{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 625,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0016,
      "grad_norm": 1.5717865228652954,
      "learning_rate": 3.1746031746031746e-06,
      "loss": 2.7077,
      "mean_token_accuracy": 0.459680899977684,
      "step": 1
    },
    {
      "epoch": 0.008,
      "grad_norm": 1.202968716621399,
      "learning_rate": 1.5873015873015872e-05,
      "loss": 2.2671,
      "mean_token_accuracy": 0.5242175050079823,
      "step": 5
    },
    {
      "epoch": 0.016,
      "grad_norm": 1.0542172193527222,
      "learning_rate": 3.1746031746031745e-05,
      "loss": 2.2698,
      "mean_token_accuracy": 0.5283941477537155,
      "step": 10
    },
    {
      "epoch": 0.024,
      "grad_norm": 0.7223343253135681,
      "learning_rate": 4.761904761904762e-05,
      "loss": 2.1156,
      "mean_token_accuracy": 0.5514968127012253,
      "step": 15
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.6670629382133484,
      "learning_rate": 6.349206349206349e-05,
      "loss": 1.8644,
      "mean_token_accuracy": 0.5947718620300293,
      "step": 20
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8882656693458557,
      "learning_rate": 7.936507936507937e-05,
      "loss": 1.6563,
      "mean_token_accuracy": 0.6210890173912048,
      "step": 25
    },
    {
      "epoch": 0.048,
      "grad_norm": 1.0315715074539185,
      "learning_rate": 9.523809523809524e-05,
      "loss": 1.4437,
      "mean_token_accuracy": 0.654322224855423,
      "step": 30
    },
    {
      "epoch": 0.056,
      "grad_norm": 1.4416899681091309,
      "learning_rate": 0.00011111111111111112,
      "loss": 1.0037,
      "mean_token_accuracy": 0.7415849208831787,
      "step": 35
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.8182677626609802,
      "learning_rate": 0.00012698412698412698,
      "loss": 0.6626,
      "mean_token_accuracy": 0.853333032131195,
      "step": 40
    },
    {
      "epoch": 0.072,
      "grad_norm": 0.4539111256599426,
      "learning_rate": 0.00014285714285714287,
      "loss": 0.4946,
      "mean_token_accuracy": 0.8977023303508759,
      "step": 45
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.3059958815574646,
      "learning_rate": 0.00015873015873015873,
      "loss": 0.4924,
      "mean_token_accuracy": 0.8881445944309234,
      "step": 50
    },
    {
      "epoch": 0.088,
      "grad_norm": 0.20895916223526,
      "learning_rate": 0.00017460317460317462,
      "loss": 0.3816,
      "mean_token_accuracy": 0.9159973561763763,
      "step": 55
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.241449773311615,
      "learning_rate": 0.00019047619047619048,
      "loss": 0.2424,
      "mean_token_accuracy": 0.9461545169353485,
      "step": 60
    },
    {
      "epoch": 0.104,
      "grad_norm": 0.13886623084545135,
      "learning_rate": 0.00019999375039475277,
      "loss": 0.2525,
      "mean_token_accuracy": 0.9451982796192169,
      "step": 65
    },
    {
      "epoch": 0.112,
      "grad_norm": 0.15808190405368805,
      "learning_rate": 0.0001999234513064475,
      "loss": 0.3108,
      "mean_token_accuracy": 0.9311837553977966,
      "step": 70
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.1368260234594345,
      "learning_rate": 0.00019977509622105233,
      "loss": 0.2217,
      "mean_token_accuracy": 0.9483723938465118,
      "step": 75
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.14972065389156342,
      "learning_rate": 0.0001995488010273198,
      "loss": 0.1707,
      "mean_token_accuracy": 0.9621350407600403,
      "step": 80
    },
    {
      "epoch": 0.136,
      "grad_norm": 0.2032196819782257,
      "learning_rate": 0.00019924474249753655,
      "loss": 0.3796,
      "mean_token_accuracy": 0.9232820510864258,
      "step": 85
    },
    {
      "epoch": 0.144,
      "grad_norm": 0.09374168515205383,
      "learning_rate": 0.00019886315814943647,
      "loss": 0.3032,
      "mean_token_accuracy": 0.9280714929103852,
      "step": 90
    },
    {
      "epoch": 0.152,
      "grad_norm": 0.1895170956850052,
      "learning_rate": 0.0001984043460606618,
      "loss": 0.2635,
      "mean_token_accuracy": 0.9419675827026367,
      "step": 95
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.17207027971744537,
      "learning_rate": 0.0001978686646359173,
      "loss": 0.2961,
      "mean_token_accuracy": 0.9313458263874054,
      "step": 100
    },
    {
      "epoch": 0.168,
      "grad_norm": 0.1757289320230484,
      "learning_rate": 0.0001972565323269996,
      "loss": 0.2087,
      "mean_token_accuracy": 0.9513824045658111,
      "step": 105
    },
    {
      "epoch": 0.176,
      "grad_norm": 0.2260744869709015,
      "learning_rate": 0.00019656842730592046,
      "loss": 0.3184,
      "mean_token_accuracy": 0.9232833862304688,
      "step": 110
    },
    {
      "epoch": 0.184,
      "grad_norm": 0.052938517183065414,
      "learning_rate": 0.0001958048870913786,
      "loss": 0.1822,
      "mean_token_accuracy": 0.9538747906684876,
      "step": 115
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.1941942721605301,
      "learning_rate": 0.0001949665081288729,
      "loss": 0.2273,
      "mean_token_accuracy": 0.9453578472137452,
      "step": 120
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.2198331207036972,
      "learning_rate": 0.00019405394532478424,
      "loss": 0.2049,
      "mean_token_accuracy": 0.9542803287506103,
      "step": 125
    },
    {
      "epoch": 0.208,
      "grad_norm": 0.24331390857696533,
      "learning_rate": 0.00019306791153479006,
      "loss": 0.2061,
      "mean_token_accuracy": 0.9513079404830933,
      "step": 130
    },
    {
      "epoch": 0.216,
      "grad_norm": 0.2335432767868042,
      "learning_rate": 0.00019200917700701176,
      "loss": 0.1368,
      "mean_token_accuracy": 0.9651266932487488,
      "step": 135
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.23610533773899078,
      "learning_rate": 0.0001908785687803289,
      "loss": 0.3442,
      "mean_token_accuracy": 0.9188908874988556,
      "step": 140
    },
    {
      "epoch": 0.232,
      "grad_norm": 0.22944550216197968,
      "learning_rate": 0.00018967697003833157,
      "loss": 0.337,
      "mean_token_accuracy": 0.9214253067970276,
      "step": 145
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.21961958706378937,
      "learning_rate": 0.0001884053194194142,
      "loss": 0.223,
      "mean_token_accuracy": 0.9437191367149353,
      "step": 150
    },
    {
      "epoch": 0.248,
      "grad_norm": 0.20730167627334595,
      "learning_rate": 0.00018706461028355104,
      "loss": 0.2741,
      "mean_token_accuracy": 0.938022255897522,
      "step": 155
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.16260486841201782,
      "learning_rate": 0.00018565588993632487,
      "loss": 0.2648,
      "mean_token_accuracy": 0.9357350587844848,
      "step": 160
    },
    {
      "epoch": 0.264,
      "grad_norm": 0.22049878537654877,
      "learning_rate": 0.0001841802588108161,
      "loss": 0.3412,
      "mean_token_accuracy": 0.9166976153850556,
      "step": 165
    },
    {
      "epoch": 0.272,
      "grad_norm": 0.19371896982192993,
      "learning_rate": 0.00018263886960799062,
      "loss": 0.2456,
      "mean_token_accuracy": 0.945724493265152,
      "step": 170
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.22933930158615112,
      "learning_rate": 0.00018103292639625837,
      "loss": 0.3841,
      "mean_token_accuracy": 0.9089005470275879,
      "step": 175
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.1793544441461563,
      "learning_rate": 0.0001793636836709057,
      "loss": 0.2999,
      "mean_token_accuracy": 0.9297621667385101,
      "step": 180
    },
    {
      "epoch": 0.296,
      "grad_norm": 0.0395968072116375,
      "learning_rate": 0.0001776324453741365,
      "loss": 0.1624,
      "mean_token_accuracy": 0.9584607481956482,
      "step": 185
    },
    {
      "epoch": 0.304,
      "grad_norm": 0.2473573386669159,
      "learning_rate": 0.00017584056387648727,
      "loss": 0.3053,
      "mean_token_accuracy": 0.9382229685783386,
      "step": 190
    },
    {
      "epoch": 0.312,
      "grad_norm": 0.04702218249440193,
      "learning_rate": 0.0001739894389204122,
      "loss": 0.172,
      "mean_token_accuracy": 0.9594997644424439,
      "step": 195
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.23656697571277618,
      "learning_rate": 0.00017208051652686335,
      "loss": 0.2276,
      "mean_token_accuracy": 0.9398577272891998,
      "step": 200
    },
    {
      "epoch": 0.328,
      "grad_norm": 0.07044972479343414,
      "learning_rate": 0.00017011528786571969,
      "loss": 0.2854,
      "mean_token_accuracy": 0.926360809803009,
      "step": 205
    },
    {
      "epoch": 0.336,
      "grad_norm": 0.03222399577498436,
      "learning_rate": 0.00016809528809094807,
      "loss": 0.2172,
      "mean_token_accuracy": 0.9438831508159637,
      "step": 210
    },
    {
      "epoch": 0.344,
      "grad_norm": 0.243390753865242,
      "learning_rate": 0.0001660220951414055,
      "loss": 0.2852,
      "mean_token_accuracy": 0.9389818787574769,
      "step": 215
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.23130089044570923,
      "learning_rate": 0.00016389732850821966,
      "loss": 0.2899,
      "mean_token_accuracy": 0.9331403017044068,
      "step": 220
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.2532561123371124,
      "learning_rate": 0.0001617226479697105,
      "loss": 0.3069,
      "mean_token_accuracy": 0.9253335416316986,
      "step": 225
    },
    {
      "epoch": 0.368,
      "grad_norm": 0.20991413295269012,
      "learning_rate": 0.00015949975229484134,
      "loss": 0.3178,
      "mean_token_accuracy": 0.9201959311962128,
      "step": 230
    },
    {
      "epoch": 0.376,
      "grad_norm": 0.20208591222763062,
      "learning_rate": 0.00015723037791621193,
      "loss": 0.18,
      "mean_token_accuracy": 0.9534633576869964,
      "step": 235
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.23704850673675537,
      "learning_rate": 0.00015491629757363032,
      "loss": 0.3864,
      "mean_token_accuracy": 0.9041174411773681,
      "step": 240
    },
    {
      "epoch": 0.392,
      "grad_norm": 0.22486881911754608,
      "learning_rate": 0.00015255931892932333,
      "loss": 0.3174,
      "mean_token_accuracy": 0.9177540898323059,
      "step": 245
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.2111813724040985,
      "learning_rate": 0.0001501612831558664,
      "loss": 0.3145,
      "mean_token_accuracy": 0.9238011240959167,
      "step": 250
    },
    {
      "epoch": 0.408,
      "grad_norm": 0.23434004187583923,
      "learning_rate": 0.00014772406349793744,
      "loss": 0.2928,
      "mean_token_accuracy": 0.929435807466507,
      "step": 255
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.25533345341682434,
      "learning_rate": 0.0001452495638090167,
      "loss": 0.3171,
      "mean_token_accuracy": 0.927526468038559,
      "step": 260
    },
    {
      "epoch": 0.424,
      "grad_norm": 0.1922113448381424,
      "learning_rate": 0.00014273971706417647,
      "loss": 0.2162,
      "mean_token_accuracy": 0.9465757727622985,
      "step": 265
    },
    {
      "epoch": 0.432,
      "grad_norm": 0.2513563334941864,
      "learning_rate": 0.00014019648385012244,
      "loss": 0.3813,
      "mean_token_accuracy": 0.908132141828537,
      "step": 270
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.25622016191482544,
      "learning_rate": 0.00013762185083366556,
      "loss": 0.3005,
      "mean_token_accuracy": 0.9310987770557404,
      "step": 275
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.20694169402122498,
      "learning_rate": 0.00013501782920982184,
      "loss": 0.1721,
      "mean_token_accuracy": 0.9563108265399933,
      "step": 280
    },
    {
      "epoch": 0.456,
      "grad_norm": 0.21768316626548767,
      "learning_rate": 0.00013238645313075104,
      "loss": 0.2444,
      "mean_token_accuracy": 0.9432082533836365,
      "step": 285
    },
    {
      "epoch": 0.464,
      "grad_norm": 0.1706106811761856,
      "learning_rate": 0.00012972977811676287,
      "loss": 0.233,
      "mean_token_accuracy": 0.9396327018737793,
      "step": 290
    },
    {
      "epoch": 0.472,
      "grad_norm": 0.2342345267534256,
      "learning_rate": 0.00012704987945063068,
      "loss": 0.2106,
      "mean_token_accuracy": 0.9517422139644622,
      "step": 295
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2374894917011261,
      "learning_rate": 0.00012434885055646823,
      "loss": 0.2222,
      "mean_token_accuracy": 0.9430564045906067,
      "step": 300
    },
    {
      "epoch": 0.488,
      "grad_norm": 0.20393089950084686,
      "learning_rate": 0.00012162880136443447,
      "loss": 0.3069,
      "mean_token_accuracy": 0.9208929121494294,
      "step": 305
    },
    {
      "epoch": 0.496,
      "grad_norm": 0.2132725864648819,
      "learning_rate": 0.00011889185666254506,
      "loss": 0.2821,
      "mean_token_accuracy": 0.9318405151367187,
      "step": 310
    },
    {
      "epoch": 0.504,
      "grad_norm": 0.15664634108543396,
      "learning_rate": 0.00011614015443687722,
      "loss": 0.16,
      "mean_token_accuracy": 0.9559766769409179,
      "step": 315
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.22459107637405396,
      "learning_rate": 0.0001133758442014651,
      "loss": 0.1543,
      "mean_token_accuracy": 0.9638924479484559,
      "step": 320
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.2619342505931854,
      "learning_rate": 0.00011060108531918971,
      "loss": 0.2865,
      "mean_token_accuracy": 0.9385590136051178,
      "step": 325
    },
    {
      "epoch": 0.528,
      "grad_norm": 0.21467000246047974,
      "learning_rate": 0.0001078180453149754,
      "loss": 0.2918,
      "mean_token_accuracy": 0.9376808702945709,
      "step": 330
    },
    {
      "epoch": 0.536,
      "grad_norm": 0.2496771514415741,
      "learning_rate": 0.00010502889818261075,
      "loss": 0.4184,
      "mean_token_accuracy": 0.8969966411590576,
      "step": 335
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.2137787938117981,
      "learning_rate": 0.00010223582268651586,
      "loss": 0.257,
      "mean_token_accuracy": 0.9423645496368408,
      "step": 340
    },
    {
      "epoch": 0.552,
      "grad_norm": 0.2176610231399536,
      "learning_rate": 9.94410006597835e-05,
      "loss": 0.3322,
      "mean_token_accuracy": 0.9129085004329681,
      "step": 345
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.21382386982440948,
      "learning_rate": 9.66466152998226e-05,
      "loss": 0.3399,
      "mean_token_accuracy": 0.9135319530963898,
      "step": 350
    },
    {
      "epoch": 0.568,
      "grad_norm": 0.22343839704990387,
      "learning_rate": 9.385484946293637e-05,
      "loss": 0.2464,
      "mean_token_accuracy": 0.9397193789482117,
      "step": 355
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.18257169425487518,
      "learning_rate": 9.106788395916678e-05,
      "loss": 0.1896,
      "mean_token_accuracy": 0.951214075088501,
      "step": 360
    },
    {
      "epoch": 0.584,
      "grad_norm": 0.04507756605744362,
      "learning_rate": 8.828789584873754e-05,
      "loss": 0.2693,
      "mean_token_accuracy": 0.9345629513263702,
      "step": 365
    },
    {
      "epoch": 0.592,
      "grad_norm": 0.23864690959453583,
      "learning_rate": 8.551705674142617e-05,
      "loss": 0.2743,
      "mean_token_accuracy": 0.9325626492500305,
      "step": 370
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.035314541310071945,
      "learning_rate": 8.275753110019367e-05,
      "loss": 0.1799,
      "mean_token_accuracy": 0.9588949978351593,
      "step": 375
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.19131699204444885,
      "learning_rate": 8.001147455039735e-05,
      "loss": 0.2339,
      "mean_token_accuracy": 0.9403933227062226,
      "step": 380
    },
    {
      "epoch": 0.616,
      "grad_norm": 0.20205603539943695,
      "learning_rate": 7.728103219590681e-05,
      "loss": 0.2315,
      "mean_token_accuracy": 0.9407491445541382,
      "step": 385
    },
    {
      "epoch": 0.624,
      "grad_norm": 0.23396767675876617,
      "learning_rate": 7.456833694343906e-05,
      "loss": 0.2248,
      "mean_token_accuracy": 0.9434651434421539,
      "step": 390
    },
    {
      "epoch": 0.632,
      "grad_norm": 0.05149530991911888,
      "learning_rate": 7.18755078364214e-05,
      "loss": 0.1834,
      "mean_token_accuracy": 0.950703501701355,
      "step": 395
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.1885431706905365,
      "learning_rate": 6.920464839968405e-05,
      "loss": 0.2292,
      "mean_token_accuracy": 0.9455470621585846,
      "step": 400
    },
    {
      "epoch": 0.648,
      "grad_norm": 0.22768881916999817,
      "learning_rate": 6.65578449962749e-05,
      "loss": 0.3276,
      "mean_token_accuracy": 0.9167207598686218,
      "step": 405
    },
    {
      "epoch": 0.656,
      "grad_norm": 0.2584383189678192,
      "learning_rate": 6.393716519768047e-05,
      "loss": 0.2714,
      "mean_token_accuracy": 0.9378273606300354,
      "step": 410
    },
    {
      "epoch": 0.664,
      "grad_norm": 0.27143898606300354,
      "learning_rate": 6.134465616872598e-05,
      "loss": 0.3238,
      "mean_token_accuracy": 0.9266472160816193,
      "step": 415
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.16849137842655182,
      "learning_rate": 5.878234306841637e-05,
      "loss": 0.1855,
      "mean_token_accuracy": 0.9501271903514862,
      "step": 420
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.23237672448158264,
      "learning_rate": 5.62522274679673e-05,
      "loss": 0.2252,
      "mean_token_accuracy": 0.9437941908836365,
      "step": 425
    },
    {
      "epoch": 0.688,
      "grad_norm": 0.19916483759880066,
      "learning_rate": 5.375628578726181e-05,
      "loss": 0.1561,
      "mean_token_accuracy": 0.9600933134555817,
      "step": 430
    },
    {
      "epoch": 0.696,
      "grad_norm": 0.22922399640083313,
      "learning_rate": 5.1296467750954314e-05,
      "loss": 0.3147,
      "mean_token_accuracy": 0.925593638420105,
      "step": 435
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.2341531664133072,
      "learning_rate": 4.8874694865427676e-05,
      "loss": 0.2745,
      "mean_token_accuracy": 0.9313132464885712,
      "step": 440
    },
    {
      "epoch": 0.712,
      "grad_norm": 0.24038153886795044,
      "learning_rate": 4.649285891779327e-05,
      "loss": 0.2954,
      "mean_token_accuracy": 0.9240333676338196,
      "step": 445
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.2073904424905777,
      "learning_rate": 4.415282049810644e-05,
      "loss": 0.2418,
      "mean_token_accuracy": 0.93941730260849,
      "step": 450
    },
    {
      "epoch": 0.728,
      "grad_norm": 0.232681542634964,
      "learning_rate": 4.1856407545951834e-05,
      "loss": 0.3575,
      "mean_token_accuracy": 0.9102162063121796,
      "step": 455
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.20729133486747742,
      "learning_rate": 3.9605413922533874e-05,
      "loss": 0.2844,
      "mean_token_accuracy": 0.9322370827198029,
      "step": 460
    },
    {
      "epoch": 0.744,
      "grad_norm": 0.20540155470371246,
      "learning_rate": 3.740159800938784e-05,
      "loss": 0.1801,
      "mean_token_accuracy": 0.9571804940700531,
      "step": 465
    },
    {
      "epoch": 0.752,
      "grad_norm": 0.27062955498695374,
      "learning_rate": 3.5246681334806175e-05,
      "loss": 0.1628,
      "mean_token_accuracy": 0.9602212190628052,
      "step": 470
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.2563348710536957,
      "learning_rate": 3.3142347229053015e-05,
      "loss": 0.2317,
      "mean_token_accuracy": 0.9395133197307587,
      "step": 475
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.23781529068946838,
      "learning_rate": 3.109023950941736e-05,
      "loss": 0.3099,
      "mean_token_accuracy": 0.9263223648071289,
      "step": 480
    },
    {
      "epoch": 0.776,
      "grad_norm": 0.04337846860289574,
      "learning_rate": 2.909196119613218e-05,
      "loss": 0.1425,
      "mean_token_accuracy": 0.9617048621177673,
      "step": 485
    },
    {
      "epoch": 0.784,
      "grad_norm": 0.20578506588935852,
      "learning_rate": 2.7149073260162416e-05,
      "loss": 0.2547,
      "mean_token_accuracy": 0.9427634000778198,
      "step": 490
    },
    {
      "epoch": 0.792,
      "grad_norm": 0.2657724618911743,
      "learning_rate": 2.5263093403840142e-05,
      "loss": 0.2759,
      "mean_token_accuracy": 0.9276242256164551,
      "step": 495
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.20405757427215576,
      "learning_rate": 2.3435494875299314e-05,
      "loss": 0.1946,
      "mean_token_accuracy": 0.9477591276168823,
      "step": 500
    },
    {
      "epoch": 0.808,
      "grad_norm": 0.2120545208454132,
      "learning_rate": 2.166770531763633e-05,
      "loss": 0.1492,
      "mean_token_accuracy": 0.9617019176483155,
      "step": 505
    },
    {
      "epoch": 0.816,
      "grad_norm": 0.1790059208869934,
      "learning_rate": 1.9961105653695266e-05,
      "loss": 0.2097,
      "mean_token_accuracy": 0.9475365877151489,
      "step": 510
    },
    {
      "epoch": 0.824,
      "grad_norm": 0.23069578409194946,
      "learning_rate": 1.8317029007349085e-05,
      "loss": 0.3262,
      "mean_token_accuracy": 0.9245599567890167,
      "step": 515
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.1952206939458847,
      "learning_rate": 1.6736759662119183e-05,
      "loss": 0.2593,
      "mean_token_accuracy": 0.9349613726139069,
      "step": 520
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.16667886078357697,
      "learning_rate": 1.5221532057947419e-05,
      "loss": 0.1832,
      "mean_token_accuracy": 0.9498234987258911,
      "step": 525
    },
    {
      "epoch": 0.848,
      "grad_norm": 0.16531656682491302,
      "learning_rate": 1.3772529826903269e-05,
      "loss": 0.3003,
      "mean_token_accuracy": 0.927387660741806,
      "step": 530
    },
    {
      "epoch": 0.856,
      "grad_norm": 0.03590136766433716,
      "learning_rate": 1.23908848685804e-05,
      "loss": 0.209,
      "mean_token_accuracy": 0.9517148792743683,
      "step": 535
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.20711013674736023,
      "learning_rate": 1.1077676465904208e-05,
      "loss": 0.229,
      "mean_token_accuracy": 0.9443170964717865,
      "step": 540
    },
    {
      "epoch": 0.872,
      "grad_norm": 0.2173730880022049,
      "learning_rate": 9.833930442041506e-06,
      "loss": 0.1711,
      "mean_token_accuracy": 0.95446497797966,
      "step": 545
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.25314000248908997,
      "learning_rate": 8.660618359070604e-06,
      "loss": 0.2478,
      "mean_token_accuracy": 0.9370902240276336,
      "step": 550
    },
    {
      "epoch": 0.888,
      "grad_norm": 0.21840238571166992,
      "learning_rate": 7.558656759037797e-06,
      "loss": 0.184,
      "mean_token_accuracy": 0.9521986067295074,
      "step": 555
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.21549563109874725,
      "learning_rate": 6.528906447993288e-06,
      "loss": 0.2443,
      "mean_token_accuracy": 0.9441956281661987,
      "step": 560
    },
    {
      "epoch": 0.904,
      "grad_norm": 0.1734277904033661,
      "learning_rate": 5.572171823565797e-06,
      "loss": 0.22,
      "mean_token_accuracy": 0.9439444482326508,
      "step": 565
    },
    {
      "epoch": 0.912,
      "grad_norm": 0.2154768407344818,
      "learning_rate": 4.689200246600867e-06,
      "loss": 0.1987,
      "mean_token_accuracy": 0.9493344485759735,
      "step": 570
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.1548367738723755,
      "learning_rate": 3.880681457354118e-06,
      "loss": 0.1942,
      "mean_token_accuracy": 0.9513484239578247,
      "step": 575
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.18485531210899353,
      "learning_rate": 3.1472470366950334e-06,
      "loss": 0.2624,
      "mean_token_accuracy": 0.9301692008972168,
      "step": 580
    },
    {
      "epoch": 0.936,
      "grad_norm": 0.24809648096561432,
      "learning_rate": 2.4894699127426367e-06,
      "loss": 0.2501,
      "mean_token_accuracy": 0.9349741101264953,
      "step": 585
    },
    {
      "epoch": 0.944,
      "grad_norm": 0.2469683289527893,
      "learning_rate": 1.907863913318153e-06,
      "loss": 0.1806,
      "mean_token_accuracy": 0.9534947514533997,
      "step": 590
    },
    {
      "epoch": 0.952,
      "grad_norm": 0.21619068086147308,
      "learning_rate": 1.4028833645643113e-06,
      "loss": 0.1114,
      "mean_token_accuracy": 0.9707520604133606,
      "step": 595
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.20197683572769165,
      "learning_rate": 9.749227360448143e-07,
      "loss": 0.202,
      "mean_token_accuracy": 0.947997921705246,
      "step": 600
    },
    {
      "epoch": 0.968,
      "grad_norm": 0.21867972612380981,
      "learning_rate": 6.243163326014267e-07,
      "loss": 0.2394,
      "mean_token_accuracy": 0.9438014030456543,
      "step": 605
    },
    {
      "epoch": 0.976,
      "grad_norm": 0.2265222817659378,
      "learning_rate": 3.5133803320896994e-07,
      "loss": 0.2911,
      "mean_token_accuracy": 0.9275119185447693,
      "step": 610
    },
    {
      "epoch": 0.984,
      "grad_norm": 0.19685231149196625,
      "learning_rate": 1.562010770326916e-07,
      "loss": 0.2725,
      "mean_token_accuracy": 0.9320066690444946,
      "step": 615
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.15336309373378754,
      "learning_rate": 3.905789685471062e-08,
      "loss": 0.3007,
      "mean_token_accuracy": 0.9265518903732299,
      "step": 620
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.23042897880077362,
      "learning_rate": 0.0,
      "loss": 0.2386,
      "mean_token_accuracy": 0.9401108622550964,
      "step": 625
    },
    {
      "epoch": 1.0,
      "step": 625,
      "total_flos": 1.5047606448160768e+16,
      "train_loss": 0.3481092903137207,
      "train_runtime": 972.7733,
      "train_samples_per_second": 5.14,
      "train_steps_per_second": 0.642
    }
  ],
  "logging_steps": 5,
  "max_steps": 625,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5047606448160768e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}