{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.1307953260459858,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003769317753486619,
      "grad_norm": 2.665037155151367,
      "learning_rate": 4.999956146783009e-05,
      "loss": 1.564,
      "num_input_tokens_seen": 23856,
      "step": 5
    },
    {
      "epoch": 0.007538635506973238,
      "grad_norm": 1.4514005184173584,
      "learning_rate": 4.9998245886705174e-05,
      "loss": 0.968,
      "num_input_tokens_seen": 47584,
      "step": 10
    },
    {
      "epoch": 0.011307953260459858,
      "grad_norm": 2.0009350776672363,
      "learning_rate": 4.999605330277923e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 70864,
      "step": 15
    },
    {
      "epoch": 0.015077271013946476,
      "grad_norm": 1.2164318561553955,
      "learning_rate": 4.999298379297376e-05,
      "loss": 1.073,
      "num_input_tokens_seen": 94192,
      "step": 20
    },
    {
      "epoch": 0.018846588767433094,
      "grad_norm": 2.2099952697753906,
      "learning_rate": 4.998903746497505e-05,
      "loss": 0.8549,
      "num_input_tokens_seen": 117472,
      "step": 25
    },
    {
      "epoch": 0.022615906520919715,
      "grad_norm": 1.6259618997573853,
      "learning_rate": 4.998421445723046e-05,
      "loss": 0.8704,
      "num_input_tokens_seen": 140704,
      "step": 30
    },
    {
      "epoch": 0.026385224274406333,
      "grad_norm": 1.7849559783935547,
      "learning_rate": 4.997851493894349e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 163680,
      "step": 35
    },
    {
      "epoch": 0.03015454202789295,
      "grad_norm": 1.8650152683258057,
      "learning_rate": 4.997193911006793e-05,
      "loss": 0.8356,
      "num_input_tokens_seen": 187072,
      "step": 40
    },
    {
      "epoch": 0.03392385978137957,
      "grad_norm": 1.681622862815857,
      "learning_rate": 4.996448720130077e-05,
      "loss": 0.778,
      "num_input_tokens_seen": 211040,
      "step": 45
    },
    {
      "epoch": 0.03769317753486619,
      "grad_norm": 1.6398649215698242,
      "learning_rate": 4.995615947407415e-05,
      "loss": 0.939,
      "num_input_tokens_seen": 234560,
      "step": 50
    },
    {
      "epoch": 0.04146249528835281,
      "grad_norm": 2.327622890472412,
      "learning_rate": 4.994695622054618e-05,
      "loss": 1.0274,
      "num_input_tokens_seen": 257632,
      "step": 55
    },
    {
      "epoch": 0.04523181304183943,
      "grad_norm": 2.6391327381134033,
      "learning_rate": 4.9936877763590664e-05,
      "loss": 0.8822,
      "num_input_tokens_seen": 281024,
      "step": 60
    },
    {
      "epoch": 0.049001130795326045,
      "grad_norm": 1.8776222467422485,
      "learning_rate": 4.992592445678582e-05,
      "loss": 0.8146,
      "num_input_tokens_seen": 304688,
      "step": 65
    },
    {
      "epoch": 0.052770448548812667,
      "grad_norm": 1.6592698097229004,
      "learning_rate": 4.991409668440185e-05,
      "loss": 0.9989,
      "num_input_tokens_seen": 327424,
      "step": 70
    },
    {
      "epoch": 0.05653976630229928,
      "grad_norm": 1.709376335144043,
      "learning_rate": 4.990139486138743e-05,
      "loss": 0.9344,
      "num_input_tokens_seen": 350528,
      "step": 75
    },
    {
      "epoch": 0.0603090840557859,
      "grad_norm": 1.3547413349151611,
      "learning_rate": 4.988781943335521e-05,
      "loss": 0.7932,
      "num_input_tokens_seen": 373280,
      "step": 80
    },
    {
      "epoch": 0.06407840180927252,
      "grad_norm": 1.3162572383880615,
      "learning_rate": 4.987337087656614e-05,
      "loss": 0.9445,
      "num_input_tokens_seen": 395856,
      "step": 85
    },
    {
      "epoch": 0.06784771956275915,
      "grad_norm": 1.8211767673492432,
      "learning_rate": 4.985804969791279e-05,
      "loss": 0.7369,
      "num_input_tokens_seen": 418704,
      "step": 90
    },
    {
      "epoch": 0.07161703731624576,
      "grad_norm": 1.552996039390564,
      "learning_rate": 4.984185643490151e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 442432,
      "step": 95
    },
    {
      "epoch": 0.07538635506973237,
      "grad_norm": 1.816767930984497,
      "learning_rate": 4.9824791655633676e-05,
      "loss": 0.7753,
      "num_input_tokens_seen": 466128,
      "step": 100
    },
    {
      "epoch": 0.079155672823219,
      "grad_norm": 1.6724802255630493,
      "learning_rate": 4.9806855958785625e-05,
      "loss": 0.8278,
      "num_input_tokens_seen": 489536,
      "step": 105
    },
    {
      "epoch": 0.08292499057670562,
      "grad_norm": 2.187622308731079,
      "learning_rate": 4.978804997358779e-05,
      "loss": 0.8432,
      "num_input_tokens_seen": 513200,
      "step": 110
    },
    {
      "epoch": 0.08669430833019223,
      "grad_norm": 1.500815987586975,
      "learning_rate": 4.9768374359802525e-05,
      "loss": 0.9649,
      "num_input_tokens_seen": 536432,
      "step": 115
    },
    {
      "epoch": 0.09046362608367886,
      "grad_norm": 1.6637320518493652,
      "learning_rate": 4.9747829807701e-05,
      "loss": 0.8249,
      "num_input_tokens_seen": 559776,
      "step": 120
    },
    {
      "epoch": 0.09423294383716548,
      "grad_norm": 1.5978686809539795,
      "learning_rate": 4.972641703803896e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 583248,
      "step": 125
    },
    {
      "epoch": 0.09800226159065209,
      "grad_norm": 1.4269095659255981,
      "learning_rate": 4.9704136802031485e-05,
      "loss": 0.885,
      "num_input_tokens_seen": 606768,
      "step": 130
    },
    {
      "epoch": 0.1017715793441387,
      "grad_norm": 1.6915644407272339,
      "learning_rate": 4.96809898813266e-05,
      "loss": 0.8736,
      "num_input_tokens_seen": 630896,
      "step": 135
    },
    {
      "epoch": 0.10554089709762533,
      "grad_norm": 1.6915837526321411,
      "learning_rate": 4.965697708797784e-05,
      "loss": 0.7312,
      "num_input_tokens_seen": 654320,
      "step": 140
    },
    {
      "epoch": 0.10931021485111195,
      "grad_norm": 1.921988606452942,
      "learning_rate": 4.963209926441581e-05,
      "loss": 0.9478,
      "num_input_tokens_seen": 677248,
      "step": 145
    },
    {
      "epoch": 0.11307953260459856,
      "grad_norm": 2.241665840148926,
      "learning_rate": 4.9606357283418575e-05,
      "loss": 0.9174,
      "num_input_tokens_seen": 700672,
      "step": 150
    },
    {
      "epoch": 0.11684885035808519,
      "grad_norm": 2.076327323913574,
      "learning_rate": 4.957975204808108e-05,
      "loss": 0.8453,
      "num_input_tokens_seen": 724480,
      "step": 155
    },
    {
      "epoch": 0.1206181681115718,
      "grad_norm": 1.5655834674835205,
      "learning_rate": 4.955228449178345e-05,
      "loss": 0.701,
      "num_input_tokens_seen": 748144,
      "step": 160
    },
    {
      "epoch": 0.12438748586505842,
      "grad_norm": 1.8062623739242554,
      "learning_rate": 4.952395557815826e-05,
      "loss": 0.7981,
      "num_input_tokens_seen": 771584,
      "step": 165
    },
    {
      "epoch": 0.12815680361854503,
      "grad_norm": 2.2940807342529297,
      "learning_rate": 4.949476630105669e-05,
      "loss": 0.8824,
      "num_input_tokens_seen": 795248,
      "step": 170
    },
    {
      "epoch": 0.13192612137203166,
      "grad_norm": 2.108461856842041,
      "learning_rate": 4.9464717684513726e-05,
      "loss": 0.8368,
      "num_input_tokens_seen": 818272,
      "step": 175
    },
    {
      "epoch": 0.1356954391255183,
      "grad_norm": 1.5639266967773438,
      "learning_rate": 4.943381078271214e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 841440,
      "step": 180
    },
    {
      "epoch": 0.1394647568790049,
      "grad_norm": 1.4479436874389648,
      "learning_rate": 4.9402046679945613e-05,
      "loss": 0.8697,
      "num_input_tokens_seen": 864640,
      "step": 185
    },
    {
      "epoch": 0.14323407463249152,
      "grad_norm": 1.8331745862960815,
      "learning_rate": 4.936942649058061e-05,
      "loss": 0.7765,
      "num_input_tokens_seen": 888032,
      "step": 190
    },
    {
      "epoch": 0.14700339238597815,
      "grad_norm": 1.6916519403457642,
      "learning_rate": 4.933595135901732e-05,
      "loss": 0.778,
      "num_input_tokens_seen": 911008,
      "step": 195
    },
    {
      "epoch": 0.15077271013946475,
      "grad_norm": 2.0202736854553223,
      "learning_rate": 4.930162245964952e-05,
      "loss": 0.8926,
      "num_input_tokens_seen": 934432,
      "step": 200
    },
    {
      "epoch": 0.15454202789295138,
      "grad_norm": 1.8288179636001587,
      "learning_rate": 4.926644099682334e-05,
      "loss": 0.805,
      "num_input_tokens_seen": 958064,
      "step": 205
    },
    {
      "epoch": 0.158311345646438,
      "grad_norm": 1.9957374334335327,
      "learning_rate": 4.9230408204795034e-05,
      "loss": 0.8433,
      "num_input_tokens_seen": 980992,
      "step": 210
    },
    {
      "epoch": 0.1620806633999246,
      "grad_norm": 1.5921601057052612,
      "learning_rate": 4.9193525347687696e-05,
      "loss": 0.8483,
      "num_input_tokens_seen": 1004736,
      "step": 215
    },
    {
      "epoch": 0.16584998115341124,
      "grad_norm": 2.1509461402893066,
      "learning_rate": 4.9155793719446863e-05,
      "loss": 0.9689,
      "num_input_tokens_seen": 1028320,
      "step": 220
    },
    {
      "epoch": 0.16961929890689786,
      "grad_norm": 1.9183369874954224,
      "learning_rate": 4.911721464379516e-05,
      "loss": 0.9025,
      "num_input_tokens_seen": 1051696,
      "step": 225
    },
    {
      "epoch": 0.17338861666038446,
      "grad_norm": 1.5253652334213257,
      "learning_rate": 4.907778947418585e-05,
      "loss": 0.8423,
      "num_input_tokens_seen": 1074640,
      "step": 230
    },
    {
      "epoch": 0.1771579344138711,
      "grad_norm": 1.917833924293518,
      "learning_rate": 4.9037519593755356e-05,
      "loss": 0.968,
      "num_input_tokens_seen": 1098096,
      "step": 235
    },
    {
      "epoch": 0.18092725216735772,
      "grad_norm": 1.8350000381469727,
      "learning_rate": 4.89964064152747e-05,
      "loss": 0.8118,
      "num_input_tokens_seen": 1121344,
      "step": 240
    },
    {
      "epoch": 0.18469656992084432,
      "grad_norm": 1.4250982999801636,
      "learning_rate": 4.895445138110001e-05,
      "loss": 0.8695,
      "num_input_tokens_seen": 1144608,
      "step": 245
    },
    {
      "epoch": 0.18846588767433095,
      "grad_norm": 3.1919894218444824,
      "learning_rate": 4.891165596312186e-05,
      "loss": 0.9135,
      "num_input_tokens_seen": 1168112,
      "step": 250
    },
    {
      "epoch": 0.19223520542781755,
      "grad_norm": 2.4486422538757324,
      "learning_rate": 4.886802166271364e-05,
      "loss": 0.798,
      "num_input_tokens_seen": 1190992,
      "step": 255
    },
    {
      "epoch": 0.19600452318130418,
      "grad_norm": 1.6712846755981445,
      "learning_rate": 4.882355001067892e-05,
      "loss": 0.8068,
      "num_input_tokens_seen": 1214832,
      "step": 260
    },
    {
      "epoch": 0.1997738409347908,
      "grad_norm": 2.27498197555542,
      "learning_rate": 4.8778242567197685e-05,
      "loss": 0.7621,
      "num_input_tokens_seen": 1238080,
      "step": 265
    },
    {
      "epoch": 0.2035431586882774,
      "grad_norm": 1.8954802751541138,
      "learning_rate": 4.873210092177167e-05,
      "loss": 0.7604,
      "num_input_tokens_seen": 1261056,
      "step": 270
    },
    {
      "epoch": 0.20731247644176404,
      "grad_norm": 2.805199146270752,
      "learning_rate": 4.868512669316855e-05,
      "loss": 0.9166,
      "num_input_tokens_seen": 1283968,
      "step": 275
    },
    {
      "epoch": 0.21108179419525067,
      "grad_norm": 1.6355040073394775,
      "learning_rate": 4.863732152936514e-05,
      "loss": 0.8487,
      "num_input_tokens_seen": 1308192,
      "step": 280
    },
    {
      "epoch": 0.21485111194873727,
      "grad_norm": 1.5382211208343506,
      "learning_rate": 4.858868710748963e-05,
      "loss": 0.9388,
      "num_input_tokens_seen": 1331648,
      "step": 285
    },
    {
      "epoch": 0.2186204297022239,
      "grad_norm": 4.8234052658081055,
      "learning_rate": 4.85392251337627e-05,
      "loss": 0.8762,
      "num_input_tokens_seen": 1354880,
      "step": 290
    },
    {
      "epoch": 0.22238974745571052,
      "grad_norm": 1.7781318426132202,
      "learning_rate": 4.848893734343769e-05,
      "loss": 0.8672,
      "num_input_tokens_seen": 1378064,
      "step": 295
    },
    {
      "epoch": 0.22615906520919712,
      "grad_norm": 1.4167485237121582,
      "learning_rate": 4.8437825500739696e-05,
      "loss": 0.9335,
      "num_input_tokens_seen": 1401568,
      "step": 300
    },
    {
      "epoch": 0.22992838296268375,
      "grad_norm": 1.112923502922058,
      "learning_rate": 4.838589139880371e-05,
      "loss": 0.7408,
      "num_input_tokens_seen": 1424960,
      "step": 305
    },
    {
      "epoch": 0.23369770071617038,
      "grad_norm": 1.789054274559021,
      "learning_rate": 4.833313685961167e-05,
      "loss": 0.7255,
      "num_input_tokens_seen": 1448560,
      "step": 310
    },
    {
      "epoch": 0.23746701846965698,
      "grad_norm": 2.056309223175049,
      "learning_rate": 4.82795637339286e-05,
      "loss": 0.8214,
      "num_input_tokens_seen": 1472224,
      "step": 315
    },
    {
      "epoch": 0.2412363362231436,
      "grad_norm": 2.14561128616333,
      "learning_rate": 4.822517390123761e-05,
      "loss": 0.7577,
      "num_input_tokens_seen": 1495232,
      "step": 320
    },
    {
      "epoch": 0.24500565397663024,
      "grad_norm": 2.092505931854248,
      "learning_rate": 4.8169969269674016e-05,
      "loss": 0.7087,
      "num_input_tokens_seen": 1518608,
      "step": 325
    },
    {
      "epoch": 0.24877497173011684,
      "grad_norm": 3.2833504676818848,
      "learning_rate": 4.811395177595836e-05,
      "loss": 0.8798,
      "num_input_tokens_seen": 1541552,
      "step": 330
    },
    {
      "epoch": 0.25254428948360347,
      "grad_norm": 3.2298083305358887,
      "learning_rate": 4.8057123385328495e-05,
      "loss": 0.8511,
      "num_input_tokens_seen": 1565136,
      "step": 335
    },
    {
      "epoch": 0.25631360723709007,
      "grad_norm": 3.187859296798706,
      "learning_rate": 4.799948609147061e-05,
      "loss": 0.8655,
      "num_input_tokens_seen": 1588624,
      "step": 340
    },
    {
      "epoch": 0.2600829249905767,
      "grad_norm": 1.7054897546768188,
      "learning_rate": 4.7941041916449316e-05,
      "loss": 0.7643,
      "num_input_tokens_seen": 1612016,
      "step": 345
    },
    {
      "epoch": 0.2638522427440633,
      "grad_norm": 2.040390729904175,
      "learning_rate": 4.788179291063667e-05,
      "loss": 0.8325,
      "num_input_tokens_seen": 1635856,
      "step": 350
    },
    {
      "epoch": 0.2676215604975499,
      "grad_norm": 1.6121920347213745,
      "learning_rate": 4.78217411526403e-05,
      "loss": 0.6917,
      "num_input_tokens_seen": 1659168,
      "step": 355
    },
    {
      "epoch": 0.2713908782510366,
      "grad_norm": 1.7065237760543823,
      "learning_rate": 4.7760888749230416e-05,
      "loss": 0.9086,
      "num_input_tokens_seen": 1682480,
      "step": 360
    },
    {
      "epoch": 0.2751601960045232,
      "grad_norm": 2.437305212020874,
      "learning_rate": 4.769923783526593e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 1705952,
      "step": 365
    },
    {
      "epoch": 0.2789295137580098,
      "grad_norm": 1.7989013195037842,
      "learning_rate": 4.7636790573619586e-05,
      "loss": 0.8712,
      "num_input_tokens_seen": 1729504,
      "step": 370
    },
    {
      "epoch": 0.28269883151149644,
      "grad_norm": 1.2867515087127686,
      "learning_rate": 4.7573549155102014e-05,
      "loss": 0.7044,
      "num_input_tokens_seen": 1752784,
      "step": 375
    },
    {
      "epoch": 0.28646814926498304,
      "grad_norm": 1.7485004663467407,
      "learning_rate": 4.7509515798384956e-05,
      "loss": 0.92,
      "num_input_tokens_seen": 1775840,
      "step": 380
    },
    {
      "epoch": 0.29023746701846964,
      "grad_norm": 1.8417341709136963,
      "learning_rate": 4.7444692749923345e-05,
      "loss": 0.7396,
      "num_input_tokens_seen": 1798944,
      "step": 385
    },
    {
      "epoch": 0.2940067847719563,
      "grad_norm": 1.755854606628418,
      "learning_rate": 4.7379082283876566e-05,
      "loss": 0.8619,
      "num_input_tokens_seen": 1822016,
      "step": 390
    },
    {
      "epoch": 0.2977761025254429,
      "grad_norm": 1.924985408782959,
      "learning_rate": 4.73126867020286e-05,
      "loss": 0.7399,
      "num_input_tokens_seen": 1845584,
      "step": 395
    },
    {
      "epoch": 0.3015454202789295,
      "grad_norm": 2.078490972518921,
      "learning_rate": 4.724550833370735e-05,
      "loss": 0.7882,
      "num_input_tokens_seen": 1868976,
      "step": 400
    },
    {
      "epoch": 0.30531473803241616,
      "grad_norm": 1.7066706418991089,
      "learning_rate": 4.717754953570286e-05,
      "loss": 0.7579,
      "num_input_tokens_seen": 1892096,
      "step": 405
    },
    {
      "epoch": 0.30908405578590276,
      "grad_norm": 1.5587912797927856,
      "learning_rate": 4.710881269218467e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 1915136,
      "step": 410
    },
    {
      "epoch": 0.31285337353938936,
      "grad_norm": 2.13066029548645,
      "learning_rate": 4.7039300214618134e-05,
      "loss": 0.8279,
      "num_input_tokens_seen": 1938464,
      "step": 415
    },
    {
      "epoch": 0.316622691292876,
      "grad_norm": 1.6272748708724976,
      "learning_rate": 4.696901454167988e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 1961696,
      "step": 420
    },
    {
      "epoch": 0.3203920090463626,
      "grad_norm": 1.9481947422027588,
      "learning_rate": 4.68979581391722e-05,
      "loss": 0.7804,
      "num_input_tokens_seen": 1985072,
      "step": 425
    },
    {
      "epoch": 0.3241613267998492,
      "grad_norm": 2.0169599056243896,
      "learning_rate": 4.682613349993655e-05,
      "loss": 0.8848,
      "num_input_tokens_seen": 2007936,
      "step": 430
    },
    {
      "epoch": 0.32793064455333587,
      "grad_norm": 2.285831928253174,
      "learning_rate": 4.675354314376614e-05,
      "loss": 0.7155,
      "num_input_tokens_seen": 2031280,
      "step": 435
    },
    {
      "epoch": 0.33169996230682247,
      "grad_norm": 1.7333121299743652,
      "learning_rate": 4.6680189617317474e-05,
      "loss": 0.7408,
      "num_input_tokens_seen": 2054176,
      "step": 440
    },
    {
      "epoch": 0.33546928006030907,
      "grad_norm": 1.5312005281448364,
      "learning_rate": 4.660607549402103e-05,
      "loss": 0.6917,
      "num_input_tokens_seen": 2077392,
      "step": 445
    },
    {
      "epoch": 0.33923859781379573,
      "grad_norm": 1.2847707271575928,
      "learning_rate": 4.6531203373991014e-05,
      "loss": 0.8376,
      "num_input_tokens_seen": 2100464,
      "step": 450
    },
    {
      "epoch": 0.34300791556728233,
      "grad_norm": 2.3927323818206787,
      "learning_rate": 4.645557588393407e-05,
      "loss": 0.7672,
      "num_input_tokens_seen": 2124240,
      "step": 455
    },
    {
      "epoch": 0.34677723332076893,
      "grad_norm": 1.467666506767273,
      "learning_rate": 4.63791956770572e-05,
      "loss": 0.7285,
      "num_input_tokens_seen": 2147584,
      "step": 460
    },
    {
      "epoch": 0.3505465510742556,
      "grad_norm": 2.08063006401062,
      "learning_rate": 4.6302065432974616e-05,
      "loss": 0.8577,
      "num_input_tokens_seen": 2170976,
      "step": 465
    },
    {
      "epoch": 0.3543158688277422,
      "grad_norm": 2.0516254901885986,
      "learning_rate": 4.6224187857613786e-05,
      "loss": 0.8195,
      "num_input_tokens_seen": 2194128,
      "step": 470
    },
    {
      "epoch": 0.3580851865812288,
      "grad_norm": 1.4173938035964966,
      "learning_rate": 4.6145565683120496e-05,
      "loss": 0.878,
      "num_input_tokens_seen": 2217056,
      "step": 475
    },
    {
      "epoch": 0.36185450433471544,
      "grad_norm": 2.968780517578125,
      "learning_rate": 4.606620166776294e-05,
      "loss": 0.8187,
      "num_input_tokens_seen": 2240800,
      "step": 480
    },
    {
      "epoch": 0.36562382208820204,
      "grad_norm": 2.426954507827759,
      "learning_rate": 4.598609859583506e-05,
      "loss": 0.7713,
      "num_input_tokens_seen": 2263984,
      "step": 485
    },
    {
      "epoch": 0.36939313984168864,
      "grad_norm": 1.7469669580459595,
      "learning_rate": 4.590525927755874e-05,
      "loss": 0.8583,
      "num_input_tokens_seen": 2287776,
      "step": 490
    },
    {
      "epoch": 0.3731624575951753,
      "grad_norm": 4.131247043609619,
      "learning_rate": 4.582368654898533e-05,
      "loss": 0.8106,
      "num_input_tokens_seen": 2311072,
      "step": 495
    },
    {
      "epoch": 0.3769317753486619,
      "grad_norm": 2.013920545578003,
      "learning_rate": 4.5741383271896094e-05,
      "loss": 0.8578,
      "num_input_tokens_seen": 2334768,
      "step": 500
    },
    {
      "epoch": 0.3807010931021485,
      "grad_norm": 2.0032143592834473,
      "learning_rate": 4.565835233370178e-05,
      "loss": 0.7192,
      "num_input_tokens_seen": 2358272,
      "step": 505
    },
    {
      "epoch": 0.3844704108556351,
      "grad_norm": 1.9632434844970703,
      "learning_rate": 4.557459664734141e-05,
      "loss": 0.8681,
      "num_input_tokens_seen": 2381440,
      "step": 510
    },
    {
      "epoch": 0.38823972860912176,
      "grad_norm": 3.1098926067352295,
      "learning_rate": 4.549011915118001e-05,
      "loss": 0.6713,
      "num_input_tokens_seen": 2403984,
      "step": 515
    },
    {
      "epoch": 0.39200904636260836,
      "grad_norm": 2.1910107135772705,
      "learning_rate": 4.540492280890555e-05,
      "loss": 0.7997,
      "num_input_tokens_seen": 2427392,
      "step": 520
    },
    {
      "epoch": 0.39577836411609496,
      "grad_norm": 2.5994174480438232,
      "learning_rate": 4.531901060942497e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 2450384,
      "step": 525
    },
    {
      "epoch": 0.3995476818695816,
      "grad_norm": 2.206372022628784,
      "learning_rate": 4.523238556675935e-05,
      "loss": 0.6608,
      "num_input_tokens_seen": 2473424,
      "step": 530
    },
    {
      "epoch": 0.4033169996230682,
      "grad_norm": 2.1349329948425293,
      "learning_rate": 4.514505071993812e-05,
      "loss": 0.7568,
      "num_input_tokens_seen": 2497232,
      "step": 535
    },
    {
      "epoch": 0.4070863173765548,
      "grad_norm": 2.0707926750183105,
      "learning_rate": 4.505700913289246e-05,
      "loss": 0.8325,
      "num_input_tokens_seen": 2520432,
      "step": 540
    },
    {
      "epoch": 0.4108556351300415,
      "grad_norm": 1.6440961360931396,
      "learning_rate": 4.496826389434784e-05,
      "loss": 0.8083,
      "num_input_tokens_seen": 2543616,
      "step": 545
    },
    {
      "epoch": 0.4146249528835281,
      "grad_norm": 2.5610768795013428,
      "learning_rate": 4.48788181177156e-05,
      "loss": 0.7702,
      "num_input_tokens_seen": 2566992,
      "step": 550
    },
    {
      "epoch": 0.4183942706370147,
      "grad_norm": 2.882587432861328,
      "learning_rate": 4.478867494098381e-05,
      "loss": 0.6993,
      "num_input_tokens_seen": 2590144,
      "step": 555
    },
    {
      "epoch": 0.42216358839050133,
      "grad_norm": 2.3161754608154297,
      "learning_rate": 4.469783752660709e-05,
      "loss": 0.8109,
      "num_input_tokens_seen": 2614240,
      "step": 560
    },
    {
      "epoch": 0.42593290614398793,
      "grad_norm": 2.578278064727783,
      "learning_rate": 4.460630906139571e-05,
      "loss": 0.7901,
      "num_input_tokens_seen": 2637696,
      "step": 565
    },
    {
      "epoch": 0.42970222389747453,
      "grad_norm": 2.362873077392578,
      "learning_rate": 4.451409275640379e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 2660768,
      "step": 570
    },
    {
      "epoch": 0.4334715416509612,
      "grad_norm": 1.8829095363616943,
      "learning_rate": 4.442119184681664e-05,
      "loss": 0.703,
      "num_input_tokens_seen": 2683792,
      "step": 575
    },
    {
      "epoch": 0.4372408594044478,
      "grad_norm": 1.8470242023468018,
      "learning_rate": 4.432760959183725e-05,
      "loss": 0.9203,
      "num_input_tokens_seen": 2707136,
      "step": 580
    },
    {
      "epoch": 0.4410101771579344,
      "grad_norm": 2.575732707977295,
      "learning_rate": 4.423334927457198e-05,
      "loss": 0.9314,
      "num_input_tokens_seen": 2729808,
      "step": 585
    },
    {
      "epoch": 0.44477949491142105,
      "grad_norm": 1.765871524810791,
      "learning_rate": 4.413841420191532e-05,
      "loss": 0.7105,
      "num_input_tokens_seen": 2752992,
      "step": 590
    },
    {
      "epoch": 0.44854881266490765,
      "grad_norm": 1.8366035223007202,
      "learning_rate": 4.404280770443398e-05,
      "loss": 0.894,
      "num_input_tokens_seen": 2776480,
      "step": 595
    },
    {
      "epoch": 0.45231813041839425,
      "grad_norm": 2.326671838760376,
      "learning_rate": 4.3946533136249926e-05,
      "loss": 0.8868,
      "num_input_tokens_seen": 2799728,
      "step": 600
    },
    {
      "epoch": 0.4560874481718809,
      "grad_norm": 1.8004709482192993,
      "learning_rate": 4.384959387492277e-05,
      "loss": 0.7679,
      "num_input_tokens_seen": 2823008,
      "step": 605
    },
    {
      "epoch": 0.4598567659253675,
      "grad_norm": 1.893351435661316,
      "learning_rate": 4.37519933213313e-05,
      "loss": 0.7895,
      "num_input_tokens_seen": 2845984,
      "step": 610
    },
    {
      "epoch": 0.4636260836788541,
      "grad_norm": 2.1309449672698975,
      "learning_rate": 4.365373489955411e-05,
      "loss": 0.8527,
      "num_input_tokens_seen": 2869024,
      "step": 615
    },
    {
      "epoch": 0.46739540143234076,
      "grad_norm": 2.4292616844177246,
      "learning_rate": 4.355482205674951e-05,
      "loss": 0.7533,
      "num_input_tokens_seen": 2892240,
      "step": 620
    },
    {
      "epoch": 0.47116471918582736,
      "grad_norm": 4.796603679656982,
      "learning_rate": 4.3455258263034605e-05,
      "loss": 0.8048,
      "num_input_tokens_seen": 2915792,
      "step": 625
    },
    {
      "epoch": 0.47493403693931396,
      "grad_norm": 2.2890474796295166,
      "learning_rate": 4.33550470113635e-05,
      "loss": 0.7072,
      "num_input_tokens_seen": 2939488,
      "step": 630
    },
    {
      "epoch": 0.4787033546928006,
      "grad_norm": 1.8012391328811646,
      "learning_rate": 4.3254191817404804e-05,
      "loss": 0.7911,
      "num_input_tokens_seen": 2962992,
      "step": 635
    },
    {
      "epoch": 0.4824726724462872,
      "grad_norm": 1.8014194965362549,
      "learning_rate": 4.3152696219418295e-05,
      "loss": 0.8293,
      "num_input_tokens_seen": 2986544,
      "step": 640
    },
    {
      "epoch": 0.4862419901997738,
      "grad_norm": 2.239366292953491,
      "learning_rate": 4.305056377813075e-05,
      "loss": 0.8835,
      "num_input_tokens_seen": 3009984,
      "step": 645
    },
    {
      "epoch": 0.4900113079532605,
      "grad_norm": 1.5583738088607788,
      "learning_rate": 4.294779807661105e-05,
      "loss": 0.7262,
      "num_input_tokens_seen": 3033520,
      "step": 650
    },
    {
      "epoch": 0.4937806257067471,
      "grad_norm": 1.7024521827697754,
      "learning_rate": 4.2844402720144496e-05,
      "loss": 0.8231,
      "num_input_tokens_seen": 3057056,
      "step": 655
    },
    {
      "epoch": 0.4975499434602337,
      "grad_norm": 3.108896493911743,
      "learning_rate": 4.274038133610628e-05,
      "loss": 0.8302,
      "num_input_tokens_seen": 3080656,
      "step": 660
    },
    {
      "epoch": 0.5013192612137203,
      "grad_norm": 3.166442394256592,
      "learning_rate": 4.263573757383427e-05,
      "loss": 0.8095,
      "num_input_tokens_seen": 3103792,
      "step": 665
    },
    {
      "epoch": 0.5050885789672069,
      "grad_norm": 2.218747854232788,
      "learning_rate": 4.2530475104500956e-05,
      "loss": 0.8756,
      "num_input_tokens_seen": 3126976,
      "step": 670
    },
    {
      "epoch": 0.5088578967206936,
      "grad_norm": 2.5289299488067627,
      "learning_rate": 4.242459762098466e-05,
      "loss": 0.7733,
      "num_input_tokens_seen": 3150224,
      "step": 675
    },
    {
      "epoch": 0.5126272144741801,
      "grad_norm": 1.674834132194519,
      "learning_rate": 4.231810883773999e-05,
      "loss": 0.7715,
      "num_input_tokens_seen": 3173296,
      "step": 680
    },
    {
      "epoch": 0.5163965322276668,
      "grad_norm": 1.7702105045318604,
      "learning_rate": 4.2211012490667524e-05,
      "loss": 0.6996,
      "num_input_tokens_seen": 3196560,
      "step": 685
    },
    {
      "epoch": 0.5201658499811534,
      "grad_norm": 2.7741544246673584,
      "learning_rate": 4.2103312336982734e-05,
      "loss": 0.8889,
      "num_input_tokens_seen": 3220432,
      "step": 690
    },
    {
      "epoch": 0.52393516773464,
      "grad_norm": 2.749328136444092,
      "learning_rate": 4.19950121550842e-05,
      "loss": 0.8717,
      "num_input_tokens_seen": 3243712,
      "step": 695
    },
    {
      "epoch": 0.5277044854881267,
      "grad_norm": 2.4472267627716064,
      "learning_rate": 4.188611574442101e-05,
      "loss": 0.6314,
      "num_input_tokens_seen": 3266976,
      "step": 700
    },
    {
      "epoch": 0.5314738032416133,
      "grad_norm": 2.1869001388549805,
      "learning_rate": 4.177662692535952e-05,
      "loss": 0.7517,
      "num_input_tokens_seen": 3290656,
      "step": 705
    },
    {
      "epoch": 0.5352431209950999,
      "grad_norm": 1.7907201051712036,
      "learning_rate": 4.166654953904926e-05,
      "loss": 0.8373,
      "num_input_tokens_seen": 3313712,
      "step": 710
    },
    {
      "epoch": 0.5390124387485865,
      "grad_norm": 1.8779516220092773,
      "learning_rate": 4.155588744728826e-05,
      "loss": 0.7025,
      "num_input_tokens_seen": 3336880,
      "step": 715
    },
    {
      "epoch": 0.5427817565020732,
      "grad_norm": 2.0460379123687744,
      "learning_rate": 4.144464453238748e-05,
      "loss": 0.6709,
      "num_input_tokens_seen": 3360416,
      "step": 720
    },
    {
      "epoch": 0.5465510742555597,
      "grad_norm": 2.096501350402832,
      "learning_rate": 4.133282469703469e-05,
      "loss": 0.6451,
      "num_input_tokens_seen": 3383408,
      "step": 725
    },
    {
      "epoch": 0.5503203920090464,
      "grad_norm": 2.441951036453247,
      "learning_rate": 4.122043186415746e-05,
      "loss": 0.754,
      "num_input_tokens_seen": 3406752,
      "step": 730
    },
    {
      "epoch": 0.554089709762533,
      "grad_norm": 2.059370517730713,
      "learning_rate": 4.110746997678565e-05,
      "loss": 0.8784,
      "num_input_tokens_seen": 3429968,
      "step": 735
    },
    {
      "epoch": 0.5578590275160196,
      "grad_norm": 1.969912052154541,
      "learning_rate": 4.0993942997912984e-05,
      "loss": 0.9384,
      "num_input_tokens_seen": 3453136,
      "step": 740
    },
    {
      "epoch": 0.5616283452695062,
      "grad_norm": 2.3595261573791504,
      "learning_rate": 4.087985491035804e-05,
      "loss": 0.7915,
      "num_input_tokens_seen": 3476384,
      "step": 745
    },
    {
      "epoch": 0.5653976630229929,
      "grad_norm": 2.187009572982788,
      "learning_rate": 4.076520971662455e-05,
      "loss": 0.8374,
      "num_input_tokens_seen": 3500176,
      "step": 750
    },
    {
      "epoch": 0.5691669807764794,
      "grad_norm": 1.7618275880813599,
      "learning_rate": 4.065001143876097e-05,
      "loss": 0.7302,
      "num_input_tokens_seen": 3524048,
      "step": 755
    },
    {
      "epoch": 0.5729362985299661,
      "grad_norm": 2.1269426345825195,
      "learning_rate": 4.053426411821934e-05,
      "loss": 0.754,
      "num_input_tokens_seen": 3547056,
      "step": 760
    },
    {
      "epoch": 0.5767056162834527,
      "grad_norm": 2.3369758129119873,
      "learning_rate": 4.0417971815713584e-05,
      "loss": 0.8208,
      "num_input_tokens_seen": 3570608,
      "step": 765
    },
    {
      "epoch": 0.5804749340369393,
      "grad_norm": 2.156280279159546,
      "learning_rate": 4.030113861107693e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 3594192,
      "step": 770
    },
    {
      "epoch": 0.5842442517904259,
      "grad_norm": 1.7685117721557617,
      "learning_rate": 4.0183768603118886e-05,
      "loss": 0.7606,
      "num_input_tokens_seen": 3617600,
      "step": 775
    },
    {
      "epoch": 0.5880135695439126,
      "grad_norm": 2.0940916538238525,
      "learning_rate": 4.0065865909481417e-05,
      "loss": 0.7283,
      "num_input_tokens_seen": 3640416,
      "step": 780
    },
    {
      "epoch": 0.5917828872973991,
      "grad_norm": 2.5771267414093018,
      "learning_rate": 3.994743466649442e-05,
      "loss": 0.6878,
      "num_input_tokens_seen": 3663696,
      "step": 785
    },
    {
      "epoch": 0.5955522050508858,
      "grad_norm": 2.336862087249756,
      "learning_rate": 3.982847902903071e-05,
      "loss": 0.8009,
      "num_input_tokens_seen": 3686880,
      "step": 790
    },
    {
      "epoch": 0.5993215228043725,
      "grad_norm": 2.91589093208313,
      "learning_rate": 3.9709003170360176e-05,
      "loss": 0.794,
      "num_input_tokens_seen": 3710224,
      "step": 795
    },
    {
      "epoch": 0.603090840557859,
      "grad_norm": 3.051378011703491,
      "learning_rate": 3.958901128200344e-05,
      "loss": 0.7532,
      "num_input_tokens_seen": 3733536,
      "step": 800
    },
    {
      "epoch": 0.6068601583113457,
      "grad_norm": 3.457322835922241,
      "learning_rate": 3.946850757358475e-05,
      "loss": 0.8581,
      "num_input_tokens_seen": 3756688,
      "step": 805
    },
    {
      "epoch": 0.6106294760648323,
      "grad_norm": 1.4198774099349976,
      "learning_rate": 3.9347496272684325e-05,
      "loss": 0.7889,
      "num_input_tokens_seen": 3779504,
      "step": 810
    },
    {
      "epoch": 0.6143987938183189,
      "grad_norm": 2.5762572288513184,
      "learning_rate": 3.922598162469003e-05,
      "loss": 0.7401,
      "num_input_tokens_seen": 3802544,
      "step": 815
    },
    {
      "epoch": 0.6181681115718055,
      "grad_norm": 2.5552995204925537,
      "learning_rate": 3.910396789264845e-05,
      "loss": 0.799,
      "num_input_tokens_seen": 3825680,
      "step": 820
    },
    {
      "epoch": 0.6219374293252922,
      "grad_norm": 1.8057173490524292,
      "learning_rate": 3.8981459357115325e-05,
      "loss": 0.8015,
      "num_input_tokens_seen": 3849040,
      "step": 825
    },
    {
      "epoch": 0.6257067470787787,
      "grad_norm": 1.8398315906524658,
      "learning_rate": 3.885846031600536e-05,
      "loss": 0.6905,
      "num_input_tokens_seen": 3872320,
      "step": 830
    },
    {
      "epoch": 0.6294760648322654,
      "grad_norm": 2.7629125118255615,
      "learning_rate": 3.8734975084441466e-05,
      "loss": 0.845,
      "num_input_tokens_seen": 3895920,
      "step": 835
    },
    {
      "epoch": 0.633245382585752,
      "grad_norm": 1.9969942569732666,
      "learning_rate": 3.8611007994603365e-05,
      "loss": 0.7317,
      "num_input_tokens_seen": 3919360,
      "step": 840
    },
    {
      "epoch": 0.6370147003392386,
      "grad_norm": 3.591285467147827,
      "learning_rate": 3.8486563395575625e-05,
      "loss": 0.724,
      "num_input_tokens_seen": 3942736,
      "step": 845
    },
    {
      "epoch": 0.6407840180927252,
      "grad_norm": 2.4714577198028564,
      "learning_rate": 3.8361645653195026e-05,
      "loss": 0.664,
      "num_input_tokens_seen": 3966144,
      "step": 850
    },
    {
      "epoch": 0.6445533358462119,
      "grad_norm": 3.081908702850342,
      "learning_rate": 3.823625914989748e-05,
      "loss": 0.5922,
      "num_input_tokens_seen": 3989472,
      "step": 855
    },
    {
      "epoch": 0.6483226535996984,
      "grad_norm": 3.5514636039733887,
      "learning_rate": 3.811040828456421e-05,
      "loss": 0.9465,
      "num_input_tokens_seen": 4012656,
      "step": 860
    },
    {
      "epoch": 0.6520919713531851,
      "grad_norm": 3.4660677909851074,
      "learning_rate": 3.798409747236745e-05,
      "loss": 0.7781,
      "num_input_tokens_seen": 4035888,
      "step": 865
    },
    {
      "epoch": 0.6558612891066717,
      "grad_norm": 3.08896541595459,
      "learning_rate": 3.7857331144615574e-05,
      "loss": 0.8309,
      "num_input_tokens_seen": 4059504,
      "step": 870
    },
    {
      "epoch": 0.6596306068601583,
      "grad_norm": 2.4300408363342285,
      "learning_rate": 3.773011374859761e-05,
      "loss": 0.7812,
      "num_input_tokens_seen": 4083440,
      "step": 875
    },
    {
      "epoch": 0.6633999246136449,
      "grad_norm": 3.5635619163513184,
      "learning_rate": 3.7602449747427204e-05,
      "loss": 0.7888,
      "num_input_tokens_seen": 4106928,
      "step": 880
    },
    {
      "epoch": 0.6671692423671316,
      "grad_norm": 1.9514297246932983,
      "learning_rate": 3.747434361988608e-05,
      "loss": 0.7584,
      "num_input_tokens_seen": 4130336,
      "step": 885
    },
    {
      "epoch": 0.6709385601206181,
      "grad_norm": 1.6948657035827637,
      "learning_rate": 3.734579986026688e-05,
      "loss": 0.6848,
      "num_input_tokens_seen": 4153872,
      "step": 890
    },
    {
      "epoch": 0.6747078778741048,
      "grad_norm": 1.9811207056045532,
      "learning_rate": 3.7216822978215514e-05,
      "loss": 0.8068,
      "num_input_tokens_seen": 4177072,
      "step": 895
    },
    {
      "epoch": 0.6784771956275915,
      "grad_norm": 3.7668049335479736,
      "learning_rate": 3.7087417498572944e-05,
      "loss": 0.8122,
      "num_input_tokens_seen": 4200256,
      "step": 900
    },
    {
      "epoch": 0.682246513381078,
      "grad_norm": 2.3684747219085693,
      "learning_rate": 3.695758796121642e-05,
      "loss": 0.5764,
      "num_input_tokens_seen": 4223264,
      "step": 905
    },
    {
      "epoch": 0.6860158311345647,
      "grad_norm": 2.597867250442505,
      "learning_rate": 3.6827338920900254e-05,
      "loss": 0.8898,
      "num_input_tokens_seen": 4247024,
      "step": 910
    },
    {
      "epoch": 0.6897851488880513,
      "grad_norm": 2.0299720764160156,
      "learning_rate": 3.6696674947095984e-05,
      "loss": 0.7741,
      "num_input_tokens_seen": 4270800,
      "step": 915
    },
    {
      "epoch": 0.6935544666415379,
      "grad_norm": 2.4363317489624023,
      "learning_rate": 3.656560062383208e-05,
      "loss": 0.8219,
      "num_input_tokens_seen": 4294352,
      "step": 920
    },
    {
      "epoch": 0.6973237843950245,
      "grad_norm": 1.752125859260559,
      "learning_rate": 3.6434120549533135e-05,
      "loss": 0.6813,
      "num_input_tokens_seen": 4318208,
      "step": 925
    },
    {
      "epoch": 0.7010931021485112,
      "grad_norm": 3.325110912322998,
      "learning_rate": 3.6302239336858545e-05,
      "loss": 0.6866,
      "num_input_tokens_seen": 4341728,
      "step": 930
    },
    {
      "epoch": 0.7048624199019977,
      "grad_norm": 1.9037253856658936,
      "learning_rate": 3.6169961612540645e-05,
      "loss": 0.8628,
      "num_input_tokens_seen": 4365392,
      "step": 935
    },
    {
      "epoch": 0.7086317376554844,
      "grad_norm": 1.9913270473480225,
      "learning_rate": 3.603729201722244e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 4389184,
      "step": 940
    },
    {
      "epoch": 0.712401055408971,
      "grad_norm": 2.668295383453369,
      "learning_rate": 3.5904235205294776e-05,
      "loss": 0.7572,
      "num_input_tokens_seen": 4412272,
      "step": 945
    },
    {
      "epoch": 0.7161703731624576,
      "grad_norm": 2.363468885421753,
      "learning_rate": 3.5770795844733035e-05,
      "loss": 0.7205,
      "num_input_tokens_seen": 4435072,
      "step": 950
    },
    {
      "epoch": 0.7199396909159442,
      "grad_norm": 2.3162343502044678,
      "learning_rate": 3.5636978616933416e-05,
      "loss": 0.7521,
      "num_input_tokens_seen": 4458272,
      "step": 955
    },
    {
      "epoch": 0.7237090086694309,
      "grad_norm": 1.9907945394515991,
      "learning_rate": 3.550278821654866e-05,
      "loss": 0.7205,
      "num_input_tokens_seen": 4481056,
      "step": 960
    },
    {
      "epoch": 0.7274783264229174,
      "grad_norm": 2.0076990127563477,
      "learning_rate": 3.536822935132336e-05,
      "loss": 0.6597,
      "num_input_tokens_seen": 4504288,
      "step": 965
    },
    {
      "epoch": 0.7312476441764041,
      "grad_norm": 1.6346359252929688,
      "learning_rate": 3.5233306741928806e-05,
      "loss": 0.7293,
      "num_input_tokens_seen": 4527552,
      "step": 970
    },
    {
      "epoch": 0.7350169619298907,
      "grad_norm": 2.227130651473999,
      "learning_rate": 3.509802512179737e-05,
      "loss": 0.7055,
      "num_input_tokens_seen": 4551024,
      "step": 975
    },
    {
      "epoch": 0.7387862796833773,
      "grad_norm": 3.003014087677002,
      "learning_rate": 3.496238923695646e-05,
      "loss": 0.763,
      "num_input_tokens_seen": 4574464,
      "step": 980
    },
    {
      "epoch": 0.742555597436864,
      "grad_norm": 2.196281671524048,
      "learning_rate": 3.4826403845861986e-05,
      "loss": 0.6831,
      "num_input_tokens_seen": 4597600,
      "step": 985
    },
    {
      "epoch": 0.7463249151903506,
      "grad_norm": 2.487607717514038,
      "learning_rate": 3.4690073719231425e-05,
      "loss": 0.6586,
      "num_input_tokens_seen": 4621424,
      "step": 990
    },
    {
      "epoch": 0.7500942329438371,
      "grad_norm": 2.686861753463745,
      "learning_rate": 3.455340363987648e-05,
      "loss": 0.7481,
      "num_input_tokens_seen": 4644400,
      "step": 995
    },
    {
      "epoch": 0.7538635506973238,
      "grad_norm": 2.511491060256958,
      "learning_rate": 3.4416398402535284e-05,
      "loss": 0.8084,
      "num_input_tokens_seen": 4667568,
      "step": 1000
    },
    {
      "epoch": 0.7576328684508103,
      "grad_norm": 3.698335886001587,
      "learning_rate": 3.427906281370414e-05,
      "loss": 0.8876,
      "num_input_tokens_seen": 4690624,
      "step": 1005
    },
    {
      "epoch": 0.761402186204297,
      "grad_norm": 4.032379150390625,
      "learning_rate": 3.414140169146896e-05,
      "loss": 0.8391,
      "num_input_tokens_seen": 4714032,
      "step": 1010
    },
    {
      "epoch": 0.7651715039577837,
      "grad_norm": 1.8396016359329224,
      "learning_rate": 3.400341986533618e-05,
      "loss": 0.6847,
      "num_input_tokens_seen": 4737280,
      "step": 1015
    },
    {
      "epoch": 0.7689408217112702,
      "grad_norm": 2.474912166595459,
      "learning_rate": 3.386512217606339e-05,
      "loss": 0.6742,
      "num_input_tokens_seen": 4760848,
      "step": 1020
    },
    {
      "epoch": 0.7727101394647569,
      "grad_norm": 3.3939056396484375,
      "learning_rate": 3.3726513475489445e-05,
      "loss": 0.8607,
      "num_input_tokens_seen": 4783888,
      "step": 1025
    },
    {
      "epoch": 0.7764794572182435,
      "grad_norm": 2.451918363571167,
      "learning_rate": 3.3587598626364294e-05,
      "loss": 0.6614,
      "num_input_tokens_seen": 4807312,
      "step": 1030
    },
    {
      "epoch": 0.7802487749717301,
      "grad_norm": 2.324479341506958,
      "learning_rate": 3.344838250217833e-05,
      "loss": 0.7078,
      "num_input_tokens_seen": 4831088,
      "step": 1035
    },
    {
      "epoch": 0.7840180927252167,
      "grad_norm": 2.647658586502075,
      "learning_rate": 3.330886998699149e-05,
      "loss": 0.7944,
      "num_input_tokens_seen": 4854608,
      "step": 1040
    },
    {
      "epoch": 0.7877874104787034,
      "grad_norm": 2.21478533744812,
      "learning_rate": 3.316906597526186e-05,
      "loss": 0.7436,
      "num_input_tokens_seen": 4877648,
      "step": 1045
    },
    {
      "epoch": 0.7915567282321899,
      "grad_norm": 2.0611255168914795,
      "learning_rate": 3.302897537167397e-05,
      "loss": 0.8234,
      "num_input_tokens_seen": 4900672,
      "step": 1050
    },
    {
      "epoch": 0.7953260459856766,
      "grad_norm": 2.0772039890289307,
      "learning_rate": 3.288860309096671e-05,
      "loss": 0.6463,
      "num_input_tokens_seen": 4924192,
      "step": 1055
    },
    {
      "epoch": 0.7990953637391632,
      "grad_norm": 3.051643133163452,
      "learning_rate": 3.2747954057760965e-05,
      "loss": 0.6872,
      "num_input_tokens_seen": 4947600,
      "step": 1060
    },
    {
      "epoch": 0.8028646814926498,
      "grad_norm": 2.400989532470703,
      "learning_rate": 3.260703320638679e-05,
      "loss": 0.6143,
      "num_input_tokens_seen": 4971888,
      "step": 1065
    },
    {
      "epoch": 0.8066339992461364,
      "grad_norm": 2.88338303565979,
      "learning_rate": 3.246584548071034e-05,
      "loss": 0.8022,
      "num_input_tokens_seen": 4995056,
      "step": 1070
    },
    {
      "epoch": 0.8104033169996231,
      "grad_norm": 3.0658576488494873,
      "learning_rate": 3.232439583396036e-05,
      "loss": 0.7144,
      "num_input_tokens_seen": 5018208,
      "step": 1075
    },
    {
      "epoch": 0.8141726347531096,
      "grad_norm": 2.397193193435669,
      "learning_rate": 3.2182689228554517e-05,
      "loss": 0.757,
      "num_input_tokens_seen": 5041584,
      "step": 1080
    },
    {
      "epoch": 0.8179419525065963,
      "grad_norm": 2.7260894775390625,
      "learning_rate": 3.204073063592522e-05,
      "loss": 0.729,
      "num_input_tokens_seen": 5064352,
      "step": 1085
    },
    {
      "epoch": 0.821711270260083,
      "grad_norm": 2.864222764968872,
      "learning_rate": 3.189852503634523e-05,
      "loss": 0.7441,
      "num_input_tokens_seen": 5087712,
      "step": 1090
    },
    {
      "epoch": 0.8254805880135695,
      "grad_norm": 2.9511120319366455,
      "learning_rate": 3.1756077418752967e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 5111104,
      "step": 1095
    },
    {
      "epoch": 0.8292499057670562,
      "grad_norm": 2.240609645843506,
      "learning_rate": 3.1613392780577455e-05,
      "loss": 0.7098,
      "num_input_tokens_seen": 5134720,
      "step": 1100
    },
    {
      "epoch": 0.8330192235205428,
      "grad_norm": 2.762382984161377,
      "learning_rate": 3.147047612756302e-05,
      "loss": 0.7099,
      "num_input_tokens_seen": 5157744,
      "step": 1105
    },
    {
      "epoch": 0.8367885412740294,
      "grad_norm": 2.2460415363311768,
      "learning_rate": 3.132733247359366e-05,
      "loss": 0.7648,
      "num_input_tokens_seen": 5180976,
      "step": 1110
    },
    {
      "epoch": 0.840557859027516,
      "grad_norm": 2.3306996822357178,
      "learning_rate": 3.118396684051714e-05,
      "loss": 0.7906,
      "num_input_tokens_seen": 5204976,
      "step": 1115
    },
    {
      "epoch": 0.8443271767810027,
      "grad_norm": 2.5235891342163086,
      "learning_rate": 3.104038425796884e-05,
      "loss": 0.6022,
      "num_input_tokens_seen": 5228048,
      "step": 1120
    },
    {
      "epoch": 0.8480964945344892,
      "grad_norm": 1.9855157136917114,
      "learning_rate": 3.089658976319528e-05,
      "loss": 0.8142,
      "num_input_tokens_seen": 5251664,
      "step": 1125
    },
    {
      "epoch": 0.8518658122879759,
      "grad_norm": 2.2459208965301514,
      "learning_rate": 3.0752588400877405e-05,
      "loss": 0.8263,
      "num_input_tokens_seen": 5274976,
      "step": 1130
    },
    {
      "epoch": 0.8556351300414625,
      "grad_norm": 2.2121737003326416,
      "learning_rate": 3.060838522295361e-05,
      "loss": 0.8581,
      "num_input_tokens_seen": 5298352,
      "step": 1135
    },
    {
      "epoch": 0.8594044477949491,
      "grad_norm": 2.092641592025757,
      "learning_rate": 3.0463985288442475e-05,
      "loss": 0.7224,
      "num_input_tokens_seen": 5321440,
      "step": 1140
    },
    {
      "epoch": 0.8631737655484357,
      "grad_norm": 2.473989963531494,
      "learning_rate": 3.031939366326535e-05,
      "loss": 0.7379,
      "num_input_tokens_seen": 5344992,
      "step": 1145
    },
    {
      "epoch": 0.8669430833019224,
      "grad_norm": 2.082732677459717,
      "learning_rate": 3.0174615420068563e-05,
      "loss": 0.6864,
      "num_input_tokens_seen": 5368448,
      "step": 1150
    },
    {
      "epoch": 0.8707124010554089,
      "grad_norm": 2.066802501678467,
      "learning_rate": 3.0029655638045496e-05,
      "loss": 0.6144,
      "num_input_tokens_seen": 5391568,
      "step": 1155
    },
    {
      "epoch": 0.8744817188088956,
      "grad_norm": 1.7655906677246094,
      "learning_rate": 2.9884519402758342e-05,
      "loss": 0.6818,
      "num_input_tokens_seen": 5414992,
      "step": 1160
    },
    {
      "epoch": 0.8782510365623822,
      "grad_norm": 2.6639342308044434,
      "learning_rate": 2.9739211805959783e-05,
      "loss": 0.8296,
      "num_input_tokens_seen": 5438368,
      "step": 1165
    },
    {
      "epoch": 0.8820203543158688,
      "grad_norm": 2.0490756034851074,
      "learning_rate": 2.9593737945414264e-05,
      "loss": 0.7764,
      "num_input_tokens_seen": 5461504,
      "step": 1170
    },
    {
      "epoch": 0.8857896720693554,
      "grad_norm": 2.066798448562622,
      "learning_rate": 2.9448102924719207e-05,
      "loss": 0.7245,
      "num_input_tokens_seen": 5484992,
      "step": 1175
    },
    {
      "epoch": 0.8895589898228421,
      "grad_norm": 2.2488503456115723,
      "learning_rate": 2.9302311853125942e-05,
      "loss": 0.8319,
      "num_input_tokens_seen": 5508480,
      "step": 1180
    },
    {
      "epoch": 0.8933283075763286,
      "grad_norm": 3.150104284286499,
      "learning_rate": 2.9156369845360467e-05,
      "loss": 0.6652,
      "num_input_tokens_seen": 5531520,
      "step": 1185
    },
    {
      "epoch": 0.8970976253298153,
      "grad_norm": 2.160212755203247,
      "learning_rate": 2.9010282021444008e-05,
      "loss": 0.5858,
      "num_input_tokens_seen": 5554480,
      "step": 1190
    },
    {
      "epoch": 0.900866943083302,
      "grad_norm": 2.795851707458496,
      "learning_rate": 2.8864053506513405e-05,
      "loss": 0.6248,
      "num_input_tokens_seen": 5577888,
      "step": 1195
    },
    {
      "epoch": 0.9046362608367885,
      "grad_norm": 2.5486669540405273,
      "learning_rate": 2.8717689430641292e-05,
      "loss": 0.8447,
      "num_input_tokens_seen": 5601424,
      "step": 1200
    },
    {
      "epoch": 0.9084055785902752,
      "grad_norm": 2.2841227054595947,
      "learning_rate": 2.857119492865613e-05,
      "loss": 0.6355,
      "num_input_tokens_seen": 5624880,
      "step": 1205
    },
    {
      "epoch": 0.9121748963437618,
      "grad_norm": 2.077192544937134,
      "learning_rate": 2.842457513996207e-05,
      "loss": 0.5075,
      "num_input_tokens_seen": 5648160,
      "step": 1210
    },
    {
      "epoch": 0.9159442140972484,
      "grad_norm": 2.1494250297546387,
      "learning_rate": 2.8277835208358637e-05,
      "loss": 0.5762,
      "num_input_tokens_seen": 5671280,
      "step": 1215
    },
    {
      "epoch": 0.919713531850735,
      "grad_norm": 2.5202853679656982,
      "learning_rate": 2.813098028186028e-05,
      "loss": 0.666,
      "num_input_tokens_seen": 5694352,
      "step": 1220
    },
    {
      "epoch": 0.9234828496042217,
      "grad_norm": 3.3707363605499268,
      "learning_rate": 2.798401551251576e-05,
      "loss": 0.6056,
      "num_input_tokens_seen": 5717536,
      "step": 1225
    },
    {
      "epoch": 0.9272521673577082,
      "grad_norm": 3.4344711303710938,
      "learning_rate": 2.7836946056227426e-05,
      "loss": 0.8095,
      "num_input_tokens_seen": 5740416,
      "step": 1230
    },
    {
      "epoch": 0.9310214851111949,
      "grad_norm": 2.2710204124450684,
      "learning_rate": 2.7689777072570287e-05,
      "loss": 0.7501,
      "num_input_tokens_seen": 5763712,
      "step": 1235
    },
    {
      "epoch": 0.9347908028646815,
      "grad_norm": 2.100459337234497,
      "learning_rate": 2.7542513724611057e-05,
      "loss": 0.5595,
      "num_input_tokens_seen": 5787168,
      "step": 1240
    },
    {
      "epoch": 0.9385601206181681,
      "grad_norm": 3.0176243782043457,
      "learning_rate": 2.739516117872697e-05,
      "loss": 0.694,
      "num_input_tokens_seen": 5810704,
      "step": 1245
    },
    {
      "epoch": 0.9423294383716547,
      "grad_norm": 2.814894676208496,
      "learning_rate": 2.7247724604424557e-05,
      "loss": 0.8521,
      "num_input_tokens_seen": 5834192,
      "step": 1250
    },
    {
      "epoch": 0.9460987561251414,
      "grad_norm": 1.729038953781128,
      "learning_rate": 2.71002091741583e-05,
      "loss": 0.7794,
      "num_input_tokens_seen": 5857344,
      "step": 1255
    },
    {
      "epoch": 0.9498680738786279,
      "grad_norm": 2.401945114135742,
      "learning_rate": 2.695262006314912e-05,
      "loss": 0.7721,
      "num_input_tokens_seen": 5880448,
      "step": 1260
    },
    {
      "epoch": 0.9536373916321146,
      "grad_norm": 2.005068302154541,
      "learning_rate": 2.680496244920287e-05,
      "loss": 0.6403,
      "num_input_tokens_seen": 5903200,
      "step": 1265
    },
    {
      "epoch": 0.9574067093856012,
      "grad_norm": 1.8433932065963745,
      "learning_rate": 2.665724151252868e-05,
      "loss": 0.738,
      "num_input_tokens_seen": 5926272,
      "step": 1270
    },
    {
      "epoch": 0.9611760271390878,
      "grad_norm": 2.32893705368042,
      "learning_rate": 2.6509462435557152e-05,
      "loss": 0.6469,
      "num_input_tokens_seen": 5949680,
      "step": 1275
    },
    {
      "epoch": 0.9649453448925744,
      "grad_norm": 2.5352189540863037,
      "learning_rate": 2.6361630402758648e-05,
      "loss": 0.6953,
      "num_input_tokens_seen": 5973088,
      "step": 1280
    },
    {
      "epoch": 0.9687146626460611,
      "grad_norm": 2.217909574508667,
      "learning_rate": 2.6213750600461334e-05,
      "loss": 0.6365,
      "num_input_tokens_seen": 5996688,
      "step": 1285
    },
    {
      "epoch": 0.9724839803995476,
      "grad_norm": 1.218327283859253,
      "learning_rate": 2.6065828216669253e-05,
      "loss": 0.6691,
      "num_input_tokens_seen": 6019744,
      "step": 1290
    },
    {
      "epoch": 0.9762532981530343,
      "grad_norm": 1.710153579711914,
      "learning_rate": 2.5917868440880317e-05,
      "loss": 0.64,
      "num_input_tokens_seen": 6042640,
      "step": 1295
    },
    {
      "epoch": 0.980022615906521,
      "grad_norm": 2.1945626735687256,
      "learning_rate": 2.5769876463904265e-05,
      "loss": 0.6674,
      "num_input_tokens_seen": 6066112,
      "step": 1300
    },
    {
      "epoch": 0.9837919336600075,
      "grad_norm": 2.905019760131836,
      "learning_rate": 2.5621857477680506e-05,
      "loss": 0.6288,
      "num_input_tokens_seen": 6089296,
      "step": 1305
    },
    {
      "epoch": 0.9875612514134942,
      "grad_norm": 2.2408697605133057,
      "learning_rate": 2.5473816675096017e-05,
      "loss": 0.8477,
      "num_input_tokens_seen": 6112784,
      "step": 1310
    },
    {
      "epoch": 0.9913305691669808,
      "grad_norm": 2.5332565307617188,
      "learning_rate": 2.5325759249803154e-05,
      "loss": 0.7663,
      "num_input_tokens_seen": 6136048,
      "step": 1315
    },
    {
      "epoch": 0.9950998869204674,
      "grad_norm": 2.0274627208709717,
      "learning_rate": 2.517769039603744e-05,
      "loss": 0.861,
      "num_input_tokens_seen": 6159920,
      "step": 1320
    },
    {
      "epoch": 0.998869204673954,
      "grad_norm": 2.1323111057281494,
      "learning_rate": 2.5029615308435338e-05,
      "loss": 0.8545,
      "num_input_tokens_seen": 6183024,
      "step": 1325
    },
    {
      "epoch": 1.0026385224274406,
      "grad_norm": 1.8671432733535767,
      "learning_rate": 2.4881539181851986e-05,
      "loss": 0.6469,
      "num_input_tokens_seen": 6206224,
      "step": 1330
    },
    {
      "epoch": 1.0064078401809273,
      "grad_norm": 1.7728124856948853,
      "learning_rate": 2.4733467211179008e-05,
      "loss": 0.7507,
      "num_input_tokens_seen": 6229936,
      "step": 1335
    },
    {
      "epoch": 1.0101771579344139,
      "grad_norm": 3.0307295322418213,
      "learning_rate": 2.4585404591162218e-05,
      "loss": 0.7181,
      "num_input_tokens_seen": 6253152,
      "step": 1340
    },
    {
      "epoch": 1.0139464756879004,
      "grad_norm": 1.6619493961334229,
      "learning_rate": 2.4437356516219358e-05,
      "loss": 0.627,
      "num_input_tokens_seen": 6276208,
      "step": 1345
    },
    {
      "epoch": 1.0177157934413872,
      "grad_norm": 2.804950714111328,
      "learning_rate": 2.4289328180257926e-05,
      "loss": 0.7266,
      "num_input_tokens_seen": 6299408,
      "step": 1350
    },
    {
      "epoch": 1.0214851111948737,
      "grad_norm": 1.9919047355651855,
      "learning_rate": 2.4141324776492915e-05,
      "loss": 0.5684,
      "num_input_tokens_seen": 6323024,
      "step": 1355
    },
    {
      "epoch": 1.0252544289483603,
      "grad_norm": 5.080316543579102,
      "learning_rate": 2.399335149726463e-05,
      "loss": 0.6196,
      "num_input_tokens_seen": 6345952,
      "step": 1360
    },
    {
      "epoch": 1.029023746701847,
      "grad_norm": 2.5290539264678955,
      "learning_rate": 2.3845413533856517e-05,
      "loss": 0.5296,
      "num_input_tokens_seen": 6369856,
      "step": 1365
    },
    {
      "epoch": 1.0327930644553336,
      "grad_norm": 2.438260555267334,
      "learning_rate": 2.3697516076313066e-05,
      "loss": 0.7537,
      "num_input_tokens_seen": 6393136,
      "step": 1370
    },
    {
      "epoch": 1.0365623822088201,
      "grad_norm": 3.0706968307495117,
      "learning_rate": 2.354966431325773e-05,
      "loss": 0.5909,
      "num_input_tokens_seen": 6416736,
      "step": 1375
    },
    {
      "epoch": 1.040331699962307,
      "grad_norm": 2.409477472305298,
      "learning_rate": 2.3401863431710863e-05,
      "loss": 0.7042,
      "num_input_tokens_seen": 6440048,
      "step": 1380
    },
    {
      "epoch": 1.0441010177157934,
      "grad_norm": 2.7514872550964355,
      "learning_rate": 2.325411861690776e-05,
      "loss": 0.7301,
      "num_input_tokens_seen": 6463504,
      "step": 1385
    },
    {
      "epoch": 1.04787033546928,
      "grad_norm": 3.314840793609619,
      "learning_rate": 2.3106435052116764e-05,
      "loss": 0.7472,
      "num_input_tokens_seen": 6486608,
      "step": 1390
    },
    {
      "epoch": 1.0516396532227668,
      "grad_norm": 3.0806994438171387,
      "learning_rate": 2.2958817918457412e-05,
      "loss": 0.5799,
      "num_input_tokens_seen": 6509760,
      "step": 1395
    },
    {
      "epoch": 1.0554089709762533,
      "grad_norm": 3.1639201641082764,
      "learning_rate": 2.2811272394718647e-05,
      "loss": 0.6512,
      "num_input_tokens_seen": 6532992,
      "step": 1400
    },
    {
      "epoch": 1.0591782887297398,
      "grad_norm": 2.3345203399658203,
      "learning_rate": 2.2663803657177173e-05,
      "loss": 0.748,
      "num_input_tokens_seen": 6556384,
      "step": 1405
    },
    {
      "epoch": 1.0629476064832266,
      "grad_norm": 6.463221549987793,
      "learning_rate": 2.2516416879415824e-05,
      "loss": 0.6635,
      "num_input_tokens_seen": 6580336,
      "step": 1410
    },
    {
      "epoch": 1.0667169242367132,
      "grad_norm": 2.5758097171783447,
      "learning_rate": 2.2369117232142077e-05,
      "loss": 0.5894,
      "num_input_tokens_seen": 6603584,
      "step": 1415
    },
    {
      "epoch": 1.0704862419901997,
      "grad_norm": 2.079308271408081,
      "learning_rate": 2.2221909883006646e-05,
      "loss": 0.5952,
      "num_input_tokens_seen": 6626864,
      "step": 1420
    },
    {
      "epoch": 1.0742555597436865,
      "grad_norm": 2.637446403503418,
      "learning_rate": 2.20747999964222e-05,
      "loss": 0.7849,
      "num_input_tokens_seen": 6650480,
      "step": 1425
    },
    {
      "epoch": 1.078024877497173,
      "grad_norm": 2.275803565979004,
      "learning_rate": 2.192779273338215e-05,
      "loss": 0.7059,
      "num_input_tokens_seen": 6673808,
      "step": 1430
    },
    {
      "epoch": 1.0817941952506596,
      "grad_norm": 3.0671417713165283,
      "learning_rate": 2.1780893251279626e-05,
      "loss": 0.7389,
      "num_input_tokens_seen": 6697232,
      "step": 1435
    },
    {
      "epoch": 1.0855635130041463,
      "grad_norm": 2.4702844619750977,
      "learning_rate": 2.163410670372652e-05,
      "loss": 0.5858,
      "num_input_tokens_seen": 6721104,
      "step": 1440
    },
    {
      "epoch": 1.0893328307576329,
      "grad_norm": 2.3466484546661377,
      "learning_rate": 2.148743824037269e-05,
      "loss": 0.6743,
      "num_input_tokens_seen": 6744624,
      "step": 1445
    },
    {
      "epoch": 1.0931021485111194,
      "grad_norm": 1.9515085220336914,
      "learning_rate": 2.1340893006725288e-05,
      "loss": 0.5893,
      "num_input_tokens_seen": 6768000,
      "step": 1450
    },
    {
      "epoch": 1.0968714662646062,
      "grad_norm": 2.264751434326172,
      "learning_rate": 2.1194476143968258e-05,
      "loss": 0.653,
      "num_input_tokens_seen": 6791296,
      "step": 1455
    },
    {
      "epoch": 1.1006407840180927,
      "grad_norm": 3.2526028156280518,
      "learning_rate": 2.1048192788781977e-05,
      "loss": 0.6829,
      "num_input_tokens_seen": 6814800,
      "step": 1460
    },
    {
      "epoch": 1.1044101017715793,
      "grad_norm": 3.7399232387542725,
      "learning_rate": 2.090204807316301e-05,
      "loss": 0.5642,
      "num_input_tokens_seen": 6838128,
      "step": 1465
    },
    {
      "epoch": 1.108179419525066,
      "grad_norm": 4.240833282470703,
      "learning_rate": 2.0756047124244095e-05,
      "loss": 0.6401,
      "num_input_tokens_seen": 6861312,
      "step": 1470
    },
    {
      "epoch": 1.1119487372785526,
      "grad_norm": 3.439541816711426,
      "learning_rate": 2.0610195064114273e-05,
      "loss": 0.6188,
      "num_input_tokens_seen": 6884128,
      "step": 1475
    },
    {
      "epoch": 1.1157180550320391,
      "grad_norm": 3.2410943508148193,
      "learning_rate": 2.0464497009639176e-05,
      "loss": 0.5929,
      "num_input_tokens_seen": 6908336,
      "step": 1480
    },
    {
      "epoch": 1.119487372785526,
      "grad_norm": 2.0210845470428467,
      "learning_rate": 2.0318958072281517e-05,
      "loss": 0.7596,
      "num_input_tokens_seen": 6931392,
      "step": 1485
    },
    {
      "epoch": 1.1232566905390124,
      "grad_norm": 3.183546781539917,
      "learning_rate": 2.017358335792178e-05,
      "loss": 0.6313,
      "num_input_tokens_seen": 6954800,
      "step": 1490
    },
    {
      "epoch": 1.127026008292499,
      "grad_norm": 3.0664725303649902,
      "learning_rate": 2.0028377966679092e-05,
      "loss": 0.6371,
      "num_input_tokens_seen": 6978384,
      "step": 1495
    },
    {
      "epoch": 1.1307953260459858,
      "grad_norm": 2.9559929370880127,
      "learning_rate": 1.9883346992732256e-05,
      "loss": 0.6982,
      "num_input_tokens_seen": 7001632,
      "step": 1500
    }
  ],
  "logging_steps": 5,
  "max_steps": 2652,
  "num_input_tokens_seen": 7001632,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.33828956912681e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}