{
  "best_metric": 3.7425780296325684,
  "best_model_checkpoint": "./results/models/checkpoint-406536",
  "epoch": 26.0,
  "eval_steps": 500,
  "global_step": 406536,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.031977487848554614,
      "grad_norm": 0.248046875,
      "learning_rate": 0.000999360450243029,
      "loss": 1.941,
      "step": 500
    },
    {
      "epoch": 0.06395497569710923,
      "grad_norm": 0.205078125,
      "learning_rate": 0.0009987209004860577,
      "loss": 1.9127,
      "step": 1000
    },
    {
      "epoch": 0.09593246354566386,
      "grad_norm": 0.189453125,
      "learning_rate": 0.0009980813507290869,
      "loss": 1.9075,
      "step": 1500
    },
    {
      "epoch": 0.12790995139421846,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0009974418009721156,
      "loss": 1.9087,
      "step": 2000
    },
    {
      "epoch": 0.1598874392427731,
      "grad_norm": 0.3984375,
      "learning_rate": 0.0009968022512151446,
      "loss": 1.9096,
      "step": 2500
    },
    {
      "epoch": 0.1918649270913277,
      "grad_norm": 0.375,
      "learning_rate": 0.0009961627014581735,
      "loss": 1.9154,
      "step": 3000
    },
    {
      "epoch": 0.22384241493988233,
      "grad_norm": 0.6953125,
      "learning_rate": 0.0009955231517012023,
      "loss": 1.9201,
      "step": 3500
    },
    {
      "epoch": 0.2558199027884369,
      "grad_norm": 0.212890625,
      "learning_rate": 0.0009948836019442312,
      "loss": 1.9187,
      "step": 4000
    },
    {
      "epoch": 0.28779739063699156,
      "grad_norm": 0.3828125,
      "learning_rate": 0.0009942440521872602,
      "loss": 1.9125,
      "step": 4500
    },
    {
      "epoch": 0.3197748784855462,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0009936045024302891,
      "loss": 1.9153,
      "step": 5000
    },
    {
      "epoch": 0.3517523663341008,
      "grad_norm": 0.29296875,
      "learning_rate": 0.000992964952673318,
      "loss": 1.9143,
      "step": 5500
    },
    {
      "epoch": 0.3837298541826554,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0009923254029163468,
      "loss": 1.9124,
      "step": 6000
    },
    {
      "epoch": 0.41570734203121,
      "grad_norm": 0.263671875,
      "learning_rate": 0.0009916858531593758,
      "loss": 1.9106,
      "step": 6500
    },
    {
      "epoch": 0.44768482987976466,
      "grad_norm": 0.419921875,
      "learning_rate": 0.0009910463034024047,
      "loss": 1.9127,
      "step": 7000
    },
    {
      "epoch": 0.47966231772831924,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0009904067536454337,
      "loss": 1.916,
      "step": 7500
    },
    {
      "epoch": 0.5116398055768738,
      "grad_norm": 0.369140625,
      "learning_rate": 0.0009897672038884624,
      "loss": 1.9145,
      "step": 8000
    },
    {
      "epoch": 0.5436172934254285,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0009891276541314914,
      "loss": 1.9158,
      "step": 8500
    },
    {
      "epoch": 0.5755947812739831,
      "grad_norm": 0.28515625,
      "learning_rate": 0.0009884881043745204,
      "loss": 1.9134,
      "step": 9000
    },
    {
      "epoch": 0.6075722691225377,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0009878485546175493,
      "loss": 1.9113,
      "step": 9500
    },
    {
      "epoch": 0.6395497569710924,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0009872090048605783,
      "loss": 1.9112,
      "step": 10000
    },
    {
      "epoch": 0.671527244819647,
      "grad_norm": 0.72265625,
      "learning_rate": 0.000986569455103607,
      "loss": 1.9132,
      "step": 10500
    },
    {
      "epoch": 0.7035047326682016,
      "grad_norm": 0.44140625,
      "learning_rate": 0.000985929905346636,
      "loss": 1.9186,
      "step": 11000
    },
    {
      "epoch": 0.7354822205167562,
      "grad_norm": 0.34375,
      "learning_rate": 0.000985290355589665,
      "loss": 1.9154,
      "step": 11500
    },
    {
      "epoch": 0.7674597083653109,
      "grad_norm": 0.7265625,
      "learning_rate": 0.0009846508058326939,
      "loss": 1.9149,
      "step": 12000
    },
    {
      "epoch": 0.7994371962138654,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0009840112560757228,
      "loss": 1.9168,
      "step": 12500
    },
    {
      "epoch": 0.83141468406242,
      "grad_norm": 0.42578125,
      "learning_rate": 0.0009833717063187516,
      "loss": 1.9159,
      "step": 13000
    },
    {
      "epoch": 0.8633921719109747,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0009827321565617805,
      "loss": 1.916,
      "step": 13500
    },
    {
      "epoch": 0.8953696597595293,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0009820926068048095,
      "loss": 1.9138,
      "step": 14000
    },
    {
      "epoch": 0.9273471476080839,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0009814530570478384,
      "loss": 1.9106,
      "step": 14500
    },
    {
      "epoch": 0.9593246354566385,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0009808135072908672,
      "loss": 1.9103,
      "step": 15000
    },
    {
      "epoch": 0.9913021233051932,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0009801739575338961,
      "loss": 1.91,
      "step": 15500
    },
    {
      "epoch": 1.0,
      "eval_loss": 3.8195321559906006,
      "eval_runtime": 6.1933,
      "eval_samples_per_second": 80.732,
      "eval_steps_per_second": 1.292,
      "step": 15636
    },
    {
      "epoch": 1.0232796111537477,
      "grad_norm": 0.43359375,
      "learning_rate": 0.000979534407776925,
      "loss": 1.9079,
      "step": 16000
    },
    {
      "epoch": 1.0552570990023025,
      "grad_norm": 0.373046875,
      "learning_rate": 0.000978894858019954,
      "loss": 1.9065,
      "step": 16500
    },
    {
      "epoch": 1.087234586850857,
      "grad_norm": 0.283203125,
      "learning_rate": 0.000978255308262983,
      "loss": 1.9056,
      "step": 17000
    },
    {
      "epoch": 1.1192120746994116,
      "grad_norm": 0.36328125,
      "learning_rate": 0.0009776157585060117,
      "loss": 1.9043,
      "step": 17500
    },
    {
      "epoch": 1.1511895625479662,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0009769762087490407,
      "loss": 1.9049,
      "step": 18000
    },
    {
      "epoch": 1.1831670503965208,
      "grad_norm": 0.361328125,
      "learning_rate": 0.0009763366589920695,
      "loss": 1.9041,
      "step": 18500
    },
    {
      "epoch": 1.2151445382450754,
      "grad_norm": 0.375,
      "learning_rate": 0.0009756971092350985,
      "loss": 1.905,
      "step": 19000
    },
    {
      "epoch": 1.24712202609363,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0009750575594781274,
      "loss": 1.9047,
      "step": 19500
    },
    {
      "epoch": 1.2790995139421848,
      "grad_norm": 0.400390625,
      "learning_rate": 0.0009744180097211564,
      "loss": 1.9053,
      "step": 20000
    },
    {
      "epoch": 1.3110770017907394,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0009737784599641852,
      "loss": 1.9066,
      "step": 20500
    },
    {
      "epoch": 1.343054489639294,
      "grad_norm": 0.453125,
      "learning_rate": 0.0009731389102072141,
      "loss": 1.9097,
      "step": 21000
    },
    {
      "epoch": 1.3750319774878486,
      "grad_norm": 0.404296875,
      "learning_rate": 0.000972499360450243,
      "loss": 1.9072,
      "step": 21500
    },
    {
      "epoch": 1.4070094653364031,
      "grad_norm": 0.328125,
      "learning_rate": 0.000971859810693272,
      "loss": 1.905,
      "step": 22000
    },
    {
      "epoch": 1.4389869531849577,
      "grad_norm": 0.578125,
      "learning_rate": 0.0009712202609363008,
      "loss": 1.904,
      "step": 22500
    },
    {
      "epoch": 1.4709644410335123,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0009705807111793298,
      "loss": 1.9029,
      "step": 23000
    },
    {
      "epoch": 1.5029419288820671,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0009699411614223587,
      "loss": 1.9024,
      "step": 23500
    },
    {
      "epoch": 1.5349194167306215,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0009693016116653875,
      "loss": 1.9005,
      "step": 24000
    },
    {
      "epoch": 1.5668969045791763,
      "grad_norm": 0.341796875,
      "learning_rate": 0.0009686620619084166,
      "loss": 1.9027,
      "step": 24500
    },
    {
      "epoch": 1.5988743924277309,
      "grad_norm": 0.3125,
      "learning_rate": 0.0009680225121514454,
      "loss": 1.9011,
      "step": 25000
    },
    {
      "epoch": 1.6308518802762855,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0009673829623944744,
      "loss": 1.9012,
      "step": 25500
    },
    {
      "epoch": 1.66282936812484,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0009667434126375032,
      "loss": 1.9013,
      "step": 26000
    },
    {
      "epoch": 1.6948068559733946,
      "grad_norm": 0.296875,
      "learning_rate": 0.0009661038628805321,
      "loss": 1.9005,
      "step": 26500
    },
    {
      "epoch": 1.7267843438219495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000965464313123561,
      "loss": 1.9015,
      "step": 27000
    },
    {
      "epoch": 1.7587618316705038,
      "grad_norm": 0.96875,
      "learning_rate": 0.00096482476336659,
      "loss": 1.9019,
      "step": 27500
    },
    {
      "epoch": 1.7907393195190586,
      "grad_norm": 0.330078125,
      "learning_rate": 0.0009641852136096188,
      "loss": 1.8994,
      "step": 28000
    },
    {
      "epoch": 1.8227168073676132,
      "grad_norm": 0.28125,
      "learning_rate": 0.0009635456638526478,
      "loss": 1.8994,
      "step": 28500
    },
    {
      "epoch": 1.8546942952161678,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0009629061140956766,
      "loss": 1.9011,
      "step": 29000
    },
    {
      "epoch": 1.8866717830647224,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0009622665643387055,
      "loss": 1.8997,
      "step": 29500
    },
    {
      "epoch": 1.918649270913277,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0009616270145817345,
      "loss": 1.8997,
      "step": 30000
    },
    {
      "epoch": 1.9506267587618318,
      "grad_norm": 0.341796875,
      "learning_rate": 0.0009609874648247634,
      "loss": 1.899,
      "step": 30500
    },
    {
      "epoch": 1.9826042466103861,
      "grad_norm": 0.40625,
      "learning_rate": 0.0009603479150677923,
      "loss": 1.8971,
      "step": 31000
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.8006832599639893,
      "eval_runtime": 6.729,
      "eval_samples_per_second": 74.305,
      "eval_steps_per_second": 1.189,
      "step": 31272
    },
    {
      "epoch": 2.014581734458941,
      "grad_norm": 0.404296875,
      "learning_rate": 0.0009597083653108212,
      "loss": 1.8987,
      "step": 31500
    },
    {
      "epoch": 2.0465592223074953,
      "grad_norm": 0.578125,
      "learning_rate": 0.00095906881555385,
      "loss": 1.898,
      "step": 32000
    },
    {
      "epoch": 2.07853671015605,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0009584292657968791,
      "loss": 1.8984,
      "step": 32500
    },
    {
      "epoch": 2.110514198004605,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0009577897160399079,
      "loss": 1.8978,
      "step": 33000
    },
    {
      "epoch": 2.1424916858531593,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0009571501662829368,
      "loss": 1.8979,
      "step": 33500
    },
    {
      "epoch": 2.174469173701714,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0009565106165259657,
      "loss": 1.8974,
      "step": 34000
    },
    {
      "epoch": 2.2064466615502685,
      "grad_norm": 0.240234375,
      "learning_rate": 0.0009558710667689946,
      "loss": 1.8975,
      "step": 34500
    },
    {
      "epoch": 2.2384241493988233,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0009552315170120236,
      "loss": 1.8977,
      "step": 35000
    },
    {
      "epoch": 2.270401637247378,
      "grad_norm": 0.546875,
      "learning_rate": 0.0009545919672550525,
      "loss": 1.8967,
      "step": 35500
    },
    {
      "epoch": 2.3023791250959325,
      "grad_norm": 0.2138671875,
      "learning_rate": 0.0009539524174980813,
      "loss": 1.897,
      "step": 36000
    },
    {
      "epoch": 2.3343566129444873,
      "grad_norm": 0.26953125,
      "learning_rate": 0.0009533128677411103,
      "loss": 1.897,
      "step": 36500
    },
    {
      "epoch": 2.3663341007930416,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0009526733179841391,
      "loss": 1.8963,
      "step": 37000
    },
    {
      "epoch": 2.3983115886415964,
      "grad_norm": 0.3984375,
      "learning_rate": 0.0009520337682271681,
      "loss": 1.8962,
      "step": 37500
    },
    {
      "epoch": 2.430289076490151,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0009513942184701971,
      "loss": 1.8957,
      "step": 38000
    },
    {
      "epoch": 2.4622665643387056,
      "grad_norm": 0.2490234375,
      "learning_rate": 0.0009507546687132259,
      "loss": 1.8945,
      "step": 38500
    },
    {
      "epoch": 2.49424405218726,
      "grad_norm": 0.326171875,
      "learning_rate": 0.0009501151189562548,
      "loss": 1.8946,
      "step": 39000
    },
    {
      "epoch": 2.526221540035815,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0009494755691992837,
      "loss": 1.8952,
      "step": 39500
    },
    {
      "epoch": 2.5581990278843696,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0009488360194423127,
      "loss": 1.8949,
      "step": 40000
    },
    {
      "epoch": 2.590176515732924,
      "grad_norm": 0.2392578125,
      "learning_rate": 0.0009481964696853416,
      "loss": 1.8943,
      "step": 40500
    },
    {
      "epoch": 2.6221540035814788,
      "grad_norm": 0.2734375,
      "learning_rate": 0.0009475569199283705,
      "loss": 1.8935,
      "step": 41000
    },
    {
      "epoch": 2.654131491430033,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0009469173701713993,
      "loss": 1.8934,
      "step": 41500
    },
    {
      "epoch": 2.686108979278588,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.0009462778204144283,
      "loss": 1.8928,
      "step": 42000
    },
    {
      "epoch": 2.7180864671271427,
      "grad_norm": 0.373046875,
      "learning_rate": 0.0009456382706574572,
      "loss": 1.8919,
      "step": 42500
    },
    {
      "epoch": 2.750063954975697,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0009449987209004861,
      "loss": 1.8921,
      "step": 43000
    },
    {
      "epoch": 2.7820414428242515,
      "grad_norm": 0.291015625,
      "learning_rate": 0.000944359171143515,
      "loss": 1.8925,
      "step": 43500
    },
    {
      "epoch": 2.8140189306728063,
      "grad_norm": 0.333984375,
      "learning_rate": 0.0009437196213865439,
      "loss": 1.8928,
      "step": 44000
    },
    {
      "epoch": 2.845996418521361,
      "grad_norm": 0.224609375,
      "learning_rate": 0.0009430800716295727,
      "loss": 1.8934,
      "step": 44500
    },
    {
      "epoch": 2.8779739063699155,
      "grad_norm": 0.412109375,
      "learning_rate": 0.0009424405218726017,
      "loss": 1.8922,
      "step": 45000
    },
    {
      "epoch": 2.9099513942184703,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0009418009721156306,
      "loss": 1.8918,
      "step": 45500
    },
    {
      "epoch": 2.9419288820670246,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0009411614223586596,
      "loss": 1.8927,
      "step": 46000
    },
    {
      "epoch": 2.9739063699155794,
      "grad_norm": 0.1923828125,
      "learning_rate": 0.0009405218726016884,
      "loss": 1.892,
      "step": 46500
    },
    {
      "epoch": 3.0,
      "eval_loss": 3.7860679626464844,
      "eval_runtime": 5.9935,
      "eval_samples_per_second": 83.424,
      "eval_steps_per_second": 1.335,
      "step": 46908
    },
    {
      "epoch": 3.0058838577641342,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0009398823228447173,
      "loss": 1.8906,
      "step": 47000
    },
    {
      "epoch": 3.0378613456126886,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0009392427730877462,
      "loss": 1.8917,
      "step": 47500
    },
    {
      "epoch": 3.0698388334612434,
      "grad_norm": 0.400390625,
      "learning_rate": 0.0009386032233307752,
      "loss": 1.8924,
      "step": 48000
    },
    {
      "epoch": 3.101816321309798,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000937963673573804,
      "loss": 1.8918,
      "step": 48500
    },
    {
      "epoch": 3.1337938091583526,
      "grad_norm": 0.279296875,
      "learning_rate": 0.000937324123816833,
      "loss": 1.8911,
      "step": 49000
    },
    {
      "epoch": 3.165771297006907,
      "grad_norm": 0.21484375,
      "learning_rate": 0.0009366845740598618,
      "loss": 1.8908,
      "step": 49500
    },
    {
      "epoch": 3.1977487848554618,
      "grad_norm": 0.36328125,
      "learning_rate": 0.0009360450243028907,
      "loss": 1.8894,
      "step": 50000
    },
    {
      "epoch": 3.2297262727040166,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0009354054745459198,
      "loss": 1.8902,
      "step": 50500
    },
    {
      "epoch": 3.261703760552571,
      "grad_norm": 0.24609375,
      "learning_rate": 0.0009347659247889486,
      "loss": 1.8909,
      "step": 51000
    },
    {
      "epoch": 3.2936812484011257,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0009341263750319776,
      "loss": 1.8917,
      "step": 51500
    },
    {
      "epoch": 3.32565873624968,
      "grad_norm": 0.31640625,
      "learning_rate": 0.0009334868252750064,
      "loss": 1.8923,
      "step": 52000
    },
    {
      "epoch": 3.357636224098235,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0009328472755180352,
      "loss": 1.8925,
      "step": 52500
    },
    {
      "epoch": 3.3896137119467893,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0009322077257610643,
      "loss": 1.8923,
      "step": 53000
    },
    {
      "epoch": 3.421591199795344,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0009315681760040932,
      "loss": 1.892,
      "step": 53500
    },
    {
      "epoch": 3.453568687643899,
      "grad_norm": 0.240234375,
      "learning_rate": 0.000930928626247122,
      "loss": 1.8916,
      "step": 54000
    },
    {
      "epoch": 3.4855461754924533,
      "grad_norm": 0.330078125,
      "learning_rate": 0.000930289076490151,
      "loss": 1.8911,
      "step": 54500
    },
    {
      "epoch": 3.517523663341008,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0009296495267331798,
      "loss": 1.8911,
      "step": 55000
    },
    {
      "epoch": 3.5495011511895624,
      "grad_norm": 2.515625,
      "learning_rate": 0.0009290099769762088,
      "loss": 1.8916,
      "step": 55500
    },
    {
      "epoch": 3.5814786390381173,
      "grad_norm": 0.341796875,
      "learning_rate": 0.0009283704272192377,
      "loss": 1.8919,
      "step": 56000
    },
    {
      "epoch": 3.613456126886672,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.0009277308774622666,
      "loss": 1.8914,
      "step": 56500
    },
    {
      "epoch": 3.6454336147352264,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0009270913277052955,
      "loss": 1.8916,
      "step": 57000
    },
    {
      "epoch": 3.677411102583781,
      "grad_norm": 0.234375,
      "learning_rate": 0.0009264517779483244,
      "loss": 1.8914,
      "step": 57500
    },
    {
      "epoch": 3.7093885904323356,
      "grad_norm": 0.408203125,
      "learning_rate": 0.0009258122281913533,
      "loss": 1.8904,
      "step": 58000
    },
    {
      "epoch": 3.7413660782808904,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0009251726784343823,
      "loss": 1.8905,
      "step": 58500
    },
    {
      "epoch": 3.7733435661294448,
      "grad_norm": 0.26171875,
      "learning_rate": 0.0009245331286774111,
      "loss": 1.8906,
      "step": 59000
    },
    {
      "epoch": 3.8053210539779996,
      "grad_norm": 0.310546875,
      "learning_rate": 0.00092389357892044,
      "loss": 1.8902,
      "step": 59500
    },
    {
      "epoch": 3.837298541826554,
      "grad_norm": 0.232421875,
      "learning_rate": 0.0009232540291634689,
      "loss": 1.8904,
      "step": 60000
    },
    {
      "epoch": 3.8692760296751088,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0009226144794064979,
      "loss": 1.8904,
      "step": 60500
    },
    {
      "epoch": 3.9012535175236636,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0009219749296495267,
      "loss": 1.8904,
      "step": 61000
    },
    {
      "epoch": 3.933231005372218,
      "grad_norm": 0.216796875,
      "learning_rate": 0.0009213353798925557,
      "loss": 1.8894,
      "step": 61500
    },
    {
      "epoch": 3.9652084932207727,
      "grad_norm": 0.296875,
      "learning_rate": 0.0009206958301355845,
      "loss": 1.8897,
      "step": 62000
    },
    {
      "epoch": 3.997185981069327,
      "grad_norm": 0.2060546875,
      "learning_rate": 0.0009200562803786135,
      "loss": 1.8892,
      "step": 62500
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.7807514667510986,
      "eval_runtime": 5.9175,
      "eval_samples_per_second": 84.495,
      "eval_steps_per_second": 1.352,
      "step": 62544
    },
    {
      "epoch": 4.029163468917882,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0009194167306216423,
      "loss": 1.8893,
      "step": 63000
    },
    {
      "epoch": 4.061140956766437,
      "grad_norm": 0.232421875,
      "learning_rate": 0.0009187771808646713,
      "loss": 1.8887,
      "step": 63500
    },
    {
      "epoch": 4.093118444614991,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0009181376311077002,
      "loss": 1.8887,
      "step": 64000
    },
    {
      "epoch": 4.1250959324635454,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0009174980813507291,
      "loss": 1.8897,
      "step": 64500
    },
    {
      "epoch": 4.1570734203121,
      "grad_norm": 1.59375,
      "learning_rate": 0.0009168585315937579,
      "loss": 1.889,
      "step": 65000
    },
    {
      "epoch": 4.189050908160655,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0009162189818367869,
      "loss": 1.8902,
      "step": 65500
    },
    {
      "epoch": 4.22102839600921,
      "grad_norm": 0.228515625,
      "learning_rate": 0.0009155794320798159,
      "loss": 1.8891,
      "step": 66000
    },
    {
      "epoch": 4.253005883857764,
      "grad_norm": 0.2421875,
      "learning_rate": 0.0009149398823228447,
      "loss": 1.8891,
      "step": 66500
    },
    {
      "epoch": 4.284983371706319,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0009143003325658737,
      "loss": 1.8886,
      "step": 67000
    },
    {
      "epoch": 4.316960859554873,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0009136607828089025,
      "loss": 1.8897,
      "step": 67500
    },
    {
      "epoch": 4.348938347403428,
      "grad_norm": 0.359375,
      "learning_rate": 0.0009130212330519315,
      "loss": 1.8891,
      "step": 68000
    },
    {
      "epoch": 4.380915835251983,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.0009123816832949604,
      "loss": 1.8888,
      "step": 68500
    },
    {
      "epoch": 4.412893323100537,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0009117421335379893,
      "loss": 1.8903,
      "step": 69000
    },
    {
      "epoch": 4.444870810949092,
      "grad_norm": 0.40625,
      "learning_rate": 0.0009111025837810182,
      "loss": 1.8891,
      "step": 69500
    },
    {
      "epoch": 4.476848298797647,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0009104630340240471,
      "loss": 1.8893,
      "step": 70000
    },
    {
      "epoch": 4.508825786646201,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0009098234842670759,
      "loss": 1.8895,
      "step": 70500
    },
    {
      "epoch": 4.540803274494756,
      "grad_norm": 0.35546875,
      "learning_rate": 0.000909183934510105,
      "loss": 1.8901,
      "step": 71000
    },
    {
      "epoch": 4.57278076234331,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0009085443847531338,
      "loss": 1.8892,
      "step": 71500
    },
    {
      "epoch": 4.604758250191865,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.0009079048349961627,
      "loss": 1.89,
      "step": 72000
    },
    {
      "epoch": 4.63673573804042,
      "grad_norm": 0.28515625,
      "learning_rate": 0.0009072652852391916,
      "loss": 1.8896,
      "step": 72500
    },
    {
      "epoch": 4.6687132258889745,
      "grad_norm": 0.25,
      "learning_rate": 0.0009066257354822205,
      "loss": 1.8895,
      "step": 73000
    },
    {
      "epoch": 4.7006907137375284,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0009059861857252495,
      "loss": 1.8892,
      "step": 73500
    },
    {
      "epoch": 4.732668201586083,
      "grad_norm": 0.294921875,
      "learning_rate": 0.0009053466359682784,
      "loss": 1.8885,
      "step": 74000
    },
    {
      "epoch": 4.764645689434638,
      "grad_norm": 0.236328125,
      "learning_rate": 0.0009047070862113072,
      "loss": 1.8874,
      "step": 74500
    },
    {
      "epoch": 4.796623177283193,
      "grad_norm": 0.220703125,
      "learning_rate": 0.0009040675364543362,
      "loss": 1.8879,
      "step": 75000
    },
    {
      "epoch": 4.828600665131747,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000903427986697365,
      "loss": 1.8879,
      "step": 75500
    },
    {
      "epoch": 4.860578152980302,
      "grad_norm": 0.443359375,
      "learning_rate": 0.000902788436940394,
      "loss": 1.887,
      "step": 76000
    },
    {
      "epoch": 4.892555640828856,
      "grad_norm": 0.337890625,
      "learning_rate": 0.0009021488871834229,
      "loss": 1.8873,
      "step": 76500
    },
    {
      "epoch": 4.924533128677411,
      "grad_norm": 0.2197265625,
      "learning_rate": 0.0009015093374264518,
      "loss": 1.8877,
      "step": 77000
    },
    {
      "epoch": 4.956510616525966,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0009008697876694807,
      "loss": 1.8881,
      "step": 77500
    },
    {
      "epoch": 4.98848810437452,
      "grad_norm": 0.2138671875,
      "learning_rate": 0.0009002302379125096,
      "loss": 1.8881,
      "step": 78000
    },
    {
      "epoch": 5.0,
      "eval_loss": 3.7772316932678223,
      "eval_runtime": 5.7326,
      "eval_samples_per_second": 87.22,
      "eval_steps_per_second": 1.396,
      "step": 78180
    },
    {
      "epoch": 5.020465592223075,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0008995906881555385,
      "loss": 1.8888,
      "step": 78500
    },
    {
      "epoch": 5.05244308007163,
      "grad_norm": 0.2177734375,
      "learning_rate": 0.0008989511383985675,
      "loss": 1.8888,
      "step": 79000
    },
    {
      "epoch": 5.084420567920184,
      "grad_norm": 0.255859375,
      "learning_rate": 0.0008983115886415963,
      "loss": 1.8888,
      "step": 79500
    },
    {
      "epoch": 5.116398055768739,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0008976720388846252,
      "loss": 1.8887,
      "step": 80000
    },
    {
      "epoch": 5.148375543617293,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0008970324891276542,
      "loss": 1.8893,
      "step": 80500
    },
    {
      "epoch": 5.180353031465848,
      "grad_norm": 0.5078125,
      "learning_rate": 0.000896392939370683,
      "loss": 1.8899,
      "step": 81000
    },
    {
      "epoch": 5.212330519314403,
      "grad_norm": 0.259765625,
      "learning_rate": 0.000895753389613712,
      "loss": 1.8886,
      "step": 81500
    },
    {
      "epoch": 5.2443080071629575,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0008951138398567409,
      "loss": 1.8879,
      "step": 82000
    },
    {
      "epoch": 5.276285495011512,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0008944742900997698,
      "loss": 1.8881,
      "step": 82500
    },
    {
      "epoch": 5.308262982860066,
      "grad_norm": 0.2451171875,
      "learning_rate": 0.0008938347403427987,
      "loss": 1.8889,
      "step": 83000
    },
    {
      "epoch": 5.340240470708621,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0008931951905858276,
      "loss": 1.8883,
      "step": 83500
    },
    {
      "epoch": 5.372217958557176,
      "grad_norm": 0.2392578125,
      "learning_rate": 0.0008925556408288565,
      "loss": 1.8897,
      "step": 84000
    },
    {
      "epoch": 5.404195446405731,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0008919160910718855,
      "loss": 1.8889,
      "step": 84500
    },
    {
      "epoch": 5.436172934254285,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.0008912765413149143,
      "loss": 1.8886,
      "step": 85000
    },
    {
      "epoch": 5.468150422102839,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0008906369915579432,
      "loss": 1.8875,
      "step": 85500
    },
    {
      "epoch": 5.500127909951394,
      "grad_norm": 0.671875,
      "learning_rate": 0.0008899974418009721,
      "loss": 1.8873,
      "step": 86000
    },
    {
      "epoch": 5.532105397799949,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0008893578920440011,
      "loss": 1.8871,
      "step": 86500
    },
    {
      "epoch": 5.564082885648504,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0008887183422870299,
      "loss": 1.8881,
      "step": 87000
    },
    {
      "epoch": 5.596060373497058,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0008880787925300589,
      "loss": 1.8875,
      "step": 87500
    },
    {
      "epoch": 5.628037861345613,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0008874392427730877,
      "loss": 1.8873,
      "step": 88000
    },
    {
      "epoch": 5.660015349194167,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0008867996930161167,
      "loss": 1.8868,
      "step": 88500
    },
    {
      "epoch": 5.691992837042722,
      "grad_norm": 0.203125,
      "learning_rate": 0.0008861601432591456,
      "loss": 1.8868,
      "step": 89000
    },
    {
      "epoch": 5.723970324891276,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.0008855205935021745,
      "loss": 1.8869,
      "step": 89500
    },
    {
      "epoch": 5.755947812739831,
      "grad_norm": 0.28125,
      "learning_rate": 0.0008848810437452034,
      "loss": 1.8865,
      "step": 90000
    },
    {
      "epoch": 5.787925300588386,
      "grad_norm": 0.26171875,
      "learning_rate": 0.0008842414939882323,
      "loss": 1.8868,
      "step": 90500
    },
    {
      "epoch": 5.8199027884369405,
      "grad_norm": 0.28125,
      "learning_rate": 0.0008836019442312611,
      "loss": 1.8864,
      "step": 91000
    },
    {
      "epoch": 5.851880276285495,
      "grad_norm": 0.244140625,
      "learning_rate": 0.0008829623944742902,
      "loss": 1.8868,
      "step": 91500
    },
    {
      "epoch": 5.883857764134049,
      "grad_norm": 0.2451171875,
      "learning_rate": 0.000882322844717319,
      "loss": 1.8863,
      "step": 92000
    },
    {
      "epoch": 5.915835251982604,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0008816832949603479,
      "loss": 1.8863,
      "step": 92500
    },
    {
      "epoch": 5.947812739831159,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0008810437452033768,
      "loss": 1.8858,
      "step": 93000
    },
    {
      "epoch": 5.979790227679714,
      "grad_norm": 0.2421875,
      "learning_rate": 0.0008804041954464057,
      "loss": 1.8862,
      "step": 93500
    },
    {
      "epoch": 6.0,
      "eval_loss": 3.7749125957489014,
      "eval_runtime": 6.8027,
      "eval_samples_per_second": 73.5,
      "eval_steps_per_second": 1.176,
      "step": 93816
    },
    {
      "epoch": 6.0117677155282685,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0008797646456894348,
      "loss": 1.8847,
      "step": 94000
    },
    {
      "epoch": 6.043745203376822,
      "grad_norm": 0.234375,
      "learning_rate": 0.0008791250959324636,
      "loss": 1.8858,
      "step": 94500
    },
    {
      "epoch": 6.075722691225377,
      "grad_norm": 0.25,
      "learning_rate": 0.0008784855461754924,
      "loss": 1.8857,
      "step": 95000
    },
    {
      "epoch": 6.107700179073932,
      "grad_norm": 0.173828125,
      "learning_rate": 0.0008778459964185214,
      "loss": 1.8855,
      "step": 95500
    },
    {
      "epoch": 6.139677666922487,
      "grad_norm": 0.314453125,
      "learning_rate": 0.0008772064466615503,
      "loss": 1.885,
      "step": 96000
    },
    {
      "epoch": 6.171655154771041,
      "grad_norm": 0.1875,
      "learning_rate": 0.0008765668969045792,
      "loss": 1.8847,
      "step": 96500
    },
    {
      "epoch": 6.203632642619596,
      "grad_norm": 0.205078125,
      "learning_rate": 0.0008759273471476082,
      "loss": 1.8847,
      "step": 97000
    },
    {
      "epoch": 6.23561013046815,
      "grad_norm": 0.26953125,
      "learning_rate": 0.000875287797390637,
      "loss": 1.8848,
      "step": 97500
    },
    {
      "epoch": 6.267587618316705,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0008746482476336659,
      "loss": 1.8847,
      "step": 98000
    },
    {
      "epoch": 6.29956510616526,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0008740086978766948,
      "loss": 1.8854,
      "step": 98500
    },
    {
      "epoch": 6.331542594013814,
      "grad_norm": 0.34765625,
      "learning_rate": 0.0008733691481197237,
      "loss": 1.8849,
      "step": 99000
    },
    {
      "epoch": 6.363520081862369,
      "grad_norm": 0.1787109375,
      "learning_rate": 0.0008727295983627527,
      "loss": 1.885,
      "step": 99500
    },
    {
      "epoch": 6.3954975697109235,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0008720900486057816,
      "loss": 1.8839,
      "step": 100000
    },
    {
      "epoch": 6.427475057559478,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0008714504988488104,
      "loss": 1.8848,
      "step": 100500
    },
    {
      "epoch": 6.459452545408033,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0008708109490918394,
      "loss": 1.8844,
      "step": 101000
    },
    {
      "epoch": 6.491430033256587,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0008701713993348682,
      "loss": 1.8847,
      "step": 101500
    },
    {
      "epoch": 6.523407521105142,
      "grad_norm": 0.265625,
      "learning_rate": 0.0008695318495778972,
      "loss": 1.885,
      "step": 102000
    },
    {
      "epoch": 6.555385008953697,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0008688922998209261,
      "loss": 1.8839,
      "step": 102500
    },
    {
      "epoch": 6.5873624968022515,
      "grad_norm": 0.703125,
      "learning_rate": 0.000868252750063955,
      "loss": 1.8833,
      "step": 103000
    },
    {
      "epoch": 6.619339984650805,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0008676132003069838,
      "loss": 1.8847,
      "step": 103500
    },
    {
      "epoch": 6.65131747249936,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0008669736505500128,
      "loss": 1.8833,
      "step": 104000
    },
    {
      "epoch": 6.683294960347915,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0008663341007930417,
      "loss": 1.8848,
      "step": 104500
    },
    {
      "epoch": 6.71527244819647,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0008656945510360707,
      "loss": 1.8849,
      "step": 105000
    },
    {
      "epoch": 6.747249936045025,
      "grad_norm": 0.2392578125,
      "learning_rate": 0.0008650550012790995,
      "loss": 1.8839,
      "step": 105500
    },
    {
      "epoch": 6.779227423893579,
      "grad_norm": 0.220703125,
      "learning_rate": 0.0008644154515221284,
      "loss": 1.8844,
      "step": 106000
    },
    {
      "epoch": 6.811204911742133,
      "grad_norm": 0.376953125,
      "learning_rate": 0.0008637759017651573,
      "loss": 1.8841,
      "step": 106500
    },
    {
      "epoch": 6.843182399590688,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0008631363520081863,
      "loss": 1.8836,
      "step": 107000
    },
    {
      "epoch": 6.875159887439243,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0008624968022512151,
      "loss": 1.8839,
      "step": 107500
    },
    {
      "epoch": 6.907137375287798,
      "grad_norm": 0.232421875,
      "learning_rate": 0.0008618572524942441,
      "loss": 1.8831,
      "step": 108000
    },
    {
      "epoch": 6.939114863136352,
      "grad_norm": 0.228515625,
      "learning_rate": 0.0008612177027372729,
      "loss": 1.8841,
      "step": 108500
    },
    {
      "epoch": 6.9710923509849065,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0008605781529803018,
      "loss": 1.8837,
      "step": 109000
    },
    {
      "epoch": 7.0,
      "eval_loss": 3.7695865631103516,
      "eval_runtime": 7.1616,
      "eval_samples_per_second": 69.817,
      "eval_steps_per_second": 1.117,
      "step": 109452
    },
    {
      "epoch": 7.003069838833461,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0008599386032233309,
      "loss": 1.884,
      "step": 109500
    },
    {
      "epoch": 7.035047326682016,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0008592990534663597,
      "loss": 1.8827,
      "step": 110000
    },
    {
      "epoch": 7.06702481453057,
      "grad_norm": 0.2216796875,
      "learning_rate": 0.0008586595037093887,
      "loss": 1.8839,
      "step": 110500
    },
    {
      "epoch": 7.099002302379125,
      "grad_norm": 0.1875,
      "learning_rate": 0.0008580199539524175,
      "loss": 1.8835,
      "step": 111000
    },
    {
      "epoch": 7.13097979022768,
      "grad_norm": 0.283203125,
      "learning_rate": 0.0008573804041954464,
      "loss": 1.8835,
      "step": 111500
    },
    {
      "epoch": 7.1629572780762345,
      "grad_norm": 0.30078125,
      "learning_rate": 0.0008567408544384754,
      "loss": 1.8834,
      "step": 112000
    },
    {
      "epoch": 7.194934765924789,
      "grad_norm": 0.2197265625,
      "learning_rate": 0.0008561013046815043,
      "loss": 1.8832,
      "step": 112500
    },
    {
      "epoch": 7.226912253773343,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0008554617549245331,
      "loss": 1.883,
      "step": 113000
    },
    {
      "epoch": 7.258889741621898,
      "grad_norm": 0.205078125,
      "learning_rate": 0.0008548222051675621,
      "loss": 1.8831,
      "step": 113500
    },
    {
      "epoch": 7.290867229470453,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0008541826554105909,
      "loss": 1.8826,
      "step": 114000
    },
    {
      "epoch": 7.322844717319008,
      "grad_norm": 0.255859375,
      "learning_rate": 0.0008535431056536199,
      "loss": 1.8836,
      "step": 114500
    },
    {
      "epoch": 7.354822205167562,
      "grad_norm": 0.1845703125,
      "learning_rate": 0.0008529035558966488,
      "loss": 1.8821,
      "step": 115000
    },
    {
      "epoch": 7.386799693016116,
      "grad_norm": 0.46875,
      "learning_rate": 0.0008522640061396777,
      "loss": 1.8823,
      "step": 115500
    },
    {
      "epoch": 7.418777180864671,
      "grad_norm": 1.53125,
      "learning_rate": 0.0008516244563827066,
      "loss": 1.8823,
      "step": 116000
    },
    {
      "epoch": 7.450754668713226,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0008509849066257355,
      "loss": 1.8824,
      "step": 116500
    },
    {
      "epoch": 7.482732156561781,
      "grad_norm": 0.2080078125,
      "learning_rate": 0.0008503453568687643,
      "loss": 1.8817,
      "step": 117000
    },
    {
      "epoch": 7.514709644410335,
      "grad_norm": 0.283203125,
      "learning_rate": 0.0008497058071117934,
      "loss": 1.8831,
      "step": 117500
    },
    {
      "epoch": 7.5466871322588895,
      "grad_norm": 0.3203125,
      "learning_rate": 0.0008490662573548222,
      "loss": 1.8819,
      "step": 118000
    },
    {
      "epoch": 7.578664620107444,
      "grad_norm": 0.2451171875,
      "learning_rate": 0.0008484267075978511,
      "loss": 1.8814,
      "step": 118500
    },
    {
      "epoch": 7.610642107955999,
      "grad_norm": 0.34375,
      "learning_rate": 0.00084778715784088,
      "loss": 1.8824,
      "step": 119000
    },
    {
      "epoch": 7.642619595804554,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.0008471476080839089,
      "loss": 1.8821,
      "step": 119500
    },
    {
      "epoch": 7.674597083653108,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0008465080583269378,
      "loss": 1.8822,
      "step": 120000
    },
    {
      "epoch": 7.706574571501663,
      "grad_norm": 0.2890625,
      "learning_rate": 0.0008458685085699668,
      "loss": 1.8826,
      "step": 120500
    },
    {
      "epoch": 7.7385520593502175,
      "grad_norm": 0.255859375,
      "learning_rate": 0.0008452289588129956,
      "loss": 1.8828,
      "step": 121000
    },
    {
      "epoch": 7.770529547198772,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0008445894090560246,
      "loss": 1.8825,
      "step": 121500
    },
    {
      "epoch": 7.802507035047327,
      "grad_norm": 0.578125,
      "learning_rate": 0.0008439498592990534,
      "loss": 1.8819,
      "step": 122000
    },
    {
      "epoch": 7.834484522895881,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0008433103095420824,
      "loss": 1.8818,
      "step": 122500
    },
    {
      "epoch": 7.866462010744436,
      "grad_norm": 0.2421875,
      "learning_rate": 0.0008426707597851114,
      "loss": 1.8827,
      "step": 123000
    },
    {
      "epoch": 7.898439498592991,
      "grad_norm": 0.2197265625,
      "learning_rate": 0.0008420312100281402,
      "loss": 1.8806,
      "step": 123500
    },
    {
      "epoch": 7.9304169864415455,
      "grad_norm": 0.173828125,
      "learning_rate": 0.000841391660271169,
      "loss": 1.8811,
      "step": 124000
    },
    {
      "epoch": 7.962394474290099,
      "grad_norm": 0.2421875,
      "learning_rate": 0.000840752110514198,
      "loss": 1.8818,
      "step": 124500
    },
    {
      "epoch": 7.994371962138654,
      "grad_norm": 0.392578125,
      "learning_rate": 0.000840112560757227,
      "loss": 1.882,
      "step": 125000
    },
    {
      "epoch": 8.0,
      "eval_loss": 3.7648119926452637,
      "eval_runtime": 7.2758,
      "eval_samples_per_second": 68.721,
      "eval_steps_per_second": 1.1,
      "step": 125088
    },
    {
      "epoch": 8.026349449987208,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0008394730110002559,
      "loss": 1.8808,
      "step": 125500
    },
    {
      "epoch": 8.058326937835764,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0008388334612432848,
      "loss": 1.8805,
      "step": 126000
    },
    {
      "epoch": 8.090304425684318,
      "grad_norm": 0.4296875,
      "learning_rate": 0.0008381939114863136,
      "loss": 1.882,
      "step": 126500
    },
    {
      "epoch": 8.122281913532873,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0008375543617293426,
      "loss": 1.8811,
      "step": 127000
    },
    {
      "epoch": 8.154259401381427,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0008369148119723715,
      "loss": 1.8798,
      "step": 127500
    },
    {
      "epoch": 8.186236889229981,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0008362752622154004,
      "loss": 1.8817,
      "step": 128000
    },
    {
      "epoch": 8.218214377078537,
      "grad_norm": 0.255859375,
      "learning_rate": 0.0008356357124584293,
      "loss": 1.881,
      "step": 128500
    },
    {
      "epoch": 8.250191864927091,
      "grad_norm": 3.3125,
      "learning_rate": 0.0008349961627014582,
      "loss": 1.8805,
      "step": 129000
    },
    {
      "epoch": 8.282169352775647,
      "grad_norm": 0.2138671875,
      "learning_rate": 0.000834356612944487,
      "loss": 1.8808,
      "step": 129500
    },
    {
      "epoch": 8.3141468406242,
      "grad_norm": 0.2734375,
      "learning_rate": 0.0008337170631875161,
      "loss": 1.881,
      "step": 130000
    },
    {
      "epoch": 8.346124328472754,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.0008330775134305449,
      "loss": 1.8805,
      "step": 130500
    },
    {
      "epoch": 8.37810181632131,
      "grad_norm": 0.1923828125,
      "learning_rate": 0.0008324379636735739,
      "loss": 1.8791,
      "step": 131000
    },
    {
      "epoch": 8.410079304169864,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0008317984139166027,
      "loss": 1.8813,
      "step": 131500
    },
    {
      "epoch": 8.44205679201842,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0008311588641596316,
      "loss": 1.8806,
      "step": 132000
    },
    {
      "epoch": 8.474034279866974,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0008305193144026606,
      "loss": 1.8803,
      "step": 132500
    },
    {
      "epoch": 8.506011767715528,
      "grad_norm": 0.21484375,
      "learning_rate": 0.0008298797646456895,
      "loss": 1.8803,
      "step": 133000
    },
    {
      "epoch": 8.537989255564083,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0008292402148887183,
      "loss": 1.8806,
      "step": 133500
    },
    {
      "epoch": 8.569966743412637,
      "grad_norm": 0.330078125,
      "learning_rate": 0.0008286006651317473,
      "loss": 1.8793,
      "step": 134000
    },
    {
      "epoch": 8.601944231261193,
      "grad_norm": 0.30078125,
      "learning_rate": 0.0008279611153747761,
      "loss": 1.8801,
      "step": 134500
    },
    {
      "epoch": 8.633921719109747,
      "grad_norm": 0.255859375,
      "learning_rate": 0.000827321565617805,
      "loss": 1.8801,
      "step": 135000
    },
    {
      "epoch": 8.6658992069583,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.000826682015860834,
      "loss": 1.8813,
      "step": 135500
    },
    {
      "epoch": 8.697876694806856,
      "grad_norm": 0.232421875,
      "learning_rate": 0.0008260424661038629,
      "loss": 1.88,
      "step": 136000
    },
    {
      "epoch": 8.72985418265541,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0008254029163468918,
      "loss": 1.8802,
      "step": 136500
    },
    {
      "epoch": 8.761831670503966,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.0008247633665899207,
      "loss": 1.8788,
      "step": 137000
    },
    {
      "epoch": 8.79380915835252,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0008241238168329495,
      "loss": 1.8803,
      "step": 137500
    },
    {
      "epoch": 8.825786646201074,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0008234842670759786,
      "loss": 1.88,
      "step": 138000
    },
    {
      "epoch": 8.85776413404963,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0008228447173190075,
      "loss": 1.8799,
      "step": 138500
    },
    {
      "epoch": 8.889741621898184,
      "grad_norm": 0.283203125,
      "learning_rate": 0.0008222051675620363,
      "loss": 1.8809,
      "step": 139000
    },
    {
      "epoch": 8.92171910974674,
      "grad_norm": 0.1640625,
      "learning_rate": 0.0008215656178050653,
      "loss": 1.8808,
      "step": 139500
    },
    {
      "epoch": 8.953696597595293,
      "grad_norm": 0.228515625,
      "learning_rate": 0.0008209260680480941,
      "loss": 1.8809,
      "step": 140000
    },
    {
      "epoch": 8.985674085443847,
      "grad_norm": 0.2177734375,
      "learning_rate": 0.0008202865182911231,
      "loss": 1.8797,
      "step": 140500
    },
    {
      "epoch": 9.0,
      "eval_loss": 3.763540744781494,
      "eval_runtime": 8.199,
      "eval_samples_per_second": 60.983,
      "eval_steps_per_second": 0.976,
      "step": 140724
    },
    {
      "epoch": 9.017651573292403,
      "grad_norm": 0.328125,
      "learning_rate": 0.000819646968534152,
      "loss": 1.8798,
      "step": 141000
    },
    {
      "epoch": 9.049629061140957,
      "grad_norm": 0.6875,
      "learning_rate": 0.0008190074187771809,
      "loss": 1.8795,
      "step": 141500
    },
    {
      "epoch": 9.08160654898951,
      "grad_norm": 0.337890625,
      "learning_rate": 0.0008183678690202098,
      "loss": 1.8794,
      "step": 142000
    },
    {
      "epoch": 9.113584036838066,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0008177283192632387,
      "loss": 1.8808,
      "step": 142500
    },
    {
      "epoch": 9.14556152468662,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0008170887695062676,
      "loss": 1.8794,
      "step": 143000
    },
    {
      "epoch": 9.177539012535176,
      "grad_norm": 0.193359375,
      "learning_rate": 0.0008164492197492966,
      "loss": 1.8799,
      "step": 143500
    },
    {
      "epoch": 9.20951650038373,
      "grad_norm": 0.828125,
      "learning_rate": 0.0008158096699923254,
      "loss": 1.88,
      "step": 144000
    },
    {
      "epoch": 9.241493988232284,
      "grad_norm": 0.208984375,
      "learning_rate": 0.0008151701202353543,
      "loss": 1.8802,
      "step": 144500
    },
    {
      "epoch": 9.27347147608084,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0008145305704783832,
      "loss": 1.8799,
      "step": 145000
    },
    {
      "epoch": 9.305448963929393,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0008138910207214122,
      "loss": 1.8791,
      "step": 145500
    },
    {
      "epoch": 9.337426451777949,
      "grad_norm": 0.248046875,
      "learning_rate": 0.000813251470964441,
      "loss": 1.8791,
      "step": 146000
    },
    {
      "epoch": 9.369403939626503,
      "grad_norm": 0.294921875,
      "learning_rate": 0.00081261192120747,
      "loss": 1.8792,
      "step": 146500
    },
    {
      "epoch": 9.401381427475057,
      "grad_norm": 0.337890625,
      "learning_rate": 0.0008119723714504988,
      "loss": 1.8796,
      "step": 147000
    },
    {
      "epoch": 9.433358915323613,
      "grad_norm": 0.328125,
      "learning_rate": 0.0008113328216935278,
      "loss": 1.8797,
      "step": 147500
    },
    {
      "epoch": 9.465336403172167,
      "grad_norm": 0.2890625,
      "learning_rate": 0.0008106932719365567,
      "loss": 1.8797,
      "step": 148000
    },
    {
      "epoch": 9.497313891020722,
      "grad_norm": 0.3828125,
      "learning_rate": 0.0008100537221795856,
      "loss": 1.8787,
      "step": 148500
    },
    {
      "epoch": 9.529291378869276,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0008094141724226145,
      "loss": 1.8798,
      "step": 149000
    },
    {
      "epoch": 9.56126886671783,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0008087746226656434,
      "loss": 1.8783,
      "step": 149500
    },
    {
      "epoch": 9.593246354566386,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0008081350729086722,
      "loss": 1.8806,
      "step": 150000
    },
    {
      "epoch": 9.62522384241494,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0008074955231517013,
      "loss": 1.8797,
      "step": 150500
    },
    {
      "epoch": 9.657201330263494,
      "grad_norm": 0.33984375,
      "learning_rate": 0.0008068559733947301,
      "loss": 1.8789,
      "step": 151000
    },
    {
      "epoch": 9.68917881811205,
      "grad_norm": 0.267578125,
      "learning_rate": 0.000806216423637759,
      "loss": 1.8792,
      "step": 151500
    },
    {
      "epoch": 9.721156305960603,
      "grad_norm": 0.349609375,
      "learning_rate": 0.000805576873880788,
      "loss": 1.88,
      "step": 152000
    },
    {
      "epoch": 9.753133793809159,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0008049373241238168,
      "loss": 1.88,
      "step": 152500
    },
    {
      "epoch": 9.785111281657713,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0008042977743668458,
      "loss": 1.8798,
      "step": 153000
    },
    {
      "epoch": 9.817088769506267,
      "grad_norm": 0.296875,
      "learning_rate": 0.0008036582246098747,
      "loss": 1.8803,
      "step": 153500
    },
    {
      "epoch": 9.849066257354822,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0008030186748529036,
      "loss": 1.8799,
      "step": 154000
    },
    {
      "epoch": 9.881043745203376,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0008023791250959325,
      "loss": 1.879,
      "step": 154500
    },
    {
      "epoch": 9.913021233051932,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0008017395753389614,
      "loss": 1.88,
      "step": 155000
    },
    {
      "epoch": 9.944998720900486,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0008011000255819902,
      "loss": 1.879,
      "step": 155500
    },
    {
      "epoch": 9.97697620874904,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.0008004604758250193,
      "loss": 1.8798,
      "step": 156000
    },
    {
      "epoch": 10.0,
      "eval_loss": 3.761446237564087,
      "eval_runtime": 6.8391,
      "eval_samples_per_second": 73.109,
      "eval_steps_per_second": 1.17,
      "step": 156360
    },
    {
      "epoch": 10.008953696597596,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0007998209260680481,
      "loss": 1.8786,
      "step": 156500
    },
    {
      "epoch": 10.04093118444615,
      "grad_norm": 0.40234375,
      "learning_rate": 0.000799181376311077,
      "loss": 1.8793,
      "step": 157000
    },
    {
      "epoch": 10.072908672294705,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0007985418265541059,
      "loss": 1.8795,
      "step": 157500
    },
    {
      "epoch": 10.10488616014326,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0007979022767971348,
      "loss": 1.88,
      "step": 158000
    },
    {
      "epoch": 10.136863647991813,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0007972627270401638,
      "loss": 1.879,
      "step": 158500
    },
    {
      "epoch": 10.168841135840369,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0007966231772831927,
      "loss": 1.879,
      "step": 159000
    },
    {
      "epoch": 10.200818623688923,
      "grad_norm": 2.609375,
      "learning_rate": 0.0007959836275262215,
      "loss": 1.8792,
      "step": 159500
    },
    {
      "epoch": 10.232796111537478,
      "grad_norm": 0.294921875,
      "learning_rate": 0.0007953440777692505,
      "loss": 1.88,
      "step": 160000
    },
    {
      "epoch": 10.264773599386032,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0007947045280122793,
      "loss": 1.8792,
      "step": 160500
    },
    {
      "epoch": 10.296751087234586,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0007940649782553083,
      "loss": 1.8785,
      "step": 161000
    },
    {
      "epoch": 10.328728575083142,
      "grad_norm": 0.30078125,
      "learning_rate": 0.0007934254284983372,
      "loss": 1.8798,
      "step": 161500
    },
    {
      "epoch": 10.360706062931696,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0007927858787413661,
      "loss": 1.8801,
      "step": 162000
    },
    {
      "epoch": 10.392683550780252,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0007921463289843949,
      "loss": 1.8792,
      "step": 162500
    },
    {
      "epoch": 10.424661038628805,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0007915067792274239,
      "loss": 1.8787,
      "step": 163000
    },
    {
      "epoch": 10.45663852647736,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0007908672294704528,
      "loss": 1.8793,
      "step": 163500
    },
    {
      "epoch": 10.488616014325915,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0007902276797134818,
      "loss": 1.8795,
      "step": 164000
    },
    {
      "epoch": 10.520593502174469,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0007895881299565106,
      "loss": 1.8785,
      "step": 164500
    },
    {
      "epoch": 10.552570990023025,
      "grad_norm": 0.24609375,
      "learning_rate": 0.0007889485801995395,
      "loss": 1.879,
      "step": 165000
    },
    {
      "epoch": 10.584548477871579,
      "grad_norm": 0.33984375,
      "learning_rate": 0.0007883090304425684,
      "loss": 1.8793,
      "step": 165500
    },
    {
      "epoch": 10.616525965720133,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0007876694806855974,
      "loss": 1.8781,
      "step": 166000
    },
    {
      "epoch": 10.648503453568688,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0007870299309286262,
      "loss": 1.8796,
      "step": 166500
    },
    {
      "epoch": 10.680480941417242,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.0007863903811716552,
      "loss": 1.879,
      "step": 167000
    },
    {
      "epoch": 10.712458429265796,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.000785750831414684,
      "loss": 1.8787,
      "step": 167500
    },
    {
      "epoch": 10.744435917114352,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0007851112816577129,
      "loss": 1.8774,
      "step": 168000
    },
    {
      "epoch": 10.776413404962906,
      "grad_norm": 0.40625,
      "learning_rate": 0.000784471731900742,
      "loss": 1.8793,
      "step": 168500
    },
    {
      "epoch": 10.808390892811461,
      "grad_norm": 0.34765625,
      "learning_rate": 0.0007838321821437708,
      "loss": 1.8796,
      "step": 169000
    },
    {
      "epoch": 10.840368380660015,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0007831926323867998,
      "loss": 1.8782,
      "step": 169500
    },
    {
      "epoch": 10.87234586850857,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0007825530826298286,
      "loss": 1.8793,
      "step": 170000
    },
    {
      "epoch": 10.904323356357125,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0007819135328728575,
      "loss": 1.8795,
      "step": 170500
    },
    {
      "epoch": 10.936300844205679,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0007812739831158864,
      "loss": 1.8787,
      "step": 171000
    },
    {
      "epoch": 10.968278332054235,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0007806344333589154,
      "loss": 1.8786,
      "step": 171500
    },
    {
      "epoch": 11.0,
      "eval_loss": 3.7612926959991455,
      "eval_runtime": 6.0782,
      "eval_samples_per_second": 82.261,
      "eval_steps_per_second": 1.316,
      "step": 171996
    },
    {
      "epoch": 11.000255819902788,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0007799948836019442,
      "loss": 1.8785,
      "step": 172000
    },
    {
      "epoch": 11.032233307751342,
      "grad_norm": 0.236328125,
      "learning_rate": 0.0007793553338449732,
      "loss": 1.8787,
      "step": 172500
    },
    {
      "epoch": 11.064210795599898,
      "grad_norm": 0.17578125,
      "learning_rate": 0.000778715784088002,
      "loss": 1.8788,
      "step": 173000
    },
    {
      "epoch": 11.096188283448452,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0007780762343310309,
      "loss": 1.8778,
      "step": 173500
    },
    {
      "epoch": 11.128165771297008,
      "grad_norm": 0.31640625,
      "learning_rate": 0.0007774366845740599,
      "loss": 1.879,
      "step": 174000
    },
    {
      "epoch": 11.160143259145562,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0007767971348170888,
      "loss": 1.8782,
      "step": 174500
    },
    {
      "epoch": 11.192120746994116,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0007761575850601177,
      "loss": 1.8773,
      "step": 175000
    },
    {
      "epoch": 11.224098234842671,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0007755180353031466,
      "loss": 1.8782,
      "step": 175500
    },
    {
      "epoch": 11.256075722691225,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0007748784855461754,
      "loss": 1.8772,
      "step": 176000
    },
    {
      "epoch": 11.28805321053978,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0007742389357892045,
      "loss": 1.8781,
      "step": 176500
    },
    {
      "epoch": 11.320030698388335,
      "grad_norm": 0.2890625,
      "learning_rate": 0.0007735993860322333,
      "loss": 1.8773,
      "step": 177000
    },
    {
      "epoch": 11.352008186236889,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0007729598362752622,
      "loss": 1.8777,
      "step": 177500
    },
    {
      "epoch": 11.383985674085444,
      "grad_norm": 0.90625,
      "learning_rate": 0.0007723202865182911,
      "loss": 1.8781,
      "step": 178000
    },
    {
      "epoch": 11.415963161933998,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.00077168073676132,
      "loss": 1.8776,
      "step": 178500
    },
    {
      "epoch": 11.447940649782552,
      "grad_norm": 0.31640625,
      "learning_rate": 0.000771041187004349,
      "loss": 1.8759,
      "step": 179000
    },
    {
      "epoch": 11.479918137631108,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0007704016372473779,
      "loss": 1.8773,
      "step": 179500
    },
    {
      "epoch": 11.511895625479662,
      "grad_norm": 0.26171875,
      "learning_rate": 0.0007697620874904067,
      "loss": 1.8777,
      "step": 180000
    },
    {
      "epoch": 11.543873113328218,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0007691225377334357,
      "loss": 1.8783,
      "step": 180500
    },
    {
      "epoch": 11.575850601176771,
      "grad_norm": 0.40625,
      "learning_rate": 0.0007684829879764645,
      "loss": 1.878,
      "step": 181000
    },
    {
      "epoch": 11.607828089025325,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0007678434382194935,
      "loss": 1.8775,
      "step": 181500
    },
    {
      "epoch": 11.639805576873881,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0007672038884625225,
      "loss": 1.8773,
      "step": 182000
    },
    {
      "epoch": 11.671783064722435,
      "grad_norm": 0.423828125,
      "learning_rate": 0.0007665643387055513,
      "loss": 1.8775,
      "step": 182500
    },
    {
      "epoch": 11.70376055257099,
      "grad_norm": 0.421875,
      "learning_rate": 0.0007659247889485802,
      "loss": 1.8768,
      "step": 183000
    },
    {
      "epoch": 11.735738040419545,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0007652852391916091,
      "loss": 1.8769,
      "step": 183500
    },
    {
      "epoch": 11.767715528268099,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0007646456894346381,
      "loss": 1.8771,
      "step": 184000
    },
    {
      "epoch": 11.799693016116654,
      "grad_norm": 0.419921875,
      "learning_rate": 0.000764006139677667,
      "loss": 1.8769,
      "step": 184500
    },
    {
      "epoch": 11.831670503965208,
      "grad_norm": 0.4375,
      "learning_rate": 0.0007633665899206959,
      "loss": 1.8776,
      "step": 185000
    },
    {
      "epoch": 11.863647991813764,
      "grad_norm": 0.26953125,
      "learning_rate": 0.0007627270401637247,
      "loss": 1.8773,
      "step": 185500
    },
    {
      "epoch": 11.895625479662318,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0007620874904067537,
      "loss": 1.8772,
      "step": 186000
    },
    {
      "epoch": 11.927602967510872,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0007614479406497826,
      "loss": 1.8779,
      "step": 186500
    },
    {
      "epoch": 11.959580455359427,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0007608083908928115,
      "loss": 1.8773,
      "step": 187000
    },
    {
      "epoch": 11.991557943207981,
      "grad_norm": 0.294921875,
      "learning_rate": 0.0007601688411358404,
      "loss": 1.8775,
      "step": 187500
    },
    {
      "epoch": 12.0,
      "eval_loss": 3.757962465286255,
      "eval_runtime": 6.1778,
      "eval_samples_per_second": 80.935,
      "eval_steps_per_second": 1.295,
      "step": 187632
    },
    {
      "epoch": 12.023535431056537,
      "grad_norm": 0.224609375,
      "learning_rate": 0.0007595292913788693,
      "loss": 1.8775,
      "step": 188000
    },
    {
      "epoch": 12.055512918905091,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.0007588897416218981,
      "loss": 1.8769,
      "step": 188500
    },
    {
      "epoch": 12.087490406753645,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0007582501918649271,
      "loss": 1.8776,
      "step": 189000
    },
    {
      "epoch": 12.1194678946022,
      "grad_norm": 0.361328125,
      "learning_rate": 0.000757610642107956,
      "loss": 1.8772,
      "step": 189500
    },
    {
      "epoch": 12.151445382450754,
      "grad_norm": 0.36328125,
      "learning_rate": 0.000756971092350985,
      "loss": 1.877,
      "step": 190000
    },
    {
      "epoch": 12.18342287029931,
      "grad_norm": 0.181640625,
      "learning_rate": 0.0007563315425940138,
      "loss": 1.8772,
      "step": 190500
    },
    {
      "epoch": 12.215400358147864,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0007556919928370427,
      "loss": 1.8777,
      "step": 191000
    },
    {
      "epoch": 12.247377845996418,
      "grad_norm": 0.796875,
      "learning_rate": 0.0007550524430800716,
      "loss": 1.8766,
      "step": 191500
    },
    {
      "epoch": 12.279355333844974,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0007544128933231006,
      "loss": 1.8777,
      "step": 192000
    },
    {
      "epoch": 12.311332821693528,
      "grad_norm": 0.373046875,
      "learning_rate": 0.0007537733435661294,
      "loss": 1.8764,
      "step": 192500
    },
    {
      "epoch": 12.343310309542082,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0007531337938091584,
      "loss": 1.8776,
      "step": 193000
    },
    {
      "epoch": 12.375287797390637,
      "grad_norm": 2.65625,
      "learning_rate": 0.0007524942440521872,
      "loss": 1.8776,
      "step": 193500
    },
    {
      "epoch": 12.407265285239191,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0007518546942952161,
      "loss": 1.8766,
      "step": 194000
    },
    {
      "epoch": 12.439242773087747,
      "grad_norm": 0.177734375,
      "learning_rate": 0.0007512151445382452,
      "loss": 1.8766,
      "step": 194500
    },
    {
      "epoch": 12.4712202609363,
      "grad_norm": 0.2294921875,
      "learning_rate": 0.000750575594781274,
      "loss": 1.8761,
      "step": 195000
    },
    {
      "epoch": 12.503197748784855,
      "grad_norm": 0.263671875,
      "learning_rate": 0.000749936045024303,
      "loss": 1.8767,
      "step": 195500
    },
    {
      "epoch": 12.53517523663341,
      "grad_norm": 0.3125,
      "learning_rate": 0.0007492964952673318,
      "loss": 1.8767,
      "step": 196000
    },
    {
      "epoch": 12.567152724481964,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.0007486569455103606,
      "loss": 1.8772,
      "step": 196500
    },
    {
      "epoch": 12.59913021233052,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0007480173957533897,
      "loss": 1.8772,
      "step": 197000
    },
    {
      "epoch": 12.631107700179074,
      "grad_norm": 0.314453125,
      "learning_rate": 0.0007473778459964186,
      "loss": 1.8769,
      "step": 197500
    },
    {
      "epoch": 12.663085188027628,
      "grad_norm": 0.3203125,
      "learning_rate": 0.0007467382962394474,
      "loss": 1.8759,
      "step": 198000
    },
    {
      "epoch": 12.695062675876184,
      "grad_norm": 0.244140625,
      "learning_rate": 0.0007460987464824764,
      "loss": 1.8762,
      "step": 198500
    },
    {
      "epoch": 12.727040163724737,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0007454591967255052,
      "loss": 1.8782,
      "step": 199000
    },
    {
      "epoch": 12.759017651573293,
      "grad_norm": 60.0,
      "learning_rate": 0.0007448196469685342,
      "loss": 1.8776,
      "step": 199500
    },
    {
      "epoch": 12.790995139421847,
      "grad_norm": 0.345703125,
      "learning_rate": 0.0007441800972115631,
      "loss": 1.8765,
      "step": 200000
    },
    {
      "epoch": 12.822972627270401,
      "grad_norm": 0.3359375,
      "learning_rate": 0.000743540547454592,
      "loss": 1.8781,
      "step": 200500
    },
    {
      "epoch": 12.854950115118957,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0007429009976976209,
      "loss": 1.8764,
      "step": 201000
    },
    {
      "epoch": 12.88692760296751,
      "grad_norm": 0.34765625,
      "learning_rate": 0.0007422614479406498,
      "loss": 1.8766,
      "step": 201500
    },
    {
      "epoch": 12.918905090816066,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0007416218981836787,
      "loss": 1.8766,
      "step": 202000
    },
    {
      "epoch": 12.95088257866462,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0007409823484267077,
      "loss": 1.8772,
      "step": 202500
    },
    {
      "epoch": 12.982860066513174,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0007403427986697365,
      "loss": 1.8765,
      "step": 203000
    },
    {
      "epoch": 13.0,
      "eval_loss": 3.757168769836426,
      "eval_runtime": 9.5405,
      "eval_samples_per_second": 52.408,
      "eval_steps_per_second": 0.839,
      "step": 203268
    },
    {
      "epoch": 13.01483755436173,
      "grad_norm": 0.326171875,
      "learning_rate": 0.0007397032489127654,
      "loss": 1.8766,
      "step": 203500
    },
    {
      "epoch": 13.046815042210284,
      "grad_norm": 0.287109375,
      "learning_rate": 0.0007390636991557943,
      "loss": 1.8753,
      "step": 204000
    },
    {
      "epoch": 13.078792530058838,
      "grad_norm": 0.326171875,
      "learning_rate": 0.0007384241493988233,
      "loss": 1.8766,
      "step": 204500
    },
    {
      "epoch": 13.110770017907393,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0007377845996418521,
      "loss": 1.8763,
      "step": 205000
    },
    {
      "epoch": 13.142747505755947,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0007371450498848811,
      "loss": 1.8769,
      "step": 205500
    },
    {
      "epoch": 13.174724993604503,
      "grad_norm": 0.5625,
      "learning_rate": 0.0007365055001279099,
      "loss": 1.8761,
      "step": 206000
    },
    {
      "epoch": 13.206702481453057,
      "grad_norm": 0.1962890625,
      "learning_rate": 0.0007358659503709389,
      "loss": 1.8761,
      "step": 206500
    },
    {
      "epoch": 13.23867996930161,
      "grad_norm": 0.2001953125,
      "learning_rate": 0.0007352264006139677,
      "loss": 1.8765,
      "step": 207000
    },
    {
      "epoch": 13.270657457150167,
      "grad_norm": 0.28515625,
      "learning_rate": 0.0007345868508569967,
      "loss": 1.8757,
      "step": 207500
    },
    {
      "epoch": 13.30263494499872,
      "grad_norm": 0.396484375,
      "learning_rate": 0.0007339473011000256,
      "loss": 1.8751,
      "step": 208000
    },
    {
      "epoch": 13.334612432847276,
      "grad_norm": 0.2734375,
      "learning_rate": 0.0007333077513430545,
      "loss": 1.8759,
      "step": 208500
    },
    {
      "epoch": 13.36658992069583,
      "grad_norm": 0.2294921875,
      "learning_rate": 0.0007326682015860833,
      "loss": 1.8766,
      "step": 209000
    },
    {
      "epoch": 13.398567408544384,
      "grad_norm": 1.09375,
      "learning_rate": 0.0007320286518291123,
      "loss": 1.876,
      "step": 209500
    },
    {
      "epoch": 13.43054489639294,
      "grad_norm": 0.2890625,
      "learning_rate": 0.0007313891020721413,
      "loss": 1.8767,
      "step": 210000
    },
    {
      "epoch": 13.462522384241494,
      "grad_norm": 0.345703125,
      "learning_rate": 0.0007307495523151701,
      "loss": 1.8757,
      "step": 210500
    },
    {
      "epoch": 13.49449987209005,
      "grad_norm": 0.203125,
      "learning_rate": 0.0007301100025581991,
      "loss": 1.8769,
      "step": 211000
    },
    {
      "epoch": 13.526477359938603,
      "grad_norm": 0.228515625,
      "learning_rate": 0.0007294704528012279,
      "loss": 1.8765,
      "step": 211500
    },
    {
      "epoch": 13.558454847787157,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0007288309030442569,
      "loss": 1.8756,
      "step": 212000
    },
    {
      "epoch": 13.590432335635713,
      "grad_norm": 0.23828125,
      "learning_rate": 0.0007281913532872858,
      "loss": 1.8754,
      "step": 212500
    },
    {
      "epoch": 13.622409823484267,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0007275518035303147,
      "loss": 1.8756,
      "step": 213000
    },
    {
      "epoch": 13.654387311332822,
      "grad_norm": 0.228515625,
      "learning_rate": 0.0007269122537733436,
      "loss": 1.8757,
      "step": 213500
    },
    {
      "epoch": 13.686364799181376,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0007262727040163725,
      "loss": 1.876,
      "step": 214000
    },
    {
      "epoch": 13.71834228702993,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0007256331542594013,
      "loss": 1.8762,
      "step": 214500
    },
    {
      "epoch": 13.750319774878486,
      "grad_norm": 0.255859375,
      "learning_rate": 0.0007249936045024304,
      "loss": 1.875,
      "step": 215000
    },
    {
      "epoch": 13.78229726272704,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0007243540547454592,
      "loss": 1.875,
      "step": 215500
    },
    {
      "epoch": 13.814274750575596,
      "grad_norm": 0.28125,
      "learning_rate": 0.0007237145049884881,
      "loss": 1.8762,
      "step": 216000
    },
    {
      "epoch": 13.84625223842415,
      "grad_norm": 0.240234375,
      "learning_rate": 0.000723074955231517,
      "loss": 1.8748,
      "step": 216500
    },
    {
      "epoch": 13.878229726272703,
      "grad_norm": 0.294921875,
      "learning_rate": 0.0007224354054745459,
      "loss": 1.8758,
      "step": 217000
    },
    {
      "epoch": 13.91020721412126,
      "grad_norm": 0.671875,
      "learning_rate": 0.0007217958557175749,
      "loss": 1.876,
      "step": 217500
    },
    {
      "epoch": 13.942184701969813,
      "grad_norm": 0.1962890625,
      "learning_rate": 0.0007211563059606038,
      "loss": 1.874,
      "step": 218000
    },
    {
      "epoch": 13.974162189818369,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0007205167562036326,
      "loss": 1.8746,
      "step": 218500
    },
    {
      "epoch": 14.0,
      "eval_loss": 3.752974033355713,
      "eval_runtime": 5.3176,
      "eval_samples_per_second": 94.027,
      "eval_steps_per_second": 1.504,
      "step": 218904
    },
    {
      "epoch": 14.006139677666923,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0007198772064466616,
      "loss": 1.8744,
      "step": 219000
    },
    {
      "epoch": 14.038117165515477,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0007192376566896904,
      "loss": 1.8739,
      "step": 219500
    },
    {
      "epoch": 14.070094653364032,
      "grad_norm": 0.296875,
      "learning_rate": 0.0007185981069327194,
      "loss": 1.8751,
      "step": 220000
    },
    {
      "epoch": 14.102072141212586,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0007179585571757483,
      "loss": 1.8755,
      "step": 220500
    },
    {
      "epoch": 14.13404962906114,
      "grad_norm": 0.2490234375,
      "learning_rate": 0.0007173190074187772,
      "loss": 1.8745,
      "step": 221000
    },
    {
      "epoch": 14.166027116909696,
      "grad_norm": 0.265625,
      "learning_rate": 0.000716679457661806,
      "loss": 1.8744,
      "step": 221500
    },
    {
      "epoch": 14.19800460475825,
      "grad_norm": 0.359375,
      "learning_rate": 0.000716039907904835,
      "loss": 1.8739,
      "step": 222000
    },
    {
      "epoch": 14.229982092606805,
      "grad_norm": 0.287109375,
      "learning_rate": 0.000715400358147864,
      "loss": 1.8738,
      "step": 222500
    },
    {
      "epoch": 14.26195958045536,
      "grad_norm": 1.25,
      "learning_rate": 0.0007147608083908929,
      "loss": 1.8747,
      "step": 223000
    },
    {
      "epoch": 14.293937068303913,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0007141212586339217,
      "loss": 1.8747,
      "step": 223500
    },
    {
      "epoch": 14.325914556152469,
      "grad_norm": 0.408203125,
      "learning_rate": 0.0007134817088769506,
      "loss": 1.8751,
      "step": 224000
    },
    {
      "epoch": 14.357892044001023,
      "grad_norm": 0.341796875,
      "learning_rate": 0.0007128421591199796,
      "loss": 1.8744,
      "step": 224500
    },
    {
      "epoch": 14.389869531849579,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0007122026093630084,
      "loss": 1.8742,
      "step": 225000
    },
    {
      "epoch": 14.421847019698133,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0007115630596060374,
      "loss": 1.8753,
      "step": 225500
    },
    {
      "epoch": 14.453824507546686,
      "grad_norm": 1.46875,
      "learning_rate": 0.0007109235098490663,
      "loss": 1.8746,
      "step": 226000
    },
    {
      "epoch": 14.485801995395242,
      "grad_norm": 0.341796875,
      "learning_rate": 0.0007102839600920952,
      "loss": 1.8737,
      "step": 226500
    },
    {
      "epoch": 14.517779483243796,
      "grad_norm": 0.263671875,
      "learning_rate": 0.000709644410335124,
      "loss": 1.8741,
      "step": 227000
    },
    {
      "epoch": 14.549756971092352,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.000709004860578153,
      "loss": 1.8745,
      "step": 227500
    },
    {
      "epoch": 14.581734458940906,
      "grad_norm": 0.18359375,
      "learning_rate": 0.0007083653108211819,
      "loss": 1.8745,
      "step": 228000
    },
    {
      "epoch": 14.61371194678946,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.0007077257610642109,
      "loss": 1.8748,
      "step": 228500
    },
    {
      "epoch": 14.645689434638015,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0007070862113072397,
      "loss": 1.8752,
      "step": 229000
    },
    {
      "epoch": 14.67766692248657,
      "grad_norm": 0.20703125,
      "learning_rate": 0.0007064466615502686,
      "loss": 1.8741,
      "step": 229500
    },
    {
      "epoch": 14.709644410335123,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0007058071117932975,
      "loss": 1.8756,
      "step": 230000
    },
    {
      "epoch": 14.741621898183679,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0007051675620363265,
      "loss": 1.8734,
      "step": 230500
    },
    {
      "epoch": 14.773599386032233,
      "grad_norm": 0.375,
      "learning_rate": 0.0007045280122793553,
      "loss": 1.8757,
      "step": 231000
    },
    {
      "epoch": 14.805576873880788,
      "grad_norm": 0.28125,
      "learning_rate": 0.0007038884625223843,
      "loss": 1.8741,
      "step": 231500
    },
    {
      "epoch": 14.837554361729342,
      "grad_norm": 0.20703125,
      "learning_rate": 0.0007032489127654131,
      "loss": 1.8734,
      "step": 232000
    },
    {
      "epoch": 14.869531849577896,
      "grad_norm": 0.236328125,
      "learning_rate": 0.0007026093630084421,
      "loss": 1.8747,
      "step": 232500
    },
    {
      "epoch": 14.901509337426452,
      "grad_norm": 0.474609375,
      "learning_rate": 0.000701969813251471,
      "loss": 1.8735,
      "step": 233000
    },
    {
      "epoch": 14.933486825275006,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0007013302634944999,
      "loss": 1.875,
      "step": 233500
    },
    {
      "epoch": 14.965464313123562,
      "grad_norm": 0.30078125,
      "learning_rate": 0.0007006907137375288,
      "loss": 1.875,
      "step": 234000
    },
    {
      "epoch": 14.997441800972116,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0007000511639805577,
      "loss": 1.8735,
      "step": 234500
    },
    {
      "epoch": 15.0,
      "eval_loss": 3.751368761062622,
      "eval_runtime": 5.6537,
      "eval_samples_per_second": 88.437,
      "eval_steps_per_second": 1.415,
      "step": 234540
    },
    {
      "epoch": 15.02941928882067,
      "grad_norm": 0.287109375,
      "learning_rate": 0.0006994116142235865,
      "loss": 1.8741,
      "step": 235000
    },
    {
      "epoch": 15.061396776669225,
      "grad_norm": 0.408203125,
      "learning_rate": 0.0006987720644666156,
      "loss": 1.8739,
      "step": 235500
    },
    {
      "epoch": 15.093374264517779,
      "grad_norm": 0.33984375,
      "learning_rate": 0.0006981325147096444,
      "loss": 1.875,
      "step": 236000
    },
    {
      "epoch": 15.125351752366335,
      "grad_norm": 0.2177734375,
      "learning_rate": 0.0006974929649526733,
      "loss": 1.8751,
      "step": 236500
    },
    {
      "epoch": 15.157329240214889,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0006968534151957022,
      "loss": 1.8744,
      "step": 237000
    },
    {
      "epoch": 15.189306728063443,
      "grad_norm": 0.26953125,
      "learning_rate": 0.0006962138654387311,
      "loss": 1.8749,
      "step": 237500
    },
    {
      "epoch": 15.221284215911998,
      "grad_norm": 0.20703125,
      "learning_rate": 0.0006955743156817602,
      "loss": 1.8731,
      "step": 238000
    },
    {
      "epoch": 15.253261703760552,
      "grad_norm": 0.466796875,
      "learning_rate": 0.000694934765924789,
      "loss": 1.8734,
      "step": 238500
    },
    {
      "epoch": 15.285239191609108,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0006942952161678179,
      "loss": 1.8747,
      "step": 239000
    },
    {
      "epoch": 15.317216679457662,
      "grad_norm": 0.26953125,
      "learning_rate": 0.0006936556664108468,
      "loss": 1.874,
      "step": 239500
    },
    {
      "epoch": 15.349194167306216,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0006930161166538757,
      "loss": 1.8741,
      "step": 240000
    },
    {
      "epoch": 15.381171655154771,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.0006923765668969046,
      "loss": 1.8743,
      "step": 240500
    },
    {
      "epoch": 15.413149143003325,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.0006917370171399336,
      "loss": 1.8749,
      "step": 241000
    },
    {
      "epoch": 15.445126630851881,
      "grad_norm": 0.2275390625,
      "learning_rate": 0.0006910974673829624,
      "loss": 1.8743,
      "step": 241500
    },
    {
      "epoch": 15.477104118700435,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.0006904579176259913,
      "loss": 1.8739,
      "step": 242000
    },
    {
      "epoch": 15.509081606548989,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0006898183678690202,
      "loss": 1.874,
      "step": 242500
    },
    {
      "epoch": 15.541059094397545,
      "grad_norm": 0.3984375,
      "learning_rate": 0.0006891788181120491,
      "loss": 1.8732,
      "step": 243000
    },
    {
      "epoch": 15.573036582246099,
      "grad_norm": 0.234375,
      "learning_rate": 0.0006885392683550781,
      "loss": 1.8735,
      "step": 243500
    },
    {
      "epoch": 15.605014070094654,
      "grad_norm": 0.3984375,
      "learning_rate": 0.000687899718598107,
      "loss": 1.8736,
      "step": 244000
    },
    {
      "epoch": 15.636991557943208,
      "grad_norm": 0.1884765625,
      "learning_rate": 0.0006872601688411358,
      "loss": 1.873,
      "step": 244500
    },
    {
      "epoch": 15.668969045791762,
      "grad_norm": 0.30078125,
      "learning_rate": 0.0006866206190841648,
      "loss": 1.8736,
      "step": 245000
    },
    {
      "epoch": 15.700946533640318,
      "grad_norm": 0.328125,
      "learning_rate": 0.0006859810693271936,
      "loss": 1.874,
      "step": 245500
    },
    {
      "epoch": 15.732924021488872,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0006853415195702226,
      "loss": 1.8727,
      "step": 246000
    },
    {
      "epoch": 15.764901509337426,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0006847019698132515,
      "loss": 1.8744,
      "step": 246500
    },
    {
      "epoch": 15.796878997185981,
      "grad_norm": 0.2392578125,
      "learning_rate": 0.0006840624200562804,
      "loss": 1.8734,
      "step": 247000
    },
    {
      "epoch": 15.828856485034535,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.0006834228702993092,
      "loss": 1.8738,
      "step": 247500
    },
    {
      "epoch": 15.860833972883091,
      "grad_norm": 0.216796875,
      "learning_rate": 0.0006827833205423382,
      "loss": 1.8718,
      "step": 248000
    },
    {
      "epoch": 15.892811460731645,
      "grad_norm": 0.26171875,
      "learning_rate": 0.0006821437707853671,
      "loss": 1.8738,
      "step": 248500
    },
    {
      "epoch": 15.924788948580199,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0006815042210283961,
      "loss": 1.8721,
      "step": 249000
    },
    {
      "epoch": 15.956766436428754,
      "grad_norm": 0.2890625,
      "learning_rate": 0.0006808646712714249,
      "loss": 1.8734,
      "step": 249500
    },
    {
      "epoch": 15.988743924277308,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0006802251215144538,
      "loss": 1.8731,
      "step": 250000
    },
    {
      "epoch": 16.0,
      "eval_loss": 3.7499747276306152,
      "eval_runtime": 5.866,
      "eval_samples_per_second": 85.238,
      "eval_steps_per_second": 1.364,
      "step": 250176
    },
    {
      "epoch": 16.020721412125862,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0006795855717574827,
      "loss": 1.8732,
      "step": 250500
    },
    {
      "epoch": 16.052698899974416,
      "grad_norm": 0.2734375,
      "learning_rate": 0.0006789460220005117,
      "loss": 1.8737,
      "step": 251000
    },
    {
      "epoch": 16.084676387822974,
      "grad_norm": 0.2451171875,
      "learning_rate": 0.0006783064722435405,
      "loss": 1.8727,
      "step": 251500
    },
    {
      "epoch": 16.116653875671528,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0006776669224865695,
      "loss": 1.8731,
      "step": 252000
    },
    {
      "epoch": 16.14863136352008,
      "grad_norm": 0.212890625,
      "learning_rate": 0.0006770273727295983,
      "loss": 1.8737,
      "step": 252500
    },
    {
      "epoch": 16.180608851368635,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0006763878229726272,
      "loss": 1.8728,
      "step": 253000
    },
    {
      "epoch": 16.21258633921719,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0006757482732156563,
      "loss": 1.8735,
      "step": 253500
    },
    {
      "epoch": 16.244563827065747,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0006751087234586851,
      "loss": 1.8722,
      "step": 254000
    },
    {
      "epoch": 16.2765413149143,
      "grad_norm": 0.240234375,
      "learning_rate": 0.0006744691737017141,
      "loss": 1.8743,
      "step": 254500
    },
    {
      "epoch": 16.308518802762855,
      "grad_norm": 0.265625,
      "learning_rate": 0.0006738296239447429,
      "loss": 1.871,
      "step": 255000
    },
    {
      "epoch": 16.34049629061141,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0006731900741877718,
      "loss": 1.8727,
      "step": 255500
    },
    {
      "epoch": 16.372473778459963,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0006725505244308008,
      "loss": 1.8746,
      "step": 256000
    },
    {
      "epoch": 16.40445126630852,
      "grad_norm": 0.328125,
      "learning_rate": 0.0006719109746738297,
      "loss": 1.8715,
      "step": 256500
    },
    {
      "epoch": 16.436428754157074,
      "grad_norm": 0.2060546875,
      "learning_rate": 0.0006712714249168585,
      "loss": 1.8731,
      "step": 257000
    },
    {
      "epoch": 16.468406242005628,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0006706318751598875,
      "loss": 1.8731,
      "step": 257500
    },
    {
      "epoch": 16.500383729854182,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0006699923254029163,
      "loss": 1.8735,
      "step": 258000
    },
    {
      "epoch": 16.532361217702736,
      "grad_norm": 0.361328125,
      "learning_rate": 0.0006693527756459453,
      "loss": 1.8731,
      "step": 258500
    },
    {
      "epoch": 16.564338705551293,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0006687132258889742,
      "loss": 1.8719,
      "step": 259000
    },
    {
      "epoch": 16.596316193399847,
      "grad_norm": 0.2451171875,
      "learning_rate": 0.0006680736761320031,
      "loss": 1.8738,
      "step": 259500
    },
    {
      "epoch": 16.6282936812484,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.000667434126375032,
      "loss": 1.8724,
      "step": 260000
    },
    {
      "epoch": 16.660271169096955,
      "grad_norm": 0.376953125,
      "learning_rate": 0.0006667945766180609,
      "loss": 1.8721,
      "step": 260500
    },
    {
      "epoch": 16.69224865694551,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0006661550268610897,
      "loss": 1.8715,
      "step": 261000
    },
    {
      "epoch": 16.724226144794066,
      "grad_norm": 0.216796875,
      "learning_rate": 0.0006655154771041188,
      "loss": 1.8727,
      "step": 261500
    },
    {
      "epoch": 16.75620363264262,
      "grad_norm": 0.1923828125,
      "learning_rate": 0.0006648759273471476,
      "loss": 1.8725,
      "step": 262000
    },
    {
      "epoch": 16.788181120491174,
      "grad_norm": 0.255859375,
      "learning_rate": 0.0006642363775901765,
      "loss": 1.8714,
      "step": 262500
    },
    {
      "epoch": 16.820158608339728,
      "grad_norm": 0.19921875,
      "learning_rate": 0.0006635968278332054,
      "loss": 1.8734,
      "step": 263000
    },
    {
      "epoch": 16.852136096188282,
      "grad_norm": 0.234375,
      "learning_rate": 0.0006629572780762343,
      "loss": 1.8716,
      "step": 263500
    },
    {
      "epoch": 16.88411358403684,
      "grad_norm": 0.236328125,
      "learning_rate": 0.0006623177283192632,
      "loss": 1.8727,
      "step": 264000
    },
    {
      "epoch": 16.916091071885393,
      "grad_norm": 0.1845703125,
      "learning_rate": 0.0006616781785622922,
      "loss": 1.8723,
      "step": 264500
    },
    {
      "epoch": 16.948068559733947,
      "grad_norm": 0.400390625,
      "learning_rate": 0.000661038628805321,
      "loss": 1.872,
      "step": 265000
    },
    {
      "epoch": 16.9800460475825,
      "grad_norm": 0.1669921875,
      "learning_rate": 0.00066039907904835,
      "loss": 1.8721,
      "step": 265500
    },
    {
      "epoch": 17.0,
      "eval_loss": 3.7480361461639404,
      "eval_runtime": 5.2443,
      "eval_samples_per_second": 95.341,
      "eval_steps_per_second": 1.525,
      "step": 265812
    },
    {
      "epoch": 17.012023535431055,
      "grad_norm": 0.208984375,
      "learning_rate": 0.0006597595292913788,
      "loss": 1.8729,
      "step": 266000
    },
    {
      "epoch": 17.044001023279613,
      "grad_norm": 0.625,
      "learning_rate": 0.0006591199795344078,
      "loss": 1.8719,
      "step": 266500
    },
    {
      "epoch": 17.075978511128167,
      "grad_norm": 0.265625,
      "learning_rate": 0.0006584804297774368,
      "loss": 1.8722,
      "step": 267000
    },
    {
      "epoch": 17.10795599897672,
      "grad_norm": 0.21875,
      "learning_rate": 0.0006578408800204656,
      "loss": 1.8714,
      "step": 267500
    },
    {
      "epoch": 17.139933486825274,
      "grad_norm": 0.23828125,
      "learning_rate": 0.0006572013302634944,
      "loss": 1.8726,
      "step": 268000
    },
    {
      "epoch": 17.17191097467383,
      "grad_norm": 0.2734375,
      "learning_rate": 0.0006565617805065234,
      "loss": 1.8719,
      "step": 268500
    },
    {
      "epoch": 17.203888462522386,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0006559222307495524,
      "loss": 1.8726,
      "step": 269000
    },
    {
      "epoch": 17.23586595037094,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0006552826809925812,
      "loss": 1.8723,
      "step": 269500
    },
    {
      "epoch": 17.267843438219494,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0006546431312356102,
      "loss": 1.8718,
      "step": 270000
    },
    {
      "epoch": 17.299820926068048,
      "grad_norm": 0.35546875,
      "learning_rate": 0.000654003581478639,
      "loss": 1.8712,
      "step": 270500
    },
    {
      "epoch": 17.3317984139166,
      "grad_norm": 0.296875,
      "learning_rate": 0.000653364031721668,
      "loss": 1.8717,
      "step": 271000
    },
    {
      "epoch": 17.36377590176516,
      "grad_norm": 0.2890625,
      "learning_rate": 0.0006527244819646969,
      "loss": 1.8718,
      "step": 271500
    },
    {
      "epoch": 17.395753389613713,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0006520849322077258,
      "loss": 1.8717,
      "step": 272000
    },
    {
      "epoch": 17.427730877462267,
      "grad_norm": 0.263671875,
      "learning_rate": 0.0006514453824507547,
      "loss": 1.8727,
      "step": 272500
    },
    {
      "epoch": 17.45970836531082,
      "grad_norm": 0.21484375,
      "learning_rate": 0.0006508058326937836,
      "loss": 1.8713,
      "step": 273000
    },
    {
      "epoch": 17.491685853159375,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0006501662829368124,
      "loss": 1.8727,
      "step": 273500
    },
    {
      "epoch": 17.523663341007932,
      "grad_norm": 0.37890625,
      "learning_rate": 0.0006495267331798415,
      "loss": 1.8708,
      "step": 274000
    },
    {
      "epoch": 17.555640828856486,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.0006488871834228703,
      "loss": 1.8717,
      "step": 274500
    },
    {
      "epoch": 17.58761831670504,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0006482476336658993,
      "loss": 1.8717,
      "step": 275000
    },
    {
      "epoch": 17.619595804553594,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0006476080839089281,
      "loss": 1.8713,
      "step": 275500
    },
    {
      "epoch": 17.651573292402148,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.000646968534151957,
      "loss": 1.8718,
      "step": 276000
    },
    {
      "epoch": 17.683550780250705,
      "grad_norm": 0.1884765625,
      "learning_rate": 0.000646328984394986,
      "loss": 1.8727,
      "step": 276500
    },
    {
      "epoch": 17.71552826809926,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0006456894346380149,
      "loss": 1.8714,
      "step": 277000
    },
    {
      "epoch": 17.747505755947813,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0006450498848810437,
      "loss": 1.8709,
      "step": 277500
    },
    {
      "epoch": 17.779483243796367,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0006444103351240727,
      "loss": 1.872,
      "step": 278000
    },
    {
      "epoch": 17.81146073164492,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0006437707853671015,
      "loss": 1.872,
      "step": 278500
    },
    {
      "epoch": 17.84343821949348,
      "grad_norm": 0.23828125,
      "learning_rate": 0.0006431312356101304,
      "loss": 1.872,
      "step": 279000
    },
    {
      "epoch": 17.875415707342032,
      "grad_norm": 0.181640625,
      "learning_rate": 0.0006424916858531594,
      "loss": 1.8723,
      "step": 279500
    },
    {
      "epoch": 17.907393195190586,
      "grad_norm": 0.21875,
      "learning_rate": 0.0006418521360961883,
      "loss": 1.8717,
      "step": 280000
    },
    {
      "epoch": 17.93937068303914,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0006412125863392172,
      "loss": 1.8721,
      "step": 280500
    },
    {
      "epoch": 17.971348170887694,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0006405730365822461,
      "loss": 1.8721,
      "step": 281000
    },
    {
      "epoch": 18.0,
      "eval_loss": 3.7465760707855225,
      "eval_runtime": 6.2329,
      "eval_samples_per_second": 80.22,
      "eval_steps_per_second": 1.284,
      "step": 281448
    },
    {
      "epoch": 18.003325658736248,
      "grad_norm": 0.330078125,
      "learning_rate": 0.0006399334868252749,
      "loss": 1.8722,
      "step": 281500
    },
    {
      "epoch": 18.035303146584805,
      "grad_norm": 0.7421875,
      "learning_rate": 0.000639293937068304,
      "loss": 1.871,
      "step": 282000
    },
    {
      "epoch": 18.06728063443336,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.0006386543873113329,
      "loss": 1.8722,
      "step": 282500
    },
    {
      "epoch": 18.099258122281913,
      "grad_norm": 0.2099609375,
      "learning_rate": 0.0006380148375543617,
      "loss": 1.8707,
      "step": 283000
    },
    {
      "epoch": 18.131235610130467,
      "grad_norm": 0.224609375,
      "learning_rate": 0.0006373752877973907,
      "loss": 1.8711,
      "step": 283500
    },
    {
      "epoch": 18.16321309797902,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0006367357380404195,
      "loss": 1.8712,
      "step": 284000
    },
    {
      "epoch": 18.19519058582758,
      "grad_norm": 0.2412109375,
      "learning_rate": 0.0006360961882834485,
      "loss": 1.871,
      "step": 284500
    },
    {
      "epoch": 18.227168073676133,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0006354566385264774,
      "loss": 1.8717,
      "step": 285000
    },
    {
      "epoch": 18.259145561524686,
      "grad_norm": 0.169921875,
      "learning_rate": 0.0006348170887695063,
      "loss": 1.8714,
      "step": 285500
    },
    {
      "epoch": 18.29112304937324,
      "grad_norm": 0.201171875,
      "learning_rate": 0.0006341775390125352,
      "loss": 1.8707,
      "step": 286000
    },
    {
      "epoch": 18.323100537221794,
      "grad_norm": 0.234375,
      "learning_rate": 0.0006335379892555641,
      "loss": 1.871,
      "step": 286500
    },
    {
      "epoch": 18.355078025070352,
      "grad_norm": 0.3515625,
      "learning_rate": 0.000632898439498593,
      "loss": 1.8718,
      "step": 287000
    },
    {
      "epoch": 18.387055512918906,
      "grad_norm": 0.228515625,
      "learning_rate": 0.000632258889741622,
      "loss": 1.8708,
      "step": 287500
    },
    {
      "epoch": 18.41903300076746,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0006316193399846508,
      "loss": 1.8713,
      "step": 288000
    },
    {
      "epoch": 18.451010488616014,
      "grad_norm": 0.30078125,
      "learning_rate": 0.0006309797902276797,
      "loss": 1.8709,
      "step": 288500
    },
    {
      "epoch": 18.482987976464567,
      "grad_norm": 0.21875,
      "learning_rate": 0.0006303402404707086,
      "loss": 1.8718,
      "step": 289000
    },
    {
      "epoch": 18.514965464313125,
      "grad_norm": 0.345703125,
      "learning_rate": 0.0006297006907137376,
      "loss": 1.8713,
      "step": 289500
    },
    {
      "epoch": 18.54694295216168,
      "grad_norm": 0.19140625,
      "learning_rate": 0.0006290611409567664,
      "loss": 1.8718,
      "step": 290000
    },
    {
      "epoch": 18.578920440010233,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0006284215911997954,
      "loss": 1.8715,
      "step": 290500
    },
    {
      "epoch": 18.610897927858787,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.0006277820414428242,
      "loss": 1.8721,
      "step": 291000
    },
    {
      "epoch": 18.64287541570734,
      "grad_norm": 0.2109375,
      "learning_rate": 0.0006271424916858532,
      "loss": 1.8716,
      "step": 291500
    },
    {
      "epoch": 18.674852903555898,
      "grad_norm": 0.2177734375,
      "learning_rate": 0.0006265029419288821,
      "loss": 1.8711,
      "step": 292000
    },
    {
      "epoch": 18.706830391404452,
      "grad_norm": 0.1962890625,
      "learning_rate": 0.000625863392171911,
      "loss": 1.8725,
      "step": 292500
    },
    {
      "epoch": 18.738807879253006,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0006252238424149399,
      "loss": 1.8719,
      "step": 293000
    },
    {
      "epoch": 18.77078536710156,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0006245842926579688,
      "loss": 1.8717,
      "step": 293500
    },
    {
      "epoch": 18.802762854950114,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0006239447429009976,
      "loss": 1.8719,
      "step": 294000
    },
    {
      "epoch": 18.83474034279867,
      "grad_norm": 0.34765625,
      "learning_rate": 0.0006233051931440267,
      "loss": 1.8718,
      "step": 294500
    },
    {
      "epoch": 18.866717830647225,
      "grad_norm": 0.2412109375,
      "learning_rate": 0.0006226656433870555,
      "loss": 1.8713,
      "step": 295000
    },
    {
      "epoch": 18.89869531849578,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0006220260936300844,
      "loss": 1.8708,
      "step": 295500
    },
    {
      "epoch": 18.930672806344333,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0006213865438731134,
      "loss": 1.8722,
      "step": 296000
    },
    {
      "epoch": 18.962650294192887,
      "grad_norm": 0.2177734375,
      "learning_rate": 0.0006207469941161422,
      "loss": 1.8709,
      "step": 296500
    },
    {
      "epoch": 18.994627782041444,
      "grad_norm": 0.388671875,
      "learning_rate": 0.0006201074443591712,
      "loss": 1.8705,
      "step": 297000
    },
    {
      "epoch": 19.0,
      "eval_loss": 3.7460944652557373,
      "eval_runtime": 6.5742,
      "eval_samples_per_second": 76.055,
      "eval_steps_per_second": 1.217,
      "step": 297084
    },
    {
      "epoch": 19.02660526989,
      "grad_norm": 0.33984375,
      "learning_rate": 0.0006194678946022001,
      "loss": 1.872,
      "step": 297500
    },
    {
      "epoch": 19.058582757738552,
      "grad_norm": 0.392578125,
      "learning_rate": 0.000618828344845229,
      "loss": 1.8707,
      "step": 298000
    },
    {
      "epoch": 19.090560245587106,
      "grad_norm": 0.34375,
      "learning_rate": 0.0006181887950882579,
      "loss": 1.8704,
      "step": 298500
    },
    {
      "epoch": 19.12253773343566,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0006175492453312868,
      "loss": 1.8708,
      "step": 299000
    },
    {
      "epoch": 19.154515221284218,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0006169096955743156,
      "loss": 1.8723,
      "step": 299500
    },
    {
      "epoch": 19.18649270913277,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.0006162701458173447,
      "loss": 1.872,
      "step": 300000
    },
    {
      "epoch": 19.218470196981325,
      "grad_norm": 0.2275390625,
      "learning_rate": 0.0006156305960603735,
      "loss": 1.8719,
      "step": 300500
    },
    {
      "epoch": 19.25044768482988,
      "grad_norm": 0.244140625,
      "learning_rate": 0.0006149910463034024,
      "loss": 1.8711,
      "step": 301000
    },
    {
      "epoch": 19.282425172678433,
      "grad_norm": 0.2060546875,
      "learning_rate": 0.0006143514965464313,
      "loss": 1.871,
      "step": 301500
    },
    {
      "epoch": 19.31440266052699,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0006137119467894602,
      "loss": 1.8707,
      "step": 302000
    },
    {
      "epoch": 19.346380148375545,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0006130723970324892,
      "loss": 1.8715,
      "step": 302500
    },
    {
      "epoch": 19.3783576362241,
      "grad_norm": 0.283203125,
      "learning_rate": 0.0006124328472755181,
      "loss": 1.8707,
      "step": 303000
    },
    {
      "epoch": 19.410335124072652,
      "grad_norm": 0.263671875,
      "learning_rate": 0.0006117932975185469,
      "loss": 1.8702,
      "step": 303500
    },
    {
      "epoch": 19.442312611921206,
      "grad_norm": 0.2001953125,
      "learning_rate": 0.0006111537477615759,
      "loss": 1.8714,
      "step": 304000
    },
    {
      "epoch": 19.47429009976976,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0006105141980046047,
      "loss": 1.871,
      "step": 304500
    },
    {
      "epoch": 19.506267587618318,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0006098746482476337,
      "loss": 1.87,
      "step": 305000
    },
    {
      "epoch": 19.53824507546687,
      "grad_norm": 0.2099609375,
      "learning_rate": 0.0006092350984906626,
      "loss": 1.8711,
      "step": 305500
    },
    {
      "epoch": 19.570222563315426,
      "grad_norm": 0.337890625,
      "learning_rate": 0.0006085955487336915,
      "loss": 1.8716,
      "step": 306000
    },
    {
      "epoch": 19.60220005116398,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0006079559989767203,
      "loss": 1.8707,
      "step": 306500
    },
    {
      "epoch": 19.634177539012533,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0006073164492197493,
      "loss": 1.8709,
      "step": 307000
    },
    {
      "epoch": 19.66615502686109,
      "grad_norm": 0.18359375,
      "learning_rate": 0.0006066768994627782,
      "loss": 1.872,
      "step": 307500
    },
    {
      "epoch": 19.698132514709645,
      "grad_norm": 0.376953125,
      "learning_rate": 0.0006060373497058072,
      "loss": 1.8714,
      "step": 308000
    },
    {
      "epoch": 19.7301100025582,
      "grad_norm": 0.291015625,
      "learning_rate": 0.000605397799948836,
      "loss": 1.8716,
      "step": 308500
    },
    {
      "epoch": 19.762087490406753,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0006047582501918649,
      "loss": 1.8706,
      "step": 309000
    },
    {
      "epoch": 19.794064978255307,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0006041187004348938,
      "loss": 1.8716,
      "step": 309500
    },
    {
      "epoch": 19.826042466103864,
      "grad_norm": 0.20703125,
      "learning_rate": 0.0006034791506779228,
      "loss": 1.8708,
      "step": 310000
    },
    {
      "epoch": 19.858019953952418,
      "grad_norm": 0.296875,
      "learning_rate": 0.0006028396009209516,
      "loss": 1.8721,
      "step": 310500
    },
    {
      "epoch": 19.889997441800972,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0006022000511639806,
      "loss": 1.8714,
      "step": 311000
    },
    {
      "epoch": 19.921974929649526,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0006015605014070095,
      "loss": 1.8714,
      "step": 311500
    },
    {
      "epoch": 19.95395241749808,
      "grad_norm": 0.330078125,
      "learning_rate": 0.0006009209516500383,
      "loss": 1.8717,
      "step": 312000
    },
    {
      "epoch": 19.985929905346637,
      "grad_norm": 0.2109375,
      "learning_rate": 0.0006002814018930674,
      "loss": 1.8707,
      "step": 312500
    },
    {
      "epoch": 20.0,
      "eval_loss": 3.745512008666992,
      "eval_runtime": 7.7064,
      "eval_samples_per_second": 64.881,
      "eval_steps_per_second": 1.038,
      "step": 312720
    },
    {
      "epoch": 20.01790739319519,
      "grad_norm": 0.3125,
      "learning_rate": 0.0005996418521360962,
      "loss": 1.8713,
      "step": 313000
    },
    {
      "epoch": 20.049884881043745,
      "grad_norm": 0.34765625,
      "learning_rate": 0.0005990023023791252,
      "loss": 1.8702,
      "step": 313500
    },
    {
      "epoch": 20.0818623688923,
      "grad_norm": 0.205078125,
      "learning_rate": 0.000598362752622154,
      "loss": 1.8713,
      "step": 314000
    },
    {
      "epoch": 20.113839856740853,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0005977232028651829,
      "loss": 1.8705,
      "step": 314500
    },
    {
      "epoch": 20.14581734458941,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0005970836531082118,
      "loss": 1.8717,
      "step": 315000
    },
    {
      "epoch": 20.177794832437964,
      "grad_norm": 0.1953125,
      "learning_rate": 0.0005964441033512408,
      "loss": 1.8715,
      "step": 315500
    },
    {
      "epoch": 20.20977232028652,
      "grad_norm": 0.2412109375,
      "learning_rate": 0.0005958045535942696,
      "loss": 1.8707,
      "step": 316000
    },
    {
      "epoch": 20.241749808135072,
      "grad_norm": 0.189453125,
      "learning_rate": 0.0005951650038372986,
      "loss": 1.8707,
      "step": 316500
    },
    {
      "epoch": 20.273727295983626,
      "grad_norm": 0.240234375,
      "learning_rate": 0.0005945254540803274,
      "loss": 1.8715,
      "step": 317000
    },
    {
      "epoch": 20.305704783832184,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0005938859043233563,
      "loss": 1.8719,
      "step": 317500
    },
    {
      "epoch": 20.337682271680738,
      "grad_norm": 0.439453125,
      "learning_rate": 0.0005932463545663853,
      "loss": 1.871,
      "step": 318000
    },
    {
      "epoch": 20.36965975952929,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0005926068048094142,
      "loss": 1.8714,
      "step": 318500
    },
    {
      "epoch": 20.401637247377845,
      "grad_norm": 0.640625,
      "learning_rate": 0.0005919672550524431,
      "loss": 1.8701,
      "step": 319000
    },
    {
      "epoch": 20.4336147352264,
      "grad_norm": 0.298828125,
      "learning_rate": 0.000591327705295472,
      "loss": 1.8707,
      "step": 319500
    },
    {
      "epoch": 20.465592223074957,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0005906881555385008,
      "loss": 1.8712,
      "step": 320000
    },
    {
      "epoch": 20.49756971092351,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0005900486057815299,
      "loss": 1.8697,
      "step": 320500
    },
    {
      "epoch": 20.529547198772065,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0005894090560245587,
      "loss": 1.8708,
      "step": 321000
    },
    {
      "epoch": 20.56152468662062,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0005887695062675876,
      "loss": 1.8712,
      "step": 321500
    },
    {
      "epoch": 20.593502174469172,
      "grad_norm": 0.2001953125,
      "learning_rate": 0.0005881299565106165,
      "loss": 1.8704,
      "step": 322000
    },
    {
      "epoch": 20.62547966231773,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0005874904067536454,
      "loss": 1.8703,
      "step": 322500
    },
    {
      "epoch": 20.657457150166284,
      "grad_norm": 0.19140625,
      "learning_rate": 0.0005868508569966745,
      "loss": 1.8709,
      "step": 323000
    },
    {
      "epoch": 20.689434638014838,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0005862113072397033,
      "loss": 1.8704,
      "step": 323500
    },
    {
      "epoch": 20.72141212586339,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0005855717574827321,
      "loss": 1.8711,
      "step": 324000
    },
    {
      "epoch": 20.753389613711946,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0005849322077257611,
      "loss": 1.8687,
      "step": 324500
    },
    {
      "epoch": 20.785367101560503,
      "grad_norm": 0.212890625,
      "learning_rate": 0.00058429265796879,
      "loss": 1.8699,
      "step": 325000
    },
    {
      "epoch": 20.817344589409057,
      "grad_norm": 0.21875,
      "learning_rate": 0.0005836531082118189,
      "loss": 1.8717,
      "step": 325500
    },
    {
      "epoch": 20.84932207725761,
      "grad_norm": 0.1884765625,
      "learning_rate": 0.0005830135584548479,
      "loss": 1.8709,
      "step": 326000
    },
    {
      "epoch": 20.881299565106165,
      "grad_norm": 1.40625,
      "learning_rate": 0.0005823740086978767,
      "loss": 1.8707,
      "step": 326500
    },
    {
      "epoch": 20.91327705295472,
      "grad_norm": 0.396484375,
      "learning_rate": 0.0005817344589409056,
      "loss": 1.8715,
      "step": 327000
    },
    {
      "epoch": 20.945254540803276,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0005810949091839345,
      "loss": 1.8705,
      "step": 327500
    },
    {
      "epoch": 20.97723202865183,
      "grad_norm": 0.2177734375,
      "learning_rate": 0.0005804553594269635,
      "loss": 1.8717,
      "step": 328000
    },
    {
      "epoch": 21.0,
      "eval_loss": 3.743771553039551,
      "eval_runtime": 6.2166,
      "eval_samples_per_second": 80.43,
      "eval_steps_per_second": 1.287,
      "step": 328356
    },
    {
      "epoch": 21.009209516500384,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.0005798158096699924,
      "loss": 1.8696,
      "step": 328500
    },
    {
      "epoch": 21.041187004348938,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0005791762599130213,
      "loss": 1.8703,
      "step": 329000
    },
    {
      "epoch": 21.073164492197492,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0005785367101560501,
      "loss": 1.8707,
      "step": 329500
    },
    {
      "epoch": 21.10514198004605,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0005778971603990791,
      "loss": 1.8707,
      "step": 330000
    },
    {
      "epoch": 21.137119467894603,
      "grad_norm": 0.2080078125,
      "learning_rate": 0.000577257610642108,
      "loss": 1.8701,
      "step": 330500
    },
    {
      "epoch": 21.169096955743157,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.0005766180608851369,
      "loss": 1.8701,
      "step": 331000
    },
    {
      "epoch": 21.20107444359171,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0005759785111281658,
      "loss": 1.8707,
      "step": 331500
    },
    {
      "epoch": 21.233051931440265,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0005753389613711947,
      "loss": 1.8692,
      "step": 332000
    },
    {
      "epoch": 21.26502941928882,
      "grad_norm": 0.216796875,
      "learning_rate": 0.0005746994116142235,
      "loss": 1.8703,
      "step": 332500
    },
    {
      "epoch": 21.297006907137376,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0005740598618572525,
      "loss": 1.8707,
      "step": 333000
    },
    {
      "epoch": 21.32898439498593,
      "grad_norm": 0.4375,
      "learning_rate": 0.0005734203121002814,
      "loss": 1.8699,
      "step": 333500
    },
    {
      "epoch": 21.360961882834484,
      "grad_norm": 0.26171875,
      "learning_rate": 0.0005727807623433104,
      "loss": 1.8706,
      "step": 334000
    },
    {
      "epoch": 21.392939370683038,
      "grad_norm": 0.171875,
      "learning_rate": 0.0005721412125863392,
      "loss": 1.8716,
      "step": 334500
    },
    {
      "epoch": 21.424916858531592,
      "grad_norm": 0.2421875,
      "learning_rate": 0.0005715016628293681,
      "loss": 1.8708,
      "step": 335000
    },
    {
      "epoch": 21.45689434638015,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.000570862113072397,
      "loss": 1.8703,
      "step": 335500
    },
    {
      "epoch": 21.488871834228704,
      "grad_norm": 0.2080078125,
      "learning_rate": 0.000570222563315426,
      "loss": 1.8701,
      "step": 336000
    },
    {
      "epoch": 21.520849322077257,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0005695830135584548,
      "loss": 1.8706,
      "step": 336500
    },
    {
      "epoch": 21.55282680992581,
      "grad_norm": 0.205078125,
      "learning_rate": 0.0005689434638014838,
      "loss": 1.8697,
      "step": 337000
    },
    {
      "epoch": 21.584804297774365,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0005683039140445126,
      "loss": 1.8709,
      "step": 337500
    },
    {
      "epoch": 21.616781785622923,
      "grad_norm": 0.19140625,
      "learning_rate": 0.0005676643642875415,
      "loss": 1.8712,
      "step": 338000
    },
    {
      "epoch": 21.648759273471477,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0005670248145305706,
      "loss": 1.8701,
      "step": 338500
    },
    {
      "epoch": 21.68073676132003,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0005663852647735994,
      "loss": 1.8693,
      "step": 339000
    },
    {
      "epoch": 21.712714249168585,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0005657457150166284,
      "loss": 1.87,
      "step": 339500
    },
    {
      "epoch": 21.74469173701714,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0005651061652596572,
      "loss": 1.8693,
      "step": 340000
    },
    {
      "epoch": 21.776669224865696,
      "grad_norm": 0.2001953125,
      "learning_rate": 0.000564466615502686,
      "loss": 1.8712,
      "step": 340500
    },
    {
      "epoch": 21.80864671271425,
      "grad_norm": 0.375,
      "learning_rate": 0.0005638270657457151,
      "loss": 1.8715,
      "step": 341000
    },
    {
      "epoch": 21.840624200562804,
      "grad_norm": 0.224609375,
      "learning_rate": 0.000563187515988744,
      "loss": 1.8704,
      "step": 341500
    },
    {
      "epoch": 21.872601688411358,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0005625479662317728,
      "loss": 1.8707,
      "step": 342000
    },
    {
      "epoch": 21.90457917625991,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0005619084164748018,
      "loss": 1.8698,
      "step": 342500
    },
    {
      "epoch": 21.93655666410847,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0005612688667178306,
      "loss": 1.8718,
      "step": 343000
    },
    {
      "epoch": 21.968534151957023,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.0005606293169608596,
      "loss": 1.8696,
      "step": 343500
    },
    {
      "epoch": 22.0,
      "eval_loss": 3.744166851043701,
      "eval_runtime": 7.4298,
      "eval_samples_per_second": 67.297,
      "eval_steps_per_second": 1.077,
      "step": 343992
    },
    {
      "epoch": 22.000511639805577,
      "grad_norm": 0.201171875,
      "learning_rate": 0.0005599897672038885,
      "loss": 1.8706,
      "step": 344000
    },
    {
      "epoch": 22.03248912765413,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0005593502174469174,
      "loss": 1.8698,
      "step": 344500
    },
    {
      "epoch": 22.064466615502685,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0005587106676899463,
      "loss": 1.8701,
      "step": 345000
    },
    {
      "epoch": 22.096444103351242,
      "grad_norm": 0.16796875,
      "learning_rate": 0.0005580711179329752,
      "loss": 1.871,
      "step": 345500
    },
    {
      "epoch": 22.128421591199796,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0005574315681760041,
      "loss": 1.8702,
      "step": 346000
    },
    {
      "epoch": 22.16039907904835,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0005567920184190331,
      "loss": 1.8708,
      "step": 346500
    },
    {
      "epoch": 22.192376566896904,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0005561524686620619,
      "loss": 1.8707,
      "step": 347000
    },
    {
      "epoch": 22.224354054745458,
      "grad_norm": 0.333984375,
      "learning_rate": 0.0005555129189050908,
      "loss": 1.871,
      "step": 347500
    },
    {
      "epoch": 22.256331542594015,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0005548733691481197,
      "loss": 1.8703,
      "step": 348000
    },
    {
      "epoch": 22.28830903044257,
      "grad_norm": 0.171875,
      "learning_rate": 0.0005542338193911487,
      "loss": 1.8709,
      "step": 348500
    },
    {
      "epoch": 22.320286518291123,
      "grad_norm": 0.2734375,
      "learning_rate": 0.0005535942696341775,
      "loss": 1.8707,
      "step": 349000
    },
    {
      "epoch": 22.352264006139677,
      "grad_norm": 0.25,
      "learning_rate": 0.0005529547198772065,
      "loss": 1.8714,
      "step": 349500
    },
    {
      "epoch": 22.38424149398823,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0005523151701202353,
      "loss": 1.871,
      "step": 350000
    },
    {
      "epoch": 22.41621898183679,
      "grad_norm": 0.2060546875,
      "learning_rate": 0.0005516756203632643,
      "loss": 1.8698,
      "step": 350500
    },
    {
      "epoch": 22.448196469685342,
      "grad_norm": 0.255859375,
      "learning_rate": 0.0005510360706062931,
      "loss": 1.8718,
      "step": 351000
    },
    {
      "epoch": 22.480173957533896,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0005503965208493221,
      "loss": 1.8701,
      "step": 351500
    },
    {
      "epoch": 22.51215144538245,
      "grad_norm": 0.220703125,
      "learning_rate": 0.000549756971092351,
      "loss": 1.871,
      "step": 352000
    },
    {
      "epoch": 22.544128933231004,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0005491174213353799,
      "loss": 1.8692,
      "step": 352500
    },
    {
      "epoch": 22.57610642107956,
      "grad_norm": 0.1796875,
      "learning_rate": 0.0005484778715784087,
      "loss": 1.8705,
      "step": 353000
    },
    {
      "epoch": 22.608083908928116,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0005478383218214377,
      "loss": 1.8705,
      "step": 353500
    },
    {
      "epoch": 22.64006139677667,
      "grad_norm": 0.1826171875,
      "learning_rate": 0.0005471987720644667,
      "loss": 1.87,
      "step": 354000
    },
    {
      "epoch": 22.672038884625223,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.0005465592223074955,
      "loss": 1.8703,
      "step": 354500
    },
    {
      "epoch": 22.704016372473777,
      "grad_norm": 0.390625,
      "learning_rate": 0.0005459196725505245,
      "loss": 1.8708,
      "step": 355000
    },
    {
      "epoch": 22.73599386032233,
      "grad_norm": 0.330078125,
      "learning_rate": 0.0005452801227935533,
      "loss": 1.8713,
      "step": 355500
    },
    {
      "epoch": 22.76797134817089,
      "grad_norm": 0.3203125,
      "learning_rate": 0.0005446405730365823,
      "loss": 1.8701,
      "step": 356000
    },
    {
      "epoch": 22.799948836019443,
      "grad_norm": 0.2490234375,
      "learning_rate": 0.0005440010232796112,
      "loss": 1.8703,
      "step": 356500
    },
    {
      "epoch": 22.831926323867997,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0005433614735226401,
      "loss": 1.8694,
      "step": 357000
    },
    {
      "epoch": 22.86390381171655,
      "grad_norm": 0.291015625,
      "learning_rate": 0.000542721923765669,
      "loss": 1.8704,
      "step": 357500
    },
    {
      "epoch": 22.895881299565104,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0005420823740086979,
      "loss": 1.8703,
      "step": 358000
    },
    {
      "epoch": 22.927858787413662,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.0005414428242517267,
      "loss": 1.8698,
      "step": 358500
    },
    {
      "epoch": 22.959836275262216,
      "grad_norm": 0.4296875,
      "learning_rate": 0.0005408032744947558,
      "loss": 1.8701,
      "step": 359000
    },
    {
      "epoch": 22.99181376311077,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0005401637247377846,
      "loss": 1.8702,
      "step": 359500
    },
    {
      "epoch": 23.0,
      "eval_loss": 3.743412971496582,
      "eval_runtime": 5.7338,
      "eval_samples_per_second": 87.202,
      "eval_steps_per_second": 1.395,
      "step": 359628
    },
    {
      "epoch": 23.023791250959324,
      "grad_norm": 0.287109375,
      "learning_rate": 0.0005395241749808135,
      "loss": 1.8714,
      "step": 360000
    },
    {
      "epoch": 23.055768738807878,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0005388846252238424,
      "loss": 1.8714,
      "step": 360500
    },
    {
      "epoch": 23.087746226656435,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0005382450754668713,
      "loss": 1.8709,
      "step": 361000
    },
    {
      "epoch": 23.11972371450499,
      "grad_norm": 0.412109375,
      "learning_rate": 0.0005376055257099003,
      "loss": 1.8698,
      "step": 361500
    },
    {
      "epoch": 23.151701202353543,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0005369659759529292,
      "loss": 1.8705,
      "step": 362000
    },
    {
      "epoch": 23.183678690202097,
      "grad_norm": 0.486328125,
      "learning_rate": 0.000536326426195958,
      "loss": 1.8708,
      "step": 362500
    },
    {
      "epoch": 23.21565617805065,
      "grad_norm": 0.291015625,
      "learning_rate": 0.000535686876438987,
      "loss": 1.8698,
      "step": 363000
    },
    {
      "epoch": 23.247633665899208,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.0005350473266820158,
      "loss": 1.87,
      "step": 363500
    },
    {
      "epoch": 23.279611153747762,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0005344077769250448,
      "loss": 1.8704,
      "step": 364000
    },
    {
      "epoch": 23.311588641596316,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0005337682271680737,
      "loss": 1.8697,
      "step": 364500
    },
    {
      "epoch": 23.34356612944487,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0005331286774111026,
      "loss": 1.8702,
      "step": 365000
    },
    {
      "epoch": 23.375543617293424,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0005324891276541314,
      "loss": 1.8702,
      "step": 365500
    },
    {
      "epoch": 23.40752110514198,
      "grad_norm": 0.197265625,
      "learning_rate": 0.0005318495778971604,
      "loss": 1.8707,
      "step": 366000
    },
    {
      "epoch": 23.439498592990535,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0005312100281401893,
      "loss": 1.8691,
      "step": 366500
    },
    {
      "epoch": 23.47147608083909,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0005305704783832183,
      "loss": 1.8707,
      "step": 367000
    },
    {
      "epoch": 23.503453568687643,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0005299309286262472,
      "loss": 1.8714,
      "step": 367500
    },
    {
      "epoch": 23.535431056536197,
      "grad_norm": 0.283203125,
      "learning_rate": 0.000529291378869276,
      "loss": 1.8707,
      "step": 368000
    },
    {
      "epoch": 23.567408544384755,
      "grad_norm": 0.17578125,
      "learning_rate": 0.000528651829112305,
      "loss": 1.8714,
      "step": 368500
    },
    {
      "epoch": 23.59938603223331,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0005280122793553338,
      "loss": 1.8717,
      "step": 369000
    },
    {
      "epoch": 23.631363520081862,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0005273727295983628,
      "loss": 1.8707,
      "step": 369500
    },
    {
      "epoch": 23.663341007930416,
      "grad_norm": 0.189453125,
      "learning_rate": 0.0005267331798413917,
      "loss": 1.8691,
      "step": 370000
    },
    {
      "epoch": 23.69531849577897,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0005260936300844206,
      "loss": 1.8697,
      "step": 370500
    },
    {
      "epoch": 23.727295983627528,
      "grad_norm": 0.1826171875,
      "learning_rate": 0.0005254540803274494,
      "loss": 1.869,
      "step": 371000
    },
    {
      "epoch": 23.75927347147608,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0005248145305704784,
      "loss": 1.869,
      "step": 371500
    },
    {
      "epoch": 23.791250959324636,
      "grad_norm": 0.232421875,
      "learning_rate": 0.0005241749808135073,
      "loss": 1.8698,
      "step": 372000
    },
    {
      "epoch": 23.82322844717319,
      "grad_norm": 0.2099609375,
      "learning_rate": 0.0005235354310565363,
      "loss": 1.8698,
      "step": 372500
    },
    {
      "epoch": 23.855205935021743,
      "grad_norm": 0.216796875,
      "learning_rate": 0.0005228958812995651,
      "loss": 1.8709,
      "step": 373000
    },
    {
      "epoch": 23.8871834228703,
      "grad_norm": 0.212890625,
      "learning_rate": 0.000522256331542594,
      "loss": 1.8706,
      "step": 373500
    },
    {
      "epoch": 23.919160910718855,
      "grad_norm": 0.26953125,
      "learning_rate": 0.0005216167817856229,
      "loss": 1.8702,
      "step": 374000
    },
    {
      "epoch": 23.95113839856741,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0005209772320286519,
      "loss": 1.8689,
      "step": 374500
    },
    {
      "epoch": 23.983115886415963,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0005203376822716807,
      "loss": 1.8702,
      "step": 375000
    },
    {
      "epoch": 24.0,
      "eval_loss": 3.7443900108337402,
      "eval_runtime": 8.2971,
      "eval_samples_per_second": 60.262,
      "eval_steps_per_second": 0.964,
      "step": 375264
    },
    {
      "epoch": 24.015093374264517,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0005196981325147097,
      "loss": 1.8703,
      "step": 375500
    },
    {
      "epoch": 24.047070862113074,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.0005190585827577385,
      "loss": 1.8703,
      "step": 376000
    },
    {
      "epoch": 24.079048349961628,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0005184190330007674,
      "loss": 1.8704,
      "step": 376500
    },
    {
      "epoch": 24.111025837810182,
      "grad_norm": 0.1875,
      "learning_rate": 0.0005177794832437964,
      "loss": 1.8706,
      "step": 377000
    },
    {
      "epoch": 24.143003325658736,
      "grad_norm": 0.18359375,
      "learning_rate": 0.0005171399334868253,
      "loss": 1.8702,
      "step": 377500
    },
    {
      "epoch": 24.17498081350729,
      "grad_norm": 0.2119140625,
      "learning_rate": 0.0005165003837298542,
      "loss": 1.8701,
      "step": 378000
    },
    {
      "epoch": 24.206958301355847,
      "grad_norm": 0.7265625,
      "learning_rate": 0.0005158608339728831,
      "loss": 1.8692,
      "step": 378500
    },
    {
      "epoch": 24.2389357892044,
      "grad_norm": 0.2197265625,
      "learning_rate": 0.0005152212842159119,
      "loss": 1.8712,
      "step": 379000
    },
    {
      "epoch": 24.270913277052955,
      "grad_norm": 0.197265625,
      "learning_rate": 0.000514581734458941,
      "loss": 1.8711,
      "step": 379500
    },
    {
      "epoch": 24.30289076490151,
      "grad_norm": 0.43359375,
      "learning_rate": 0.0005139421847019698,
      "loss": 1.8714,
      "step": 380000
    },
    {
      "epoch": 24.334868252750063,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.0005133026349449987,
      "loss": 1.8695,
      "step": 380500
    },
    {
      "epoch": 24.36684574059862,
      "grad_norm": 0.1875,
      "learning_rate": 0.0005126630851880276,
      "loss": 1.8699,
      "step": 381000
    },
    {
      "epoch": 24.398823228447174,
      "grad_norm": 0.1875,
      "learning_rate": 0.0005120235354310565,
      "loss": 1.8697,
      "step": 381500
    },
    {
      "epoch": 24.430800716295728,
      "grad_norm": 0.16796875,
      "learning_rate": 0.0005113839856740856,
      "loss": 1.8709,
      "step": 382000
    },
    {
      "epoch": 24.462778204144282,
      "grad_norm": 0.26171875,
      "learning_rate": 0.0005107444359171144,
      "loss": 1.8705,
      "step": 382500
    },
    {
      "epoch": 24.494755691992836,
      "grad_norm": 0.2138671875,
      "learning_rate": 0.0005101048861601433,
      "loss": 1.8701,
      "step": 383000
    },
    {
      "epoch": 24.526733179841393,
      "grad_norm": 0.1826171875,
      "learning_rate": 0.0005094653364031722,
      "loss": 1.8693,
      "step": 383500
    },
    {
      "epoch": 24.558710667689947,
      "grad_norm": 0.173828125,
      "learning_rate": 0.000508825786646201,
      "loss": 1.8695,
      "step": 384000
    },
    {
      "epoch": 24.5906881555385,
      "grad_norm": 0.2041015625,
      "learning_rate": 0.00050818623688923,
      "loss": 1.8711,
      "step": 384500
    },
    {
      "epoch": 24.622665643387055,
      "grad_norm": 0.287109375,
      "learning_rate": 0.000507546687132259,
      "loss": 1.8694,
      "step": 385000
    },
    {
      "epoch": 24.65464313123561,
      "grad_norm": 0.2373046875,
      "learning_rate": 0.0005069071373752878,
      "loss": 1.8703,
      "step": 385500
    },
    {
      "epoch": 24.686620619084163,
      "grad_norm": 3.65625,
      "learning_rate": 0.0005062675876183167,
      "loss": 1.8688,
      "step": 386000
    },
    {
      "epoch": 24.71859810693272,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.0005056280378613456,
      "loss": 1.8695,
      "step": 386500
    },
    {
      "epoch": 24.750575594781274,
      "grad_norm": 0.216796875,
      "learning_rate": 0.0005049884881043745,
      "loss": 1.8696,
      "step": 387000
    },
    {
      "epoch": 24.78255308262983,
      "grad_norm": 0.21875,
      "learning_rate": 0.0005043489383474035,
      "loss": 1.8702,
      "step": 387500
    },
    {
      "epoch": 24.814530570478382,
      "grad_norm": 0.1865234375,
      "learning_rate": 0.0005037093885904324,
      "loss": 1.8698,
      "step": 388000
    },
    {
      "epoch": 24.846508058326936,
      "grad_norm": 0.1904296875,
      "learning_rate": 0.0005030698388334612,
      "loss": 1.8698,
      "step": 388500
    },
    {
      "epoch": 24.878485546175494,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0005024302890764902,
      "loss": 1.8706,
      "step": 389000
    },
    {
      "epoch": 24.910463034024048,
      "grad_norm": 0.2001953125,
      "learning_rate": 0.000501790739319519,
      "loss": 1.8703,
      "step": 389500
    },
    {
      "epoch": 24.9424405218726,
      "grad_norm": 0.259765625,
      "learning_rate": 0.000501151189562548,
      "loss": 1.8702,
      "step": 390000
    },
    {
      "epoch": 24.974418009721155,
      "grad_norm": 0.3984375,
      "learning_rate": 0.0005005116398055769,
      "loss": 1.8707,
      "step": 390500
    },
    {
      "epoch": 25.0,
      "eval_loss": 3.7427737712860107,
      "eval_runtime": 6.6939,
      "eval_samples_per_second": 74.695,
      "eval_steps_per_second": 1.195,
      "step": 390900
    },
    {
      "epoch": 25.00639549756971,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.0004998720900486058,
      "loss": 1.8704,
      "step": 391000
    },
    {
      "epoch": 25.038372985418267,
      "grad_norm": 0.2109375,
      "learning_rate": 0.0004992325402916346,
      "loss": 1.8698,
      "step": 391500
    },
    {
      "epoch": 25.07035047326682,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0004985929905346636,
      "loss": 1.8708,
      "step": 392000
    },
    {
      "epoch": 25.102327961115375,
      "grad_norm": 0.2138671875,
      "learning_rate": 0.0004979534407776925,
      "loss": 1.8701,
      "step": 392500
    },
    {
      "epoch": 25.13430544896393,
      "grad_norm": 0.201171875,
      "learning_rate": 0.0004973138910207215,
      "loss": 1.8697,
      "step": 393000
    },
    {
      "epoch": 25.166282936812483,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0004966743412637503,
      "loss": 1.87,
      "step": 393500
    },
    {
      "epoch": 25.19826042466104,
      "grad_norm": 0.255859375,
      "learning_rate": 0.0004960347915067792,
      "loss": 1.8701,
      "step": 394000
    },
    {
      "epoch": 25.230237912509594,
      "grad_norm": 0.1953125,
      "learning_rate": 0.0004953952417498081,
      "loss": 1.8699,
      "step": 394500
    },
    {
      "epoch": 25.262215400358148,
      "grad_norm": 0.26171875,
      "learning_rate": 0.000494755691992837,
      "loss": 1.8695,
      "step": 395000
    },
    {
      "epoch": 25.2941928882067,
      "grad_norm": 0.671875,
      "learning_rate": 0.0004941161422358659,
      "loss": 1.869,
      "step": 395500
    },
    {
      "epoch": 25.326170376055256,
      "grad_norm": 0.359375,
      "learning_rate": 0.0004934765924788949,
      "loss": 1.8712,
      "step": 396000
    },
    {
      "epoch": 25.358147863903813,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0004928370427219237,
      "loss": 1.8696,
      "step": 396500
    },
    {
      "epoch": 25.390125351752367,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0004921974929649527,
      "loss": 1.8695,
      "step": 397000
    },
    {
      "epoch": 25.42210283960092,
      "grad_norm": 0.15234375,
      "learning_rate": 0.0004915579432079816,
      "loss": 1.8709,
      "step": 397500
    },
    {
      "epoch": 25.454080327449475,
      "grad_norm": 0.1591796875,
      "learning_rate": 0.0004909183934510105,
      "loss": 1.8705,
      "step": 398000
    },
    {
      "epoch": 25.48605781529803,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0004902788436940395,
      "loss": 1.8699,
      "step": 398500
    },
    {
      "epoch": 25.518035303146586,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0004896392939370683,
      "loss": 1.8689,
      "step": 399000
    },
    {
      "epoch": 25.55001279099514,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0004889997441800973,
      "loss": 1.8699,
      "step": 399500
    },
    {
      "epoch": 25.581990278843694,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.0004883601944231261,
      "loss": 1.8703,
      "step": 400000
    },
    {
      "epoch": 25.613967766692248,
      "grad_norm": 0.28515625,
      "learning_rate": 0.000487720644666155,
      "loss": 1.8704,
      "step": 400500
    },
    {
      "epoch": 25.645945254540802,
      "grad_norm": 0.189453125,
      "learning_rate": 0.00048708109490918397,
      "loss": 1.869,
      "step": 401000
    },
    {
      "epoch": 25.67792274238936,
      "grad_norm": 0.3125,
      "learning_rate": 0.0004864415451522128,
      "loss": 1.8699,
      "step": 401500
    },
    {
      "epoch": 25.709900230237913,
      "grad_norm": 0.21484375,
      "learning_rate": 0.00048580199539524177,
      "loss": 1.8696,
      "step": 402000
    },
    {
      "epoch": 25.741877718086467,
      "grad_norm": 0.302734375,
      "learning_rate": 0.00048516244563827067,
      "loss": 1.8697,
      "step": 402500
    },
    {
      "epoch": 25.77385520593502,
      "grad_norm": 0.1748046875,
      "learning_rate": 0.00048452289588129957,
      "loss": 1.8692,
      "step": 403000
    },
    {
      "epoch": 25.805832693783575,
      "grad_norm": 0.244140625,
      "learning_rate": 0.0004838833461243285,
      "loss": 1.8699,
      "step": 403500
    },
    {
      "epoch": 25.837810181632133,
      "grad_norm": 0.1806640625,
      "learning_rate": 0.0004832437963673574,
      "loss": 1.8706,
      "step": 404000
    },
    {
      "epoch": 25.869787669480687,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.00048260424661038633,
      "loss": 1.87,
      "step": 404500
    },
    {
      "epoch": 25.90176515732924,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.00048196469685341523,
      "loss": 1.8707,
      "step": 405000
    },
    {
      "epoch": 25.933742645177794,
      "grad_norm": 0.2138671875,
      "learning_rate": 0.0004813251470964441,
      "loss": 1.8704,
      "step": 405500
    },
    {
      "epoch": 25.96572013302635,
      "grad_norm": 0.3125,
      "learning_rate": 0.00048068559733947303,
      "loss": 1.8698,
      "step": 406000
    },
    {
      "epoch": 25.997697620874906,
      "grad_norm": 0.375,
      "learning_rate": 0.00048004604758250194,
      "loss": 1.8698,
      "step": 406500
    },
    {
      "epoch": 26.0,
      "eval_loss": 3.7425780296325684,
      "eval_runtime": 5.0921,
      "eval_samples_per_second": 98.191,
      "eval_steps_per_second": 1.571,
      "step": 406536
    }
  ],
  "logging_steps": 500,
  "max_steps": 781800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.064315931639603e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}