{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 1000,
  "global_step": 9753,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03076804738279297,
      "grad_norm": 2.239741325378418,
      "learning_rate": 4.85e-05,
      "loss": 2.6548,
      "step": 100
    },
    {
      "epoch": 0.06153609476558594,
      "grad_norm": 1.6382789611816406,
      "learning_rate": 4.94975655236714e-05,
      "loss": 2.3258,
      "step": 200
    },
    {
      "epoch": 0.09230414214837891,
      "grad_norm": 1.4356640577316284,
      "learning_rate": 4.89795918367347e-05,
      "loss": 2.2618,
      "step": 300
    },
    {
      "epoch": 0.12307218953117188,
      "grad_norm": 1.5954153537750244,
      "learning_rate": 4.846161814979799e-05,
      "loss": 2.2213,
      "step": 400
    },
    {
      "epoch": 0.15384023691396484,
      "grad_norm": 1.5558085441589355,
      "learning_rate": 4.794364446286129e-05,
      "loss": 2.2169,
      "step": 500
    },
    {
      "epoch": 0.18460828429675782,
      "grad_norm": 1.3300689458847046,
      "learning_rate": 4.7425670775924586e-05,
      "loss": 2.1835,
      "step": 600
    },
    {
      "epoch": 0.21537633167955078,
      "grad_norm": 1.5879923105239868,
      "learning_rate": 4.690769708898788e-05,
      "loss": 2.1786,
      "step": 700
    },
    {
      "epoch": 0.24614437906234377,
      "grad_norm": 1.4449992179870605,
      "learning_rate": 4.638972340205118e-05,
      "loss": 2.17,
      "step": 800
    },
    {
      "epoch": 0.27691242644513675,
      "grad_norm": 1.660819411277771,
      "learning_rate": 4.5871749715114475e-05,
      "loss": 2.1397,
      "step": 900
    },
    {
      "epoch": 0.3076804738279297,
      "grad_norm": 1.3502899408340454,
      "learning_rate": 4.535377602817777e-05,
      "loss": 2.1476,
      "step": 1000
    },
    {
      "epoch": 0.3076804738279297,
      "eval_loss": 2.0527570247650146,
      "eval_runtime": 4.7056,
      "eval_samples_per_second": 42.502,
      "eval_steps_per_second": 21.251,
      "step": 1000
    },
    {
      "epoch": 0.33844852121072266,
      "grad_norm": 1.5296121835708618,
      "learning_rate": 4.483580234124107e-05,
      "loss": 2.1347,
      "step": 1100
    },
    {
      "epoch": 0.36921656859351565,
      "grad_norm": 1.4386141300201416,
      "learning_rate": 4.4317828654304364e-05,
      "loss": 2.1128,
      "step": 1200
    },
    {
      "epoch": 0.3999846159763086,
      "grad_norm": 1.7359758615493774,
      "learning_rate": 4.379985496736766e-05,
      "loss": 2.1256,
      "step": 1300
    },
    {
      "epoch": 0.43075266335910156,
      "grad_norm": 1.2721539735794067,
      "learning_rate": 4.3281881280430956e-05,
      "loss": 2.0986,
      "step": 1400
    },
    {
      "epoch": 0.46152071074189455,
      "grad_norm": 1.507233738899231,
      "learning_rate": 4.276390759349425e-05,
      "loss": 2.1203,
      "step": 1500
    },
    {
      "epoch": 0.49228875812468753,
      "grad_norm": 1.5739669799804688,
      "learning_rate": 4.224593390655755e-05,
      "loss": 2.0817,
      "step": 1600
    },
    {
      "epoch": 0.5230568055074805,
      "grad_norm": 1.7889518737792969,
      "learning_rate": 4.1727960219620845e-05,
      "loss": 2.0951,
      "step": 1700
    },
    {
      "epoch": 0.5538248528902735,
      "grad_norm": 1.3344364166259766,
      "learning_rate": 4.120998653268414e-05,
      "loss": 2.0856,
      "step": 1800
    },
    {
      "epoch": 0.5845929002730664,
      "grad_norm": 1.4601390361785889,
      "learning_rate": 4.069201284574744e-05,
      "loss": 2.0758,
      "step": 1900
    },
    {
      "epoch": 0.6153609476558594,
      "grad_norm": 1.1662238836288452,
      "learning_rate": 4.0174039158810734e-05,
      "loss": 2.0747,
      "step": 2000
    },
    {
      "epoch": 0.6153609476558594,
      "eval_loss": 1.9568997621536255,
      "eval_runtime": 4.6999,
      "eval_samples_per_second": 42.554,
      "eval_steps_per_second": 21.277,
      "step": 2000
    },
    {
      "epoch": 0.6461289950386524,
      "grad_norm": 1.5872595310211182,
      "learning_rate": 3.965606547187403e-05,
      "loss": 2.0822,
      "step": 2100
    },
    {
      "epoch": 0.6768970424214453,
      "grad_norm": 1.1704156398773193,
      "learning_rate": 3.913809178493733e-05,
      "loss": 2.0962,
      "step": 2200
    },
    {
      "epoch": 0.7076650898042383,
      "grad_norm": 1.6314693689346313,
      "learning_rate": 3.8620118098000623e-05,
      "loss": 2.0639,
      "step": 2300
    },
    {
      "epoch": 0.7384331371870313,
      "grad_norm": 1.4110997915267944,
      "learning_rate": 3.810214441106392e-05,
      "loss": 2.0581,
      "step": 2400
    },
    {
      "epoch": 0.7692011845698242,
      "grad_norm": 1.2487901449203491,
      "learning_rate": 3.7584170724127216e-05,
      "loss": 2.0588,
      "step": 2500
    },
    {
      "epoch": 0.7999692319526172,
      "grad_norm": 1.5325045585632324,
      "learning_rate": 3.706619703719051e-05,
      "loss": 2.0731,
      "step": 2600
    },
    {
      "epoch": 0.8307372793354102,
      "grad_norm": 1.2173811197280884,
      "learning_rate": 3.654822335025381e-05,
      "loss": 2.0459,
      "step": 2700
    },
    {
      "epoch": 0.8615053267182031,
      "grad_norm": 1.6108065843582153,
      "learning_rate": 3.6030249663317105e-05,
      "loss": 2.0776,
      "step": 2800
    },
    {
      "epoch": 0.8922733741009962,
      "grad_norm": 1.7993882894515991,
      "learning_rate": 3.55122759763804e-05,
      "loss": 2.0391,
      "step": 2900
    },
    {
      "epoch": 0.9230414214837891,
      "grad_norm": 1.4659514427185059,
      "learning_rate": 3.49943022894437e-05,
      "loss": 2.0292,
      "step": 3000
    },
    {
      "epoch": 0.9230414214837891,
      "eval_loss": 1.8916079998016357,
      "eval_runtime": 4.8218,
      "eval_samples_per_second": 41.478,
      "eval_steps_per_second": 20.739,
      "step": 3000
    },
    {
      "epoch": 0.953809468866582,
      "grad_norm": 1.5378167629241943,
      "learning_rate": 3.4476328602506994e-05,
      "loss": 2.0525,
      "step": 3100
    },
    {
      "epoch": 0.9845775162493751,
      "grad_norm": 1.4396274089813232,
      "learning_rate": 3.395835491557029e-05,
      "loss": 2.0479,
      "step": 3200
    },
    {
      "epoch": 1.0150763432175685,
      "grad_norm": 1.1797999143600464,
      "learning_rate": 3.344038122863359e-05,
      "loss": 1.9838,
      "step": 3300
    },
    {
      "epoch": 1.0458443906003616,
      "grad_norm": 1.2820570468902588,
      "learning_rate": 3.292240754169688e-05,
      "loss": 1.9482,
      "step": 3400
    },
    {
      "epoch": 1.0766124379831545,
      "grad_norm": 1.4676374197006226,
      "learning_rate": 3.240443385476018e-05,
      "loss": 1.9538,
      "step": 3500
    },
    {
      "epoch": 1.1073804853659475,
      "grad_norm": 1.5519450902938843,
      "learning_rate": 3.1886460167823476e-05,
      "loss": 1.9727,
      "step": 3600
    },
    {
      "epoch": 1.1381485327487404,
      "grad_norm": 1.3620473146438599,
      "learning_rate": 3.136848648088677e-05,
      "loss": 1.96,
      "step": 3700
    },
    {
      "epoch": 1.1689165801315333,
      "grad_norm": 1.3301316499710083,
      "learning_rate": 3.085051279395007e-05,
      "loss": 1.9463,
      "step": 3800
    },
    {
      "epoch": 1.1996846275143263,
      "grad_norm": 1.3053226470947266,
      "learning_rate": 3.0332539107013365e-05,
      "loss": 1.9709,
      "step": 3900
    },
    {
      "epoch": 1.2304526748971194,
      "grad_norm": 1.2289018630981445,
      "learning_rate": 2.981456542007666e-05,
      "loss": 1.9847,
      "step": 4000
    },
    {
      "epoch": 1.2304526748971194,
      "eval_loss": 1.8471519947052002,
      "eval_runtime": 4.6843,
      "eval_samples_per_second": 42.696,
      "eval_steps_per_second": 21.348,
      "step": 4000
    },
    {
      "epoch": 1.2612207222799123,
      "grad_norm": 1.319795846939087,
      "learning_rate": 2.9296591733139954e-05,
      "loss": 1.9524,
      "step": 4100
    },
    {
      "epoch": 1.2919887696627053,
      "grad_norm": 1.1395128965377808,
      "learning_rate": 2.8778618046203254e-05,
      "loss": 1.9531,
      "step": 4200
    },
    {
      "epoch": 1.3227568170454982,
      "grad_norm": 1.2818642854690552,
      "learning_rate": 2.8260644359266553e-05,
      "loss": 1.946,
      "step": 4300
    },
    {
      "epoch": 1.3535248644282913,
      "grad_norm": 1.404370903968811,
      "learning_rate": 2.7742670672329846e-05,
      "loss": 1.9611,
      "step": 4400
    },
    {
      "epoch": 1.3842929118110843,
      "grad_norm": 1.3051358461380005,
      "learning_rate": 2.7224696985393146e-05,
      "loss": 1.9345,
      "step": 4500
    },
    {
      "epoch": 1.4150609591938772,
      "grad_norm": 1.2720344066619873,
      "learning_rate": 2.670672329845644e-05,
      "loss": 1.9596,
      "step": 4600
    },
    {
      "epoch": 1.4458290065766701,
      "grad_norm": 1.1316062211990356,
      "learning_rate": 2.6188749611519735e-05,
      "loss": 1.9655,
      "step": 4700
    },
    {
      "epoch": 1.476597053959463,
      "grad_norm": 1.1589127779006958,
      "learning_rate": 2.56759556614524e-05,
      "loss": 1.9336,
      "step": 4800
    },
    {
      "epoch": 1.507365101342256,
      "grad_norm": 1.418484091758728,
      "learning_rate": 2.5157981974515692e-05,
      "loss": 1.9538,
      "step": 4900
    },
    {
      "epoch": 1.538133148725049,
      "grad_norm": 1.1148117780685425,
      "learning_rate": 2.464000828757899e-05,
      "loss": 1.9373,
      "step": 5000
    },
    {
      "epoch": 1.538133148725049,
      "eval_loss": 1.8176252841949463,
      "eval_runtime": 4.6868,
      "eval_samples_per_second": 42.673,
      "eval_steps_per_second": 21.337,
      "step": 5000
    },
    {
      "epoch": 1.5689011961078418,
      "grad_norm": 1.409507155418396,
      "learning_rate": 2.4122034600642288e-05,
      "loss": 1.9487,
      "step": 5100
    },
    {
      "epoch": 1.599669243490635,
      "grad_norm": 1.4098600149154663,
      "learning_rate": 2.3604060913705588e-05,
      "loss": 1.9158,
      "step": 5200
    },
    {
      "epoch": 1.630437290873428,
      "grad_norm": 1.2400621175765991,
      "learning_rate": 2.308608722676888e-05,
      "loss": 1.9257,
      "step": 5300
    },
    {
      "epoch": 1.661205338256221,
      "grad_norm": 1.3149056434631348,
      "learning_rate": 2.2568113539832177e-05,
      "loss": 1.9254,
      "step": 5400
    },
    {
      "epoch": 1.691973385639014,
      "grad_norm": 1.306784749031067,
      "learning_rate": 2.2050139852895473e-05,
      "loss": 1.9446,
      "step": 5500
    },
    {
      "epoch": 1.722741433021807,
      "grad_norm": 1.3898966312408447,
      "learning_rate": 2.153216616595877e-05,
      "loss": 1.9151,
      "step": 5600
    },
    {
      "epoch": 1.7535094804045999,
      "grad_norm": 1.2994170188903809,
      "learning_rate": 2.1014192479022066e-05,
      "loss": 1.9464,
      "step": 5700
    },
    {
      "epoch": 1.7842775277873928,
      "grad_norm": 1.3611286878585815,
      "learning_rate": 2.0496218792085366e-05,
      "loss": 1.9206,
      "step": 5800
    },
    {
      "epoch": 1.8150455751701857,
      "grad_norm": 1.406435489654541,
      "learning_rate": 1.997824510514866e-05,
      "loss": 1.9303,
      "step": 5900
    },
    {
      "epoch": 1.8458136225529787,
      "grad_norm": 1.4985857009887695,
      "learning_rate": 1.9460271418211955e-05,
      "loss": 1.9306,
      "step": 6000
    },
    {
      "epoch": 1.8458136225529787,
      "eval_loss": 1.7914341688156128,
      "eval_runtime": 4.8041,
      "eval_samples_per_second": 41.631,
      "eval_steps_per_second": 20.816,
      "step": 6000
    },
    {
      "epoch": 1.8765816699357716,
      "grad_norm": 1.3169234991073608,
      "learning_rate": 1.894229773127525e-05,
      "loss": 1.9435,
      "step": 6100
    },
    {
      "epoch": 1.9073497173185645,
      "grad_norm": 1.4807860851287842,
      "learning_rate": 1.8424324044338548e-05,
      "loss": 1.9108,
      "step": 6200
    },
    {
      "epoch": 1.9381177647013577,
      "grad_norm": 1.5189799070358276,
      "learning_rate": 1.7906350357401844e-05,
      "loss": 1.9274,
      "step": 6300
    },
    {
      "epoch": 1.9688858120841506,
      "grad_norm": 1.37026846408844,
      "learning_rate": 1.7388376670465144e-05,
      "loss": 1.9307,
      "step": 6400
    },
    {
      "epoch": 1.9996538594669437,
      "grad_norm": 1.2851979732513428,
      "learning_rate": 1.687040298352844e-05,
      "loss": 1.9324,
      "step": 6500
    },
    {
      "epoch": 2.030152686435137,
      "grad_norm": 1.5312321186065674,
      "learning_rate": 1.6352429296591733e-05,
      "loss": 1.8818,
      "step": 6600
    },
    {
      "epoch": 2.0609207338179303,
      "grad_norm": 1.2317149639129639,
      "learning_rate": 1.583445560965503e-05,
      "loss": 1.8666,
      "step": 6700
    },
    {
      "epoch": 2.091688781200723,
      "grad_norm": 1.4043223857879639,
      "learning_rate": 1.5316481922718326e-05,
      "loss": 1.8588,
      "step": 6800
    },
    {
      "epoch": 2.122456828583516,
      "grad_norm": 1.3751603364944458,
      "learning_rate": 1.4798508235781622e-05,
      "loss": 1.873,
      "step": 6900
    },
    {
      "epoch": 2.153224875966309,
      "grad_norm": 1.605711579322815,
      "learning_rate": 1.428053454884492e-05,
      "loss": 1.8861,
      "step": 7000
    },
    {
      "epoch": 2.153224875966309,
      "eval_loss": 1.770365595817566,
      "eval_runtime": 4.8144,
      "eval_samples_per_second": 41.542,
      "eval_steps_per_second": 20.771,
      "step": 7000
    },
    {
      "epoch": 2.183992923349102,
      "grad_norm": 1.642946720123291,
      "learning_rate": 1.3762560861908216e-05,
      "loss": 1.8608,
      "step": 7100
    },
    {
      "epoch": 2.214760970731895,
      "grad_norm": 1.3048804998397827,
      "learning_rate": 1.3244587174971513e-05,
      "loss": 1.8653,
      "step": 7200
    },
    {
      "epoch": 2.245529018114688,
      "grad_norm": 1.3654470443725586,
      "learning_rate": 1.2726613488034809e-05,
      "loss": 1.8803,
      "step": 7300
    },
    {
      "epoch": 2.276297065497481,
      "grad_norm": 1.377312421798706,
      "learning_rate": 1.2208639801098105e-05,
      "loss": 1.8783,
      "step": 7400
    },
    {
      "epoch": 2.3070651128802737,
      "grad_norm": 1.2883929014205933,
      "learning_rate": 1.1690666114161402e-05,
      "loss": 1.8612,
      "step": 7500
    },
    {
      "epoch": 2.3378331602630666,
      "grad_norm": 1.3439315557479858,
      "learning_rate": 1.1172692427224698e-05,
      "loss": 1.8627,
      "step": 7600
    },
    {
      "epoch": 2.3686012076458596,
      "grad_norm": 1.1767522096633911,
      "learning_rate": 1.0654718740287994e-05,
      "loss": 1.8713,
      "step": 7700
    },
    {
      "epoch": 2.3993692550286525,
      "grad_norm": 1.2367137670516968,
      "learning_rate": 1.0141924790220656e-05,
      "loss": 1.8753,
      "step": 7800
    },
    {
      "epoch": 2.430137302411446,
      "grad_norm": 1.218923568725586,
      "learning_rate": 9.623951103283954e-06,
      "loss": 1.8924,
      "step": 7900
    },
    {
      "epoch": 2.460905349794239,
      "grad_norm": 1.481924057006836,
      "learning_rate": 9.10597741634725e-06,
      "loss": 1.875,
      "step": 8000
    },
    {
      "epoch": 2.460905349794239,
      "eval_loss": 1.7587065696716309,
      "eval_runtime": 4.6695,
      "eval_samples_per_second": 42.831,
      "eval_steps_per_second": 21.416,
      "step": 8000
    },
    {
      "epoch": 2.4916733971770317,
      "grad_norm": 1.5611909627914429,
      "learning_rate": 8.588003729410545e-06,
      "loss": 1.8847,
      "step": 8100
    },
    {
      "epoch": 2.5224414445598247,
      "grad_norm": 1.573014736175537,
      "learning_rate": 8.070030042473843e-06,
      "loss": 1.8645,
      "step": 8200
    },
    {
      "epoch": 2.5532094919426176,
      "grad_norm": 1.4010423421859741,
      "learning_rate": 7.5520563555371394e-06,
      "loss": 1.8716,
      "step": 8300
    },
    {
      "epoch": 2.5839775393254105,
      "grad_norm": 1.4731730222702026,
      "learning_rate": 7.034082668600435e-06,
      "loss": 1.8545,
      "step": 8400
    },
    {
      "epoch": 2.6147455867082035,
      "grad_norm": 1.5401034355163574,
      "learning_rate": 6.516108981663732e-06,
      "loss": 1.8701,
      "step": 8500
    },
    {
      "epoch": 2.6455136340909964,
      "grad_norm": 1.4150763750076294,
      "learning_rate": 5.998135294727028e-06,
      "loss": 1.8781,
      "step": 8600
    },
    {
      "epoch": 2.6762816814737893,
      "grad_norm": 1.2187877893447876,
      "learning_rate": 5.480161607790325e-06,
      "loss": 1.8554,
      "step": 8700
    },
    {
      "epoch": 2.7070497288565827,
      "grad_norm": 1.2359226942062378,
      "learning_rate": 4.962187920853621e-06,
      "loss": 1.8655,
      "step": 8800
    },
    {
      "epoch": 2.7378177762393756,
      "grad_norm": 1.5639677047729492,
      "learning_rate": 4.444214233916917e-06,
      "loss": 1.8702,
      "step": 8900
    },
    {
      "epoch": 2.7685858236221685,
      "grad_norm": 1.3015304803848267,
      "learning_rate": 3.926240546980214e-06,
      "loss": 1.8724,
      "step": 9000
    },
    {
      "epoch": 2.7685858236221685,
      "eval_loss": 1.7501640319824219,
      "eval_runtime": 4.6364,
      "eval_samples_per_second": 43.137,
      "eval_steps_per_second": 21.568,
      "step": 9000
    },
    {
      "epoch": 2.7993538710049615,
      "grad_norm": 1.2435898780822754,
      "learning_rate": 3.40826686004351e-06,
      "loss": 1.8699,
      "step": 9100
    },
    {
      "epoch": 2.8301219183877544,
      "grad_norm": 1.4046730995178223,
      "learning_rate": 2.890293173106806e-06,
      "loss": 1.874,
      "step": 9200
    },
    {
      "epoch": 2.8608899657705473,
      "grad_norm": 1.4343743324279785,
      "learning_rate": 2.3723194861701027e-06,
      "loss": 1.8844,
      "step": 9300
    },
    {
      "epoch": 2.8916580131533403,
      "grad_norm": 1.506251573562622,
      "learning_rate": 1.8543457992333989e-06,
      "loss": 1.8738,
      "step": 9400
    },
    {
      "epoch": 2.922426060536133,
      "grad_norm": 1.1848888397216797,
      "learning_rate": 1.3363721122966954e-06,
      "loss": 1.8549,
      "step": 9500
    },
    {
      "epoch": 2.953194107918926,
      "grad_norm": 1.301543951034546,
      "learning_rate": 8.183984253599917e-07,
      "loss": 1.8691,
      "step": 9600
    },
    {
      "epoch": 2.983962155301719,
      "grad_norm": 1.4360523223876953,
      "learning_rate": 3.004247384232881e-07,
      "loss": 1.8902,
      "step": 9700
    }
  ],
  "logging_steps": 100,
  "max_steps": 9753,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.0763124744192e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}