{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.999356582164458,
  "eval_steps": 1000,
  "global_step": 19425,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.012868356710848025,
      "grad_norm": 5.15623140335083,
      "learning_rate": 4.987387387387388e-05,
      "loss": 1.7328,
      "step": 50
    },
    {
      "epoch": 0.02573671342169605,
      "grad_norm": 4.769791603088379,
      "learning_rate": 4.9745173745173744e-05,
      "loss": 1.3827,
      "step": 100
    },
    {
      "epoch": 0.03860507013254407,
      "grad_norm": 3.983968734741211,
      "learning_rate": 4.9616473616473616e-05,
      "loss": 1.2993,
      "step": 150
    },
    {
      "epoch": 0.0514734268433921,
      "grad_norm": 4.153149127960205,
      "learning_rate": 4.9487773487773494e-05,
      "loss": 1.2884,
      "step": 200
    },
    {
      "epoch": 0.06434178355424013,
      "grad_norm": 3.8115487098693848,
      "learning_rate": 4.935907335907336e-05,
      "loss": 1.2803,
      "step": 250
    },
    {
      "epoch": 0.07721014026508814,
      "grad_norm": 4.397896766662598,
      "learning_rate": 4.923037323037323e-05,
      "loss": 1.2442,
      "step": 300
    },
    {
      "epoch": 0.09007849697593617,
      "grad_norm": 4.619286060333252,
      "learning_rate": 4.91016731016731e-05,
      "loss": 1.2187,
      "step": 350
    },
    {
      "epoch": 0.1029468536867842,
      "grad_norm": 4.834469795227051,
      "learning_rate": 4.8972972972972974e-05,
      "loss": 1.1784,
      "step": 400
    },
    {
      "epoch": 0.11581521039763222,
      "grad_norm": 3.5152344703674316,
      "learning_rate": 4.8844272844272846e-05,
      "loss": 1.162,
      "step": 450
    },
    {
      "epoch": 0.12868356710848025,
      "grad_norm": 4.991385459899902,
      "learning_rate": 4.871557271557272e-05,
      "loss": 1.2053,
      "step": 500
    },
    {
      "epoch": 0.14155192381932827,
      "grad_norm": 3.806436538696289,
      "learning_rate": 4.8589446589446594e-05,
      "loss": 1.1554,
      "step": 550
    },
    {
      "epoch": 0.15442028053017628,
      "grad_norm": 4.663485050201416,
      "learning_rate": 4.8460746460746466e-05,
      "loss": 1.135,
      "step": 600
    },
    {
      "epoch": 0.16728863724102433,
      "grad_norm": 4.0664238929748535,
      "learning_rate": 4.833204633204633e-05,
      "loss": 1.1372,
      "step": 650
    },
    {
      "epoch": 0.18015699395187235,
      "grad_norm": 3.8850207328796387,
      "learning_rate": 4.82033462033462e-05,
      "loss": 1.0818,
      "step": 700
    },
    {
      "epoch": 0.19302535066272036,
      "grad_norm": 3.971874713897705,
      "learning_rate": 4.807464607464608e-05,
      "loss": 1.1086,
      "step": 750
    },
    {
      "epoch": 0.2058937073735684,
      "grad_norm": 3.79402232170105,
      "learning_rate": 4.7945945945945946e-05,
      "loss": 1.1449,
      "step": 800
    },
    {
      "epoch": 0.21876206408441642,
      "grad_norm": 4.102041721343994,
      "learning_rate": 4.781724581724582e-05,
      "loss": 1.1066,
      "step": 850
    },
    {
      "epoch": 0.23163042079526444,
      "grad_norm": 4.198806285858154,
      "learning_rate": 4.768854568854569e-05,
      "loss": 1.1604,
      "step": 900
    },
    {
      "epoch": 0.24449877750611246,
      "grad_norm": 3.196410655975342,
      "learning_rate": 4.755984555984556e-05,
      "loss": 1.1193,
      "step": 950
    },
    {
      "epoch": 0.2573671342169605,
      "grad_norm": 4.250946521759033,
      "learning_rate": 4.743114543114543e-05,
      "loss": 1.1378,
      "step": 1000
    },
    {
      "epoch": 0.2573671342169605,
      "eval_loss": 1.026713252067566,
      "eval_runtime": 189.3511,
      "eval_samples_per_second": 102.592,
      "eval_steps_per_second": 10.261,
      "step": 1000
    },
    {
      "epoch": 0.2702354909278085,
      "grad_norm": 3.319427251815796,
      "learning_rate": 4.7302445302445304e-05,
      "loss": 1.0153,
      "step": 1050
    },
    {
      "epoch": 0.28310384763865654,
      "grad_norm": 3.8477160930633545,
      "learning_rate": 4.7173745173745175e-05,
      "loss": 1.0782,
      "step": 1100
    },
    {
      "epoch": 0.29597220434950455,
      "grad_norm": 3.674345016479492,
      "learning_rate": 4.704504504504505e-05,
      "loss": 1.0935,
      "step": 1150
    },
    {
      "epoch": 0.30884056106035257,
      "grad_norm": 3.0566294193267822,
      "learning_rate": 4.691634491634492e-05,
      "loss": 1.1357,
      "step": 1200
    },
    {
      "epoch": 0.32170891777120064,
      "grad_norm": 3.81984806060791,
      "learning_rate": 4.678764478764479e-05,
      "loss": 1.0254,
      "step": 1250
    },
    {
      "epoch": 0.33457727448204866,
      "grad_norm": 4.312713623046875,
      "learning_rate": 4.665894465894466e-05,
      "loss": 1.0809,
      "step": 1300
    },
    {
      "epoch": 0.3474456311928967,
      "grad_norm": 4.260924339294434,
      "learning_rate": 4.653024453024453e-05,
      "loss": 1.1019,
      "step": 1350
    },
    {
      "epoch": 0.3603139879037447,
      "grad_norm": 3.334003448486328,
      "learning_rate": 4.6401544401544405e-05,
      "loss": 1.0271,
      "step": 1400
    },
    {
      "epoch": 0.3731823446145927,
      "grad_norm": 3.480570077896118,
      "learning_rate": 4.6272844272844277e-05,
      "loss": 1.0714,
      "step": 1450
    },
    {
      "epoch": 0.3860507013254407,
      "grad_norm": 3.051344871520996,
      "learning_rate": 4.614414414414415e-05,
      "loss": 1.0437,
      "step": 1500
    },
    {
      "epoch": 0.39891905803628874,
      "grad_norm": 3.3421928882598877,
      "learning_rate": 4.601544401544402e-05,
      "loss": 1.0486,
      "step": 1550
    },
    {
      "epoch": 0.4117874147471368,
      "grad_norm": 3.3062002658843994,
      "learning_rate": 4.588674388674389e-05,
      "loss": 1.067,
      "step": 1600
    },
    {
      "epoch": 0.42465577145798483,
      "grad_norm": 3.173940420150757,
      "learning_rate": 4.575804375804376e-05,
      "loss": 1.0349,
      "step": 1650
    },
    {
      "epoch": 0.43752412816883285,
      "grad_norm": 3.4759039878845215,
      "learning_rate": 4.562934362934363e-05,
      "loss": 1.0471,
      "step": 1700
    },
    {
      "epoch": 0.45039248487968087,
      "grad_norm": 3.4710917472839355,
      "learning_rate": 4.55006435006435e-05,
      "loss": 1.0891,
      "step": 1750
    },
    {
      "epoch": 0.4632608415905289,
      "grad_norm": 3.515437602996826,
      "learning_rate": 4.537194337194338e-05,
      "loss": 1.0072,
      "step": 1800
    },
    {
      "epoch": 0.4761291983013769,
      "grad_norm": 3.6845717430114746,
      "learning_rate": 4.524324324324324e-05,
      "loss": 1.0141,
      "step": 1850
    },
    {
      "epoch": 0.4889975550122249,
      "grad_norm": 4.084773540496826,
      "learning_rate": 4.5114543114543114e-05,
      "loss": 1.003,
      "step": 1900
    },
    {
      "epoch": 0.5018659117230729,
      "grad_norm": 3.49133038520813,
      "learning_rate": 4.4985842985842986e-05,
      "loss": 1.056,
      "step": 1950
    },
    {
      "epoch": 0.514734268433921,
      "grad_norm": 3.321465015411377,
      "learning_rate": 4.485714285714286e-05,
      "loss": 1.0146,
      "step": 2000
    },
    {
      "epoch": 0.514734268433921,
      "eval_loss": 0.9642198085784912,
      "eval_runtime": 189.3709,
      "eval_samples_per_second": 102.582,
      "eval_steps_per_second": 10.26,
      "step": 2000
    },
    {
      "epoch": 0.527602625144769,
      "grad_norm": 4.292445182800293,
      "learning_rate": 4.472844272844273e-05,
      "loss": 1.079,
      "step": 2050
    },
    {
      "epoch": 0.540470981855617,
      "grad_norm": 3.905832529067993,
      "learning_rate": 4.45997425997426e-05,
      "loss": 0.9615,
      "step": 2100
    },
    {
      "epoch": 0.5533393385664651,
      "grad_norm": 3.985164165496826,
      "learning_rate": 4.447104247104247e-05,
      "loss": 1.0643,
      "step": 2150
    },
    {
      "epoch": 0.5662076952773131,
      "grad_norm": 3.4493820667266846,
      "learning_rate": 4.4342342342342344e-05,
      "loss": 1.0075,
      "step": 2200
    },
    {
      "epoch": 0.5790760519881611,
      "grad_norm": 3.188607931137085,
      "learning_rate": 4.4213642213642216e-05,
      "loss": 0.9822,
      "step": 2250
    },
    {
      "epoch": 0.5919444086990091,
      "grad_norm": 4.343353748321533,
      "learning_rate": 4.408494208494209e-05,
      "loss": 1.0451,
      "step": 2300
    },
    {
      "epoch": 0.6048127654098572,
      "grad_norm": 5.567046642303467,
      "learning_rate": 4.395624195624196e-05,
      "loss": 0.9741,
      "step": 2350
    },
    {
      "epoch": 0.6176811221207051,
      "grad_norm": 3.7193870544433594,
      "learning_rate": 4.382754182754183e-05,
      "loss": 1.0067,
      "step": 2400
    },
    {
      "epoch": 0.6305494788315532,
      "grad_norm": 3.9916462898254395,
      "learning_rate": 4.36988416988417e-05,
      "loss": 1.0139,
      "step": 2450
    },
    {
      "epoch": 0.6434178355424013,
      "grad_norm": 2.9249072074890137,
      "learning_rate": 4.3570141570141574e-05,
      "loss": 0.9773,
      "step": 2500
    },
    {
      "epoch": 0.6562861922532492,
      "grad_norm": 3.220916748046875,
      "learning_rate": 4.3441441441441445e-05,
      "loss": 0.9802,
      "step": 2550
    },
    {
      "epoch": 0.6691545489640973,
      "grad_norm": 3.6743297576904297,
      "learning_rate": 4.331274131274132e-05,
      "loss": 0.9568,
      "step": 2600
    },
    {
      "epoch": 0.6820229056749453,
      "grad_norm": 3.846135377883911,
      "learning_rate": 4.318404118404119e-05,
      "loss": 1.0341,
      "step": 2650
    },
    {
      "epoch": 0.6948912623857934,
      "grad_norm": 3.933671236038208,
      "learning_rate": 4.305534105534106e-05,
      "loss": 1.0272,
      "step": 2700
    },
    {
      "epoch": 0.7077596190966413,
      "grad_norm": 4.287942409515381,
      "learning_rate": 4.2926640926640925e-05,
      "loss": 0.9458,
      "step": 2750
    },
    {
      "epoch": 0.7206279758074894,
      "grad_norm": 3.385679006576538,
      "learning_rate": 4.2797940797940796e-05,
      "loss": 1.013,
      "step": 2800
    },
    {
      "epoch": 0.7334963325183375,
      "grad_norm": 3.283371925354004,
      "learning_rate": 4.2669240669240675e-05,
      "loss": 0.9173,
      "step": 2850
    },
    {
      "epoch": 0.7463646892291854,
      "grad_norm": 3.90366268157959,
      "learning_rate": 4.254054054054054e-05,
      "loss": 0.9605,
      "step": 2900
    },
    {
      "epoch": 0.7592330459400335,
      "grad_norm": 3.5078847408294678,
      "learning_rate": 4.241184041184041e-05,
      "loss": 1.0341,
      "step": 2950
    },
    {
      "epoch": 0.7721014026508815,
      "grad_norm": 3.1368582248687744,
      "learning_rate": 4.228314028314029e-05,
      "loss": 0.9344,
      "step": 3000
    },
    {
      "epoch": 0.7721014026508815,
      "eval_loss": 0.9136499762535095,
      "eval_runtime": 189.3685,
      "eval_samples_per_second": 102.583,
      "eval_steps_per_second": 10.26,
      "step": 3000
    },
    {
      "epoch": 0.7849697593617295,
      "grad_norm": 3.6094772815704346,
      "learning_rate": 4.2154440154440154e-05,
      "loss": 0.9633,
      "step": 3050
    },
    {
      "epoch": 0.7978381160725775,
      "grad_norm": 3.157749652862549,
      "learning_rate": 4.2025740025740026e-05,
      "loss": 0.9851,
      "step": 3100
    },
    {
      "epoch": 0.8107064727834256,
      "grad_norm": 2.6654834747314453,
      "learning_rate": 4.18970398970399e-05,
      "loss": 0.9317,
      "step": 3150
    },
    {
      "epoch": 0.8235748294942736,
      "grad_norm": 3.4694721698760986,
      "learning_rate": 4.176833976833977e-05,
      "loss": 0.9166,
      "step": 3200
    },
    {
      "epoch": 0.8364431862051216,
      "grad_norm": 4.163283824920654,
      "learning_rate": 4.163963963963964e-05,
      "loss": 1.0012,
      "step": 3250
    },
    {
      "epoch": 0.8493115429159697,
      "grad_norm": 2.8416895866394043,
      "learning_rate": 4.151093951093951e-05,
      "loss": 0.9063,
      "step": 3300
    },
    {
      "epoch": 0.8621798996268176,
      "grad_norm": 2.7988712787628174,
      "learning_rate": 4.1382239382239384e-05,
      "loss": 0.9155,
      "step": 3350
    },
    {
      "epoch": 0.8750482563376657,
      "grad_norm": 2.930269241333008,
      "learning_rate": 4.1253539253539256e-05,
      "loss": 0.9897,
      "step": 3400
    },
    {
      "epoch": 0.8879166130485137,
      "grad_norm": 3.0751140117645264,
      "learning_rate": 4.112483912483913e-05,
      "loss": 0.9355,
      "step": 3450
    },
    {
      "epoch": 0.9007849697593617,
      "grad_norm": 4.277087688446045,
      "learning_rate": 4.0996138996139e-05,
      "loss": 0.945,
      "step": 3500
    },
    {
      "epoch": 0.9136533264702098,
      "grad_norm": 4.510545253753662,
      "learning_rate": 4.086743886743887e-05,
      "loss": 0.97,
      "step": 3550
    },
    {
      "epoch": 0.9265216831810578,
      "grad_norm": 3.1346421241760254,
      "learning_rate": 4.073873873873874e-05,
      "loss": 0.9602,
      "step": 3600
    },
    {
      "epoch": 0.9393900398919058,
      "grad_norm": 2.9139790534973145,
      "learning_rate": 4.0610038610038614e-05,
      "loss": 0.939,
      "step": 3650
    },
    {
      "epoch": 0.9522583966027538,
      "grad_norm": 3.7701892852783203,
      "learning_rate": 4.0481338481338485e-05,
      "loss": 0.9806,
      "step": 3700
    },
    {
      "epoch": 0.9651267533136019,
      "grad_norm": 3.3391294479370117,
      "learning_rate": 4.035263835263836e-05,
      "loss": 0.9835,
      "step": 3750
    },
    {
      "epoch": 0.9779951100244498,
      "grad_norm": 4.155078887939453,
      "learning_rate": 4.022393822393822e-05,
      "loss": 0.9921,
      "step": 3800
    },
    {
      "epoch": 0.9908634667352979,
      "grad_norm": 3.9210457801818848,
      "learning_rate": 4.00952380952381e-05,
      "loss": 0.9831,
      "step": 3850
    },
    {
      "epoch": 1.0037318234461459,
      "grad_norm": 2.388692855834961,
      "learning_rate": 3.996653796653797e-05,
      "loss": 0.8888,
      "step": 3900
    },
    {
      "epoch": 1.016600180156994,
      "grad_norm": 3.42979097366333,
      "learning_rate": 3.983783783783784e-05,
      "loss": 0.76,
      "step": 3950
    },
    {
      "epoch": 1.029468536867842,
      "grad_norm": 3.304460048675537,
      "learning_rate": 3.970913770913771e-05,
      "loss": 0.752,
      "step": 4000
    },
    {
      "epoch": 1.029468536867842,
      "eval_loss": 0.8883536458015442,
      "eval_runtime": 189.2819,
      "eval_samples_per_second": 102.63,
      "eval_steps_per_second": 10.265,
      "step": 4000
    },
    {
      "epoch": 1.04233689357869,
      "grad_norm": 3.8564417362213135,
      "learning_rate": 3.9580437580437587e-05,
      "loss": 0.7679,
      "step": 4050
    },
    {
      "epoch": 1.055205250289538,
      "grad_norm": 3.680621385574341,
      "learning_rate": 3.945173745173745e-05,
      "loss": 0.7563,
      "step": 4100
    },
    {
      "epoch": 1.068073607000386,
      "grad_norm": 3.4106550216674805,
      "learning_rate": 3.932303732303732e-05,
      "loss": 0.7149,
      "step": 4150
    },
    {
      "epoch": 1.080941963711234,
      "grad_norm": 3.5690340995788574,
      "learning_rate": 3.9194337194337195e-05,
      "loss": 0.7185,
      "step": 4200
    },
    {
      "epoch": 1.0938103204220821,
      "grad_norm": 3.215930700302124,
      "learning_rate": 3.9065637065637066e-05,
      "loss": 0.762,
      "step": 4250
    },
    {
      "epoch": 1.1066786771329302,
      "grad_norm": 3.48146390914917,
      "learning_rate": 3.893693693693694e-05,
      "loss": 0.744,
      "step": 4300
    },
    {
      "epoch": 1.119547033843778,
      "grad_norm": 3.667818069458008,
      "learning_rate": 3.880823680823681e-05,
      "loss": 0.7564,
      "step": 4350
    },
    {
      "epoch": 1.1324153905546261,
      "grad_norm": 4.029824256896973,
      "learning_rate": 3.867953667953668e-05,
      "loss": 0.7631,
      "step": 4400
    },
    {
      "epoch": 1.1452837472654742,
      "grad_norm": 3.8039183616638184,
      "learning_rate": 3.855083655083655e-05,
      "loss": 0.7635,
      "step": 4450
    },
    {
      "epoch": 1.1581521039763223,
      "grad_norm": 2.8013832569122314,
      "learning_rate": 3.8422136422136424e-05,
      "loss": 0.6891,
      "step": 4500
    },
    {
      "epoch": 1.1710204606871701,
      "grad_norm": 3.721989631652832,
      "learning_rate": 3.8293436293436296e-05,
      "loss": 0.763,
      "step": 4550
    },
    {
      "epoch": 1.1838888173980182,
      "grad_norm": 2.802401304244995,
      "learning_rate": 3.816473616473617e-05,
      "loss": 0.7624,
      "step": 4600
    },
    {
      "epoch": 1.1967571741088663,
      "grad_norm": 3.3529865741729736,
      "learning_rate": 3.803603603603604e-05,
      "loss": 0.7673,
      "step": 4650
    },
    {
      "epoch": 1.2096255308197144,
      "grad_norm": 3.866051197052002,
      "learning_rate": 3.790733590733591e-05,
      "loss": 0.7713,
      "step": 4700
    },
    {
      "epoch": 1.2224938875305624,
      "grad_norm": 3.5920650959014893,
      "learning_rate": 3.777863577863578e-05,
      "loss": 0.7484,
      "step": 4750
    },
    {
      "epoch": 1.2353622442414103,
      "grad_norm": 4.17499303817749,
      "learning_rate": 3.7649935649935654e-05,
      "loss": 0.7279,
      "step": 4800
    },
    {
      "epoch": 1.2482306009522584,
      "grad_norm": 4.160298824310303,
      "learning_rate": 3.752123552123552e-05,
      "loss": 0.7442,
      "step": 4850
    },
    {
      "epoch": 1.2610989576631064,
      "grad_norm": 5.012088775634766,
      "learning_rate": 3.73925353925354e-05,
      "loss": 0.7322,
      "step": 4900
    },
    {
      "epoch": 1.2739673143739545,
      "grad_norm": 2.925732135772705,
      "learning_rate": 3.726383526383527e-05,
      "loss": 0.7284,
      "step": 4950
    },
    {
      "epoch": 1.2868356710848023,
      "grad_norm": 3.843261241912842,
      "learning_rate": 3.7135135135135134e-05,
      "loss": 0.7816,
      "step": 5000
    },
    {
      "epoch": 1.2868356710848023,
      "eval_loss": 0.8717387914657593,
      "eval_runtime": 189.4604,
      "eval_samples_per_second": 102.533,
      "eval_steps_per_second": 10.255,
      "step": 5000
    },
    {
      "epoch": 1.2997040277956504,
      "grad_norm": 3.363952398300171,
      "learning_rate": 3.7006435006435005e-05,
      "loss": 0.7719,
      "step": 5050
    },
    {
      "epoch": 1.3125723845064985,
      "grad_norm": 3.3399648666381836,
      "learning_rate": 3.6877734877734884e-05,
      "loss": 0.7282,
      "step": 5100
    },
    {
      "epoch": 1.3254407412173466,
      "grad_norm": 3.1839489936828613,
      "learning_rate": 3.674903474903475e-05,
      "loss": 0.7269,
      "step": 5150
    },
    {
      "epoch": 1.3383090979281946,
      "grad_norm": 4.2268266677856445,
      "learning_rate": 3.662033462033462e-05,
      "loss": 0.7438,
      "step": 5200
    },
    {
      "epoch": 1.3511774546390427,
      "grad_norm": 3.201995611190796,
      "learning_rate": 3.649163449163449e-05,
      "loss": 0.7396,
      "step": 5250
    },
    {
      "epoch": 1.3640458113498906,
      "grad_norm": 2.455885887145996,
      "learning_rate": 3.636293436293436e-05,
      "loss": 0.7292,
      "step": 5300
    },
    {
      "epoch": 1.3769141680607386,
      "grad_norm": 3.5545082092285156,
      "learning_rate": 3.6234234234234235e-05,
      "loss": 0.7118,
      "step": 5350
    },
    {
      "epoch": 1.3897825247715867,
      "grad_norm": 3.1645827293395996,
      "learning_rate": 3.6105534105534106e-05,
      "loss": 0.7368,
      "step": 5400
    },
    {
      "epoch": 1.4026508814824348,
      "grad_norm": 4.050566673278809,
      "learning_rate": 3.597683397683398e-05,
      "loss": 0.7368,
      "step": 5450
    },
    {
      "epoch": 1.4155192381932826,
      "grad_norm": 3.393502950668335,
      "learning_rate": 3.584813384813385e-05,
      "loss": 0.7496,
      "step": 5500
    },
    {
      "epoch": 1.4283875949041307,
      "grad_norm": 2.807429790496826,
      "learning_rate": 3.571943371943372e-05,
      "loss": 0.7224,
      "step": 5550
    },
    {
      "epoch": 1.4412559516149788,
      "grad_norm": 3.8620431423187256,
      "learning_rate": 3.559073359073359e-05,
      "loss": 0.7135,
      "step": 5600
    },
    {
      "epoch": 1.4541243083258268,
      "grad_norm": 3.6501224040985107,
      "learning_rate": 3.5462033462033465e-05,
      "loss": 0.7158,
      "step": 5650
    },
    {
      "epoch": 1.466992665036675,
      "grad_norm": 3.178494453430176,
      "learning_rate": 3.5333333333333336e-05,
      "loss": 0.7122,
      "step": 5700
    },
    {
      "epoch": 1.4798610217475228,
      "grad_norm": 3.163177251815796,
      "learning_rate": 3.520463320463321e-05,
      "loss": 0.7268,
      "step": 5750
    },
    {
      "epoch": 1.4927293784583708,
      "grad_norm": 3.3929202556610107,
      "learning_rate": 3.507593307593308e-05,
      "loss": 0.7099,
      "step": 5800
    },
    {
      "epoch": 1.505597735169219,
      "grad_norm": 4.053615570068359,
      "learning_rate": 3.494723294723295e-05,
      "loss": 0.7291,
      "step": 5850
    },
    {
      "epoch": 1.5184660918800668,
      "grad_norm": 3.453657865524292,
      "learning_rate": 3.4818532818532816e-05,
      "loss": 0.7373,
      "step": 5900
    },
    {
      "epoch": 1.5313344485909148,
      "grad_norm": 3.537903070449829,
      "learning_rate": 3.4689832689832694e-05,
      "loss": 0.7311,
      "step": 5950
    },
    {
      "epoch": 1.544202805301763,
      "grad_norm": 3.1102797985076904,
      "learning_rate": 3.4561132561132566e-05,
      "loss": 0.7702,
      "step": 6000
    },
    {
      "epoch": 1.544202805301763,
      "eval_loss": 0.856028139591217,
      "eval_runtime": 189.5266,
      "eval_samples_per_second": 102.497,
      "eval_steps_per_second": 10.252,
      "step": 6000
    },
    {
      "epoch": 1.557071162012611,
      "grad_norm": 3.3603110313415527,
      "learning_rate": 3.443243243243243e-05,
      "loss": 0.748,
      "step": 6050
    },
    {
      "epoch": 1.569939518723459,
      "grad_norm": 3.154524326324463,
      "learning_rate": 3.43037323037323e-05,
      "loss": 0.7241,
      "step": 6100
    },
    {
      "epoch": 1.5828078754343071,
      "grad_norm": 3.641023874282837,
      "learning_rate": 3.417503217503218e-05,
      "loss": 0.7564,
      "step": 6150
    },
    {
      "epoch": 1.5956762321451552,
      "grad_norm": 2.77715802192688,
      "learning_rate": 3.4046332046332045e-05,
      "loss": 0.709,
      "step": 6200
    },
    {
      "epoch": 1.608544588856003,
      "grad_norm": 3.438999652862549,
      "learning_rate": 3.391763191763192e-05,
      "loss": 0.7545,
      "step": 6250
    },
    {
      "epoch": 1.6214129455668511,
      "grad_norm": 4.376018524169922,
      "learning_rate": 3.378893178893179e-05,
      "loss": 0.7435,
      "step": 6300
    },
    {
      "epoch": 1.6342813022776992,
      "grad_norm": 3.45723819732666,
      "learning_rate": 3.366023166023166e-05,
      "loss": 0.7191,
      "step": 6350
    },
    {
      "epoch": 1.647149658988547,
      "grad_norm": 3.0947272777557373,
      "learning_rate": 3.353153153153153e-05,
      "loss": 0.7376,
      "step": 6400
    },
    {
      "epoch": 1.660018015699395,
      "grad_norm": 4.219300746917725,
      "learning_rate": 3.3402831402831403e-05,
      "loss": 0.7567,
      "step": 6450
    },
    {
      "epoch": 1.6728863724102432,
      "grad_norm": 3.1860275268554688,
      "learning_rate": 3.3274131274131275e-05,
      "loss": 0.7771,
      "step": 6500
    },
    {
      "epoch": 1.6857547291210913,
      "grad_norm": 2.766604423522949,
      "learning_rate": 3.314543114543115e-05,
      "loss": 0.7504,
      "step": 6550
    },
    {
      "epoch": 1.6986230858319393,
      "grad_norm": 3.745673894882202,
      "learning_rate": 3.301673101673102e-05,
      "loss": 0.7389,
      "step": 6600
    },
    {
      "epoch": 1.7114914425427874,
      "grad_norm": 3.8440420627593994,
      "learning_rate": 3.288803088803089e-05,
      "loss": 0.7298,
      "step": 6650
    },
    {
      "epoch": 1.7243597992536355,
      "grad_norm": 3.5810177326202393,
      "learning_rate": 3.276190476190477e-05,
      "loss": 0.7602,
      "step": 6700
    },
    {
      "epoch": 1.7372281559644833,
      "grad_norm": 3.4311060905456543,
      "learning_rate": 3.263320463320463e-05,
      "loss": 0.74,
      "step": 6750
    },
    {
      "epoch": 1.7500965126753314,
      "grad_norm": 2.345510721206665,
      "learning_rate": 3.250450450450451e-05,
      "loss": 0.733,
      "step": 6800
    },
    {
      "epoch": 1.7629648693861792,
      "grad_norm": 3.014230251312256,
      "learning_rate": 3.2375804375804375e-05,
      "loss": 0.7469,
      "step": 6850
    },
    {
      "epoch": 1.7758332260970273,
      "grad_norm": 2.7299680709838867,
      "learning_rate": 3.224710424710425e-05,
      "loss": 0.7183,
      "step": 6900
    },
    {
      "epoch": 1.7887015828078754,
      "grad_norm": 3.163944721221924,
      "learning_rate": 3.2118404118404125e-05,
      "loss": 0.763,
      "step": 6950
    },
    {
      "epoch": 1.8015699395187235,
      "grad_norm": 3.4352288246154785,
      "learning_rate": 3.198970398970399e-05,
      "loss": 0.7653,
      "step": 7000
    },
    {
      "epoch": 1.8015699395187235,
      "eval_loss": 0.8297092914581299,
      "eval_runtime": 189.396,
      "eval_samples_per_second": 102.568,
      "eval_steps_per_second": 10.259,
      "step": 7000
    },
    {
      "epoch": 1.8144382962295715,
      "grad_norm": 4.164456844329834,
      "learning_rate": 3.186100386100386e-05,
      "loss": 0.7168,
      "step": 7050
    },
    {
      "epoch": 1.8273066529404196,
      "grad_norm": 3.9557299613952637,
      "learning_rate": 3.173230373230373e-05,
      "loss": 0.7689,
      "step": 7100
    },
    {
      "epoch": 1.8401750096512677,
      "grad_norm": 3.7688467502593994,
      "learning_rate": 3.1603603603603605e-05,
      "loss": 0.7101,
      "step": 7150
    },
    {
      "epoch": 1.8530433663621155,
      "grad_norm": 3.025460720062256,
      "learning_rate": 3.1474903474903476e-05,
      "loss": 0.7311,
      "step": 7200
    },
    {
      "epoch": 1.8659117230729636,
      "grad_norm": 3.369191884994507,
      "learning_rate": 3.134620334620335e-05,
      "loss": 0.7063,
      "step": 7250
    },
    {
      "epoch": 1.8787800797838115,
      "grad_norm": 3.8419950008392334,
      "learning_rate": 3.121750321750322e-05,
      "loss": 0.7228,
      "step": 7300
    },
    {
      "epoch": 1.8916484364946595,
      "grad_norm": 4.265018939971924,
      "learning_rate": 3.108880308880309e-05,
      "loss": 0.7699,
      "step": 7350
    },
    {
      "epoch": 1.9045167932055076,
      "grad_norm": 3.287287712097168,
      "learning_rate": 3.096010296010296e-05,
      "loss": 0.7496,
      "step": 7400
    },
    {
      "epoch": 1.9173851499163557,
      "grad_norm": 4.200809001922607,
      "learning_rate": 3.0831402831402834e-05,
      "loss": 0.7039,
      "step": 7450
    },
    {
      "epoch": 1.9302535066272037,
      "grad_norm": 2.8199522495269775,
      "learning_rate": 3.07027027027027e-05,
      "loss": 0.7166,
      "step": 7500
    },
    {
      "epoch": 1.9431218633380518,
      "grad_norm": 3.1593542098999023,
      "learning_rate": 3.057400257400258e-05,
      "loss": 0.7288,
      "step": 7550
    },
    {
      "epoch": 1.9559902200488999,
      "grad_norm": 3.7583165168762207,
      "learning_rate": 3.044530244530245e-05,
      "loss": 0.7184,
      "step": 7600
    },
    {
      "epoch": 1.9688585767597477,
      "grad_norm": 2.7919719219207764,
      "learning_rate": 3.0316602316602317e-05,
      "loss": 0.7414,
      "step": 7650
    },
    {
      "epoch": 1.9817269334705958,
      "grad_norm": 3.475163459777832,
      "learning_rate": 3.018790218790219e-05,
      "loss": 0.6855,
      "step": 7700
    },
    {
      "epoch": 1.9945952901814439,
      "grad_norm": 3.133477210998535,
      "learning_rate": 3.0059202059202064e-05,
      "loss": 0.715,
      "step": 7750
    },
    {
      "epoch": 2.0074636468922917,
      "grad_norm": 2.9434854984283447,
      "learning_rate": 2.9930501930501932e-05,
      "loss": 0.6191,
      "step": 7800
    },
    {
      "epoch": 2.02033200360314,
      "grad_norm": 3.8272597789764404,
      "learning_rate": 2.9801801801801804e-05,
      "loss": 0.5678,
      "step": 7850
    },
    {
      "epoch": 2.033200360313988,
      "grad_norm": 3.036975860595703,
      "learning_rate": 2.9673101673101672e-05,
      "loss": 0.5251,
      "step": 7900
    },
    {
      "epoch": 2.046068717024836,
      "grad_norm": 3.401226758956909,
      "learning_rate": 2.9544401544401547e-05,
      "loss": 0.554,
      "step": 7950
    },
    {
      "epoch": 2.058937073735684,
      "grad_norm": 3.0268912315368652,
      "learning_rate": 2.941570141570142e-05,
      "loss": 0.5595,
      "step": 8000
    },
    {
      "epoch": 2.058937073735684,
      "eval_loss": 0.8340552449226379,
      "eval_runtime": 189.3362,
      "eval_samples_per_second": 102.601,
      "eval_steps_per_second": 10.262,
      "step": 8000
    },
    {
      "epoch": 2.071805430446532,
      "grad_norm": 3.2826826572418213,
      "learning_rate": 2.9287001287001287e-05,
      "loss": 0.537,
      "step": 8050
    },
    {
      "epoch": 2.08467378715738,
      "grad_norm": 3.539940357208252,
      "learning_rate": 2.915830115830116e-05,
      "loss": 0.5392,
      "step": 8100
    },
    {
      "epoch": 2.0975421438682282,
      "grad_norm": 3.780297040939331,
      "learning_rate": 2.9032175032175036e-05,
      "loss": 0.5642,
      "step": 8150
    },
    {
      "epoch": 2.110410500579076,
      "grad_norm": 3.1283602714538574,
      "learning_rate": 2.8903474903474904e-05,
      "loss": 0.5538,
      "step": 8200
    },
    {
      "epoch": 2.123278857289924,
      "grad_norm": 2.545053243637085,
      "learning_rate": 2.8774774774774775e-05,
      "loss": 0.4843,
      "step": 8250
    },
    {
      "epoch": 2.136147214000772,
      "grad_norm": 3.5739200115203857,
      "learning_rate": 2.864607464607465e-05,
      "loss": 0.5477,
      "step": 8300
    },
    {
      "epoch": 2.14901557071162,
      "grad_norm": 2.904505491256714,
      "learning_rate": 2.851737451737452e-05,
      "loss": 0.5377,
      "step": 8350
    },
    {
      "epoch": 2.161883927422468,
      "grad_norm": 4.482571601867676,
      "learning_rate": 2.838867438867439e-05,
      "loss": 0.5404,
      "step": 8400
    },
    {
      "epoch": 2.1747522841333162,
      "grad_norm": 3.0093982219696045,
      "learning_rate": 2.825997425997426e-05,
      "loss": 0.5446,
      "step": 8450
    },
    {
      "epoch": 2.1876206408441643,
      "grad_norm": 5.678364276885986,
      "learning_rate": 2.8131274131274134e-05,
      "loss": 0.5632,
      "step": 8500
    },
    {
      "epoch": 2.2004889975550124,
      "grad_norm": 3.4848239421844482,
      "learning_rate": 2.8002574002574005e-05,
      "loss": 0.5451,
      "step": 8550
    },
    {
      "epoch": 2.2133573542658604,
      "grad_norm": 3.782041549682617,
      "learning_rate": 2.7873873873873873e-05,
      "loss": 0.5654,
      "step": 8600
    },
    {
      "epoch": 2.226225710976708,
      "grad_norm": 3.711773157119751,
      "learning_rate": 2.774517374517375e-05,
      "loss": 0.5549,
      "step": 8650
    },
    {
      "epoch": 2.239094067687556,
      "grad_norm": 3.105684518814087,
      "learning_rate": 2.761647361647362e-05,
      "loss": 0.5741,
      "step": 8700
    },
    {
      "epoch": 2.251962424398404,
      "grad_norm": 3.3493452072143555,
      "learning_rate": 2.7487773487773488e-05,
      "loss": 0.5474,
      "step": 8750
    },
    {
      "epoch": 2.2648307811092523,
      "grad_norm": 2.916020631790161,
      "learning_rate": 2.7359073359073363e-05,
      "loss": 0.5704,
      "step": 8800
    },
    {
      "epoch": 2.2776991378201004,
      "grad_norm": 2.867391347885132,
      "learning_rate": 2.7230373230373228e-05,
      "loss": 0.5116,
      "step": 8850
    },
    {
      "epoch": 2.2905674945309484,
      "grad_norm": 3.349109649658203,
      "learning_rate": 2.7101673101673103e-05,
      "loss": 0.5523,
      "step": 8900
    },
    {
      "epoch": 2.3034358512417965,
      "grad_norm": 3.198293685913086,
      "learning_rate": 2.6972972972972978e-05,
      "loss": 0.5437,
      "step": 8950
    },
    {
      "epoch": 2.3163042079526446,
      "grad_norm": 3.9013428688049316,
      "learning_rate": 2.6844272844272843e-05,
      "loss": 0.5501,
      "step": 9000
    },
    {
      "epoch": 2.3163042079526446,
      "eval_loss": 0.8361700773239136,
      "eval_runtime": 189.4593,
      "eval_samples_per_second": 102.534,
      "eval_steps_per_second": 10.256,
      "step": 9000
    },
    {
      "epoch": 2.3291725646634927,
      "grad_norm": 2.5401031970977783,
      "learning_rate": 2.6715572715572718e-05,
      "loss": 0.5441,
      "step": 9050
    },
    {
      "epoch": 2.3420409213743403,
      "grad_norm": 3.330003261566162,
      "learning_rate": 2.6586872586872586e-05,
      "loss": 0.5292,
      "step": 9100
    },
    {
      "epoch": 2.3549092780851884,
      "grad_norm": 3.724376916885376,
      "learning_rate": 2.6458172458172458e-05,
      "loss": 0.5555,
      "step": 9150
    },
    {
      "epoch": 2.3677776347960364,
      "grad_norm": 3.3693745136260986,
      "learning_rate": 2.6329472329472333e-05,
      "loss": 0.5638,
      "step": 9200
    },
    {
      "epoch": 2.3806459915068845,
      "grad_norm": 3.3780782222747803,
      "learning_rate": 2.62007722007722e-05,
      "loss": 0.5724,
      "step": 9250
    },
    {
      "epoch": 2.3935143482177326,
      "grad_norm": 3.949384927749634,
      "learning_rate": 2.6072072072072072e-05,
      "loss": 0.5548,
      "step": 9300
    },
    {
      "epoch": 2.4063827049285806,
      "grad_norm": 3.7752134799957275,
      "learning_rate": 2.5943371943371947e-05,
      "loss": 0.5401,
      "step": 9350
    },
    {
      "epoch": 2.4192510616394287,
      "grad_norm": 4.018200874328613,
      "learning_rate": 2.5814671814671816e-05,
      "loss": 0.5154,
      "step": 9400
    },
    {
      "epoch": 2.432119418350277,
      "grad_norm": 2.818852186203003,
      "learning_rate": 2.5685971685971687e-05,
      "loss": 0.5317,
      "step": 9450
    },
    {
      "epoch": 2.444987775061125,
      "grad_norm": 3.676013946533203,
      "learning_rate": 2.5557271557271556e-05,
      "loss": 0.5683,
      "step": 9500
    },
    {
      "epoch": 2.4578561317719725,
      "grad_norm": 3.144169807434082,
      "learning_rate": 2.542857142857143e-05,
      "loss": 0.5491,
      "step": 9550
    },
    {
      "epoch": 2.4707244884828206,
      "grad_norm": 3.605506658554077,
      "learning_rate": 2.5299871299871302e-05,
      "loss": 0.5733,
      "step": 9600
    },
    {
      "epoch": 2.4835928451936686,
      "grad_norm": 3.2025864124298096,
      "learning_rate": 2.517117117117117e-05,
      "loss": 0.5581,
      "step": 9650
    },
    {
      "epoch": 2.4964612019045167,
      "grad_norm": 3.0987846851348877,
      "learning_rate": 2.5042471042471045e-05,
      "loss": 0.5246,
      "step": 9700
    },
    {
      "epoch": 2.5093295586153648,
      "grad_norm": 2.3385045528411865,
      "learning_rate": 2.4913770913770914e-05,
      "loss": 0.5643,
      "step": 9750
    },
    {
      "epoch": 2.522197915326213,
      "grad_norm": 3.4975156784057617,
      "learning_rate": 2.4785070785070785e-05,
      "loss": 0.5793,
      "step": 9800
    },
    {
      "epoch": 2.535066272037061,
      "grad_norm": 3.657727003097534,
      "learning_rate": 2.465637065637066e-05,
      "loss": 0.561,
      "step": 9850
    },
    {
      "epoch": 2.547934628747909,
      "grad_norm": 3.558779001235962,
      "learning_rate": 2.452767052767053e-05,
      "loss": 0.5442,
      "step": 9900
    },
    {
      "epoch": 2.560802985458757,
      "grad_norm": 3.5756425857543945,
      "learning_rate": 2.43989703989704e-05,
      "loss": 0.5467,
      "step": 9950
    },
    {
      "epoch": 2.5736713421696047,
      "grad_norm": 4.302220821380615,
      "learning_rate": 2.427027027027027e-05,
      "loss": 0.5322,
      "step": 10000
    },
    {
      "epoch": 2.5736713421696047,
      "eval_loss": 0.8265257477760315,
      "eval_runtime": 189.4664,
      "eval_samples_per_second": 102.53,
      "eval_steps_per_second": 10.255,
      "step": 10000
    },
    {
      "epoch": 2.586539698880453,
      "grad_norm": 3.3012359142303467,
      "learning_rate": 2.4141570141570143e-05,
      "loss": 0.542,
      "step": 10050
    },
    {
      "epoch": 2.599408055591301,
      "grad_norm": 3.0096490383148193,
      "learning_rate": 2.4012870012870015e-05,
      "loss": 0.5641,
      "step": 10100
    },
    {
      "epoch": 2.612276412302149,
      "grad_norm": 3.803586006164551,
      "learning_rate": 2.3884169884169886e-05,
      "loss": 0.5375,
      "step": 10150
    },
    {
      "epoch": 2.625144769012997,
      "grad_norm": 3.69391131401062,
      "learning_rate": 2.3755469755469755e-05,
      "loss": 0.5579,
      "step": 10200
    },
    {
      "epoch": 2.638013125723845,
      "grad_norm": 3.184783458709717,
      "learning_rate": 2.362676962676963e-05,
      "loss": 0.5185,
      "step": 10250
    },
    {
      "epoch": 2.650881482434693,
      "grad_norm": 3.191938638687134,
      "learning_rate": 2.34980694980695e-05,
      "loss": 0.5644,
      "step": 10300
    },
    {
      "epoch": 2.663749839145541,
      "grad_norm": 2.8819658756256104,
      "learning_rate": 2.336936936936937e-05,
      "loss": 0.5318,
      "step": 10350
    },
    {
      "epoch": 2.6766181958563893,
      "grad_norm": 2.8010735511779785,
      "learning_rate": 2.324066924066924e-05,
      "loss": 0.5286,
      "step": 10400
    },
    {
      "epoch": 2.689486552567237,
      "grad_norm": 3.6001551151275635,
      "learning_rate": 2.3111969111969113e-05,
      "loss": 0.5699,
      "step": 10450
    },
    {
      "epoch": 2.7023549092780854,
      "grad_norm": 3.6811742782592773,
      "learning_rate": 2.2983268983268984e-05,
      "loss": 0.5693,
      "step": 10500
    },
    {
      "epoch": 2.715223265988933,
      "grad_norm": 3.8935017585754395,
      "learning_rate": 2.2854568854568856e-05,
      "loss": 0.5471,
      "step": 10550
    },
    {
      "epoch": 2.728091622699781,
      "grad_norm": 4.385620594024658,
      "learning_rate": 2.2725868725868727e-05,
      "loss": 0.5496,
      "step": 10600
    },
    {
      "epoch": 2.740959979410629,
      "grad_norm": 3.178663730621338,
      "learning_rate": 2.2597168597168596e-05,
      "loss": 0.5669,
      "step": 10650
    },
    {
      "epoch": 2.7538283361214773,
      "grad_norm": 2.5699925422668457,
      "learning_rate": 2.246846846846847e-05,
      "loss": 0.5445,
      "step": 10700
    },
    {
      "epoch": 2.7666966928323253,
      "grad_norm": 3.409348964691162,
      "learning_rate": 2.2339768339768342e-05,
      "loss": 0.5195,
      "step": 10750
    },
    {
      "epoch": 2.7795650495431734,
      "grad_norm": 3.578507900238037,
      "learning_rate": 2.221106821106821e-05,
      "loss": 0.5664,
      "step": 10800
    },
    {
      "epoch": 2.7924334062540215,
      "grad_norm": 3.959843873977661,
      "learning_rate": 2.2082368082368082e-05,
      "loss": 0.5509,
      "step": 10850
    },
    {
      "epoch": 2.8053017629648695,
      "grad_norm": 4.0397868156433105,
      "learning_rate": 2.1953667953667957e-05,
      "loss": 0.5287,
      "step": 10900
    },
    {
      "epoch": 2.8181701196757176,
      "grad_norm": 4.056464195251465,
      "learning_rate": 2.1824967824967825e-05,
      "loss": 0.5493,
      "step": 10950
    },
    {
      "epoch": 2.8310384763865653,
      "grad_norm": 2.7274129390716553,
      "learning_rate": 2.1696267696267697e-05,
      "loss": 0.5536,
      "step": 11000
    },
    {
      "epoch": 2.8310384763865653,
      "eval_loss": 0.8111644983291626,
      "eval_runtime": 189.4072,
      "eval_samples_per_second": 102.562,
      "eval_steps_per_second": 10.258,
      "step": 11000
    },
    {
      "epoch": 2.8439068330974133,
      "grad_norm": 4.217979907989502,
      "learning_rate": 2.156756756756757e-05,
      "loss": 0.5206,
      "step": 11050
    },
    {
      "epoch": 2.8567751898082614,
      "grad_norm": 3.4003894329071045,
      "learning_rate": 2.143886743886744e-05,
      "loss": 0.5389,
      "step": 11100
    },
    {
      "epoch": 2.8696435465191095,
      "grad_norm": 3.1953563690185547,
      "learning_rate": 2.1310167310167312e-05,
      "loss": 0.5533,
      "step": 11150
    },
    {
      "epoch": 2.8825119032299575,
      "grad_norm": 4.147376537322998,
      "learning_rate": 2.1181467181467183e-05,
      "loss": 0.5537,
      "step": 11200
    },
    {
      "epoch": 2.8953802599408056,
      "grad_norm": 3.726901054382324,
      "learning_rate": 2.105276705276705e-05,
      "loss": 0.527,
      "step": 11250
    },
    {
      "epoch": 2.9082486166516537,
      "grad_norm": 3.210848331451416,
      "learning_rate": 2.0924066924066927e-05,
      "loss": 0.5561,
      "step": 11300
    },
    {
      "epoch": 2.9211169733625018,
      "grad_norm": 4.512522220611572,
      "learning_rate": 2.0795366795366798e-05,
      "loss": 0.5642,
      "step": 11350
    },
    {
      "epoch": 2.93398533007335,
      "grad_norm": 3.112410545349121,
      "learning_rate": 2.0666666666666666e-05,
      "loss": 0.5678,
      "step": 11400
    },
    {
      "epoch": 2.9468536867841975,
      "grad_norm": 3.0859549045562744,
      "learning_rate": 2.0537966537966538e-05,
      "loss": 0.5203,
      "step": 11450
    },
    {
      "epoch": 2.9597220434950455,
      "grad_norm": 2.518927812576294,
      "learning_rate": 2.0409266409266413e-05,
      "loss": 0.5326,
      "step": 11500
    },
    {
      "epoch": 2.9725904002058936,
      "grad_norm": 3.686985492706299,
      "learning_rate": 2.028056628056628e-05,
      "loss": 0.5419,
      "step": 11550
    },
    {
      "epoch": 2.9854587569167417,
      "grad_norm": 3.1635522842407227,
      "learning_rate": 2.0151866151866153e-05,
      "loss": 0.5593,
      "step": 11600
    },
    {
      "epoch": 2.9983271136275897,
      "grad_norm": 3.734297037124634,
      "learning_rate": 2.0023166023166024e-05,
      "loss": 0.5233,
      "step": 11650
    },
    {
      "epoch": 3.011195470338438,
      "grad_norm": 4.363897800445557,
      "learning_rate": 1.9894465894465893e-05,
      "loss": 0.3853,
      "step": 11700
    },
    {
      "epoch": 3.024063827049286,
      "grad_norm": 3.380840539932251,
      "learning_rate": 1.9765765765765768e-05,
      "loss": 0.4048,
      "step": 11750
    },
    {
      "epoch": 3.036932183760134,
      "grad_norm": 2.8301613330841064,
      "learning_rate": 1.963706563706564e-05,
      "loss": 0.4104,
      "step": 11800
    },
    {
      "epoch": 3.049800540470982,
      "grad_norm": 3.5483977794647217,
      "learning_rate": 1.9508365508365508e-05,
      "loss": 0.4002,
      "step": 11850
    },
    {
      "epoch": 3.0626688971818297,
      "grad_norm": 3.0657596588134766,
      "learning_rate": 1.937966537966538e-05,
      "loss": 0.3955,
      "step": 11900
    },
    {
      "epoch": 3.0755372538926777,
      "grad_norm": 3.3583223819732666,
      "learning_rate": 1.9250965250965254e-05,
      "loss": 0.4258,
      "step": 11950
    },
    {
      "epoch": 3.088405610603526,
      "grad_norm": 2.8227546215057373,
      "learning_rate": 1.9122265122265122e-05,
      "loss": 0.4279,
      "step": 12000
    },
    {
      "epoch": 3.088405610603526,
      "eval_loss": 0.8435577154159546,
      "eval_runtime": 189.2953,
      "eval_samples_per_second": 102.623,
      "eval_steps_per_second": 10.264,
      "step": 12000
    },
    {
      "epoch": 3.101273967314374,
      "grad_norm": 3.359844923019409,
      "learning_rate": 1.8993564993564994e-05,
      "loss": 0.4204,
      "step": 12050
    },
    {
      "epoch": 3.114142324025222,
      "grad_norm": 3.8054656982421875,
      "learning_rate": 1.8864864864864866e-05,
      "loss": 0.419,
      "step": 12100
    },
    {
      "epoch": 3.12701068073607,
      "grad_norm": 3.396296977996826,
      "learning_rate": 1.8736164736164737e-05,
      "loss": 0.4167,
      "step": 12150
    },
    {
      "epoch": 3.139879037446918,
      "grad_norm": 3.7082102298736572,
      "learning_rate": 1.860746460746461e-05,
      "loss": 0.4047,
      "step": 12200
    },
    {
      "epoch": 3.152747394157766,
      "grad_norm": 3.589115619659424,
      "learning_rate": 1.847876447876448e-05,
      "loss": 0.4245,
      "step": 12250
    },
    {
      "epoch": 3.1656157508686142,
      "grad_norm": 4.928313732147217,
      "learning_rate": 1.835006435006435e-05,
      "loss": 0.4054,
      "step": 12300
    },
    {
      "epoch": 3.178484107579462,
      "grad_norm": 3.3157694339752197,
      "learning_rate": 1.8221364221364224e-05,
      "loss": 0.3828,
      "step": 12350
    },
    {
      "epoch": 3.19135246429031,
      "grad_norm": 2.832195997238159,
      "learning_rate": 1.8092664092664095e-05,
      "loss": 0.3985,
      "step": 12400
    },
    {
      "epoch": 3.204220821001158,
      "grad_norm": 3.6316025257110596,
      "learning_rate": 1.7963963963963963e-05,
      "loss": 0.4009,
      "step": 12450
    },
    {
      "epoch": 3.217089177712006,
      "grad_norm": 2.810399293899536,
      "learning_rate": 1.7835263835263835e-05,
      "loss": 0.3942,
      "step": 12500
    },
    {
      "epoch": 3.229957534422854,
      "grad_norm": 4.077607154846191,
      "learning_rate": 1.770656370656371e-05,
      "loss": 0.4139,
      "step": 12550
    },
    {
      "epoch": 3.2428258911337022,
      "grad_norm": 4.511129856109619,
      "learning_rate": 1.7577863577863578e-05,
      "loss": 0.4151,
      "step": 12600
    },
    {
      "epoch": 3.2556942478445503,
      "grad_norm": 4.129230976104736,
      "learning_rate": 1.744916344916345e-05,
      "loss": 0.4174,
      "step": 12650
    },
    {
      "epoch": 3.2685626045553984,
      "grad_norm": 2.6795713901519775,
      "learning_rate": 1.732046332046332e-05,
      "loss": 0.4305,
      "step": 12700
    },
    {
      "epoch": 3.2814309612662464,
      "grad_norm": 3.7392494678497314,
      "learning_rate": 1.7191763191763193e-05,
      "loss": 0.4221,
      "step": 12750
    },
    {
      "epoch": 3.294299317977094,
      "grad_norm": 3.319235324859619,
      "learning_rate": 1.7063063063063065e-05,
      "loss": 0.4506,
      "step": 12800
    },
    {
      "epoch": 3.3071676746879426,
      "grad_norm": 3.564242362976074,
      "learning_rate": 1.6934362934362936e-05,
      "loss": 0.4287,
      "step": 12850
    },
    {
      "epoch": 3.32003603139879,
      "grad_norm": 4.851574420928955,
      "learning_rate": 1.6805662805662805e-05,
      "loss": 0.4002,
      "step": 12900
    },
    {
      "epoch": 3.3329043881096383,
      "grad_norm": 3.396198034286499,
      "learning_rate": 1.667696267696268e-05,
      "loss": 0.3979,
      "step": 12950
    },
    {
      "epoch": 3.3457727448204864,
      "grad_norm": 2.8444738388061523,
      "learning_rate": 1.654826254826255e-05,
      "loss": 0.4015,
      "step": 13000
    },
    {
      "epoch": 3.3457727448204864,
      "eval_loss": 0.8406257033348083,
      "eval_runtime": 189.5238,
      "eval_samples_per_second": 102.499,
      "eval_steps_per_second": 10.252,
      "step": 13000
    },
    {
      "epoch": 3.3586411015313344,
      "grad_norm": 3.6319420337677,
      "learning_rate": 1.641956241956242e-05,
      "loss": 0.4175,
      "step": 13050
    },
    {
      "epoch": 3.3715094582421825,
      "grad_norm": 3.819413900375366,
      "learning_rate": 1.629086229086229e-05,
      "loss": 0.4132,
      "step": 13100
    },
    {
      "epoch": 3.3843778149530306,
      "grad_norm": 3.5537872314453125,
      "learning_rate": 1.6162162162162163e-05,
      "loss": 0.4347,
      "step": 13150
    },
    {
      "epoch": 3.3972461716638787,
      "grad_norm": 3.193889617919922,
      "learning_rate": 1.6033462033462034e-05,
      "loss": 0.4082,
      "step": 13200
    },
    {
      "epoch": 3.4101145283747267,
      "grad_norm": 2.0629472732543945,
      "learning_rate": 1.5904761904761906e-05,
      "loss": 0.4157,
      "step": 13250
    },
    {
      "epoch": 3.422982885085575,
      "grad_norm": 3.1376850605010986,
      "learning_rate": 1.5776061776061777e-05,
      "loss": 0.4063,
      "step": 13300
    },
    {
      "epoch": 3.4358512417964224,
      "grad_norm": 3.6547293663024902,
      "learning_rate": 1.5647361647361646e-05,
      "loss": 0.4006,
      "step": 13350
    },
    {
      "epoch": 3.4487195985072705,
      "grad_norm": 3.7299160957336426,
      "learning_rate": 1.551866151866152e-05,
      "loss": 0.4103,
      "step": 13400
    },
    {
      "epoch": 3.4615879552181186,
      "grad_norm": 4.016064167022705,
      "learning_rate": 1.5389961389961392e-05,
      "loss": 0.4128,
      "step": 13450
    },
    {
      "epoch": 3.4744563119289666,
      "grad_norm": 3.9914636611938477,
      "learning_rate": 1.526126126126126e-05,
      "loss": 0.4417,
      "step": 13500
    },
    {
      "epoch": 3.4873246686398147,
      "grad_norm": 3.342193126678467,
      "learning_rate": 1.5132561132561132e-05,
      "loss": 0.4369,
      "step": 13550
    },
    {
      "epoch": 3.500193025350663,
      "grad_norm": 3.600562810897827,
      "learning_rate": 1.5003861003861005e-05,
      "loss": 0.4186,
      "step": 13600
    },
    {
      "epoch": 3.513061382061511,
      "grad_norm": 2.986332654953003,
      "learning_rate": 1.4875160875160877e-05,
      "loss": 0.4109,
      "step": 13650
    },
    {
      "epoch": 3.525929738772359,
      "grad_norm": 3.3687069416046143,
      "learning_rate": 1.4746460746460747e-05,
      "loss": 0.4193,
      "step": 13700
    },
    {
      "epoch": 3.538798095483207,
      "grad_norm": 3.519691228866577,
      "learning_rate": 1.4617760617760617e-05,
      "loss": 0.4046,
      "step": 13750
    },
    {
      "epoch": 3.5516664521940546,
      "grad_norm": 2.984351396560669,
      "learning_rate": 1.448906048906049e-05,
      "loss": 0.4013,
      "step": 13800
    },
    {
      "epoch": 3.5645348089049027,
      "grad_norm": 4.0301899909973145,
      "learning_rate": 1.4360360360360362e-05,
      "loss": 0.425,
      "step": 13850
    },
    {
      "epoch": 3.577403165615751,
      "grad_norm": 3.606745481491089,
      "learning_rate": 1.4231660231660232e-05,
      "loss": 0.395,
      "step": 13900
    },
    {
      "epoch": 3.590271522326599,
      "grad_norm": 3.490335702896118,
      "learning_rate": 1.4102960102960103e-05,
      "loss": 0.4104,
      "step": 13950
    },
    {
      "epoch": 3.603139879037447,
      "grad_norm": 4.385608673095703,
      "learning_rate": 1.3974259974259976e-05,
      "loss": 0.4235,
      "step": 14000
    },
    {
      "epoch": 3.603139879037447,
      "eval_loss": 0.8373395800590515,
      "eval_runtime": 189.5425,
      "eval_samples_per_second": 102.489,
      "eval_steps_per_second": 10.251,
      "step": 14000
    },
    {
      "epoch": 3.616008235748295,
      "grad_norm": 3.1057236194610596,
      "learning_rate": 1.3845559845559846e-05,
      "loss": 0.4055,
      "step": 14050
    },
    {
      "epoch": 3.628876592459143,
      "grad_norm": 3.3940112590789795,
      "learning_rate": 1.3716859716859718e-05,
      "loss": 0.3916,
      "step": 14100
    },
    {
      "epoch": 3.641744949169991,
      "grad_norm": 3.4843504428863525,
      "learning_rate": 1.3588159588159588e-05,
      "loss": 0.4483,
      "step": 14150
    },
    {
      "epoch": 3.654613305880839,
      "grad_norm": 4.122684001922607,
      "learning_rate": 1.3462033462033463e-05,
      "loss": 0.4241,
      "step": 14200
    },
    {
      "epoch": 3.667481662591687,
      "grad_norm": 3.975159168243408,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.417,
      "step": 14250
    },
    {
      "epoch": 3.680350019302535,
      "grad_norm": 4.270761966705322,
      "learning_rate": 1.3204633204633205e-05,
      "loss": 0.4137,
      "step": 14300
    },
    {
      "epoch": 3.693218376013383,
      "grad_norm": 2.448530673980713,
      "learning_rate": 1.307850707850708e-05,
      "loss": 0.4277,
      "step": 14350
    },
    {
      "epoch": 3.706086732724231,
      "grad_norm": 3.0755138397216797,
      "learning_rate": 1.294980694980695e-05,
      "loss": 0.4003,
      "step": 14400
    },
    {
      "epoch": 3.718955089435079,
      "grad_norm": 3.795004367828369,
      "learning_rate": 1.282110682110682e-05,
      "loss": 0.4128,
      "step": 14450
    },
    {
      "epoch": 3.731823446145927,
      "grad_norm": 3.655588150024414,
      "learning_rate": 1.2692406692406693e-05,
      "loss": 0.4243,
      "step": 14500
    },
    {
      "epoch": 3.7446918028567753,
      "grad_norm": 3.4732940196990967,
      "learning_rate": 1.2563706563706565e-05,
      "loss": 0.432,
      "step": 14550
    },
    {
      "epoch": 3.7575601595676233,
      "grad_norm": 3.737543821334839,
      "learning_rate": 1.2435006435006435e-05,
      "loss": 0.4214,
      "step": 14600
    },
    {
      "epoch": 3.7704285162784714,
      "grad_norm": 4.61754035949707,
      "learning_rate": 1.2306306306306308e-05,
      "loss": 0.4075,
      "step": 14650
    },
    {
      "epoch": 3.783296872989319,
      "grad_norm": 3.1483190059661865,
      "learning_rate": 1.2177606177606178e-05,
      "loss": 0.4136,
      "step": 14700
    },
    {
      "epoch": 3.7961652297001676,
      "grad_norm": 3.9902536869049072,
      "learning_rate": 1.204890604890605e-05,
      "loss": 0.399,
      "step": 14750
    },
    {
      "epoch": 3.809033586411015,
      "grad_norm": 3.6237611770629883,
      "learning_rate": 1.1920205920205921e-05,
      "loss": 0.4144,
      "step": 14800
    },
    {
      "epoch": 3.8219019431218633,
      "grad_norm": 4.140763282775879,
      "learning_rate": 1.1791505791505791e-05,
      "loss": 0.4009,
      "step": 14850
    },
    {
      "epoch": 3.8347702998327113,
      "grad_norm": 3.700617790222168,
      "learning_rate": 1.1662805662805663e-05,
      "loss": 0.3993,
      "step": 14900
    },
    {
      "epoch": 3.8476386565435594,
      "grad_norm": 3.43810772895813,
      "learning_rate": 1.1534105534105535e-05,
      "loss": 0.4153,
      "step": 14950
    },
    {
      "epoch": 3.8605070132544075,
      "grad_norm": 3.04858136177063,
      "learning_rate": 1.1405405405405406e-05,
      "loss": 0.4056,
      "step": 15000
    },
    {
      "epoch": 3.8605070132544075,
      "eval_loss": 0.8282376527786255,
      "eval_runtime": 189.5165,
      "eval_samples_per_second": 102.503,
      "eval_steps_per_second": 10.252,
      "step": 15000
    },
    {
      "epoch": 3.8733753699652556,
      "grad_norm": 3.8784329891204834,
      "learning_rate": 1.1276705276705276e-05,
      "loss": 0.3866,
      "step": 15050
    },
    {
      "epoch": 3.8862437266761036,
      "grad_norm": 3.622274398803711,
      "learning_rate": 1.114800514800515e-05,
      "loss": 0.4208,
      "step": 15100
    },
    {
      "epoch": 3.8991120833869513,
      "grad_norm": 3.2869110107421875,
      "learning_rate": 1.101930501930502e-05,
      "loss": 0.4302,
      "step": 15150
    },
    {
      "epoch": 3.9119804400977998,
      "grad_norm": 2.9988114833831787,
      "learning_rate": 1.0890604890604891e-05,
      "loss": 0.4153,
      "step": 15200
    },
    {
      "epoch": 3.9248487968086474,
      "grad_norm": 3.2975871562957764,
      "learning_rate": 1.0761904761904763e-05,
      "loss": 0.4178,
      "step": 15250
    },
    {
      "epoch": 3.9377171535194955,
      "grad_norm": 2.9260878562927246,
      "learning_rate": 1.0633204633204634e-05,
      "loss": 0.4048,
      "step": 15300
    },
    {
      "epoch": 3.9505855102303435,
      "grad_norm": 3.6288230419158936,
      "learning_rate": 1.0504504504504504e-05,
      "loss": 0.4007,
      "step": 15350
    },
    {
      "epoch": 3.9634538669411916,
      "grad_norm": 4.0753173828125,
      "learning_rate": 1.0375804375804377e-05,
      "loss": 0.397,
      "step": 15400
    },
    {
      "epoch": 3.9763222236520397,
      "grad_norm": 3.9990575313568115,
      "learning_rate": 1.0247104247104247e-05,
      "loss": 0.4043,
      "step": 15450
    },
    {
      "epoch": 3.9891905803628878,
      "grad_norm": 3.895235776901245,
      "learning_rate": 1.0118404118404119e-05,
      "loss": 0.4147,
      "step": 15500
    },
    {
      "epoch": 4.002058937073736,
      "grad_norm": 3.6188671588897705,
      "learning_rate": 9.98970398970399e-06,
      "loss": 0.4038,
      "step": 15550
    },
    {
      "epoch": 4.0149272937845835,
      "grad_norm": 2.461876392364502,
      "learning_rate": 9.861003861003862e-06,
      "loss": 0.3183,
      "step": 15600
    },
    {
      "epoch": 4.027795650495432,
      "grad_norm": 3.0858776569366455,
      "learning_rate": 9.732303732303732e-06,
      "loss": 0.3477,
      "step": 15650
    },
    {
      "epoch": 4.04066400720628,
      "grad_norm": 3.3975493907928467,
      "learning_rate": 9.603603603603605e-06,
      "loss": 0.3225,
      "step": 15700
    },
    {
      "epoch": 4.053532363917128,
      "grad_norm": 3.6678659915924072,
      "learning_rate": 9.474903474903475e-06,
      "loss": 0.3294,
      "step": 15750
    },
    {
      "epoch": 4.066400720627976,
      "grad_norm": 3.5393424034118652,
      "learning_rate": 9.346203346203347e-06,
      "loss": 0.3421,
      "step": 15800
    },
    {
      "epoch": 4.079269077338824,
      "grad_norm": 3.039139747619629,
      "learning_rate": 9.217503217503218e-06,
      "loss": 0.328,
      "step": 15850
    },
    {
      "epoch": 4.092137434049672,
      "grad_norm": 4.036282062530518,
      "learning_rate": 9.08880308880309e-06,
      "loss": 0.3287,
      "step": 15900
    },
    {
      "epoch": 4.1050057907605195,
      "grad_norm": 3.0405406951904297,
      "learning_rate": 8.96010296010296e-06,
      "loss": 0.3335,
      "step": 15950
    },
    {
      "epoch": 4.117874147471368,
      "grad_norm": 2.893115520477295,
      "learning_rate": 8.831402831402833e-06,
      "loss": 0.3092,
      "step": 16000
    },
    {
      "epoch": 4.117874147471368,
      "eval_loss": 0.8520528674125671,
      "eval_runtime": 189.4809,
      "eval_samples_per_second": 102.522,
      "eval_steps_per_second": 10.254,
      "step": 16000
    },
    {
      "epoch": 4.130742504182216,
      "grad_norm": 4.632415771484375,
      "learning_rate": 8.702702702702703e-06,
      "loss": 0.3281,
      "step": 16050
    },
    {
      "epoch": 4.143610860893064,
      "grad_norm": 3.4265694618225098,
      "learning_rate": 8.574002574002575e-06,
      "loss": 0.3147,
      "step": 16100
    },
    {
      "epoch": 4.156479217603912,
      "grad_norm": 2.336419105529785,
      "learning_rate": 8.445302445302446e-06,
      "loss": 0.3013,
      "step": 16150
    },
    {
      "epoch": 4.16934757431476,
      "grad_norm": 3.3826658725738525,
      "learning_rate": 8.316602316602316e-06,
      "loss": 0.3124,
      "step": 16200
    },
    {
      "epoch": 4.182215931025608,
      "grad_norm": 2.8319571018218994,
      "learning_rate": 8.187902187902188e-06,
      "loss": 0.3503,
      "step": 16250
    },
    {
      "epoch": 4.1950842877364565,
      "grad_norm": 5.399038314819336,
      "learning_rate": 8.05920205920206e-06,
      "loss": 0.3394,
      "step": 16300
    },
    {
      "epoch": 4.207952644447304,
      "grad_norm": 2.59490966796875,
      "learning_rate": 7.930501930501931e-06,
      "loss": 0.3317,
      "step": 16350
    },
    {
      "epoch": 4.220821001158152,
      "grad_norm": 3.371151924133301,
      "learning_rate": 7.801801801801801e-06,
      "loss": 0.3123,
      "step": 16400
    },
    {
      "epoch": 4.233689357869,
      "grad_norm": 3.447014808654785,
      "learning_rate": 7.673101673101674e-06,
      "loss": 0.3348,
      "step": 16450
    },
    {
      "epoch": 4.246557714579848,
      "grad_norm": 4.9008893966674805,
      "learning_rate": 7.544401544401544e-06,
      "loss": 0.34,
      "step": 16500
    },
    {
      "epoch": 4.259426071290696,
      "grad_norm": 3.1242849826812744,
      "learning_rate": 7.415701415701416e-06,
      "loss": 0.3064,
      "step": 16550
    },
    {
      "epoch": 4.272294428001544,
      "grad_norm": 4.190210342407227,
      "learning_rate": 7.287001287001287e-06,
      "loss": 0.3316,
      "step": 16600
    },
    {
      "epoch": 4.2851627847123925,
      "grad_norm": 4.833367824554443,
      "learning_rate": 7.158301158301159e-06,
      "loss": 0.3151,
      "step": 16650
    },
    {
      "epoch": 4.29803114142324,
      "grad_norm": 3.561443328857422,
      "learning_rate": 7.02960102960103e-06,
      "loss": 0.3284,
      "step": 16700
    },
    {
      "epoch": 4.310899498134089,
      "grad_norm": 3.202888011932373,
      "learning_rate": 6.903474903474904e-06,
      "loss": 0.3361,
      "step": 16750
    },
    {
      "epoch": 4.323767854844936,
      "grad_norm": 4.517894744873047,
      "learning_rate": 6.774774774774775e-06,
      "loss": 0.3371,
      "step": 16800
    },
    {
      "epoch": 4.336636211555784,
      "grad_norm": 3.958711862564087,
      "learning_rate": 6.646074646074646e-06,
      "loss": 0.3236,
      "step": 16850
    },
    {
      "epoch": 4.3495045682666325,
      "grad_norm": 2.8120596408843994,
      "learning_rate": 6.517374517374518e-06,
      "loss": 0.3128,
      "step": 16900
    },
    {
      "epoch": 4.36237292497748,
      "grad_norm": 2.320220947265625,
      "learning_rate": 6.388674388674388e-06,
      "loss": 0.3079,
      "step": 16950
    },
    {
      "epoch": 4.375241281688329,
      "grad_norm": 2.833205223083496,
      "learning_rate": 6.259974259974261e-06,
      "loss": 0.3127,
      "step": 17000
    },
    {
      "epoch": 4.375241281688329,
      "eval_loss": 0.8534464836120605,
      "eval_runtime": 189.5459,
      "eval_samples_per_second": 102.487,
      "eval_steps_per_second": 10.251,
      "step": 17000
    },
    {
      "epoch": 4.388109638399176,
      "grad_norm": 4.218430519104004,
      "learning_rate": 6.1312741312741316e-06,
      "loss": 0.3375,
      "step": 17050
    },
    {
      "epoch": 4.400977995110025,
      "grad_norm": 3.2581334114074707,
      "learning_rate": 6.002574002574003e-06,
      "loss": 0.3081,
      "step": 17100
    },
    {
      "epoch": 4.413846351820872,
      "grad_norm": 3.2041099071502686,
      "learning_rate": 5.873873873873874e-06,
      "loss": 0.3368,
      "step": 17150
    },
    {
      "epoch": 4.426714708531721,
      "grad_norm": 4.340978622436523,
      "learning_rate": 5.7451737451737455e-06,
      "loss": 0.3404,
      "step": 17200
    },
    {
      "epoch": 4.4395830652425685,
      "grad_norm": 4.946235179901123,
      "learning_rate": 5.616473616473616e-06,
      "loss": 0.3199,
      "step": 17250
    },
    {
      "epoch": 4.452451421953416,
      "grad_norm": 3.671445846557617,
      "learning_rate": 5.487773487773488e-06,
      "loss": 0.3137,
      "step": 17300
    },
    {
      "epoch": 4.465319778664265,
      "grad_norm": 4.509451389312744,
      "learning_rate": 5.3590733590733595e-06,
      "loss": 0.332,
      "step": 17350
    },
    {
      "epoch": 4.478188135375112,
      "grad_norm": 3.3261749744415283,
      "learning_rate": 5.23037323037323e-06,
      "loss": 0.3339,
      "step": 17400
    },
    {
      "epoch": 4.491056492085961,
      "grad_norm": 3.5157628059387207,
      "learning_rate": 5.101673101673102e-06,
      "loss": 0.3256,
      "step": 17450
    },
    {
      "epoch": 4.503924848796808,
      "grad_norm": 3.7138776779174805,
      "learning_rate": 4.9729729729729735e-06,
      "loss": 0.3278,
      "step": 17500
    },
    {
      "epoch": 4.516793205507657,
      "grad_norm": 3.2979509830474854,
      "learning_rate": 4.844272844272844e-06,
      "loss": 0.3249,
      "step": 17550
    },
    {
      "epoch": 4.529661562218505,
      "grad_norm": 3.117856740951538,
      "learning_rate": 4.715572715572716e-06,
      "loss": 0.3174,
      "step": 17600
    },
    {
      "epoch": 4.542529918929353,
      "grad_norm": 3.44389009475708,
      "learning_rate": 4.586872586872587e-06,
      "loss": 0.3072,
      "step": 17650
    },
    {
      "epoch": 4.555398275640201,
      "grad_norm": 4.093995094299316,
      "learning_rate": 4.458172458172458e-06,
      "loss": 0.3118,
      "step": 17700
    },
    {
      "epoch": 4.568266632351049,
      "grad_norm": 2.943690061569214,
      "learning_rate": 4.32947232947233e-06,
      "loss": 0.3123,
      "step": 17750
    },
    {
      "epoch": 4.581134989061897,
      "grad_norm": 3.0897843837738037,
      "learning_rate": 4.200772200772201e-06,
      "loss": 0.3194,
      "step": 17800
    },
    {
      "epoch": 4.5940033457727445,
      "grad_norm": 3.8259048461914062,
      "learning_rate": 4.072072072072072e-06,
      "loss": 0.3321,
      "step": 17850
    },
    {
      "epoch": 4.606871702483593,
      "grad_norm": 4.632271766662598,
      "learning_rate": 3.943371943371944e-06,
      "loss": 0.3223,
      "step": 17900
    },
    {
      "epoch": 4.619740059194441,
      "grad_norm": 3.449892282485962,
      "learning_rate": 3.814671814671815e-06,
      "loss": 0.3152,
      "step": 17950
    },
    {
      "epoch": 4.632608415905289,
      "grad_norm": 4.951483726501465,
      "learning_rate": 3.685971685971686e-06,
      "loss": 0.3333,
      "step": 18000
    },
    {
      "epoch": 4.632608415905289,
      "eval_loss": 0.8536800742149353,
      "eval_runtime": 189.5498,
      "eval_samples_per_second": 102.485,
      "eval_steps_per_second": 10.251,
      "step": 18000
    },
    {
      "epoch": 4.645476772616137,
      "grad_norm": 2.879723310470581,
      "learning_rate": 3.5572715572715578e-06,
      "loss": 0.3311,
      "step": 18050
    },
    {
      "epoch": 4.658345129326985,
      "grad_norm": 2.263468027114868,
      "learning_rate": 3.428571428571429e-06,
      "loss": 0.3288,
      "step": 18100
    },
    {
      "epoch": 4.671213486037833,
      "grad_norm": 4.122782230377197,
      "learning_rate": 3.2998712998713e-06,
      "loss": 0.2881,
      "step": 18150
    },
    {
      "epoch": 4.684081842748681,
      "grad_norm": 2.713583469390869,
      "learning_rate": 3.1711711711711718e-06,
      "loss": 0.335,
      "step": 18200
    },
    {
      "epoch": 4.696950199459529,
      "grad_norm": 4.133466720581055,
      "learning_rate": 3.0424710424710425e-06,
      "loss": 0.3389,
      "step": 18250
    },
    {
      "epoch": 4.709818556170377,
      "grad_norm": 3.080181121826172,
      "learning_rate": 2.9137709137709137e-06,
      "loss": 0.3049,
      "step": 18300
    },
    {
      "epoch": 4.722686912881225,
      "grad_norm": 3.2087597846984863,
      "learning_rate": 2.7850707850707853e-06,
      "loss": 0.3021,
      "step": 18350
    },
    {
      "epoch": 4.735555269592073,
      "grad_norm": 3.037031888961792,
      "learning_rate": 2.6563706563706565e-06,
      "loss": 0.3343,
      "step": 18400
    },
    {
      "epoch": 4.748423626302921,
      "grad_norm": 3.936049699783325,
      "learning_rate": 2.5276705276705277e-06,
      "loss": 0.351,
      "step": 18450
    },
    {
      "epoch": 4.761291983013769,
      "grad_norm": 2.863168239593506,
      "learning_rate": 2.3989703989703993e-06,
      "loss": 0.3062,
      "step": 18500
    },
    {
      "epoch": 4.7741603397246175,
      "grad_norm": 3.105039358139038,
      "learning_rate": 2.2702702702702705e-06,
      "loss": 0.3139,
      "step": 18550
    },
    {
      "epoch": 4.787028696435465,
      "grad_norm": 4.5207390785217285,
      "learning_rate": 2.1415701415701416e-06,
      "loss": 0.3295,
      "step": 18600
    },
    {
      "epoch": 4.799897053146314,
      "grad_norm": 4.3474345207214355,
      "learning_rate": 2.0128700128700133e-06,
      "loss": 0.3256,
      "step": 18650
    },
    {
      "epoch": 4.812765409857161,
      "grad_norm": 3.58796763420105,
      "learning_rate": 1.8841698841698844e-06,
      "loss": 0.32,
      "step": 18700
    },
    {
      "epoch": 4.825633766568009,
      "grad_norm": 2.583491325378418,
      "learning_rate": 1.7554697554697556e-06,
      "loss": 0.3409,
      "step": 18750
    },
    {
      "epoch": 4.838502123278857,
      "grad_norm": 2.490678071975708,
      "learning_rate": 1.6267696267696266e-06,
      "loss": 0.3164,
      "step": 18800
    },
    {
      "epoch": 4.851370479989705,
      "grad_norm": 2.505882740020752,
      "learning_rate": 1.4980694980694982e-06,
      "loss": 0.2992,
      "step": 18850
    },
    {
      "epoch": 4.864238836700554,
      "grad_norm": 3.8381474018096924,
      "learning_rate": 1.371943371943372e-06,
      "loss": 0.3178,
      "step": 18900
    },
    {
      "epoch": 4.877107193411401,
      "grad_norm": 4.065976619720459,
      "learning_rate": 1.2432432432432434e-06,
      "loss": 0.312,
      "step": 18950
    },
    {
      "epoch": 4.88997555012225,
      "grad_norm": 2.3685247898101807,
      "learning_rate": 1.1145431145431146e-06,
      "loss": 0.3103,
      "step": 19000
    },
    {
      "epoch": 4.88997555012225,
      "eval_loss": 0.8551310896873474,
      "eval_runtime": 189.5403,
      "eval_samples_per_second": 102.49,
      "eval_steps_per_second": 10.251,
      "step": 19000
    },
    {
      "epoch": 4.902843906833097,
      "grad_norm": 1.829939365386963,
      "learning_rate": 9.85842985842986e-07,
      "loss": 0.3012,
      "step": 19050
    },
    {
      "epoch": 4.915712263543945,
      "grad_norm": 3.9715230464935303,
      "learning_rate": 8.571428571428572e-07,
      "loss": 0.3258,
      "step": 19100
    },
    {
      "epoch": 4.9285806202547935,
      "grad_norm": 3.2242319583892822,
      "learning_rate": 7.284427284427284e-07,
      "loss": 0.319,
      "step": 19150
    },
    {
      "epoch": 4.941448976965641,
      "grad_norm": 2.4675443172454834,
      "learning_rate": 5.997425997425998e-07,
      "loss": 0.3183,
      "step": 19200
    },
    {
      "epoch": 4.95431733367649,
      "grad_norm": 4.066344738006592,
      "learning_rate": 4.710424710424711e-07,
      "loss": 0.3368,
      "step": 19250
    },
    {
      "epoch": 4.967185690387337,
      "grad_norm": 3.9160056114196777,
      "learning_rate": 3.423423423423424e-07,
      "loss": 0.3002,
      "step": 19300
    },
    {
      "epoch": 4.980054047098186,
      "grad_norm": 3.2470641136169434,
      "learning_rate": 2.1364221364221366e-07,
      "loss": 0.3168,
      "step": 19350
    },
    {
      "epoch": 4.992922403809033,
      "grad_norm": 4.328388690948486,
      "learning_rate": 8.494208494208495e-08,
      "loss": 0.3105,
      "step": 19400
    }
  ],
  "logging_steps": 50,
  "max_steps": 19425,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 9.343651794926285e+17,
  "train_batch_size": 10,
  "trial_name": null,
  "trial_params": null
}