "grad_norm": 0.9971503389642156,
      "learning_rate": 1.7500255665513112e-05,
      "loss": 0.7279,
      "step": 830
    },
    {
      "epoch": 0.30911614993058767,
      "grad_norm": 1.169901204414489,
      "learning_rate": 1.7457344588544018e-05,
      "loss": 0.7755,
      "step": 835
    },
    {
      "epoch": 0.3109671448403517,
      "grad_norm": 1.0595523773603162,
      "learning_rate": 1.7414121902637083e-05,
      "loss": 0.7673,
      "step": 840
    },
    {
      "epoch": 0.3128181397501157,
      "grad_norm": 1.1751687012571783,
      "learning_rate": 1.7370589413874226e-05,
      "loss": 0.741,
      "step": 845
    },
    {
      "epoch": 0.31466913465987967,
      "grad_norm": 1.1835719361646306,
      "learning_rate": 1.7326748941282638e-05,
      "loss": 0.7788,
      "step": 850
    },
    {
      "epoch": 0.3165201295696437,
      "grad_norm": 1.0111560919225886,
      "learning_rate": 1.7282602316758774e-05,
      "loss": 0.7678,
      "step": 855
    },
    {
      "epoch": 0.3183711244794077,
      "grad_norm": 1.0835946252135704,
      "learning_rate": 1.72381513849918e-05,
      "loss": 0.7476,
      "step": 860
    },
    {
      "epoch": 0.32022211938917167,
      "grad_norm": 1.1099985814534863,
      "learning_rate": 1.7193398003386514e-05,
      "loss": 0.779,
      "step": 865
    },
    {
      "epoch": 0.3220731142989357,
      "grad_norm": 0.9066789182044317,
      "learning_rate": 1.7148344041985736e-05,
      "loss": 0.6901,
      "step": 870
    },
    {
      "epoch": 0.3239241092086997,
      "grad_norm": 1.1422072479983323,
      "learning_rate": 1.710299138339217e-05,
      "loss": 0.806,
      "step": 875
    },
    {
      "epoch": 0.32577510411846367,
      "grad_norm": 0.9742950243183045,
      "learning_rate": 1.7057341922689733e-05,
      "loss": 0.7739,
      "step": 880
    },
    {
      "epoch": 0.32762609902822765,
      "grad_norm": 1.0227221722499922,
      "learning_rate": 1.701139756736436e-05,
      "loss": 0.7406,
      "step": 885
    },
    {
      "epoch": 0.3294770939379917,
      "grad_norm": 1.0785612634689687,
      "learning_rate": 1.696516023722431e-05,
      "loss": 0.7728,
      "step": 890
    },
    {
      "epoch": 0.33132808884775566,
      "grad_norm": 1.0880856661959288,
      "learning_rate": 1.691863186431996e-05,
      "loss": 0.7462,
      "step": 895
    },
    {
      "epoch": 0.33317908375751965,
      "grad_norm": 1.0817764375919297,
      "learning_rate": 1.6871814392863037e-05,
      "loss": 0.8385,
      "step": 900
    },
    {
      "epoch": 0.33317908375751965,
      "eval_loss": 0.7849360704421997,
      "eval_runtime": 125.8514,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.509,
      "step": 900
    },
    {
      "epoch": 0.3350300786672837,
      "grad_norm": 1.0848147512517947,
      "learning_rate": 1.682470977914541e-05,
      "loss": 0.7653,
      "step": 905
    },
    {
      "epoch": 0.33688107357704766,
      "grad_norm": 1.1609482279651488,
      "learning_rate": 1.6777319991457325e-05,
      "loss": 0.786,
      "step": 910
    },
    {
      "epoch": 0.33873206848681164,
      "grad_norm": 1.2171008251999773,
      "learning_rate": 1.6729647010005175e-05,
      "loss": 0.8124,
      "step": 915
    },
    {
      "epoch": 0.3405830633965757,
      "grad_norm": 0.9446888141489528,
      "learning_rate": 1.6681692826828743e-05,
      "loss": 0.8249,
      "step": 920
    },
    {
      "epoch": 0.34243405830633966,
      "grad_norm": 1.099582146653851,
      "learning_rate": 1.6633459445717973e-05,
      "loss": 0.7901,
      "step": 925
    },
    {
      "epoch": 0.34428505321610364,
      "grad_norm": 0.9771996099399858,
      "learning_rate": 1.6584948882129238e-05,
      "loss": 0.759,
      "step": 930
    },
    {
      "epoch": 0.3461360481258677,
      "grad_norm": 1.0185009245061947,
      "learning_rate": 1.653616316310112e-05,
      "loss": 0.7295,
      "step": 935
    },
    {
      "epoch": 0.34798704303563166,
      "grad_norm": 1.0804952434925557,
      "learning_rate": 1.6487104327169702e-05,
      "loss": 0.755,
      "step": 940
    },
    {
      "epoch": 0.34983803794539564,
      "grad_norm": 1.0894047826173676,
      "learning_rate": 1.6437774424283414e-05,
      "loss": 0.7916,
      "step": 945
    },
    {
      "epoch": 0.3516890328551596,
      "grad_norm": 0.9805181725379623,
      "learning_rate": 1.6388175515717336e-05,
      "loss": 0.7285,
      "step": 950
    },
    {
      "epoch": 0.35354002776492366,
      "grad_norm": 1.356858685263938,
      "learning_rate": 1.63383096739871e-05,
      "loss": 0.7864,
      "step": 955
    },
    {
      "epoch": 0.35539102267468764,
      "grad_norm": 0.9525556743311706,
      "learning_rate": 1.6288178982762287e-05,
      "loss": 0.742,
      "step": 960
    },
    {
      "epoch": 0.3572420175844516,
      "grad_norm": 1.0831204564195782,
      "learning_rate": 1.6237785536779322e-05,
      "loss": 0.7386,
      "step": 965
    },
    {
      "epoch": 0.35909301249421566,
      "grad_norm": 0.9806687465833773,
      "learning_rate": 1.618713144175399e-05,
      "loss": 0.7399,
      "step": 970
    },
    {
      "epoch": 0.36094400740397964,
      "grad_norm": 0.9492684447869085,
      "learning_rate": 1.6136218814293422e-05,
      "loss": 0.763,
      "step": 975
    },
    {
      "epoch": 0.3627950023137436,
      "grad_norm": 1.05964205886519,
      "learning_rate": 1.6085049781807656e-05,
      "loss": 0.7658,
      "step": 980
    },
    {
      "epoch": 0.36464599722350766,
      "grad_norm": 1.007330323233069,
      "learning_rate": 1.603362648242076e-05,
      "loss": 0.7371,
      "step": 985
    },
    {
      "epoch": 0.36649699213327164,
      "grad_norm": 0.9531551141111623,
      "learning_rate": 1.5981951064881456e-05,
      "loss": 0.7538,
      "step": 990
    },
    {
      "epoch": 0.3683479870430356,
      "grad_norm": 1.0737293889710073,
      "learning_rate": 1.5930025688473353e-05,
      "loss": 0.7896,
      "step": 995
    },
    {
      "epoch": 0.37019898195279966,
      "grad_norm": 1.0915409073010398,
      "learning_rate": 1.5877852522924733e-05,
      "loss": 0.7926,
      "step": 1000
    },
    {
      "epoch": 0.37019898195279966,
      "eval_loss": 0.7791726589202881,
      "eval_runtime": 125.9827,
      "eval_samples_per_second": 1.016,
      "eval_steps_per_second": 0.508,
      "step": 1000
    },
    {
      "epoch": 0.37204997686256364,
      "grad_norm": 1.1973665058190375,
      "learning_rate": 1.5825433748317857e-05,
      "loss": 0.7215,
      "step": 1005
    },
    {
      "epoch": 0.3739009717723276,
      "grad_norm": 1.0701603196497034,
      "learning_rate": 1.577277155499789e-05,
      "loss": 0.7396,
      "step": 1010
    },
    {
      "epoch": 0.3757519666820916,
      "grad_norm": 1.0987041230123638,
      "learning_rate": 1.5719868143481385e-05,
      "loss": 0.7429,
      "step": 1015
    },
    {
      "epoch": 0.37760296159185563,
      "grad_norm": 1.0534343239222719,
      "learning_rate": 1.5666725724364296e-05,
      "loss": 0.7545,
      "step": 1020
    },
    {
      "epoch": 0.3794539565016196,
      "grad_norm": 1.028797489541622,
      "learning_rate": 1.5613346518229652e-05,
      "loss": 0.7264,
      "step": 1025
    },
    {
      "epoch": 0.3813049514113836,
      "grad_norm": 1.0217316773582072,
      "learning_rate": 1.5559732755554734e-05,
      "loss": 0.7177,
      "step": 1030
    },
    {
      "epoch": 0.38315594632114763,
      "grad_norm": 1.047042319865492,
      "learning_rate": 1.5505886676617902e-05,
      "loss": 0.7714,
      "step": 1035
    },
    {
      "epoch": 0.3850069412309116,
      "grad_norm": 1.0810668413004525,
      "learning_rate": 1.545181053140495e-05,
      "loss": 0.715,
      "step": 1040
    },
    {
      "epoch": 0.3868579361406756,
      "grad_norm": 1.0363170821457612,
      "learning_rate": 1.539750657951513e-05,
      "loss": 0.7664,
      "step": 1045
    },
    {
      "epoch": 0.38870893105043963,
      "grad_norm": 1.0452365868797338,
      "learning_rate": 1.5342977090066696e-05,
      "loss": 0.7236,
      "step": 1050
    },
    {
      "epoch": 0.3905599259602036,
      "grad_norm": 0.9982499977285321,
      "learning_rate": 1.52882243416021e-05,
      "loss": 0.7731,
      "step": 1055
    },
    {
      "epoch": 0.3924109208699676,
      "grad_norm": 1.030850130209849,
      "learning_rate": 1.523325062199281e-05,
      "loss": 0.7617,
      "step": 1060
    },
    {
      "epoch": 0.39426191577973163,
      "grad_norm": 1.1101345309616182,
      "learning_rate": 1.5178058228343662e-05,
      "loss": 0.7552,
      "step": 1065
    },
    {
      "epoch": 0.3961129106894956,
      "grad_norm": 0.9346869642461403,
      "learning_rate": 1.5122649466896899e-05,
      "loss": 0.7606,
      "step": 1070
    },
    {
      "epoch": 0.3979639055992596,
      "grad_norm": 0.9976085317381093,
      "learning_rate": 1.5067026652935823e-05,
      "loss": 0.7303,
      "step": 1075
    },
    {
      "epoch": 0.3998149005090236,
      "grad_norm": 0.8951591117484915,
      "learning_rate": 1.5011192110688008e-05,
      "loss": 0.6998,
      "step": 1080
    },
    {
      "epoch": 0.4016658954187876,
      "grad_norm": 1.0337985910400393,
      "learning_rate": 1.4955148173228209e-05,
      "loss": 0.7814,
      "step": 1085
    },
    {
      "epoch": 0.4035168903285516,
      "grad_norm": 1.0299336976844324,
      "learning_rate": 1.4898897182380872e-05,
      "loss": 0.7532,
      "step": 1090
    },
    {
      "epoch": 0.4053678852383156,
      "grad_norm": 0.9541582795109266,
      "learning_rate": 1.4842441488622266e-05,
      "loss": 0.7609,
      "step": 1095
    },
    {
      "epoch": 0.4072188801480796,
      "grad_norm": 1.1428468537841259,
      "learning_rate": 1.4785783450982276e-05,
      "loss": 0.7463,
      "step": 1100
    },
    {
      "epoch": 0.4072188801480796,
      "eval_loss": 0.7768411636352539,
      "eval_runtime": 125.9173,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 1100
    },
    {
      "epoch": 0.4090698750578436,
      "grad_norm": 0.9432900818713729,
      "learning_rate": 1.4728925436945838e-05,
      "loss": 0.7225,
      "step": 1105
    },
    {
      "epoch": 0.41092086996760757,
      "grad_norm": 0.9568114894842158,
      "learning_rate": 1.4671869822353993e-05,
      "loss": 0.7414,
      "step": 1110
    },
    {
      "epoch": 0.4127718648773716,
      "grad_norm": 1.1008061214147924,
      "learning_rate": 1.4614618991304625e-05,
      "loss": 0.78,
      "step": 1115
    },
    {
      "epoch": 0.4146228597871356,
      "grad_norm": 1.0503790052814121,
      "learning_rate": 1.4557175336052844e-05,
      "loss": 0.7223,
      "step": 1120
    },
    {
      "epoch": 0.41647385469689957,
      "grad_norm": 0.9583451918719977,
      "learning_rate": 1.4499541256911008e-05,
      "loss": 0.7495,
      "step": 1125
    },
    {
      "epoch": 0.4183248496066636,
      "grad_norm": 1.0275134379520088,
      "learning_rate": 1.4441719162148433e-05,
      "loss": 0.7653,
      "step": 1130
    },
    {
      "epoch": 0.4201758445164276,
      "grad_norm": 1.0425892623904864,
      "learning_rate": 1.4383711467890776e-05,
      "loss": 0.7055,
      "step": 1135
    },
    {
      "epoch": 0.42202683942619157,
      "grad_norm": 1.069154922576832,
      "learning_rate": 1.4325520598019049e-05,
      "loss": 0.7837,
      "step": 1140
    },
    {
      "epoch": 0.42387783433595555,
      "grad_norm": 0.9896521412449755,
      "learning_rate": 1.4267148984068359e-05,
      "loss": 0.7702,
      "step": 1145
    },
    {
      "epoch": 0.4257288292457196,
      "grad_norm": 1.025563581357004,
      "learning_rate": 1.4208599065126292e-05,
      "loss": 0.7604,
      "step": 1150
    },
    {
      "epoch": 0.42757982415548357,
      "grad_norm": 1.001282739654851,
      "learning_rate": 1.4149873287731005e-05,
      "loss": 0.7874,
      "step": 1155
    },
    {
      "epoch": 0.42943081906524755,
      "grad_norm": 1.0858912388862354,
      "learning_rate": 1.4090974105768977e-05,
      "loss": 0.8066,
      "step": 1160
    },
    {
      "epoch": 0.4312818139750116,
      "grad_norm": 0.9891342220670727,
      "learning_rate": 1.4031903980372503e-05,
      "loss": 0.7532,
      "step": 1165
    },
    {
      "epoch": 0.43313280888477557,
      "grad_norm": 1.1089360136811275,
      "learning_rate": 1.3972665379816821e-05,
      "loss": 0.7344,
      "step": 1170
    },
    {
      "epoch": 0.43498380379453955,
      "grad_norm": 1.0804929458235695,
      "learning_rate": 1.3913260779416999e-05,
      "loss": 0.7127,
      "step": 1175
    },
    {
      "epoch": 0.4368347987043036,
      "grad_norm": 0.9487754961773092,
      "learning_rate": 1.3853692661424485e-05,
      "loss": 0.7544,
      "step": 1180
    },
    {
      "epoch": 0.43868579361406757,
      "grad_norm": 1.1689376731656378,
      "learning_rate": 1.3793963514923398e-05,
      "loss": 0.7571,
      "step": 1185
    },
    {
      "epoch": 0.44053678852383155,
      "grad_norm": 1.1079751308984782,
      "learning_rate": 1.3734075835726515e-05,
      "loss": 0.7632,
      "step": 1190
    },
    {
      "epoch": 0.4423877834335956,
      "grad_norm": 0.9701136035154755,
      "learning_rate": 1.3674032126270982e-05,
      "loss": 0.7155,
      "step": 1195
    },
    {
      "epoch": 0.44423877834335956,
      "grad_norm": 1.0783370064152846,
      "learning_rate": 1.3613834895513748e-05,
      "loss": 0.8017,
      "step": 1200
    },
    {
      "epoch": 0.44423877834335956,
      "eval_loss": 0.7741294503211975,
      "eval_runtime": 125.9108,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 1200
    },
    {
      "epoch": 0.44608977325312354,
      "grad_norm": 1.039233538517332,
      "learning_rate": 1.355348665882673e-05,
      "loss": 0.7332,
      "step": 1205
    },
    {
      "epoch": 0.4479407681628875,
      "grad_norm": 1.1306337428774784,
      "learning_rate": 1.3492989937891694e-05,
      "loss": 0.7475,
      "step": 1210
    },
    {
      "epoch": 0.44979176307265156,
      "grad_norm": 1.0587177357723303,
      "learning_rate": 1.3432347260594911e-05,
      "loss": 0.7603,
      "step": 1215
    },
    {
      "epoch": 0.45164275798241554,
      "grad_norm": 1.0290685165693072,
      "learning_rate": 1.3371561160921507e-05,
      "loss": 0.7646,
      "step": 1220
    },
    {
      "epoch": 0.4534937528921795,
      "grad_norm": 1.34640973701702,
      "learning_rate": 1.3310634178849583e-05,
      "loss": 0.6996,
      "step": 1225
    },
    {
      "epoch": 0.45534474780194356,
      "grad_norm": 0.9941580643005229,
      "learning_rate": 1.3249568860244081e-05,
      "loss": 0.7254,
      "step": 1230
    },
    {
      "epoch": 0.45719574271170754,
      "grad_norm": 1.0409471890052142,
      "learning_rate": 1.3188367756750413e-05,
      "loss": 0.7774,
      "step": 1235
    },
    {
      "epoch": 0.4590467376214715,
      "grad_norm": 0.9723468054375276,
      "learning_rate": 1.312703342568782e-05,
      "loss": 0.7169,
      "step": 1240
    },
    {
      "epoch": 0.46089773253123556,
      "grad_norm": 0.9902905902420863,
      "learning_rate": 1.306556842994254e-05,
      "loss": 0.7297,
      "step": 1245
    },
    {
      "epoch": 0.46274872744099954,
      "grad_norm": 0.9601290745180161,
      "learning_rate": 1.3003975337860686e-05,
      "loss": 0.7538,
      "step": 1250
    },
    {
      "epoch": 0.4645997223507635,
      "grad_norm": 0.9053050963379172,
      "learning_rate": 1.2942256723140951e-05,
      "loss": 0.7109,
      "step": 1255
    },
    {
      "epoch": 0.46645071726052756,
      "grad_norm": 1.0874571881287778,
      "learning_rate": 1.2880415164727058e-05,
      "loss": 0.7919,
      "step": 1260
    },
    {
      "epoch": 0.46830171217029154,
      "grad_norm": 1.1451121612817037,
      "learning_rate": 1.2818453246699981e-05,
      "loss": 0.7391,
      "step": 1265
    },
    {
      "epoch": 0.4701527070800555,
      "grad_norm": 0.9681114397766631,
      "learning_rate": 1.2756373558169992e-05,
      "loss": 0.743,
      "step": 1270
    },
    {
      "epoch": 0.4720037019898195,
      "grad_norm": 0.9223359776123135,
      "learning_rate": 1.2694178693168468e-05,
      "loss": 0.72,
      "step": 1275
    },
    {
      "epoch": 0.47385469689958354,
      "grad_norm": 1.0469506863925069,
      "learning_rate": 1.2631871250539478e-05,
      "loss": 0.728,
      "step": 1280
    },
    {
      "epoch": 0.4757056918093475,
      "grad_norm": 1.0014451232715824,
      "learning_rate": 1.2569453833831222e-05,
      "loss": 0.717,
      "step": 1285
    },
    {
      "epoch": 0.4775566867191115,
      "grad_norm": 0.9639634997081045,
      "learning_rate": 1.250692905118721e-05,
      "loss": 0.7507,
      "step": 1290
    },
    {
      "epoch": 0.47940768162887554,
      "grad_norm": 0.9929137343346536,
      "learning_rate": 1.2444299515237298e-05,
      "loss": 0.7643,
      "step": 1295
    },
    {
      "epoch": 0.4812586765386395,
      "grad_norm": 1.0240687855090735,
      "learning_rate": 1.238156784298851e-05,
      "loss": 0.7072,
      "step": 1300
    },
    {
      "epoch": 0.4812586765386395,
      "eval_loss": 0.7685373425483704,
      "eval_runtime": 125.8483,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.509,
      "step": 1300
    },
    {
      "epoch": 0.4831096714484035,
      "grad_norm": 1.0317937768510643,
      "learning_rate": 1.2318736655715689e-05,
      "loss": 0.7638,
      "step": 1305
    },
    {
      "epoch": 0.48496066635816754,
      "grad_norm": 0.9255922281298357,
      "learning_rate": 1.2255808578851958e-05,
      "loss": 0.7467,
      "step": 1310
    },
    {
      "epoch": 0.4868116612679315,
      "grad_norm": 0.9721165058296256,
      "learning_rate": 1.2192786241879033e-05,
      "loss": 0.7546,
      "step": 1315
    },
    {
      "epoch": 0.4886626561776955,
      "grad_norm": 0.9644367734588111,
      "learning_rate": 1.2129672278217328e-05,
      "loss": 0.7169,
      "step": 1320
    },
    {
      "epoch": 0.49051365108745953,
      "grad_norm": 1.0130538871266401,
      "learning_rate": 1.2066469325115927e-05,
      "loss": 0.7369,
      "step": 1325
    },
    {
      "epoch": 0.4923646459972235,
      "grad_norm": 0.9834288208571581,
      "learning_rate": 1.2003180023542375e-05,
      "loss": 0.7325,
      "step": 1330
    },
    {
      "epoch": 0.4942156409069875,
      "grad_norm": 0.960205484882582,
      "learning_rate": 1.1939807018072345e-05,
      "loss": 0.7277,
      "step": 1335
    },
    {
      "epoch": 0.4960666358167515,
      "grad_norm": 1.000713485591568,
      "learning_rate": 1.1876352956779114e-05,
      "loss": 0.7671,
      "step": 1340
    },
    {
      "epoch": 0.4979176307265155,
      "grad_norm": 0.9972457796952376,
      "learning_rate": 1.1812820491122918e-05,
      "loss": 0.7496,
      "step": 1345
    },
    {
      "epoch": 0.4997686256362795,
      "grad_norm": 0.9176925753783508,
      "learning_rate": 1.1749212275840159e-05,
      "loss": 0.7719,
      "step": 1350
    },
    {
      "epoch": 0.5016196205460435,
      "grad_norm": 0.9486097980237569,
      "learning_rate": 1.1685530968832474e-05,
      "loss": 0.7286,
      "step": 1355
    },
    {
      "epoch": 0.5034706154558075,
      "grad_norm": 1.0669908596524535,
      "learning_rate": 1.1621779231055677e-05,
      "loss": 0.7279,
      "step": 1360
    },
    {
      "epoch": 0.5053216103655715,
      "grad_norm": 1.0656718017579419,
      "learning_rate": 1.1557959726408567e-05,
      "loss": 0.7338,
      "step": 1365
    },
    {
      "epoch": 0.5071726052753355,
      "grad_norm": 1.2390741577593356,
      "learning_rate": 1.1494075121621618e-05,
      "loss": 0.7997,
      "step": 1370
    },
    {
      "epoch": 0.5090236001850995,
      "grad_norm": 1.0260656127221468,
      "learning_rate": 1.1430128086145542e-05,
      "loss": 0.7257,
      "step": 1375
    },
    {
      "epoch": 0.5108745950948634,
      "grad_norm": 0.981403662935476,
      "learning_rate": 1.1366121292039756e-05,
      "loss": 0.6885,
      "step": 1380
    },
    {
      "epoch": 0.5127255900046275,
      "grad_norm": 0.9506854040681342,
      "learning_rate": 1.1302057413860717e-05,
      "loss": 0.7523,
      "step": 1385
    },
    {
      "epoch": 0.5145765849143915,
      "grad_norm": 0.9571399024989548,
      "learning_rate": 1.1237939128550167e-05,
      "loss": 0.7406,
      "step": 1390
    },
    {
      "epoch": 0.5164275798241554,
      "grad_norm": 0.9379988174297818,
      "learning_rate": 1.1173769115323276e-05,
      "loss": 0.7152,
      "step": 1395
    },
    {
      "epoch": 0.5182785747339195,
      "grad_norm": 0.9799140133111384,
      "learning_rate": 1.1109550055556704e-05,
      "loss": 0.6862,
      "step": 1400
    },
    {
      "epoch": 0.5182785747339195,
      "eval_loss": 0.762523889541626,
      "eval_runtime": 125.8824,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 1400
    },
    {
      "epoch": 0.5201295696436835,
      "grad_norm": 1.0128868126138035,
      "learning_rate": 1.1045284632676535e-05,
      "loss": 0.7491,
      "step": 1405
    },
    {
      "epoch": 0.5219805645534474,
      "grad_norm": 1.1054871656808112,
      "learning_rate": 1.098097553204616e-05,
      "loss": 0.7193,
      "step": 1410
    },
    {
      "epoch": 0.5238315594632115,
      "grad_norm": 1.0534251043687004,
      "learning_rate": 1.0916625440854074e-05,
      "loss": 0.7263,
      "step": 1415
    },
    {
      "epoch": 0.5256825543729755,
      "grad_norm": 1.0586326058499738,
      "learning_rate": 1.0852237048001568e-05,
      "loss": 0.771,
      "step": 1420
    },
    {
      "epoch": 0.5275335492827394,
      "grad_norm": 1.0689916226823393,
      "learning_rate": 1.0787813043990405e-05,
      "loss": 0.7936,
      "step": 1425
    },
    {
      "epoch": 0.5293845441925035,
      "grad_norm": 0.9447698094104972,
      "learning_rate": 1.0723356120810367e-05,
      "loss": 0.7594,
      "step": 1430
    },
    {
      "epoch": 0.5312355391022675,
      "grad_norm": 1.0832488716625996,
      "learning_rate": 1.0658868971826785e-05,
      "loss": 0.7371,
      "step": 1435
    },
    {
      "epoch": 0.5330865340120314,
      "grad_norm": 1.0138170678877134,
      "learning_rate": 1.0594354291667995e-05,
      "loss": 0.7389,
      "step": 1440
    },
    {
      "epoch": 0.5349375289217955,
      "grad_norm": 1.0133195189163617,
      "learning_rate": 1.0529814776112724e-05,
      "loss": 0.7443,
      "step": 1445
    },
    {
      "epoch": 0.5367885238315595,
      "grad_norm": 1.062091180379715,
      "learning_rate": 1.046525312197747e-05,
      "loss": 0.7549,
      "step": 1450
    },
    {
      "epoch": 0.5386395187413234,
      "grad_norm": 0.9945798468589808,
      "learning_rate": 1.0400672027003795e-05,
      "loss": 0.7033,
      "step": 1455
    },
    {
      "epoch": 0.5404905136510875,
      "grad_norm": 1.0798176885934982,
      "learning_rate": 1.0336074189745617e-05,
      "loss": 0.7873,
      "step": 1460
    },
    {
      "epoch": 0.5423415085608515,
      "grad_norm": 0.949930038696299,
      "learning_rate": 1.027146230945643e-05,
      "loss": 0.7655,
      "step": 1465
    },
    {
      "epoch": 0.5441925034706154,
      "grad_norm": 0.9128607712688908,
      "learning_rate": 1.0206839085976528e-05,
      "loss": 0.7045,
      "step": 1470
    },
    {
      "epoch": 0.5460434983803795,
      "grad_norm": 0.9894531926372522,
      "learning_rate": 1.014220721962018e-05,
      "loss": 0.7288,
      "step": 1475
    },
    {
      "epoch": 0.5478944932901435,
      "grad_norm": 1.0508653514368567,
      "learning_rate": 1.0077569411062804e-05,
      "loss": 0.7393,
      "step": 1480
    },
    {
      "epoch": 0.5497454881999074,
      "grad_norm": 1.0144705488797956,
      "learning_rate": 1.001292836122812e-05,
      "loss": 0.7388,
      "step": 1485
    },
    {
      "epoch": 0.5515964831096715,
      "grad_norm": 1.0862811301549016,
      "learning_rate": 9.948286771175288e-06,
      "loss": 0.7666,
      "step": 1490
    },
    {
      "epoch": 0.5534474780194355,
      "grad_norm": 0.9773752334499499,
      "learning_rate": 9.883647341986032e-06,
      "loss": 0.7572,
      "step": 1495
    },
    {
      "epoch": 0.5552984729291994,
      "grad_norm": 0.965789789651361,
      "learning_rate": 9.819012774651788e-06,
      "loss": 0.7466,
      "step": 1500
    },
    {
      "epoch": 0.5552984729291994,
      "eval_loss": 0.759286105632782,
      "eval_runtime": 125.877,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 1500
    },
    {
      "epoch": 0.5571494678389635,
      "grad_norm": 1.0286512648148671,
      "learning_rate": 9.75438576996084e-06,
      "loss": 0.7419,
      "step": 1505
    },
    {
      "epoch": 0.5590004627487274,
      "grad_norm": 0.9779662642913994,
      "learning_rate": 9.689769028385463e-06,
      "loss": 0.7594,
      "step": 1510
    },
    {
      "epoch": 0.5608514576584914,
      "grad_norm": 0.9878313389871681,
      "learning_rate": 9.625165249969075e-06,
      "loss": 0.7236,
      "step": 1515
    },
    {
      "epoch": 0.5627024525682555,
      "grad_norm": 1.020571296588044,
      "learning_rate": 9.56057713421343e-06,
      "loss": 0.7498,
      "step": 1520
    },
    {
      "epoch": 0.5645534474780194,
      "grad_norm": 0.9995998586225845,
      "learning_rate": 9.496007379965801e-06,
      "loss": 0.7606,
      "step": 1525
    },
    {
      "epoch": 0.5664044423877834,
      "grad_norm": 1.060622141268455,
      "learning_rate": 9.431458685306227e-06,
      "loss": 0.6984,
      "step": 1530
    },
    {
      "epoch": 0.5682554372975475,
      "grad_norm": 0.9493673992150904,
      "learning_rate": 9.366933747434758e-06,
      "loss": 0.6911,
      "step": 1535
    },
    {
      "epoch": 0.5701064322073114,
      "grad_norm": 1.0167667579449835,
      "learning_rate": 9.302435262558748e-06,
      "loss": 0.6778,
      "step": 1540
    },
    {
      "epoch": 0.5719574271170754,
      "grad_norm": 1.0392115654580312,
      "learning_rate": 9.237965925780207e-06,
      "loss": 0.7691,
      "step": 1545
    },
    {
      "epoch": 0.5738084220268395,
      "grad_norm": 1.0593025980336248,
      "learning_rate": 9.173528430983167e-06,
      "loss": 0.738,
      "step": 1550
    },
    {
      "epoch": 0.5756594169366034,
      "grad_norm": 0.971968443552632,
      "learning_rate": 9.109125470721141e-06,
      "loss": 0.7421,
      "step": 1555
    },
    {
      "epoch": 0.5775104118463674,
      "grad_norm": 0.932647291828516,
      "learning_rate": 9.044759736104584e-06,
      "loss": 0.7055,
      "step": 1560
    },
    {
      "epoch": 0.5793614067561315,
      "grad_norm": 0.8632798438231859,
      "learning_rate": 8.98043391668847e-06,
      "loss": 0.6961,
      "step": 1565
    },
    {
      "epoch": 0.5812124016658954,
      "grad_norm": 0.9441485921057498,
      "learning_rate": 8.916150700359896e-06,
      "loss": 0.7468,
      "step": 1570
    },
    {
      "epoch": 0.5830633965756594,
      "grad_norm": 0.9939474443131038,
      "learning_rate": 8.85191277322575e-06,
      "loss": 0.7149,
      "step": 1575
    },
    {
      "epoch": 0.5849143914854235,
      "grad_norm": 1.0783126296675078,
      "learning_rate": 8.787722819500513e-06,
      "loss": 0.7135,
      "step": 1580
    },
    {
      "epoch": 0.5867653863951874,
      "grad_norm": 0.9705292869514378,
      "learning_rate": 8.723583521394054e-06,
      "loss": 0.7109,
      "step": 1585
    },
    {
      "epoch": 0.5886163813049514,
      "grad_norm": 0.9367179465636712,
      "learning_rate": 8.659497558999579e-06,
      "loss": 0.7114,
      "step": 1590
    },
    {
      "epoch": 0.5904673762147155,
      "grad_norm": 1.0408124857679764,
      "learning_rate": 8.595467610181638e-06,
      "loss": 0.7161,
      "step": 1595
    },
    {
      "epoch": 0.5923183711244794,
      "grad_norm": 0.971797951572304,
      "learning_rate": 8.53149635046421e-06,
      "loss": 0.7508,
      "step": 1600
    },
    {
      "epoch": 0.5923183711244794,
      "eval_loss": 0.7539975643157959,
      "eval_runtime": 125.8675,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 1600
    },
    {
      "epoch": 0.5941693660342434,
      "grad_norm": 0.9800435183445545,
      "learning_rate": 8.467586452918942e-06,
      "loss": 0.715,
      "step": 1605
    },
    {
      "epoch": 0.5960203609440075,
      "grad_norm": 1.0113663434618827,
      "learning_rate": 8.403740588053408e-06,
      "loss": 0.6989,
      "step": 1610
    },
    {
      "epoch": 0.5978713558537714,
      "grad_norm": 1.0355947843868953,
      "learning_rate": 8.339961423699563e-06,
      "loss": 0.7193,
      "step": 1615
    },
    {
      "epoch": 0.5997223507635354,
      "grad_norm": 1.0175465647016573,
      "learning_rate": 8.276251624902235e-06,
      "loss": 0.7059,
      "step": 1620
    },
    {
      "epoch": 0.6015733456732995,
      "grad_norm": 1.0190101018280562,
      "learning_rate": 8.212613853807783e-06,
      "loss": 0.7302,
      "step": 1625
    },
    {
      "epoch": 0.6034243405830634,
      "grad_norm": 1.053793043329184,
      "learning_rate": 8.149050769552856e-06,
      "loss": 0.7198,
      "step": 1630
    },
    {
      "epoch": 0.6052753354928274,
      "grad_norm": 1.0096414981283535,
      "learning_rate": 8.085565028153261e-06,
      "loss": 0.7581,
      "step": 1635
    },
    {
      "epoch": 0.6071263304025913,
      "grad_norm": 0.9812300345845569,
      "learning_rate": 8.02215928239301e-06,
      "loss": 0.7472,
      "step": 1640
    },
    {
      "epoch": 0.6089773253123554,
      "grad_norm": 1.0007016748169104,
      "learning_rate": 7.958836181713445e-06,
      "loss": 0.7582,
      "step": 1645
    },
    {
      "epoch": 0.6108283202221194,
      "grad_norm": 0.9909841301661106,
      "learning_rate": 7.895598372102547e-06,
      "loss": 0.7247,
      "step": 1650
    },
    {
      "epoch": 0.6126793151318833,
      "grad_norm": 1.0341673384965404,
      "learning_rate": 7.832448495984368e-06,
      "loss": 0.6798,
      "step": 1655
    },
    {
      "epoch": 0.6145303100416474,
      "grad_norm": 0.9331106658706707,
      "learning_rate": 7.769389192108608e-06,
      "loss": 0.7123,
      "step": 1660
    },
    {
      "epoch": 0.6163813049514114,
      "grad_norm": 0.9435545677177092,
      "learning_rate": 7.706423095440367e-06,
      "loss": 0.7243,
      "step": 1665
    },
    {
      "epoch": 0.6182322998611753,
      "grad_norm": 0.991710730848484,
      "learning_rate": 7.643552837050026e-06,
      "loss": 0.7433,
      "step": 1670
    },
    {
      "epoch": 0.6200832947709394,
      "grad_norm": 1.0668190323716082,
      "learning_rate": 7.580781044003324e-06,
      "loss": 0.7375,
      "step": 1675
    },
    {
      "epoch": 0.6219342896807034,
      "grad_norm": 1.0279605405376717,
      "learning_rate": 7.518110339251568e-06,
      "loss": 0.7234,
      "step": 1680
    },
    {
      "epoch": 0.6237852845904673,
      "grad_norm": 0.9640733339122612,
      "learning_rate": 7.455543341522042e-06,
      "loss": 0.7168,
      "step": 1685
    },
    {
      "epoch": 0.6256362795002314,
      "grad_norm": 0.9907817877992142,
      "learning_rate": 7.393082665208587e-06,
      "loss": 0.7053,
      "step": 1690
    },
    {
      "epoch": 0.6274872744099954,
      "grad_norm": 1.0177580646696514,
      "learning_rate": 7.33073092026233e-06,
      "loss": 0.7772,
      "step": 1695
    },
    {
      "epoch": 0.6293382693197593,
      "grad_norm": 1.0511047158796323,
      "learning_rate": 7.268490712082667e-06,
      "loss": 0.7027,
      "step": 1700
    },
    {
      "epoch": 0.6293382693197593,
      "eval_loss": 0.7508743405342102,
      "eval_runtime": 125.8608,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 1700
    },
    {
      "epoch": 0.6311892642295234,
      "grad_norm": 0.9786722954600374,
      "learning_rate": 7.206364641408358e-06,
      "loss": 0.754,
      "step": 1705
    },
    {
      "epoch": 0.6330402591392874,
      "grad_norm": 1.001046944711236,
      "learning_rate": 7.144355304208877e-06,
      "loss": 0.7351,
      "step": 1710
    },
    {
      "epoch": 0.6348912540490513,
      "grad_norm": 0.9068727644526208,
      "learning_rate": 7.082465291575928e-06,
      "loss": 0.7213,
      "step": 1715
    },
    {
      "epoch": 0.6367422489588154,
      "grad_norm": 0.9817053840255833,
      "learning_rate": 7.02069718961518e-06,
      "loss": 0.7405,
      "step": 1720
    },
    {
      "epoch": 0.6385932438685794,
      "grad_norm": 0.9963090460429577,
      "learning_rate": 6.959053579338202e-06,
      "loss": 0.7329,
      "step": 1725
    },
    {
      "epoch": 0.6404442387783433,
      "grad_norm": 1.0063439847490308,
      "learning_rate": 6.8975370365546055e-06,
      "loss": 0.7466,
      "step": 1730
    },
    {
      "epoch": 0.6422952336881074,
      "grad_norm": 1.0756259763530167,
      "learning_rate": 6.836150131764434e-06,
      "loss": 0.7328,
      "step": 1735
    },
    {
      "epoch": 0.6441462285978714,
      "grad_norm": 1.0006380504007437,
      "learning_rate": 6.7748954300507405e-06,
      "loss": 0.7291,
      "step": 1740
    },
    {
      "epoch": 0.6459972235076353,
      "grad_norm": 1.0476443249218472,
      "learning_rate": 6.713775490972396e-06,
      "loss": 0.722,
      "step": 1745
    },
    {
      "epoch": 0.6478482184173994,
      "grad_norm": 0.9299325627763796,
      "learning_rate": 6.652792868457159e-06,
      "loss": 0.7259,
      "step": 1750
    },
    {
      "epoch": 0.6496992133271634,
      "grad_norm": 1.0107437601362794,
      "learning_rate": 6.591950110694928e-06,
      "loss": 0.7313,
      "step": 1755
    },
    {
      "epoch": 0.6515502082369273,
      "grad_norm": 1.0116149889736052,
      "learning_rate": 6.531249760031304e-06,
      "loss": 0.7495,
      "step": 1760
    },
    {
      "epoch": 0.6534012031466914,
      "grad_norm": 1.060084766737352,
      "learning_rate": 6.4706943528613135e-06,
      "loss": 0.7406,
      "step": 1765
    },
    {
      "epoch": 0.6552521980564553,
      "grad_norm": 0.9893182231305682,
      "learning_rate": 6.410286419523457e-06,
      "loss": 0.7236,
      "step": 1770
    },
    {
      "epoch": 0.6571031929662193,
      "grad_norm": 1.0249584716250184,
      "learning_rate": 6.350028484193971e-06,
      "loss": 0.7735,
      "step": 1775
    },
    {
      "epoch": 0.6589541878759834,
      "grad_norm": 0.9922083712983304,
      "learning_rate": 6.2899230647813315e-06,
      "loss": 0.7092,
      "step": 1780
    },
    {
      "epoch": 0.6608051827857473,
      "grad_norm": 1.079323750602494,
      "learning_rate": 6.229972672821081e-06,
      "loss": 0.6999,
      "step": 1785
    },
    {
      "epoch": 0.6626561776955113,
      "grad_norm": 0.9678468916444759,
      "learning_rate": 6.170179813370838e-06,
      "loss": 0.7215,
      "step": 1790
    },
    {
      "epoch": 0.6645071726052754,
      "grad_norm": 0.9400337665409387,
      "learning_rate": 6.110546984905661e-06,
      "loss": 0.7189,
      "step": 1795
    },
    {
      "epoch": 0.6663581675150393,
      "grad_norm": 0.9282443073802316,
      "learning_rate": 6.051076679213632e-06,
      "loss": 0.7228,
      "step": 1800
    },
    {
      "epoch": 0.6663581675150393,
      "eval_loss": 0.7472089529037476,
      "eval_runtime": 125.9422,
      "eval_samples_per_second": 1.016,
      "eval_steps_per_second": 0.508,
      "step": 1800
    },
    {
      "epoch": 0.6682091624248033,
      "grad_norm": 0.9554957232923886,
      "learning_rate": 5.991771381291727e-06,
      "loss": 0.7502,
      "step": 1805
    },
    {
      "epoch": 0.6700601573345674,
      "grad_norm": 1.0695199061812564,
      "learning_rate": 5.932633569242e-06,
      "loss": 0.7725,
      "step": 1810
    },
    {
      "epoch": 0.6719111522443313,
      "grad_norm": 0.9775812687514566,
      "learning_rate": 5.8736657141680066e-06,
      "loss": 0.7201,
      "step": 1815
    },
    {
      "epoch": 0.6737621471540953,
      "grad_norm": 1.0498313112511333,
      "learning_rate": 5.814870280071581e-06,
      "loss": 0.7066,
      "step": 1820
    },
    {
      "epoch": 0.6756131420638594,
      "grad_norm": 0.8996562202890858,
      "learning_rate": 5.756249723749847e-06,
      "loss": 0.6839,
      "step": 1825
    },
    {
      "epoch": 0.6774641369736233,
      "grad_norm": 0.9343357642622071,
      "learning_rate": 5.697806494692575e-06,
      "loss": 0.7006,
      "step": 1830
    },
    {
      "epoch": 0.6793151318833873,
      "grad_norm": 0.9122261783600852,
      "learning_rate": 5.6395430349798376e-06,
      "loss": 0.71,
      "step": 1835
    },
    {
      "epoch": 0.6811661267931514,
      "grad_norm": 1.0003802308196195,
      "learning_rate": 5.581461779179924e-06,
      "loss": 0.7352,
      "step": 1840
    },
    {
      "epoch": 0.6830171217029153,
      "grad_norm": 0.9577445500508659,
      "learning_rate": 5.5235651542476745e-06,
      "loss": 0.7187,
      "step": 1845
    },
    {
      "epoch": 0.6848681166126793,
      "grad_norm": 0.9048825285696591,
      "learning_rate": 5.465855579423012e-06,
      "loss": 0.6786,
      "step": 1850
    },
    {
      "epoch": 0.6867191115224434,
      "grad_norm": 0.9658805694805159,
      "learning_rate": 5.4083354661298816e-06,
      "loss": 0.7012,
      "step": 1855
    },
    {
      "epoch": 0.6885701064322073,
      "grad_norm": 0.9198866529754813,
      "learning_rate": 5.351007217875493e-06,
      "loss": 0.7193,
      "step": 1860
    },
    {
      "epoch": 0.6904211013419713,
      "grad_norm": 0.9860185880901673,
      "learning_rate": 5.293873230149851e-06,
      "loss": 0.6928,
      "step": 1865
    },
    {
      "epoch": 0.6922720962517354,
      "grad_norm": 0.9158051945254762,
      "learning_rate": 5.236935890325717e-06,
      "loss": 0.7283,
      "step": 1870
    },
    {
      "epoch": 0.6941230911614993,
      "grad_norm": 1.0229541633123165,
      "learning_rate": 5.180197577558792e-06,
      "loss": 0.6934,
      "step": 1875
    },
    {
      "epoch": 0.6959740860712633,
      "grad_norm": 1.058766779890904,
      "learning_rate": 5.123660662688352e-06,
      "loss": 0.7485,
      "step": 1880
    },
    {
      "epoch": 0.6978250809810274,
      "grad_norm": 0.8745098195983263,
      "learning_rate": 5.067327508138148e-06,
      "loss": 0.7205,
      "step": 1885
    },
    {
      "epoch": 0.6996760758907913,
      "grad_norm": 0.9669286665541216,
      "learning_rate": 5.0112004678177e-06,
      "loss": 0.7461,
      "step": 1890
    },
    {
      "epoch": 0.7015270708005553,
      "grad_norm": 0.9439248557044554,
      "learning_rate": 4.955281887023955e-06,
      "loss": 0.7075,
      "step": 1895
    },
    {
      "epoch": 0.7033780657103192,
      "grad_norm": 0.9612004193161535,
      "learning_rate": 4.899574102343247e-06,
      "loss": 0.7206,
      "step": 1900
    },
    {
      "epoch": 0.7033780657103192,
      "eval_loss": 0.7448270320892334,
      "eval_runtime": 125.9022,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 1900
    },
    {
      "epoch": 0.7052290606200833,
      "grad_norm": 0.9770375482265843,
      "learning_rate": 4.844079441553717e-06,
      "loss": 0.7312,
      "step": 1905
    },
    {
      "epoch": 0.7070800555298473,
      "grad_norm": 0.9346202648428155,
      "learning_rate": 4.7888002235279915e-06,
      "loss": 0.7138,
      "step": 1910
    },
    {
      "epoch": 0.7089310504396112,
      "grad_norm": 1.043008859124202,
      "learning_rate": 4.733738758136327e-06,
      "loss": 0.7201,
      "step": 1915
    },
    {
      "epoch": 0.7107820453493753,
      "grad_norm": 0.9898196140644209,
      "learning_rate": 4.678897346150067e-06,
      "loss": 0.666,
      "step": 1920
    },
    {
      "epoch": 0.7126330402591393,
      "grad_norm": 1.030447829309688,
      "learning_rate": 4.624278279145509e-06,
      "loss": 0.6968,
      "step": 1925
    },
    {
      "epoch": 0.7144840351689032,
      "grad_norm": 1.0002038057708351,
      "learning_rate": 4.56988383940817e-06,
      "loss": 0.6815,
      "step": 1930
    },
    {
      "epoch": 0.7163350300786673,
      "grad_norm": 0.8768349207452324,
      "learning_rate": 4.515716299837376e-06,
      "loss": 0.7143,
      "step": 1935
    },
    {
      "epoch": 0.7181860249884313,
      "grad_norm": 0.9922854914505618,
      "learning_rate": 4.461777923851337e-06,
      "loss": 0.7168,
      "step": 1940
    },
    {
      "epoch": 0.7200370198981952,
      "grad_norm": 0.9051677770762211,
      "learning_rate": 4.408070965292534e-06,
      "loss": 0.6871,
      "step": 1945
    },
    {
      "epoch": 0.7218880148079593,
      "grad_norm": 0.9724233713765718,
      "learning_rate": 4.354597668333551e-06,
      "loss": 0.7079,
      "step": 1950
    },
    {
      "epoch": 0.7237390097177233,
      "grad_norm": 1.0146841138791294,
      "learning_rate": 4.30136026738332e-06,
      "loss": 0.754,
      "step": 1955
    },
    {
      "epoch": 0.7255900046274872,
      "grad_norm": 0.9260856251952512,
      "learning_rate": 4.2483609869937115e-06,
      "loss": 0.7366,
      "step": 1960
    },
    {
      "epoch": 0.7274409995372513,
      "grad_norm": 0.9416530579594775,
      "learning_rate": 4.195602041766638e-06,
      "loss": 0.6743,
      "step": 1965
    },
    {
      "epoch": 0.7292919944470153,
      "grad_norm": 0.9416273470678079,
      "learning_rate": 4.143085636261452e-06,
      "loss": 0.7548,
      "step": 1970
    },
    {
      "epoch": 0.7311429893567792,
      "grad_norm": 0.9500826070023658,
      "learning_rate": 4.090813964902889e-06,
      "loss": 0.7318,
      "step": 1975
    },
    {
      "epoch": 0.7329939842665433,
      "grad_norm": 1.044688650303936,
      "learning_rate": 4.038789211889329e-06,
      "loss": 0.691,
      "step": 1980
    },
    {
      "epoch": 0.7348449791763073,
      "grad_norm": 0.9723998488803949,
      "learning_rate": 3.987013551101543e-06,
      "loss": 0.7205,
      "step": 1985
    },
    {
      "epoch": 0.7366959740860712,
      "grad_norm": 1.0125026598661098,
      "learning_rate": 3.9354891460118695e-06,
      "loss": 0.7348,
      "step": 1990
    },
    {
      "epoch": 0.7385469689958353,
      "grad_norm": 0.9786383589933646,
      "learning_rate": 3.884218149593776e-06,
      "loss": 0.6945,
      "step": 1995
    },
    {
      "epoch": 0.7403979639055993,
      "grad_norm": 1.0768006764875049,
      "learning_rate": 3.833202704231944e-06,
      "loss": 0.7419,
      "step": 2000
    },
    {
      "epoch": 0.7403979639055993,
      "eval_loss": 0.7422940135002136,
      "eval_runtime": 125.9877,
      "eval_samples_per_second": 1.016,
      "eval_steps_per_second": 0.508,
      "step": 2000
    },
    {
      "epoch": 0.7422489588153632,
      "grad_norm": 0.9714783620843155,
      "learning_rate": 3.7824449416327123e-06,
      "loss": 0.7486,
      "step": 2005
    },
    {
      "epoch": 0.7440999537251273,
      "grad_norm": 0.9621418965839199,
      "learning_rate": 3.7319469827350117e-06,
      "loss": 0.7492,
      "step": 2010
    },
    {
      "epoch": 0.7459509486348913,
      "grad_norm": 0.8838654671554304,
      "learning_rate": 3.6817109376217574e-06,
      "loss": 0.6973,
      "step": 2015
    },
    {
      "epoch": 0.7478019435446552,
      "grad_norm": 1.0141612675604508,
      "learning_rate": 3.631738905431641e-06,
      "loss": 0.633,
      "step": 2020
    },
    {
      "epoch": 0.7496529384544193,
      "grad_norm": 1.2478799462658503,
      "learning_rate": 3.5820329742714666e-06,
      "loss": 0.723,
      "step": 2025
    },
    {
      "epoch": 0.7515039333641832,
      "grad_norm": 1.2113929317378798,
      "learning_rate": 3.532595221128843e-06,
      "loss": 0.7189,
      "step": 2030
    },
    {
      "epoch": 0.7533549282739472,
      "grad_norm": 1.076623825378696,
      "learning_rate": 3.483427711785449e-06,
      "loss": 0.6935,
      "step": 2035
    },
    {
      "epoch": 0.7552059231837113,
      "grad_norm": 1.0039638646776863,
      "learning_rate": 3.4345325007306752e-06,
      "loss": 0.7032,
      "step": 2040
    },
    {
      "epoch": 0.7570569180934752,
      "grad_norm": 1.0481522906376972,
      "learning_rate": 3.38591163107579e-06,
      "loss": 0.7537,
      "step": 2045
    },
    {
      "epoch": 0.7589079130032392,
      "grad_norm": 1.0167538112300836,
      "learning_rate": 3.337567134468579e-06,
      "loss": 0.7492,
      "step": 2050
    },
    {
      "epoch": 0.7607589079130033,
      "grad_norm": 1.11799756175693,
      "learning_rate": 3.2895010310084174e-06,
      "loss": 0.687,
      "step": 2055
    },
    {
      "epoch": 0.7626099028227672,
      "grad_norm": 1.026751398004723,
      "learning_rate": 3.241715329161903e-06,
      "loss": 0.7888,
      "step": 2060
    },
    {
      "epoch": 0.7644608977325312,
      "grad_norm": 0.898992232633679,
      "learning_rate": 3.1942120256788966e-06,
      "loss": 0.7056,
      "step": 2065
    },
    {
      "epoch": 0.7663118926422953,
      "grad_norm": 0.9452341682541248,
      "learning_rate": 3.146993105509104e-06,
      "loss": 0.6935,
      "step": 2070
    },
    {
      "epoch": 0.7681628875520592,
      "grad_norm": 1.0381440177041223,
      "learning_rate": 3.10006054171913e-06,
      "loss": 0.6718,
      "step": 2075
    },
    {
      "epoch": 0.7700138824618232,
      "grad_norm": 0.90151266081136,
      "learning_rate": 3.0534162954100264e-06,
      "loss": 0.734,
      "step": 2080
    },
    {
      "epoch": 0.7718648773715873,
      "grad_norm": 0.9562956991505516,
      "learning_rate": 3.0070623156353685e-06,
      "loss": 0.7341,
      "step": 2085
    },
    {
      "epoch": 0.7737158722813512,
      "grad_norm": 0.8801270739522954,
      "learning_rate": 2.9610005393197707e-06,
      "loss": 0.6526,
      "step": 2090
    },
    {
      "epoch": 0.7755668671911152,
      "grad_norm": 0.9441163162257625,
      "learning_rate": 2.9152328911780027e-06,
      "loss": 0.6828,
      "step": 2095
    },
    {
      "epoch": 0.7774178621008793,
      "grad_norm": 0.9634570912343101,
      "learning_rate": 2.869761283634526e-06,
      "loss": 0.7299,
      "step": 2100
    },
    {
      "epoch": 0.7774178621008793,
      "eval_loss": 0.7405505180358887,
      "eval_runtime": 125.9141,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 2100
    },
    {
      "epoch": 0.7792688570106432,
      "grad_norm": 0.9315181675274804,
      "learning_rate": 2.8245876167435924e-06,
      "loss": 0.7134,
      "step": 2105
    },
    {
      "epoch": 0.7811198519204072,
      "grad_norm": 0.899902372156068,
      "learning_rate": 2.779713778109867e-06,
      "loss": 0.7112,
      "step": 2110
    },
    {
      "epoch": 0.7829708468301713,
      "grad_norm": 0.9552279346547553,
      "learning_rate": 2.7351416428095157e-06,
      "loss": 0.7195,
      "step": 2115
    },
    {
      "epoch": 0.7848218417399352,
      "grad_norm": 0.9400225445949076,
      "learning_rate": 2.6908730733119025e-06,
      "loss": 0.7465,
      "step": 2120
    },
    {
      "epoch": 0.7866728366496992,
      "grad_norm": 0.9800712358563762,
      "learning_rate": 2.6469099194017144e-06,
      "loss": 0.739,
      "step": 2125
    },
    {
      "epoch": 0.7885238315594633,
      "grad_norm": 0.9495123514350187,
      "learning_rate": 2.603254018101715e-06,
      "loss": 0.7729,
      "step": 2130
    },
    {
      "epoch": 0.7903748264692272,
      "grad_norm": 0.981224001693797,
      "learning_rate": 2.5599071935959495e-06,
      "loss": 0.6977,
      "step": 2135
    },
    {
      "epoch": 0.7922258213789912,
      "grad_norm": 0.979970394782048,
      "learning_rate": 2.5168712571535305e-06,
      "loss": 0.7,
      "step": 2140
    },
    {
      "epoch": 0.7940768162887553,
      "grad_norm": 1.0931405844387208,
      "learning_rate": 2.4741480070529657e-06,
      "loss": 0.7212,
      "step": 2145
    },
    {
      "epoch": 0.7959278111985192,
      "grad_norm": 0.9525278212316626,
      "learning_rate": 2.4317392285069885e-06,
      "loss": 0.7107,
      "step": 2150
    },
    {
      "epoch": 0.7977788061082832,
      "grad_norm": 0.9587323014669643,
      "learning_rate": 2.3896466935879957e-06,
      "loss": 0.6903,
      "step": 2155
    },
    {
      "epoch": 0.7996298010180471,
      "grad_norm": 0.8878613939635316,
      "learning_rate": 2.3478721611539712e-06,
      "loss": 0.755,
      "step": 2160
    },
    {
      "epoch": 0.8014807959278112,
      "grad_norm": 0.9724146216671814,
      "learning_rate": 2.3064173767750055e-06,
      "loss": 0.695,
      "step": 2165
    },
    {
      "epoch": 0.8033317908375752,
      "grad_norm": 1.0468722007497135,
      "learning_rate": 2.265284072660362e-06,
      "loss": 0.7233,
      "step": 2170
    },
    {
      "epoch": 0.8051827857473391,
      "grad_norm": 1.0316210745516678,
      "learning_rate": 2.224473967586073e-06,
      "loss": 0.7607,
      "step": 2175
    },
    {
      "epoch": 0.8070337806571032,
      "grad_norm": 1.0548409137457446,
      "learning_rate": 2.1839887668231486e-06,
      "loss": 0.697,
      "step": 2180
    },
    {
      "epoch": 0.8088847755668672,
      "grad_norm": 1.0039170894073437,
      "learning_rate": 2.1438301620662994e-06,
      "loss": 0.7324,
      "step": 2185
    },
    {
      "epoch": 0.8107357704766311,
      "grad_norm": 0.9588651745318834,
      "learning_rate": 2.103999831363258e-06,
      "loss": 0.6748,
      "step": 2190
    },
    {
      "epoch": 0.8125867653863952,
      "grad_norm": 0.8996370636189378,
      "learning_rate": 2.064499439044657e-06,
      "loss": 0.73,
      "step": 2195
    },
    {
      "epoch": 0.8144377602961592,
      "grad_norm": 0.9239380519912793,
      "learning_rate": 2.0253306356544843e-06,
      "loss": 0.7057,
      "step": 2200
    },
    {
      "epoch": 0.8144377602961592,
      "eval_loss": 0.738842248916626,
      "eval_runtime": 125.9296,
      "eval_samples_per_second": 1.016,
      "eval_steps_per_second": 0.508,
      "step": 2200
    },
    {
      "epoch": 0.8162887552059231,
      "grad_norm": 0.9429244883085828,
      "learning_rate": 1.986495057881126e-06,
      "loss": 0.6772,
      "step": 2205
    },
    {
      "epoch": 0.8181397501156872,
      "grad_norm": 1.0195318731984253,
      "learning_rate": 1.947994328488949e-06,
      "loss": 0.7356,
      "step": 2210
    },
    {
      "epoch": 0.8199907450254512,
      "grad_norm": 0.9756633823952066,
      "learning_rate": 1.9098300562505266e-06,
      "loss": 0.7717,
      "step": 2215
    },
    {
      "epoch": 0.8218417399352151,
      "grad_norm": 0.9865779961666936,
      "learning_rate": 1.872003835879389e-06,
      "loss": 0.6844,
      "step": 2220
    },
    {
      "epoch": 0.8236927348449792,
      "grad_norm": 1.1283985725895185,
      "learning_rate": 1.8345172479633977e-06,
      "loss": 0.6513,
      "step": 2225
    },
    {
      "epoch": 0.8255437297547432,
      "grad_norm": 1.008863839428332,
      "learning_rate": 1.79737185889871e-06,
      "loss": 0.6812,
      "step": 2230
    },
    {
      "epoch": 0.8273947246645071,
      "grad_norm": 0.9300553415281679,
      "learning_rate": 1.7605692208242953e-06,
      "loss": 0.6909,
      "step": 2235
    },
    {
      "epoch": 0.8292457195742712,
      "grad_norm": 0.9710044483167964,
      "learning_rate": 1.7241108715571197e-06,
      "loss": 0.718,
      "step": 2240
    },
    {
      "epoch": 0.8310967144840352,
      "grad_norm": 1.015816894080233,
      "learning_rate": 1.6879983345278528e-06,
      "loss": 0.6821,
      "step": 2245
    },
    {
      "epoch": 0.8329477093937991,
      "grad_norm": 0.9542344503119925,
      "learning_rate": 1.652233118717229e-06,
      "loss": 0.719,
      "step": 2250
    },
    {
      "epoch": 0.8347987043035632,
      "grad_norm": 0.9708933601771149,
      "learning_rate": 1.6168167185929883e-06,
      "loss": 0.7041,
      "step": 2255
    },
    {
      "epoch": 0.8366496992133272,
      "grad_norm": 1.005762211417554,
      "learning_rate": 1.5817506140474248e-06,
      "loss": 0.7491,
      "step": 2260
    },
    {
      "epoch": 0.8385006941230911,
      "grad_norm": 1.1324500609372297,
      "learning_rate": 1.5470362703355678e-06,
      "loss": 0.7599,
      "step": 2265
    },
    {
      "epoch": 0.8403516890328552,
      "grad_norm": 1.0774483423911148,
      "learning_rate": 1.5126751380139203e-06,
      "loss": 0.7304,
      "step": 2270
    },
    {
      "epoch": 0.8422026839426192,
      "grad_norm": 0.9746075541108256,
      "learning_rate": 1.4786686528798878e-06,
      "loss": 0.7859,
      "step": 2275
    },
    {
      "epoch": 0.8440536788523831,
      "grad_norm": 0.9132206341810313,
      "learning_rate": 1.4450182359117493e-06,
      "loss": 0.6567,
      "step": 2280
    },
    {
      "epoch": 0.8459046737621472,
      "grad_norm": 0.9843969970761268,
      "learning_rate": 1.4117252932093007e-06,
      "loss": 0.7079,
      "step": 2285
    },
    {
      "epoch": 0.8477556686719111,
      "grad_norm": 0.909900305880051,
      "learning_rate": 1.3787912159350903e-06,
      "loss": 0.7092,
      "step": 2290
    },
    {
      "epoch": 0.8496066635816751,
      "grad_norm": 0.9862835059860517,
      "learning_rate": 1.3462173802562917e-06,
      "loss": 0.688,
      "step": 2295
    },
    {
      "epoch": 0.8514576584914392,
      "grad_norm": 0.9817682414409103,
      "learning_rate": 1.3140051472872062e-06,
      "loss": 0.7146,
      "step": 2300
    },
    {
      "epoch": 0.8514576584914392,
      "eval_loss": 0.7375185489654541,
      "eval_runtime": 125.8616,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 2300
    },
    {
      "epoch": 0.8533086534012031,
      "grad_norm": 0.97635505354809,
      "learning_rate": 1.282155863032377e-06,
      "loss": 0.6785,
      "step": 2305
    },
    {
      "epoch": 0.8551596483109671,
      "grad_norm": 0.9432652702625532,
      "learning_rate": 1.2506708583303495e-06,
      "loss": 0.6829,
      "step": 2310
    },
    {
      "epoch": 0.8570106432207312,
      "grad_norm": 1.0500105672657278,
      "learning_rate": 1.2195514487980643e-06,
      "loss": 0.7097,
      "step": 2315
    },
    {
      "epoch": 0.8588616381304951,
      "grad_norm": 0.9677123461437018,
      "learning_rate": 1.188798934775881e-06,
      "loss": 0.7271,
      "step": 2320
    },
    {
      "epoch": 0.8607126330402591,
      "grad_norm": 0.9822692297793124,
      "learning_rate": 1.158414601273251e-06,
      "loss": 0.74,
      "step": 2325
    },
    {
      "epoch": 0.8625636279500232,
      "grad_norm": 1.02828865110176,
      "learning_rate": 1.1283997179149987e-06,
      "loss": 0.7406,
      "step": 2330
    },
    {
      "epoch": 0.8644146228597871,
      "grad_norm": 1.0596259638715508,
      "learning_rate": 1.0987555388883042e-06,
      "loss": 0.6883,
      "step": 2335
    },
    {
      "epoch": 0.8662656177695511,
      "grad_norm": 0.9898076031142375,
      "learning_rate": 1.0694833028902686e-06,
      "loss": 0.7012,
      "step": 2340
    },
    {
      "epoch": 0.8681166126793152,
      "grad_norm": 0.9639230942285071,
      "learning_rate": 1.0405842330761651e-06,
      "loss": 0.7491,
      "step": 2345
    },
    {
      "epoch": 0.8699676075890791,
      "grad_norm": 0.9865850178730345,
      "learning_rate": 1.012059537008332e-06,
      "loss": 0.7295,
      "step": 2350
    },
    {
      "epoch": 0.8718186024988431,
      "grad_norm": 0.9302298993217906,
      "learning_rate": 9.839104066057025e-07,
      "loss": 0.7505,
      "step": 2355
    },
    {
      "epoch": 0.8736695974086072,
      "grad_norm": 0.9246047712950533,
      "learning_rate": 9.561380180940182e-07,
      "loss": 0.7165,
      "step": 2360
    },
    {
      "epoch": 0.8755205923183711,
      "grad_norm": 0.9341328111976939,
      "learning_rate": 9.287435319566618e-07,
      "loss": 0.6928,
      "step": 2365
    },
    {
      "epoch": 0.8773715872281351,
      "grad_norm": 1.0029919467384116,
      "learning_rate": 9.017280928861727e-07,
      "loss": 0.7503,
      "step": 2370
    },
    {
      "epoch": 0.8792225821378992,
      "grad_norm": 0.924763857124508,
      "learning_rate": 8.750928297364192e-07,
      "loss": 0.6788,
      "step": 2375
    },
    {
      "epoch": 0.8810735770476631,
      "grad_norm": 0.9276137090660882,
      "learning_rate": 8.488388554754223e-07,
      "loss": 0.6975,
      "step": 2380
    },
    {
      "epoch": 0.8829245719574271,
      "grad_norm": 0.9511441358772591,
      "learning_rate": 8.229672671388578e-07,
      "loss": 0.7379,
      "step": 2385
    },
    {
      "epoch": 0.8847755668671912,
      "grad_norm": 0.9703292177752127,
      "learning_rate": 7.974791457842012e-07,
      "loss": 0.6863,
      "step": 2390
    },
    {
      "epoch": 0.8866265617769551,
      "grad_norm": 0.9665966427045252,
      "learning_rate": 7.723755564455771e-07,
      "loss": 0.7569,
      "step": 2395
    },
    {
      "epoch": 0.8884775566867191,
      "grad_norm": 0.9199955573458448,
      "learning_rate": 7.476575480892357e-07,
      "loss": 0.7246,
      "step": 2400
    },
    {
      "epoch": 0.8884775566867191,
      "eval_loss": 0.7366934418678284,
      "eval_runtime": 125.8212,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.509,
      "step": 2400
    },
    {
      "epoch": 0.8903285515964832,
      "grad_norm": 0.9866170842794822,
      "learning_rate": 7.233261535697333e-07,
      "loss": 0.6831,
      "step": 2405
    },
    {
      "epoch": 0.8921795465062471,
      "grad_norm": 0.9020922896178015,
      "learning_rate": 6.99382389586769e-07,
      "loss": 0.7543,
      "step": 2410
    },
    {
      "epoch": 0.8940305414160111,
      "grad_norm": 0.8891310119602659,
      "learning_rate": 6.758272566427027e-07,
      "loss": 0.6992,
      "step": 2415
    },
    {
      "epoch": 0.895881536325775,
      "grad_norm": 1.0258150846316885,
      "learning_rate": 6.526617390007506e-07,
      "loss": 0.7159,
      "step": 2420
    },
    {
      "epoch": 0.8977325312355391,
      "grad_norm": 0.9315994997755779,
      "learning_rate": 6.298868046438533e-07,
      "loss": 0.7339,
      "step": 2425
    },
    {
      "epoch": 0.8995835261453031,
      "grad_norm": 0.8794790525509596,
      "learning_rate": 6.075034052342288e-07,
      "loss": 0.6877,
      "step": 2430
    },
    {
      "epoch": 0.901434521055067,
      "grad_norm": 1.0216931119406394,
      "learning_rate": 5.855124760736119e-07,
      "loss": 0.7273,
      "step": 2435
    },
    {
      "epoch": 0.9032855159648311,
      "grad_norm": 0.9162022348716465,
      "learning_rate": 5.63914936064165e-07,
      "loss": 0.6129,
      "step": 2440
    },
    {
      "epoch": 0.9051365108745951,
      "grad_norm": 1.0548470483887789,
      "learning_rate": 5.427116876700877e-07,
      "loss": 0.7462,
      "step": 2445
    },
    {
      "epoch": 0.906987505784359,
      "grad_norm": 0.9119109915842329,
      "learning_rate": 5.219036168798986e-07,
      "loss": 0.7091,
      "step": 2450
    },
    {
      "epoch": 0.9088385006941231,
      "grad_norm": 0.9773112801791846,
      "learning_rate": 5.014915931694253e-07,
      "loss": 0.7402,
      "step": 2455
    },
    {
      "epoch": 0.9106894956038871,
      "grad_norm": 0.967120980381721,
      "learning_rate": 4.81476469465465e-07,
      "loss": 0.7563,
      "step": 2460
    },
    {
      "epoch": 0.912540490513651,
      "grad_norm": 1.0900216384603985,
      "learning_rate": 4.618590821101432e-07,
      "loss": 0.7257,
      "step": 2465
    },
    {
      "epoch": 0.9143914854234151,
      "grad_norm": 0.9410281325856403,
      "learning_rate": 4.4264025082597084e-07,
      "loss": 0.726,
      "step": 2470
    },
    {
      "epoch": 0.9162424803331791,
      "grad_norm": 0.9555645068242209,
      "learning_rate": 4.2382077868159286e-07,
      "loss": 0.6977,
      "step": 2475
    },
    {
      "epoch": 0.918093475242943,
      "grad_norm": 1.1057620347138868,
      "learning_rate": 4.054014520582283e-07,
      "loss": 0.7611,
      "step": 2480
    },
    {
      "epoch": 0.9199444701527071,
      "grad_norm": 0.9023240417887544,
      "learning_rate": 3.8738304061681107e-07,
      "loss": 0.7097,
      "step": 2485
    },
    {
      "epoch": 0.9217954650624711,
      "grad_norm": 0.8961027836341736,
      "learning_rate": 3.6976629726583115e-07,
      "loss": 0.7533,
      "step": 2490
    },
    {
      "epoch": 0.923646459972235,
      "grad_norm": 0.9463370594210976,
      "learning_rate": 3.525519581298731e-07,
      "loss": 0.7657,
      "step": 2495
    },
    {
      "epoch": 0.9254974548819991,
      "grad_norm": 0.8857551112046717,
      "learning_rate": 3.357407425188541e-07,
      "loss": 0.6637,
      "step": 2500
    },
    {
      "epoch": 0.9254974548819991,
      "eval_loss": 0.736148476600647,
      "eval_runtime": 125.8161,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.509,
      "step": 2500
    },
    {
      "epoch": 0.9273484497917631,
      "grad_norm": 0.9557036435732803,
      "learning_rate": 3.1933335289797364e-07,
      "loss": 0.7211,
      "step": 2505
    },
    {
      "epoch": 0.929199444701527,
      "grad_norm": 0.9411968516873396,
      "learning_rate": 3.033304748583543e-07,
      "loss": 0.6711,
      "step": 2510
    },
    {
      "epoch": 0.9310504396112911,
      "grad_norm": 0.9529612292006595,
      "learning_rate": 2.877327770883964e-07,
      "loss": 0.7352,
      "step": 2515
    },
    {
      "epoch": 0.9329014345210551,
      "grad_norm": 0.9072504944314087,
      "learning_rate": 2.7254091134583995e-07,
      "loss": 0.702,
      "step": 2520
    },
    {
      "epoch": 0.934752429430819,
      "grad_norm": 1.0216263771503715,
      "learning_rate": 2.577555124305209e-07,
      "loss": 0.6984,
      "step": 2525
    },
    {
      "epoch": 0.9366034243405831,
      "grad_norm": 0.9271796888513735,
      "learning_rate": 2.433771981578581e-07,
      "loss": 0.7594,
      "step": 2530
    },
    {
      "epoch": 0.9384544192503471,
      "grad_norm": 0.9325885923721386,
      "learning_rate": 2.2940656933302607e-07,
      "loss": 0.721,
      "step": 2535
    },
    {
      "epoch": 0.940305414160111,
      "grad_norm": 1.0530858890236963,
      "learning_rate": 2.1584420972586174e-07,
      "loss": 0.7108,
      "step": 2540
    },
    {
      "epoch": 0.9421564090698751,
      "grad_norm": 1.9244924879459369,
      "learning_rate": 2.0269068604646058e-07,
      "loss": 0.7149,
      "step": 2545
    },
    {
      "epoch": 0.944007403979639,
      "grad_norm": 1.0361671451409022,
      "learning_rate": 1.8994654792150125e-07,
      "loss": 0.7285,
      "step": 2550
    },
    {
      "epoch": 0.945858398889403,
      "grad_norm": 0.9757871793671443,
      "learning_rate": 1.7761232787127936e-07,
      "loss": 0.6992,
      "step": 2555
    },
    {
      "epoch": 0.9477093937991671,
      "grad_norm": 0.9675283656651317,
      "learning_rate": 1.6568854128745537e-07,
      "loss": 0.7044,
      "step": 2560
    },
    {
      "epoch": 0.949560388708931,
      "grad_norm": 0.9846806445819078,
      "learning_rate": 1.5417568641151848e-07,
      "loss": 0.6704,
      "step": 2565
    },
    {
      "epoch": 0.951411383618695,
      "grad_norm": 0.9281209052786342,
      "learning_rate": 1.4307424431396654e-07,
      "loss": 0.7176,
      "step": 2570
    },
    {
      "epoch": 0.9532623785284591,
      "grad_norm": 0.8999935681059046,
      "learning_rate": 1.323846788742078e-07,
      "loss": 0.7003,
      "step": 2575
    },
    {
      "epoch": 0.955113373438223,
      "grad_norm": 0.9213878579995839,
      "learning_rate": 1.2210743676117188e-07,
      "loss": 0.6906,
      "step": 2580
    },
    {
      "epoch": 0.956964368347987,
      "grad_norm": 0.9250556197998173,
      "learning_rate": 1.1224294741464914e-07,
      "loss": 0.6794,
      "step": 2585
    },
    {
      "epoch": 0.9588153632577511,
      "grad_norm": 0.9221148649933307,
      "learning_rate": 1.0279162302734624e-07,
      "loss": 0.6645,
      "step": 2590
    },
    {
      "epoch": 0.960666358167515,
      "grad_norm": 0.9308662826828088,
      "learning_rate": 9.375385852765983e-08,
      "loss": 0.7139,
      "step": 2595
    },
    {
      "epoch": 0.962517353077279,
      "grad_norm": 0.9640193787486492,
      "learning_rate": 8.513003156317978e-08,
      "loss": 0.7416,
      "step": 2600
    },
    {
      "epoch": 0.962517353077279,
      "eval_loss": 0.7359679937362671,
      "eval_runtime": 125.8349,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.509,
      "step": 2600
    },
    {
      "epoch": 0.9643683479870431,
      "grad_norm": 0.9534219021183425,
      "learning_rate": 7.692050248490291e-08,
      "loss": 0.7339,
      "step": 2605
    },
    {
      "epoch": 0.966219342896807,
      "grad_norm": 0.9352372053413327,
      "learning_rate": 6.912561433217946e-08,
      "loss": 0.7221,
      "step": 2610
    },
    {
      "epoch": 0.968070337806571,
      "grad_norm": 0.9840909103193446,
      "learning_rate": 6.174569281837573e-08,
      "loss": 0.7012,
      "step": 2615
    },
    {
      "epoch": 0.9699213327163351,
      "grad_norm": 0.9413628052352356,
      "learning_rate": 5.4781046317267103e-08,
      "loss": 0.7021,
      "step": 2620
    },
    {
      "epoch": 0.971772327626099,
      "grad_norm": 0.9862336464985173,
      "learning_rate": 4.823196585015066e-08,
      "loss": 0.6618,
      "step": 2625
    },
    {
      "epoch": 0.973623322535863,
      "grad_norm": 0.9484184901249422,
      "learning_rate": 4.209872507368706e-08,
      "loss": 0.6719,
      "step": 2630
    },
    {
      "epoch": 0.9754743174456271,
      "grad_norm": 1.0274897071656908,
      "learning_rate": 3.6381580268463056e-08,
      "loss": 0.7218,
      "step": 2635
    },
    {
      "epoch": 0.977325312355391,
      "grad_norm": 0.9664338709751774,
      "learning_rate": 3.108077032828116e-08,
      "loss": 0.6997,
      "step": 2640
    },
    {
      "epoch": 0.979176307265155,
      "grad_norm": 1.1238393315125719,
      "learning_rate": 2.6196516750183198e-08,
      "loss": 0.7338,
      "step": 2645
    },
    {
      "epoch": 0.9810273021749191,
      "grad_norm": 0.8933118725313647,
      "learning_rate": 2.1729023625189916e-08,
      "loss": 0.7035,
      "step": 2650
    },
    {
      "epoch": 0.982878297084683,
      "grad_norm": 0.9704540461821111,
      "learning_rate": 1.767847762977337e-08,
      "loss": 0.7668,
      "step": 2655
    },
    {
      "epoch": 0.984729291994447,
      "grad_norm": 0.9466080459329079,
      "learning_rate": 1.4045048018059837e-08,
      "loss": 0.6771,
      "step": 2660
    },
    {
      "epoch": 0.9865802869042111,
      "grad_norm": 0.9644726251473568,
      "learning_rate": 1.0828886614754342e-08,
      "loss": 0.7539,
      "step": 2665
    },
    {
      "epoch": 0.988431281813975,
      "grad_norm": 0.8818176818424777,
      "learning_rate": 8.030127808797972e-09,
      "loss": 0.719,
      "step": 2670
    },
    {
      "epoch": 0.990282276723739,
      "grad_norm": 0.8961966562928385,
      "learning_rate": 5.648888547750142e-09,
      "loss": 0.697,
      "step": 2675
    },
    {
      "epoch": 0.992133271633503,
      "grad_norm": 1.2004341644367225,
      "learning_rate": 3.6852683329058336e-09,
      "loss": 0.7457,
      "step": 2680
    },
    {
      "epoch": 0.993984266543267,
      "grad_norm": 0.9729347850621183,
      "learning_rate": 2.1393492151333684e-09,
      "loss": 0.6743,
      "step": 2685
    },
    {
      "epoch": 0.995835261453031,
      "grad_norm": 0.9956973839014348,
      "learning_rate": 1.0111957914515914e-09,
      "loss": 0.6976,
      "step": 2690
    },
    {
      "epoch": 0.997686256362795,
      "grad_norm": 0.9334887689919787,
      "learning_rate": 3.008552023242572e-10,
      "loss": 0.7241,
      "step": 2695
    },
    {
      "epoch": 0.999537251272559,
      "grad_norm": 0.9781490083976876,
      "learning_rate": 8.357129693825628e-12,
      "loss": 0.7024,
      "step": 2700
    },
    {
      "epoch": 0.999537251272559,
      "eval_loss": 0.7359436750411987,
      "eval_runtime": 125.9053,
      "eval_samples_per_second": 1.017,
      "eval_steps_per_second": 0.508,
      "step": 2700
    },
    {
      "epoch": 0.9999074502545118,
      "step": 2701,
      "total_flos": 76938190258176.0,
      "train_loss": 0.7557382643200741,
      "train_runtime": 83163.2816,
      "train_samples_per_second": 0.26,
      "train_steps_per_second": 0.032
    }
  ],
  "logging_steps": 5,
  "max_steps": 2701,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 76938190258176.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}