{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1395,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.021528525296017224,
      "grad_norm": 6.515303134918213,
      "learning_rate": 2.9996919063517692e-05,
      "loss": 2.2396,
      "num_input_tokens_seen": 33088,
      "step": 10,
      "train_runtime": 18.2558,
      "train_tokens_per_second": 1812.464
    },
    {
      "epoch": 0.04305705059203445,
      "grad_norm": 6.753899574279785,
      "learning_rate": 2.9986270537572827e-05,
      "loss": 0.2351,
      "num_input_tokens_seen": 65248,
      "step": 20,
      "train_runtime": 28.2312,
      "train_tokens_per_second": 2311.202
    },
    {
      "epoch": 0.06458557588805167,
      "grad_norm": 2.492377758026123,
      "learning_rate": 2.9968021784983436e-05,
      "loss": 0.2052,
      "num_input_tokens_seen": 98240,
      "step": 30,
      "train_runtime": 38.3616,
      "train_tokens_per_second": 2560.892
    },
    {
      "epoch": 0.0861141011840689,
      "grad_norm": 5.00617790222168,
      "learning_rate": 2.994218206053077e-05,
      "loss": 0.1693,
      "num_input_tokens_seen": 132352,
      "step": 40,
      "train_runtime": 49.3169,
      "train_tokens_per_second": 2683.707
    },
    {
      "epoch": 0.10764262648008611,
      "grad_norm": 4.012996673583984,
      "learning_rate": 2.9908764468726846e-05,
      "loss": 0.222,
      "num_input_tokens_seen": 164672,
      "step": 50,
      "train_runtime": 59.2768,
      "train_tokens_per_second": 2778.015
    },
    {
      "epoch": 0.12917115177610333,
      "grad_norm": 4.810372829437256,
      "learning_rate": 2.986778595716853e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 197792,
      "step": 60,
      "train_runtime": 69.642,
      "train_tokens_per_second": 2840.127
    },
    {
      "epoch": 0.15069967707212056,
      "grad_norm": 3.4019181728363037,
      "learning_rate": 2.981926730794264e-05,
      "loss": 0.2081,
      "num_input_tokens_seen": 230400,
      "step": 70,
      "train_runtime": 79.7644,
      "train_tokens_per_second": 2888.508
    },
    {
      "epoch": 0.1722282023681378,
      "grad_norm": 3.6833176612854004,
      "learning_rate": 2.9763233127086394e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 263200,
      "step": 80,
      "train_runtime": 89.8251,
      "train_tokens_per_second": 2930.138
    },
    {
      "epoch": 0.193756727664155,
      "grad_norm": 4.234188079833984,
      "learning_rate": 2.9699711832108555e-05,
      "loss": 0.1342,
      "num_input_tokens_seen": 295200,
      "step": 90,
      "train_runtime": 99.8419,
      "train_tokens_per_second": 2956.676
    },
    {
      "epoch": 0.21528525296017223,
      "grad_norm": 5.0137763023376465,
      "learning_rate": 2.9628735637577612e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 329536,
      "step": 100,
      "train_runtime": 110.4244,
      "train_tokens_per_second": 2984.268
    },
    {
      "epoch": 0.23681377825618946,
      "grad_norm": 4.475228309631348,
      "learning_rate": 2.955034053878428e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 362656,
      "step": 110,
      "train_runtime": 120.5983,
      "train_tokens_per_second": 3007.141
    },
    {
      "epoch": 0.25834230355220666,
      "grad_norm": 5.462347507476807,
      "learning_rate": 2.9464566293486653e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 395584,
      "step": 120,
      "train_runtime": 130.9243,
      "train_tokens_per_second": 3021.472
    },
    {
      "epoch": 0.2798708288482239,
      "grad_norm": 2.540166139602661,
      "learning_rate": 2.9371456401747193e-05,
      "loss": 0.1289,
      "num_input_tokens_seen": 428544,
      "step": 130,
      "train_runtime": 141.2073,
      "train_tokens_per_second": 3034.857
    },
    {
      "epoch": 0.3013993541442411,
      "grad_norm": 3.8444888591766357,
      "learning_rate": 2.9271058083871864e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 461632,
      "step": 140,
      "train_runtime": 151.3919,
      "train_tokens_per_second": 3049.252
    },
    {
      "epoch": 0.32292787944025836,
      "grad_norm": 6.5897603034973145,
      "learning_rate": 2.916342225646254e-05,
      "loss": 0.119,
      "num_input_tokens_seen": 496064,
      "step": 150,
      "train_runtime": 161.8879,
      "train_tokens_per_second": 3064.244
    },
    {
      "epoch": 0.3444564047362756,
      "grad_norm": 2.109292507171631,
      "learning_rate": 2.904860350659487e-05,
      "loss": 0.1208,
      "num_input_tokens_seen": 529472,
      "step": 160,
      "train_runtime": 172.271,
      "train_tokens_per_second": 3073.482
    },
    {
      "epoch": 0.36598493003229277,
      "grad_norm": 4.811038494110107,
      "learning_rate": 2.8926660064134637e-05,
      "loss": 0.1742,
      "num_input_tokens_seen": 562048,
      "step": 170,
      "train_runtime": 182.4602,
      "train_tokens_per_second": 3080.387
    },
    {
      "epoch": 0.38751345532831,
      "grad_norm": 2.0955300331115723,
      "learning_rate": 2.8797653772206726e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 594624,
      "step": 180,
      "train_runtime": 192.6154,
      "train_tokens_per_second": 3087.105
    },
    {
      "epoch": 0.40904198062432723,
      "grad_norm": 7.870272159576416,
      "learning_rate": 2.8661650055831645e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 627744,
      "step": 190,
      "train_runtime": 202.956,
      "train_tokens_per_second": 3093.005
    },
    {
      "epoch": 0.43057050592034446,
      "grad_norm": 4.0759663581848145,
      "learning_rate": 2.8518717888745486e-05,
      "loss": 0.1208,
      "num_input_tokens_seen": 661408,
      "step": 200,
      "train_runtime": 213.3508,
      "train_tokens_per_second": 3100.096
    },
    {
      "epoch": 0.4520990312163617,
      "grad_norm": 2.4314427375793457,
      "learning_rate": 2.836892975842015e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 694304,
      "step": 210,
      "train_runtime": 223.6031,
      "train_tokens_per_second": 3105.073
    },
    {
      "epoch": 0.4736275565123789,
      "grad_norm": 4.522472858428955,
      "learning_rate": 2.8212361629301655e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 727456,
      "step": 220,
      "train_runtime": 233.9109,
      "train_tokens_per_second": 3109.97
    },
    {
      "epoch": 0.4951560818083961,
      "grad_norm": 3.0568742752075195,
      "learning_rate": 2.8049092904285033e-05,
      "loss": 0.1138,
      "num_input_tokens_seen": 759584,
      "step": 230,
      "train_runtime": 243.9911,
      "train_tokens_per_second": 3113.163
    },
    {
      "epoch": 0.5166846071044133,
      "grad_norm": 3.61202335357666,
      "learning_rate": 2.7879206384445526e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 792992,
      "step": 240,
      "train_runtime": 254.4041,
      "train_tokens_per_second": 3117.056
    },
    {
      "epoch": 0.5382131324004306,
      "grad_norm": 0.9548876881599426,
      "learning_rate": 2.7702788227046286e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 825440,
      "step": 250,
      "train_runtime": 264.5793,
      "train_tokens_per_second": 3119.821
    },
    {
      "epoch": 0.5597416576964478,
      "grad_norm": 2.212982654571533,
      "learning_rate": 2.751992790184413e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 860320,
      "step": 260,
      "train_runtime": 275.3133,
      "train_tokens_per_second": 3124.876
    },
    {
      "epoch": 0.581270182992465,
      "grad_norm": 2.7812600135803223,
      "learning_rate": 2.733071814571527e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 893504,
      "step": 270,
      "train_runtime": 285.663,
      "train_tokens_per_second": 3127.825
    },
    {
      "epoch": 0.6027987082884823,
      "grad_norm": 2.8942534923553467,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 0.1447,
      "num_input_tokens_seen": 927104,
      "step": 280,
      "train_runtime": 296.1241,
      "train_tokens_per_second": 3130.795
    },
    {
      "epoch": 0.6243272335844995,
      "grad_norm": 3.4795053005218506,
      "learning_rate": 2.693363733995954e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 961088,
      "step": 290,
      "train_runtime": 306.6611,
      "train_tokens_per_second": 3134.04
    },
    {
      "epoch": 0.6458557588805167,
      "grad_norm": 3.9791197776794434,
      "learning_rate": 2.672596766826138e-05,
      "loss": 0.1085,
      "num_input_tokens_seen": 993536,
      "step": 300,
      "train_runtime": 316.7871,
      "train_tokens_per_second": 3136.289
    },
    {
      "epoch": 0.667384284176534,
      "grad_norm": 2.2527575492858887,
      "learning_rate": 2.6512351219365927e-05,
      "loss": 0.1018,
      "num_input_tokens_seen": 1026048,
      "step": 310,
      "train_runtime": 326.9673,
      "train_tokens_per_second": 3138.076
    },
    {
      "epoch": 0.6889128094725512,
      "grad_norm": 3.495659828186035,
      "learning_rate": 2.6292896327993467e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 1059456,
      "step": 320,
      "train_runtime": 337.4196,
      "train_tokens_per_second": 3139.877
    },
    {
      "epoch": 0.7104413347685683,
      "grad_norm": 2.321333169937134,
      "learning_rate": 2.606771428980681e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 1092832,
      "step": 330,
      "train_runtime": 347.7566,
      "train_tokens_per_second": 3142.52
    },
    {
      "epoch": 0.7319698600645855,
      "grad_norm": 1.52814781665802,
      "learning_rate": 2.5836919304968184e-05,
      "loss": 0.111,
      "num_input_tokens_seen": 1125984,
      "step": 340,
      "train_runtime": 358.0891,
      "train_tokens_per_second": 3144.424
    },
    {
      "epoch": 0.7534983853606028,
      "grad_norm": 1.921821117401123,
      "learning_rate": 2.5600628420223057e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 1161376,
      "step": 350,
      "train_runtime": 368.9727,
      "train_tokens_per_second": 3147.593
    },
    {
      "epoch": 0.77502691065662,
      "grad_norm": 1.9971706867218018,
      "learning_rate": 2.535896146954037e-05,
      "loss": 0.1096,
      "num_input_tokens_seen": 1194784,
      "step": 360,
      "train_runtime": 379.3554,
      "train_tokens_per_second": 3149.511
    },
    {
      "epoch": 0.7965554359526372,
      "grad_norm": 2.59256649017334,
      "learning_rate": 2.5112041013339212e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 1229888,
      "step": 370,
      "train_runtime": 390.2257,
      "train_tokens_per_second": 3151.735
    },
    {
      "epoch": 0.8180839612486545,
      "grad_norm": 1.8118033409118652,
      "learning_rate": 2.48599922763328e-05,
      "loss": 0.0873,
      "num_input_tokens_seen": 1262592,
      "step": 380,
      "train_runtime": 400.4862,
      "train_tokens_per_second": 3152.648
    },
    {
      "epoch": 0.8396124865446717,
      "grad_norm": 1.6229225397109985,
      "learning_rate": 2.4602943084021286e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 1294560,
      "step": 390,
      "train_runtime": 410.4193,
      "train_tokens_per_second": 3154.237
    },
    {
      "epoch": 0.8611410118406889,
      "grad_norm": 1.7495685815811157,
      "learning_rate": 2.434102379786553e-05,
      "loss": 0.086,
      "num_input_tokens_seen": 1326720,
      "step": 400,
      "train_runtime": 420.4429,
      "train_tokens_per_second": 3155.53
    },
    {
      "epoch": 0.8826695371367062,
      "grad_norm": 4.961688041687012,
      "learning_rate": 2.4074367249174864e-05,
      "loss": 0.1014,
      "num_input_tokens_seen": 1360032,
      "step": 410,
      "train_runtime": 430.7016,
      "train_tokens_per_second": 3157.713
    },
    {
      "epoch": 0.9041980624327234,
      "grad_norm": 2.859628915786743,
      "learning_rate": 2.380310867174219e-05,
      "loss": 0.1049,
      "num_input_tokens_seen": 1393280,
      "step": 420,
      "train_runtime": 440.9192,
      "train_tokens_per_second": 3159.944
    },
    {
      "epoch": 0.9257265877287406,
      "grad_norm": 0.8578975200653076,
      "learning_rate": 2.3527385633260724e-05,
      "loss": 0.1001,
      "num_input_tokens_seen": 1426176,
      "step": 430,
      "train_runtime": 451.2521,
      "train_tokens_per_second": 3160.486
    },
    {
      "epoch": 0.9472551130247578,
      "grad_norm": 3.2294187545776367,
      "learning_rate": 2.3247337965557097e-05,
      "loss": 0.0954,
      "num_input_tokens_seen": 1456800,
      "step": 440,
      "train_runtime": 461.036,
      "train_tokens_per_second": 3159.84
    },
    {
      "epoch": 0.9687836383207751,
      "grad_norm": 0.6617746353149414,
      "learning_rate": 2.2963107693676192e-05,
      "loss": 0.0957,
      "num_input_tokens_seen": 1489280,
      "step": 450,
      "train_runtime": 471.2544,
      "train_tokens_per_second": 3160.246
    },
    {
      "epoch": 0.9903121636167922,
      "grad_norm": 2.0861568450927734,
      "learning_rate": 2.2674838963853738e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 1521376,
      "step": 460,
      "train_runtime": 481.2188,
      "train_tokens_per_second": 3161.505
    },
    {
      "epoch": 1.0107642626480087,
      "grad_norm": 2.1552188396453857,
      "learning_rate": 2.2382677970413097e-05,
      "loss": 0.075,
      "num_input_tokens_seen": 1553344,
      "step": 470,
      "train_runtime": 491.1326,
      "train_tokens_per_second": 3162.779
    },
    {
      "epoch": 1.0322927879440258,
      "grad_norm": 1.7718193531036377,
      "learning_rate": 2.2086772881623455e-05,
      "loss": 0.0848,
      "num_input_tokens_seen": 1585824,
      "step": 480,
      "train_runtime": 501.3506,
      "train_tokens_per_second": 3163.104
    },
    {
      "epoch": 1.0538213132400431,
      "grad_norm": 2.7055087089538574,
      "learning_rate": 2.1787273764556827e-05,
      "loss": 0.0596,
      "num_input_tokens_seen": 1619616,
      "step": 490,
      "train_runtime": 511.8465,
      "train_tokens_per_second": 3164.261
    },
    {
      "epoch": 1.0753498385360603,
      "grad_norm": 3.2319579124450684,
      "learning_rate": 2.1484332508982167e-05,
      "loss": 0.0997,
      "num_input_tokens_seen": 1654080,
      "step": 500,
      "train_runtime": 522.5,
      "train_tokens_per_second": 3165.703
    },
    {
      "epoch": 1.0968783638320776,
      "grad_norm": 1.0526902675628662,
      "learning_rate": 2.11781027503351e-05,
      "loss": 0.0558,
      "num_input_tokens_seen": 1688192,
      "step": 510,
      "train_runtime": 533.003,
      "train_tokens_per_second": 3167.322
    },
    {
      "epoch": 1.1184068891280947,
      "grad_norm": 3.2765936851501465,
      "learning_rate": 2.0868739791802312e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 1721280,
      "step": 520,
      "train_runtime": 543.3656,
      "train_tokens_per_second": 3167.812
    },
    {
      "epoch": 1.1399354144241118,
      "grad_norm": 4.38949728012085,
      "learning_rate": 2.0556400525560168e-05,
      "loss": 0.0544,
      "num_input_tokens_seen": 1753536,
      "step": 530,
      "train_runtime": 553.5793,
      "train_tokens_per_second": 3167.633
    },
    {
      "epoch": 1.1614639397201292,
      "grad_norm": 3.4056718349456787,
      "learning_rate": 2.0241243353207484e-05,
      "loss": 0.1175,
      "num_input_tokens_seen": 1786464,
      "step": 540,
      "train_runtime": 563.7086,
      "train_tokens_per_second": 3169.127
    },
    {
      "epoch": 1.1829924650161463,
      "grad_norm": 2.933256149291992,
      "learning_rate": 1.9923428105432796e-05,
      "loss": 0.0816,
      "num_input_tokens_seen": 1820864,
      "step": 550,
      "train_runtime": 574.3634,
      "train_tokens_per_second": 3170.23
    },
    {
      "epoch": 1.2045209903121636,
      "grad_norm": 3.2740089893341064,
      "learning_rate": 1.9603115960956852e-05,
      "loss": 0.0827,
      "num_input_tokens_seen": 1853792,
      "step": 560,
      "train_runtime": 584.615,
      "train_tokens_per_second": 3170.962
    },
    {
      "epoch": 1.2260495156081808,
      "grad_norm": 1.8570072650909424,
      "learning_rate": 1.9280469364791505e-05,
      "loss": 0.0839,
      "num_input_tokens_seen": 1885664,
      "step": 570,
      "train_runtime": 594.6595,
      "train_tokens_per_second": 3170.998
    },
    {
      "epoch": 1.247578040904198,
      "grad_norm": 2.4209187030792236,
      "learning_rate": 1.8955651945856354e-05,
      "loss": 0.0918,
      "num_input_tokens_seen": 1919232,
      "step": 580,
      "train_runtime": 605.0097,
      "train_tokens_per_second": 3172.233
    },
    {
      "epoch": 1.2691065662002152,
      "grad_norm": 2.2689578533172607,
      "learning_rate": 1.8628828433995013e-05,
      "loss": 0.0808,
      "num_input_tokens_seen": 1951488,
      "step": 590,
      "train_runtime": 615.2203,
      "train_tokens_per_second": 3172.015
    },
    {
      "epoch": 1.2906350914962326,
      "grad_norm": 3.088000535964966,
      "learning_rate": 1.8300164576433035e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 1983520,
      "step": 600,
      "train_runtime": 625.172,
      "train_tokens_per_second": 3172.759
    },
    {
      "epoch": 1.3121636167922497,
      "grad_norm": 3.993255615234375,
      "learning_rate": 1.7969827053719862e-05,
      "loss": 0.1019,
      "num_input_tokens_seen": 2017408,
      "step": 610,
      "train_runtime": 635.5091,
      "train_tokens_per_second": 3174.475
    },
    {
      "epoch": 1.333692142088267,
      "grad_norm": 2.3626177310943604,
      "learning_rate": 1.7637983395197465e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 2048768,
      "step": 620,
      "train_runtime": 645.3543,
      "train_tokens_per_second": 3174.64
    },
    {
      "epoch": 1.3552206673842842,
      "grad_norm": 3.4464800357818604,
      "learning_rate": 1.7304801894038522e-05,
      "loss": 0.0782,
      "num_input_tokens_seen": 2082336,
      "step": 630,
      "train_runtime": 655.8583,
      "train_tokens_per_second": 3174.979
    },
    {
      "epoch": 1.3767491926803013,
      "grad_norm": 2.0285677909851074,
      "learning_rate": 1.6970451521897198e-05,
      "loss": 0.0671,
      "num_input_tokens_seen": 2115680,
      "step": 640,
      "train_runtime": 666.2674,
      "train_tokens_per_second": 3175.422
    },
    {
      "epoch": 1.3982777179763186,
      "grad_norm": 2.756606340408325,
      "learning_rate": 1.663510184321587e-05,
      "loss": 0.0956,
      "num_input_tokens_seen": 2147744,
      "step": 650,
      "train_runtime": 676.3665,
      "train_tokens_per_second": 3175.414
    },
    {
      "epoch": 1.419806243272336,
      "grad_norm": 2.5752294063568115,
      "learning_rate": 1.6298922929231194e-05,
      "loss": 0.0848,
      "num_input_tokens_seen": 2180032,
      "step": 660,
      "train_runtime": 686.2314,
      "train_tokens_per_second": 3176.817
    },
    {
      "epoch": 1.441334768568353,
      "grad_norm": 4.53647518157959,
      "learning_rate": 1.5962085271723158e-05,
      "loss": 0.0677,
      "num_input_tokens_seen": 2213280,
      "step": 670,
      "train_runtime": 696.5495,
      "train_tokens_per_second": 3177.491
    },
    {
      "epoch": 1.4628632938643702,
      "grad_norm": 3.155588150024414,
      "learning_rate": 1.5624759696550874e-05,
      "loss": 0.096,
      "num_input_tokens_seen": 2247904,
      "step": 680,
      "train_runtime": 707.1757,
      "train_tokens_per_second": 3178.707
    },
    {
      "epoch": 1.4843918191603875,
      "grad_norm": 2.9506263732910156,
      "learning_rate": 1.528711727701893e-05,
      "loss": 0.0658,
      "num_input_tokens_seen": 2280672,
      "step": 690,
      "train_runtime": 717.8603,
      "train_tokens_per_second": 3177.041
    },
    {
      "epoch": 1.5059203444564049,
      "grad_norm": 6.473484992980957,
      "learning_rate": 1.4949329247118258e-05,
      "loss": 0.0628,
      "num_input_tokens_seen": 2312896,
      "step": 700,
      "train_runtime": 728.0643,
      "train_tokens_per_second": 3176.774
    },
    {
      "epoch": 1.527448869752422,
      "grad_norm": 7.623786449432373,
      "learning_rate": 1.4611566914685489e-05,
      "loss": 0.0644,
      "num_input_tokens_seen": 2346080,
      "step": 710,
      "train_runtime": 738.4137,
      "train_tokens_per_second": 3177.189
    },
    {
      "epoch": 1.5489773950484391,
      "grad_norm": 7.29047966003418,
      "learning_rate": 1.4274001574524894e-05,
      "loss": 0.089,
      "num_input_tokens_seen": 2379584,
      "step": 720,
      "train_runtime": 748.9795,
      "train_tokens_per_second": 3177.102
    },
    {
      "epoch": 1.5705059203444565,
      "grad_norm": 4.531313896179199,
      "learning_rate": 1.3936804421536906e-05,
      "loss": 0.0497,
      "num_input_tokens_seen": 2410368,
      "step": 730,
      "train_runtime": 758.6861,
      "train_tokens_per_second": 3177.029
    },
    {
      "epoch": 1.5920344456404736,
      "grad_norm": 0.5232702493667603,
      "learning_rate": 1.3600146463897327e-05,
      "loss": 0.0773,
      "num_input_tokens_seen": 2442816,
      "step": 740,
      "train_runtime": 768.8836,
      "train_tokens_per_second": 3177.095
    },
    {
      "epoch": 1.6135629709364907,
      "grad_norm": 1.2642621994018555,
      "learning_rate": 1.3264198436331193e-05,
      "loss": 0.1227,
      "num_input_tokens_seen": 2475904,
      "step": 750,
      "train_runtime": 779.131,
      "train_tokens_per_second": 3177.776
    },
    {
      "epoch": 1.635091496232508,
      "grad_norm": 0.5201441645622253,
      "learning_rate": 1.2929130713525396e-05,
      "loss": 0.0825,
      "num_input_tokens_seen": 2508448,
      "step": 760,
      "train_runtime": 789.1919,
      "train_tokens_per_second": 3178.502
    },
    {
      "epoch": 1.6566200215285254,
      "grad_norm": 3.5326004028320312,
      "learning_rate": 1.2595113223723788e-05,
      "loss": 0.1,
      "num_input_tokens_seen": 2542976,
      "step": 770,
      "train_runtime": 799.8144,
      "train_tokens_per_second": 3179.458
    },
    {
      "epoch": 1.6781485468245425,
      "grad_norm": 4.060391902923584,
      "learning_rate": 1.2262315362548825e-05,
      "loss": 0.0743,
      "num_input_tokens_seen": 2577120,
      "step": 780,
      "train_runtime": 810.3719,
      "train_tokens_per_second": 3180.169
    },
    {
      "epoch": 1.6996770721205596,
      "grad_norm": 2.6704506874084473,
      "learning_rate": 1.1930905907093196e-05,
      "loss": 0.0631,
      "num_input_tokens_seen": 2607296,
      "step": 790,
      "train_runtime": 819.8655,
      "train_tokens_per_second": 3180.151
    },
    {
      "epoch": 1.721205597416577,
      "grad_norm": 4.3308563232421875,
      "learning_rate": 1.1601052930325236e-05,
      "loss": 0.0712,
      "num_input_tokens_seen": 2641152,
      "step": 800,
      "train_runtime": 830.2604,
      "train_tokens_per_second": 3181.113
    },
    {
      "epoch": 1.7427341227125943,
      "grad_norm": 3.4263782501220703,
      "learning_rate": 1.127292371585141e-05,
      "loss": 0.0861,
      "num_input_tokens_seen": 2674912,
      "step": 810,
      "train_runtime": 840.6083,
      "train_tokens_per_second": 3182.115
    },
    {
      "epoch": 1.7642626480086114,
      "grad_norm": 3.3427581787109375,
      "learning_rate": 1.0946684673079074e-05,
      "loss": 0.0977,
      "num_input_tokens_seen": 2708352,
      "step": 820,
      "train_runtime": 850.9805,
      "train_tokens_per_second": 3182.625
    },
    {
      "epoch": 1.7857911733046286,
      "grad_norm": 1.1789891719818115,
      "learning_rate": 1.0622501252822628e-05,
      "loss": 0.0551,
      "num_input_tokens_seen": 2741408,
      "step": 830,
      "train_runtime": 861.244,
      "train_tokens_per_second": 3183.08
    },
    {
      "epoch": 1.807319698600646,
      "grad_norm": 4.690176010131836,
      "learning_rate": 1.0300537863395823e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 2774976,
      "step": 840,
      "train_runtime": 871.6874,
      "train_tokens_per_second": 3183.453
    },
    {
      "epoch": 1.8288482238966632,
      "grad_norm": 1.8957003355026245,
      "learning_rate": 9.98095778723271e-06,
      "loss": 0.0713,
      "num_input_tokens_seen": 2807584,
      "step": 850,
      "train_runtime": 881.9692,
      "train_tokens_per_second": 3183.313
    },
    {
      "epoch": 1.8503767491926801,
      "grad_norm": 2.290959596633911,
      "learning_rate": 9.663923098079656e-06,
      "loss": 0.0659,
      "num_input_tokens_seen": 2841280,
      "step": 860,
      "train_runtime": 892.3117,
      "train_tokens_per_second": 3184.179
    },
    {
      "epoch": 1.8719052744886975,
      "grad_norm": 0.3664212226867676,
      "learning_rate": 9.349594578800259e-06,
      "loss": 0.097,
      "num_input_tokens_seen": 2874752,
      "step": 870,
      "train_runtime": 902.679,
      "train_tokens_per_second": 3184.689
    },
    {
      "epoch": 1.8934337997847148,
      "grad_norm": 1.9637306928634644,
      "learning_rate": 9.038131639834977e-06,
      "loss": 0.1038,
      "num_input_tokens_seen": 2907392,
      "step": 880,
      "train_runtime": 912.95,
      "train_tokens_per_second": 3184.613
    },
    {
      "epoch": 1.914962325080732,
      "grad_norm": 0.4590908885002136,
      "learning_rate": 8.7296922383568e-06,
      "loss": 0.0701,
      "num_input_tokens_seen": 2939712,
      "step": 890,
      "train_runtime": 923.0695,
      "train_tokens_per_second": 3184.714
    },
    {
      "epoch": 1.936490850376749,
      "grad_norm": 2.476637601852417,
      "learning_rate": 8.424432798163838e-06,
      "loss": 0.0915,
      "num_input_tokens_seen": 2971488,
      "step": 900,
      "train_runtime": 933.0579,
      "train_tokens_per_second": 3184.677
    },
    {
      "epoch": 1.9580193756727664,
      "grad_norm": 1.1158137321472168,
      "learning_rate": 8.122508130349711e-06,
      "loss": 0.0644,
      "num_input_tokens_seen": 3003840,
      "step": 910,
      "train_runtime": 943.2889,
      "train_tokens_per_second": 3184.433
    },
    {
      "epoch": 1.9795479009687837,
      "grad_norm": 3.5732078552246094,
      "learning_rate": 7.82407135479164e-06,
      "loss": 0.1249,
      "num_input_tokens_seen": 3037120,
      "step": 920,
      "train_runtime": 953.5701,
      "train_tokens_per_second": 3184.999
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.18006207048892975,
      "learning_rate": 7.529273822496379e-06,
      "loss": 0.0571,
      "num_input_tokens_seen": 3069408,
      "step": 930,
      "train_runtime": 963.4444,
      "train_tokens_per_second": 3185.869
    },
    {
      "epoch": 2.0215285252960173,
      "grad_norm": 4.495100975036621,
      "learning_rate": 7.238265038843143e-06,
      "loss": 0.064,
      "num_input_tokens_seen": 3102272,
      "step": 940,
      "train_runtime": 973.8287,
      "train_tokens_per_second": 3185.644
    },
    {
      "epoch": 2.0430570505920342,
      "grad_norm": 1.6752815246582031,
      "learning_rate": 6.951192587762555e-06,
      "loss": 0.0555,
      "num_input_tokens_seen": 3135008,
      "step": 950,
      "train_runtime": 983.9767,
      "train_tokens_per_second": 3186.059
    },
    {
      "epoch": 2.0645855758880516,
      "grad_norm": 0.5070778727531433,
      "learning_rate": 6.668202056890052e-06,
      "loss": 0.0341,
      "num_input_tokens_seen": 3167616,
      "step": 960,
      "train_runtime": 994.2128,
      "train_tokens_per_second": 3186.054
    },
    {
      "epoch": 2.086114101184069,
      "grad_norm": 4.361125469207764,
      "learning_rate": 6.389436963731707e-06,
      "loss": 0.0535,
      "num_input_tokens_seen": 3200576,
      "step": 970,
      "train_runtime": 1004.3748,
      "train_tokens_per_second": 3186.635
    },
    {
      "epoch": 2.1076426264800863,
      "grad_norm": 2.0541720390319824,
      "learning_rate": 6.115038682879904e-06,
      "loss": 0.0644,
      "num_input_tokens_seen": 3234208,
      "step": 980,
      "train_runtime": 1014.8971,
      "train_tokens_per_second": 3186.735
    },
    {
      "epoch": 2.129171151776103,
      "grad_norm": 2.3540420532226562,
      "learning_rate": 5.845146374315781e-06,
      "loss": 0.0335,
      "num_input_tokens_seen": 3267264,
      "step": 990,
      "train_runtime": 1025.1177,
      "train_tokens_per_second": 3187.209
    },
    {
      "epoch": 2.1506996770721205,
      "grad_norm": 2.877788782119751,
      "learning_rate": 5.579896912834807e-06,
      "loss": 0.0705,
      "num_input_tokens_seen": 3301888,
      "step": 1000,
      "train_runtime": 1035.6657,
      "train_tokens_per_second": 3188.179
    },
    {
      "epoch": 2.172228202368138,
      "grad_norm": 6.80332612991333,
      "learning_rate": 5.319424818631293e-06,
      "loss": 0.0719,
      "num_input_tokens_seen": 3336512,
      "step": 1010,
      "train_runtime": 1047.1738,
      "train_tokens_per_second": 3186.207
    },
    {
      "epoch": 2.193756727664155,
      "grad_norm": 4.557567596435547,
      "learning_rate": 5.063862189076988e-06,
      "loss": 0.039,
      "num_input_tokens_seen": 3370144,
      "step": 1020,
      "train_runtime": 1057.6528,
      "train_tokens_per_second": 3186.437
    },
    {
      "epoch": 2.215285252960172,
      "grad_norm": 1.0814592838287354,
      "learning_rate": 4.813338631728471e-06,
      "loss": 0.054,
      "num_input_tokens_seen": 3403328,
      "step": 1030,
      "train_runtime": 1067.9677,
      "train_tokens_per_second": 3186.733
    },
    {
      "epoch": 2.2368137782561894,
      "grad_norm": 4.169713973999023,
      "learning_rate": 4.567981198597168e-06,
      "loss": 0.0521,
      "num_input_tokens_seen": 3437760,
      "step": 1040,
      "train_runtime": 1078.5913,
      "train_tokens_per_second": 3187.268
    },
    {
      "epoch": 2.2583423035522068,
      "grad_norm": 2.4980621337890625,
      "learning_rate": 4.327914321715459e-06,
      "loss": 0.0563,
      "num_input_tokens_seen": 3470400,
      "step": 1050,
      "train_runtime": 1088.7636,
      "train_tokens_per_second": 3187.469
    },
    {
      "epoch": 2.2798708288482237,
      "grad_norm": 1.6621730327606201,
      "learning_rate": 4.09325975003146e-06,
      "loss": 0.0533,
      "num_input_tokens_seen": 3502560,
      "step": 1060,
      "train_runtime": 1098.7801,
      "train_tokens_per_second": 3187.681
    },
    {
      "epoch": 2.301399354144241,
      "grad_norm": 1.9265707731246948,
      "learning_rate": 3.864136487664538e-06,
      "loss": 0.0528,
      "num_input_tokens_seen": 3534784,
      "step": 1070,
      "train_runtime": 1108.8391,
      "train_tokens_per_second": 3187.824
    },
    {
      "epoch": 2.3229278794402584,
      "grad_norm": 3.3840229511260986,
      "learning_rate": 3.640660733552863e-06,
      "loss": 0.0346,
      "num_input_tokens_seen": 3567072,
      "step": 1080,
      "train_runtime": 1118.9836,
      "train_tokens_per_second": 3187.779
    },
    {
      "epoch": 2.3444564047362757,
      "grad_norm": 2.759793281555176,
      "learning_rate": 3.4229458225235696e-06,
      "loss": 0.0259,
      "num_input_tokens_seen": 3602336,
      "step": 1090,
      "train_runtime": 1129.9023,
      "train_tokens_per_second": 3188.184
    },
    {
      "epoch": 2.3659849300322926,
      "grad_norm": 2.270747661590576,
      "learning_rate": 3.211102167815469e-06,
      "loss": 0.0304,
      "num_input_tokens_seen": 3634496,
      "step": 1100,
      "train_runtime": 1140.0897,
      "train_tokens_per_second": 3187.904
    },
    {
      "epoch": 2.38751345532831,
      "grad_norm": 5.217435359954834,
      "learning_rate": 3.0052372050834025e-06,
      "loss": 0.0598,
      "num_input_tokens_seen": 3669088,
      "step": 1110,
      "train_runtime": 1150.8397,
      "train_tokens_per_second": 3188.183
    },
    {
      "epoch": 2.4090419806243273,
      "grad_norm": 2.080349922180176,
      "learning_rate": 2.805455337912722e-06,
      "loss": 0.0422,
      "num_input_tokens_seen": 3702464,
      "step": 1120,
      "train_runtime": 1161.0909,
      "train_tokens_per_second": 3188.781
    },
    {
      "epoch": 2.4305705059203446,
      "grad_norm": 1.8415871858596802,
      "learning_rate": 2.611857884871379e-06,
      "loss": 0.049,
      "num_input_tokens_seen": 3735424,
      "step": 1130,
      "train_runtime": 1171.1753,
      "train_tokens_per_second": 3189.466
    },
    {
      "epoch": 2.4520990312163615,
      "grad_norm": 0.11488506197929382,
      "learning_rate": 2.424543028126661e-06,
      "loss": 0.0541,
      "num_input_tokens_seen": 3768160,
      "step": 1140,
      "train_runtime": 1181.3802,
      "train_tokens_per_second": 3189.625
    },
    {
      "epoch": 2.473627556512379,
      "grad_norm": 2.6189522743225098,
      "learning_rate": 2.243605763652489e-06,
      "loss": 0.0527,
      "num_input_tokens_seen": 3802368,
      "step": 1150,
      "train_runtime": 1191.8927,
      "train_tokens_per_second": 3190.193
    },
    {
      "epoch": 2.495156081808396,
      "grad_norm": 7.0923943519592285,
      "learning_rate": 2.0691378530525933e-06,
      "loss": 0.0648,
      "num_input_tokens_seen": 3834496,
      "step": 1160,
      "train_runtime": 1201.9519,
      "train_tokens_per_second": 3190.224
    },
    {
      "epoch": 2.5166846071044136,
      "grad_norm": 3.9085617065429688,
      "learning_rate": 1.9012277770239829e-06,
      "loss": 0.0749,
      "num_input_tokens_seen": 3867232,
      "step": 1170,
      "train_runtime": 1212.1832,
      "train_tokens_per_second": 3190.303
    },
    {
      "epoch": 2.5382131324004304,
      "grad_norm": 1.4265445470809937,
      "learning_rate": 1.7399606904843224e-06,
      "loss": 0.0241,
      "num_input_tokens_seen": 3899744,
      "step": 1180,
      "train_runtime": 1222.393,
      "train_tokens_per_second": 3190.254
    },
    {
      "epoch": 2.559741657696448,
      "grad_norm": 4.786421298980713,
      "learning_rate": 1.5854183793859577e-06,
      "loss": 0.0522,
      "num_input_tokens_seen": 3934176,
      "step": 1190,
      "train_runtime": 1232.9604,
      "train_tokens_per_second": 3190.837
    },
    {
      "epoch": 2.581270182992465,
      "grad_norm": 3.693969964981079,
      "learning_rate": 1.437679219238498e-06,
      "loss": 0.0393,
      "num_input_tokens_seen": 3967328,
      "step": 1200,
      "train_runtime": 1243.3574,
      "train_tokens_per_second": 3190.819
    },
    {
      "epoch": 2.602798708288482,
      "grad_norm": 5.84261417388916,
      "learning_rate": 1.2968181353609854e-06,
      "loss": 0.0657,
      "num_input_tokens_seen": 4000768,
      "step": 1210,
      "train_runtime": 1253.7857,
      "train_tokens_per_second": 3190.95
    },
    {
      "epoch": 2.6243272335844994,
      "grad_norm": 3.2120585441589355,
      "learning_rate": 1.1629065648838495e-06,
      "loss": 0.0652,
      "num_input_tokens_seen": 4032096,
      "step": 1220,
      "train_runtime": 1263.5974,
      "train_tokens_per_second": 3190.966
    },
    {
      "epoch": 2.6458557588805167,
      "grad_norm": 1.206513524055481,
      "learning_rate": 1.0360124205198102e-06,
      "loss": 0.0328,
      "num_input_tokens_seen": 4065376,
      "step": 1230,
      "train_runtime": 1273.9879,
      "train_tokens_per_second": 3191.063
    },
    {
      "epoch": 2.667384284176534,
      "grad_norm": 4.7513227462768555,
      "learning_rate": 9.162000561222672e-07,
      "loss": 0.0663,
      "num_input_tokens_seen": 4099904,
      "step": 1240,
      "train_runtime": 1284.6957,
      "train_tokens_per_second": 3191.343
    },
    {
      "epoch": 2.6889128094725514,
      "grad_norm": 3.0120952129364014,
      "learning_rate": 8.035302340484646e-07,
      "loss": 0.052,
      "num_input_tokens_seen": 4131968,
      "step": 1250,
      "train_runtime": 1294.799,
      "train_tokens_per_second": 3191.204
    },
    {
      "epoch": 2.7104413347685683,
      "grad_norm": 4.013816833496094,
      "learning_rate": 6.980600943441273e-07,
      "loss": 0.0275,
      "num_input_tokens_seen": 4164416,
      "step": 1260,
      "train_runtime": 1305.0758,
      "train_tokens_per_second": 3190.938
    },
    {
      "epoch": 2.7319698600645856,
      "grad_norm": 1.4241931438446045,
      "learning_rate": 5.998431257651132e-07,
      "loss": 0.0557,
      "num_input_tokens_seen": 4197824,
      "step": 1270,
      "train_runtime": 1315.3489,
      "train_tokens_per_second": 3191.415
    },
    {
      "epoch": 2.7534983853606025,
      "grad_norm": 5.286342620849609,
      "learning_rate": 5.089291386508055e-07,
      "loss": 0.0491,
      "num_input_tokens_seen": 4230656,
      "step": 1280,
      "train_runtime": 1325.6172,
      "train_tokens_per_second": 3191.461
    },
    {
      "epoch": 2.77502691065662,
      "grad_norm": 4.148273944854736,
      "learning_rate": 4.2536423966299256e-07,
      "loss": 0.0422,
      "num_input_tokens_seen": 4263456,
      "step": 1290,
      "train_runtime": 1335.8602,
      "train_tokens_per_second": 3191.544
    },
    {
      "epoch": 2.7965554359526372,
      "grad_norm": 3.012843132019043,
      "learning_rate": 3.491908084030776e-07,
      "loss": 0.0596,
      "num_input_tokens_seen": 4297056,
      "step": 1300,
      "train_runtime": 1346.2641,
      "train_tokens_per_second": 3191.837
    },
    {
      "epoch": 2.8180839612486546,
      "grad_norm": 2.8270840644836426,
      "learning_rate": 2.804474759194176e-07,
      "loss": 0.0557,
      "num_input_tokens_seen": 4329824,
      "step": 1310,
      "train_runtime": 1356.4598,
      "train_tokens_per_second": 3192.003
    },
    {
      "epoch": 2.839612486544672,
      "grad_norm": 2.706238031387329,
      "learning_rate": 2.1916910511575627e-07,
      "loss": 0.0569,
      "num_input_tokens_seen": 4362176,
      "step": 1320,
      "train_runtime": 1366.5476,
      "train_tokens_per_second": 3192.114
    },
    {
      "epoch": 2.861141011840689,
      "grad_norm": 2.350955009460449,
      "learning_rate": 1.6538677307062933e-07,
      "loss": 0.0508,
      "num_input_tokens_seen": 4395264,
      "step": 1330,
      "train_runtime": 1376.8485,
      "train_tokens_per_second": 3192.264
    },
    {
      "epoch": 2.882669537136706,
      "grad_norm": 7.152644157409668,
      "learning_rate": 1.1912775527676278e-07,
      "loss": 0.0416,
      "num_input_tokens_seen": 4427392,
      "step": 1340,
      "train_runtime": 1386.9034,
      "train_tokens_per_second": 3192.286
    },
    {
      "epoch": 2.9041980624327235,
      "grad_norm": 2.857088565826416,
      "learning_rate": 8.04155118084099e-08,
      "loss": 0.0631,
      "num_input_tokens_seen": 4459488,
      "step": 1350,
      "train_runtime": 1397.0078,
      "train_tokens_per_second": 3192.171
    },
    {
      "epoch": 2.9257265877287404,
      "grad_norm": 0.2651897966861725,
      "learning_rate": 4.9269675423671556e-08,
      "loss": 0.0476,
      "num_input_tokens_seen": 4492640,
      "step": 1360,
      "train_runtime": 1407.3439,
      "train_tokens_per_second": 3192.283
    },
    {
      "epoch": 2.9472551130247577,
      "grad_norm": 2.6711535453796387,
      "learning_rate": 2.570604160781753e-08,
      "loss": 0.0412,
      "num_input_tokens_seen": 4525376,
      "step": 1370,
      "train_runtime": 1417.5421,
      "train_tokens_per_second": 3192.41
    },
    {
      "epoch": 2.968783638320775,
      "grad_norm": 3.344942808151245,
      "learning_rate": 9.736560562675912e-09,
      "loss": 0.0514,
      "num_input_tokens_seen": 4558656,
      "step": 1380,
      "train_runtime": 1427.8337,
      "train_tokens_per_second": 3192.708
    },
    {
      "epoch": 2.9903121636167924,
      "grad_norm": 1.608618974685669,
      "learning_rate": 1.3693311461326908e-09,
      "loss": 0.0776,
      "num_input_tokens_seen": 4591648,
      "step": 1390,
      "train_runtime": 1438.1289,
      "train_tokens_per_second": 3192.793
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 4606272,
      "step": 1395,
      "total_flos": 1.971883506354094e+17,
      "train_loss": 0.1050504026553964,
      "train_runtime": 1443.4932,
      "train_samples_per_second": 7.723,
      "train_steps_per_second": 0.966
    }
  ],
  "logging_steps": 10,
  "max_steps": 1395,
  "num_input_tokens_seen": 4606272,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.971883506354094e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}