{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 330,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04561003420752566,
      "grad_norm": 1.0824084281921387,
      "learning_rate": 4.9981876195011844e-05,
      "loss": 0.4811,
      "num_input_tokens_seen": 110848,
      "step": 5,
      "train_runtime": 30.3319,
      "train_tokens_per_second": 3654.501
    },
    {
      "epoch": 0.09122006841505131,
      "grad_norm": 0.6754106879234314,
      "learning_rate": 4.9908293271567286e-05,
      "loss": 0.2124,
      "num_input_tokens_seen": 220880,
      "step": 10,
      "train_runtime": 60.3932,
      "train_tokens_per_second": 3657.363
    },
    {
      "epoch": 0.13683010262257697,
      "grad_norm": 0.39871442317962646,
      "learning_rate": 4.977828505250903e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 332160,
      "step": 15,
      "train_runtime": 90.9477,
      "train_tokens_per_second": 3652.209
    },
    {
      "epoch": 0.18244013683010263,
      "grad_norm": 0.3741743564605713,
      "learning_rate": 4.959214604826831e-05,
      "loss": 0.1043,
      "num_input_tokens_seen": 442608,
      "step": 20,
      "train_runtime": 121.2095,
      "train_tokens_per_second": 3651.596
    },
    {
      "epoch": 0.22805017103762829,
      "grad_norm": 0.43203699588775635,
      "learning_rate": 4.935029792355834e-05,
      "loss": 0.0953,
      "num_input_tokens_seen": 552768,
      "step": 25,
      "train_runtime": 151.3115,
      "train_tokens_per_second": 3653.179
    },
    {
      "epoch": 0.27366020524515394,
      "grad_norm": 0.40135976672172546,
      "learning_rate": 4.9053288542168185e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 663056,
      "step": 30,
      "train_runtime": 181.4665,
      "train_tokens_per_second": 3653.876
    },
    {
      "epoch": 0.31927023945267957,
      "grad_norm": 0.33862611651420593,
      "learning_rate": 4.870179072587499e-05,
      "loss": 0.0625,
      "num_input_tokens_seen": 773936,
      "step": 35,
      "train_runtime": 211.7267,
      "train_tokens_per_second": 3655.353
    },
    {
      "epoch": 0.36488027366020526,
      "grad_norm": 0.41058799624443054,
      "learning_rate": 4.829660073028631e-05,
      "loss": 0.0727,
      "num_input_tokens_seen": 884832,
      "step": 40,
      "train_runtime": 241.9754,
      "train_tokens_per_second": 3656.702
    },
    {
      "epoch": 0.4104903078677309,
      "grad_norm": 0.5706074833869934,
      "learning_rate": 4.783863644106502e-05,
      "loss": 0.0727,
      "num_input_tokens_seen": 995744,
      "step": 45,
      "train_runtime": 272.301,
      "train_tokens_per_second": 3656.777
    },
    {
      "epoch": 0.45610034207525657,
      "grad_norm": 0.31977686285972595,
      "learning_rate": 4.73289352946231e-05,
      "loss": 0.0676,
      "num_input_tokens_seen": 1106256,
      "step": 50,
      "train_runtime": 302.3755,
      "train_tokens_per_second": 3658.551
    },
    {
      "epoch": 0.5017103762827823,
      "grad_norm": 0.29765012860298157,
      "learning_rate": 4.6768651927994434e-05,
      "loss": 0.0635,
      "num_input_tokens_seen": 1217840,
      "step": 55,
      "train_runtime": 332.9163,
      "train_tokens_per_second": 3658.097
    },
    {
      "epoch": 0.5473204104903079,
      "grad_norm": 0.3982709050178528,
      "learning_rate": 4.6159055563210604e-05,
      "loss": 0.0586,
      "num_input_tokens_seen": 1328544,
      "step": 60,
      "train_runtime": 363.1815,
      "train_tokens_per_second": 3658.072
    },
    {
      "epoch": 0.5929304446978335,
      "grad_norm": 0.24241122603416443,
      "learning_rate": 4.550152713210478e-05,
      "loss": 0.0572,
      "num_input_tokens_seen": 1438848,
      "step": 65,
      "train_runtime": 393.3602,
      "train_tokens_per_second": 3657.838
    },
    {
      "epoch": 0.6385404789053591,
      "grad_norm": 0.2856714427471161,
      "learning_rate": 4.479755614805688e-05,
      "loss": 0.054,
      "num_input_tokens_seen": 1547904,
      "step": 70,
      "train_runtime": 423.1722,
      "train_tokens_per_second": 3657.858
    },
    {
      "epoch": 0.6841505131128849,
      "grad_norm": 0.3355858325958252,
      "learning_rate": 4.404873733176678e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 1658736,
      "step": 75,
      "train_runtime": 453.4813,
      "train_tokens_per_second": 3657.782
    },
    {
      "epoch": 0.7297605473204105,
      "grad_norm": 0.3131825029850006,
      "learning_rate": 4.3256766998698936e-05,
      "loss": 0.0471,
      "num_input_tokens_seen": 1769248,
      "step": 80,
      "train_runtime": 483.7293,
      "train_tokens_per_second": 3657.517
    },
    {
      "epoch": 0.7753705815279361,
      "grad_norm": 0.2504868805408478,
      "learning_rate": 4.242343921638234e-05,
      "loss": 0.0408,
      "num_input_tokens_seen": 1879024,
      "step": 85,
      "train_runtime": 513.7413,
      "train_tokens_per_second": 3657.529
    },
    {
      "epoch": 0.8209806157354618,
      "grad_norm": 0.4044341742992401,
      "learning_rate": 4.155064174027047e-05,
      "loss": 0.0398,
      "num_input_tokens_seen": 1991024,
      "step": 90,
      "train_runtime": 544.3342,
      "train_tokens_per_second": 3657.724
    },
    {
      "epoch": 0.8665906499429875,
      "grad_norm": 0.3132927119731903,
      "learning_rate": 4.064035173736804e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 2102064,
      "step": 95,
      "train_runtime": 574.6371,
      "train_tokens_per_second": 3658.072
    },
    {
      "epoch": 0.9122006841505131,
      "grad_norm": 0.3075104057788849,
      "learning_rate": 3.969463130731183e-05,
      "loss": 0.0408,
      "num_input_tokens_seen": 2212432,
      "step": 100,
      "train_runtime": 604.7406,
      "train_tokens_per_second": 3658.481
    },
    {
      "epoch": 0.9578107183580388,
      "grad_norm": 0.3352231979370117,
      "learning_rate": 3.871562281105175e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 2322032,
      "step": 105,
      "train_runtime": 635.488,
      "train_tokens_per_second": 3653.936
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.36991724371910095,
      "learning_rate": 3.770554401771423e-05,
      "loss": 0.0356,
      "num_input_tokens_seen": 2423872,
      "step": 110,
      "train_runtime": 663.2539,
      "train_tokens_per_second": 3654.516
    },
    {
      "epoch": 1.0456100342075256,
      "grad_norm": 0.31266242265701294,
      "learning_rate": 3.6666683080641846e-05,
      "loss": 0.0358,
      "num_input_tokens_seen": 2534752,
      "step": 115,
      "train_runtime": 693.4736,
      "train_tokens_per_second": 3655.153
    },
    {
      "epoch": 1.0912200684150513,
      "grad_norm": 0.29535725712776184,
      "learning_rate": 3.5601393353990046e-05,
      "loss": 0.0388,
      "num_input_tokens_seen": 2644464,
      "step": 120,
      "train_runtime": 723.4647,
      "train_tokens_per_second": 3655.277
    },
    {
      "epoch": 1.1368301026225769,
      "grad_norm": 0.37989553809165955,
      "learning_rate": 3.4512088061623075e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 2755568,
      "step": 125,
      "train_runtime": 753.8099,
      "train_tokens_per_second": 3655.521
    },
    {
      "epoch": 1.1824401368301025,
      "grad_norm": 0.2450297474861145,
      "learning_rate": 3.3401234830385756e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 2865744,
      "step": 130,
      "train_runtime": 783.9325,
      "train_tokens_per_second": 3655.6
    },
    {
      "epoch": 1.2280501710376284,
      "grad_norm": 0.35642582178115845,
      "learning_rate": 3.2271350100134975e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 2976304,
      "step": 135,
      "train_runtime": 814.2192,
      "train_tokens_per_second": 3655.409
    },
    {
      "epoch": 1.273660205245154,
      "grad_norm": 0.34205862879753113,
      "learning_rate": 3.11249934231941e-05,
      "loss": 0.0285,
      "num_input_tokens_seen": 3086224,
      "step": 140,
      "train_runtime": 844.255,
      "train_tokens_per_second": 3655.559
    },
    {
      "epoch": 1.3192702394526796,
      "grad_norm": 0.4949014484882355,
      "learning_rate": 2.996476166614364e-05,
      "loss": 0.0292,
      "num_input_tokens_seen": 3196352,
      "step": 145,
      "train_runtime": 874.2926,
      "train_tokens_per_second": 3655.929
    },
    {
      "epoch": 1.3648802736602053,
      "grad_norm": 0.3114229142665863,
      "learning_rate": 2.8793283127083292e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 3307744,
      "step": 150,
      "train_runtime": 904.5859,
      "train_tokens_per_second": 3656.639
    },
    {
      "epoch": 1.4104903078677309,
      "grad_norm": 0.3953075706958771,
      "learning_rate": 2.761321158169134e-05,
      "loss": 0.0345,
      "num_input_tokens_seen": 3418016,
      "step": 155,
      "train_runtime": 934.6229,
      "train_tokens_per_second": 3657.107
    },
    {
      "epoch": 1.4561003420752565,
      "grad_norm": 0.5266784429550171,
      "learning_rate": 2.6427220271569203e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 3528544,
      "step": 160,
      "train_runtime": 964.715,
      "train_tokens_per_second": 3657.602
    },
    {
      "epoch": 1.5017103762827824,
      "grad_norm": 0.4755711853504181,
      "learning_rate": 2.523799584848942e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 3639056,
      "step": 165,
      "train_runtime": 994.801,
      "train_tokens_per_second": 3658.074
    },
    {
      "epoch": 1.547320410490308,
      "grad_norm": 0.4198484420776367,
      "learning_rate": 2.4048232288265253e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 3749824,
      "step": 170,
      "train_runtime": 1025.0451,
      "train_tokens_per_second": 3658.204
    },
    {
      "epoch": 1.5929304446978336,
      "grad_norm": 0.40441790223121643,
      "learning_rate": 2.2860624788029013e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 3859776,
      "step": 175,
      "train_runtime": 1055.0684,
      "train_tokens_per_second": 3658.318
    },
    {
      "epoch": 1.6385404789053593,
      "grad_norm": 0.7088226079940796,
      "learning_rate": 2.167786366074365e-05,
      "loss": 0.0296,
      "num_input_tokens_seen": 3970576,
      "step": 180,
      "train_runtime": 1085.3469,
      "train_tokens_per_second": 3658.347
    },
    {
      "epoch": 1.6841505131128849,
      "grad_norm": 0.40556278824806213,
      "learning_rate": 2.0502628240778655e-05,
      "loss": 0.0319,
      "num_input_tokens_seen": 4081568,
      "step": 185,
      "train_runtime": 1115.6872,
      "train_tokens_per_second": 3658.344
    },
    {
      "epoch": 1.7297605473204105,
      "grad_norm": 0.32311075925827026,
      "learning_rate": 1.9337580814355888e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 4191264,
      "step": 190,
      "train_runtime": 1145.6081,
      "train_tokens_per_second": 3658.55
    },
    {
      "epoch": 1.7753705815279361,
      "grad_norm": 0.3284579813480377,
      "learning_rate": 1.8185360588615058e-05,
      "loss": 0.0337,
      "num_input_tokens_seen": 4301440,
      "step": 195,
      "train_runtime": 1175.6929,
      "train_tokens_per_second": 3658.642
    },
    {
      "epoch": 1.8209806157354618,
      "grad_norm": 0.2036353498697281,
      "learning_rate": 1.7048577712960627e-05,
      "loss": 0.0356,
      "num_input_tokens_seen": 4412000,
      "step": 200,
      "train_runtime": 1206.0273,
      "train_tokens_per_second": 3658.292
    },
    {
      "epoch": 1.8665906499429874,
      "grad_norm": 0.2839590907096863,
      "learning_rate": 1.5929807366233977e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 4521984,
      "step": 205,
      "train_runtime": 1237.1367,
      "train_tokens_per_second": 3655.202
    },
    {
      "epoch": 1.912200684150513,
      "grad_norm": 0.3934095799922943,
      "learning_rate": 1.4831583923104999e-05,
      "loss": 0.0372,
      "num_input_tokens_seen": 4632176,
      "step": 210,
      "train_runtime": 1267.3175,
      "train_tokens_per_second": 3655.103
    },
    {
      "epoch": 1.9578107183580387,
      "grad_norm": 0.30704551935195923,
      "learning_rate": 1.3756395212898359e-05,
      "loss": 0.0261,
      "num_input_tokens_seen": 4743552,
      "step": 215,
      "train_runtime": 1297.7646,
      "train_tokens_per_second": 3655.171
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2414654642343521,
      "learning_rate": 1.2706676883859903e-05,
      "loss": 0.0256,
      "num_input_tokens_seen": 4845664,
      "step": 220,
      "train_runtime": 1325.6139,
      "train_tokens_per_second": 3655.411
    },
    {
      "epoch": 2.0456100342075256,
      "grad_norm": 0.271589457988739,
      "learning_rate": 1.1684806885630004e-05,
      "loss": 0.024,
      "num_input_tokens_seen": 4955440,
      "step": 225,
      "train_runtime": 1355.6355,
      "train_tokens_per_second": 3655.437
    },
    {
      "epoch": 2.0912200684150513,
      "grad_norm": 0.3638203740119934,
      "learning_rate": 1.0693100082422763e-05,
      "loss": 0.022,
      "num_input_tokens_seen": 5065744,
      "step": 230,
      "train_runtime": 1385.6557,
      "train_tokens_per_second": 3655.846
    },
    {
      "epoch": 2.136830102622577,
      "grad_norm": 0.22529903054237366,
      "learning_rate": 9.733803009114045e-06,
      "loss": 0.0223,
      "num_input_tokens_seen": 5177424,
      "step": 235,
      "train_runtime": 1416.0752,
      "train_tokens_per_second": 3656.179
    },
    {
      "epoch": 2.1824401368301025,
      "grad_norm": 0.41837063431739807,
      "learning_rate": 8.809088782117452e-06,
      "loss": 0.0251,
      "num_input_tokens_seen": 5287824,
      "step": 240,
      "train_runtime": 1446.1526,
      "train_tokens_per_second": 3656.477
    },
    {
      "epoch": 2.228050171037628,
      "grad_norm": 0.38458940386772156,
      "learning_rate": 7.921052176576644e-06,
      "loss": 0.0223,
      "num_input_tokens_seen": 5398032,
      "step": 245,
      "train_runtime": 1476.1489,
      "train_tokens_per_second": 3656.834
    },
    {
      "epoch": 2.2736602052451538,
      "grad_norm": 0.3993557095527649,
      "learning_rate": 7.071704881025915e-06,
      "loss": 0.0255,
      "num_input_tokens_seen": 5509248,
      "step": 250,
      "train_runtime": 1506.4711,
      "train_tokens_per_second": 3657.055
    },
    {
      "epoch": 2.3192702394526794,
      "grad_norm": 0.311338871717453,
      "learning_rate": 6.2629709402686535e-06,
      "loss": 0.0211,
      "num_input_tokens_seen": 5619920,
      "step": 255,
      "train_runtime": 1536.6024,
      "train_tokens_per_second": 3657.368
    },
    {
      "epoch": 2.364880273660205,
      "grad_norm": 0.3177950382232666,
      "learning_rate": 5.49668239679699e-06,
      "loss": 0.0235,
      "num_input_tokens_seen": 5729536,
      "step": 260,
      "train_runtime": 1566.4697,
      "train_tokens_per_second": 3657.61
    },
    {
      "epoch": 2.4104903078677307,
      "grad_norm": 0.2807183265686035,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 0.0249,
      "num_input_tokens_seen": 5840848,
      "step": 265,
      "train_runtime": 1596.7614,
      "train_tokens_per_second": 3657.934
    },
    {
      "epoch": 2.4561003420752567,
      "grad_norm": 0.3168954849243164,
      "learning_rate": 4.098284976946101e-06,
      "loss": 0.0228,
      "num_input_tokens_seen": 5951840,
      "step": 270,
      "train_runtime": 1626.9716,
      "train_tokens_per_second": 3658.232
    },
    {
      "epoch": 2.5017103762827824,
      "grad_norm": 0.5031352639198303,
      "learning_rate": 3.4693439204949858e-06,
      "loss": 0.0304,
      "num_input_tokens_seen": 6062160,
      "step": 275,
      "train_runtime": 1656.9898,
      "train_tokens_per_second": 3658.538
    },
    {
      "epoch": 2.547320410490308,
      "grad_norm": 0.28105977177619934,
      "learning_rate": 2.889176725054643e-06,
      "loss": 0.023,
      "num_input_tokens_seen": 6172976,
      "step": 280,
      "train_runtime": 1687.1404,
      "train_tokens_per_second": 3658.84
    },
    {
      "epoch": 2.5929304446978336,
      "grad_norm": 0.3535321056842804,
      "learning_rate": 2.3590976559242278e-06,
      "loss": 0.0263,
      "num_input_tokens_seen": 6283600,
      "step": 285,
      "train_runtime": 1717.249,
      "train_tokens_per_second": 3659.108
    },
    {
      "epoch": 2.6385404789053593,
      "grad_norm": 0.3748933970928192,
      "learning_rate": 1.8803075126867715e-06,
      "loss": 0.0259,
      "num_input_tokens_seen": 6394176,
      "step": 290,
      "train_runtime": 1747.3641,
      "train_tokens_per_second": 3659.327
    },
    {
      "epoch": 2.684150513112885,
      "grad_norm": 0.2608453035354614,
      "learning_rate": 1.4538909090118846e-06,
      "loss": 0.0217,
      "num_input_tokens_seen": 6503616,
      "step": 295,
      "train_runtime": 1777.1238,
      "train_tokens_per_second": 3659.63
    },
    {
      "epoch": 2.7297605473204105,
      "grad_norm": 0.2588518559932709,
      "learning_rate": 1.0808138156569614e-06,
      "loss": 0.0259,
      "num_input_tokens_seen": 6614736,
      "step": 300,
      "train_runtime": 1807.3109,
      "train_tokens_per_second": 3659.988
    },
    {
      "epoch": 2.775370581527936,
      "grad_norm": 0.38416263461112976,
      "learning_rate": 7.619213722327185e-07,
      "loss": 0.0259,
      "num_input_tokens_seen": 6725088,
      "step": 305,
      "train_runtime": 1838.354,
      "train_tokens_per_second": 3658.212
    },
    {
      "epoch": 2.8209806157354618,
      "grad_norm": 0.3531897962093353,
      "learning_rate": 4.979359726901639e-07,
      "loss": 0.0279,
      "num_input_tokens_seen": 6834816,
      "step": 310,
      "train_runtime": 1868.2865,
      "train_tokens_per_second": 3658.334
    },
    {
      "epoch": 2.8665906499429874,
      "grad_norm": 0.4269830584526062,
      "learning_rate": 2.894556288659395e-07,
      "loss": 0.0221,
      "num_input_tokens_seen": 6945024,
      "step": 315,
      "train_runtime": 1898.3866,
      "train_tokens_per_second": 3658.382
    },
    {
      "epoch": 2.912200684150513,
      "grad_norm": 0.5062984228134155,
      "learning_rate": 1.3695261579316777e-07,
      "loss": 0.0246,
      "num_input_tokens_seen": 7055120,
      "step": 320,
      "train_runtime": 1928.4784,
      "train_tokens_per_second": 3658.387
    },
    {
      "epoch": 2.9578107183580387,
      "grad_norm": 0.20750342309474945,
      "learning_rate": 4.07724018466088e-08,
      "loss": 0.0189,
      "num_input_tokens_seen": 7166944,
      "step": 325,
      "train_runtime": 1959.0815,
      "train_tokens_per_second": 3658.318
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6835935115814209,
      "learning_rate": 1.132866145678313e-09,
      "loss": 0.0228,
      "num_input_tokens_seen": 7269760,
      "step": 330,
      "train_runtime": 1987.0937,
      "train_tokens_per_second": 3658.489
    }
  ],
  "logging_steps": 5,
  "max_steps": 330,
  "num_input_tokens_seen": 7269760,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.282687546045235e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}