{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9994747767801315,
  "eval_steps": 500,
  "global_step": 14991,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0010004251807017982,
      "grad_norm": 24.82086753845215,
      "learning_rate": 4.9999986275753206e-05,
      "loss": 6.6438,
      "num_input_tokens_seen": 16320,
      "step": 5
    },
    {
      "epoch": 0.0020008503614035965,
      "grad_norm": 19.489648818969727,
      "learning_rate": 4.99999451030279e-05,
      "loss": 2.22,
      "num_input_tokens_seen": 33088,
      "step": 10
    },
    {
      "epoch": 0.0030012755421053947,
      "grad_norm": 7.6068010330200195,
      "learning_rate": 4.999987648186929e-05,
      "loss": 1.1072,
      "num_input_tokens_seen": 49248,
      "step": 15
    },
    {
      "epoch": 0.004001700722807193,
      "grad_norm": 12.884929656982422,
      "learning_rate": 4.99997804123527e-05,
      "loss": 0.6484,
      "num_input_tokens_seen": 66496,
      "step": 20
    },
    {
      "epoch": 0.005002125903508991,
      "grad_norm": 4.620934009552002,
      "learning_rate": 4.999965689458363e-05,
      "loss": 0.3944,
      "num_input_tokens_seen": 83328,
      "step": 25
    },
    {
      "epoch": 0.0060025510842107895,
      "grad_norm": 5.337153434753418,
      "learning_rate": 4.9999505928697686e-05,
      "loss": 0.3009,
      "num_input_tokens_seen": 100480,
      "step": 30
    },
    {
      "epoch": 0.007002976264912588,
      "grad_norm": 3.612947702407837,
      "learning_rate": 4.9999327514860616e-05,
      "loss": 0.1965,
      "num_input_tokens_seen": 117856,
      "step": 35
    },
    {
      "epoch": 0.008003401445614386,
      "grad_norm": 3.843050956726074,
      "learning_rate": 4.999912165326831e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 134656,
      "step": 40
    },
    {
      "epoch": 0.009003826626316184,
      "grad_norm": 5.87134313583374,
      "learning_rate": 4.999888834414679e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 151168,
      "step": 45
    },
    {
      "epoch": 0.010004251807017982,
      "grad_norm": 2.828153610229492,
      "learning_rate": 4.9998627587752225e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 168800,
      "step": 50
    },
    {
      "epoch": 0.01100467698771978,
      "grad_norm": 3.484650135040283,
      "learning_rate": 4.99983393843709e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 187296,
      "step": 55
    },
    {
      "epoch": 0.012005102168421579,
      "grad_norm": 1.3379253149032593,
      "learning_rate": 4.999802373431924e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 203008,
      "step": 60
    },
    {
      "epoch": 0.013005527349123377,
      "grad_norm": 2.5667777061462402,
      "learning_rate": 4.999768063794383e-05,
      "loss": 0.0907,
      "num_input_tokens_seen": 219168,
      "step": 65
    },
    {
      "epoch": 0.014005952529825175,
      "grad_norm": 3.9828338623046875,
      "learning_rate": 4.999731009562135e-05,
      "loss": 0.1304,
      "num_input_tokens_seen": 235040,
      "step": 70
    },
    {
      "epoch": 0.015006377710526974,
      "grad_norm": 2.727963924407959,
      "learning_rate": 4.999691210775864e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 251616,
      "step": 75
    },
    {
      "epoch": 0.016006802891228772,
      "grad_norm": 4.61973762512207,
      "learning_rate": 4.999648667479266e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 268608,
      "step": 80
    },
    {
      "epoch": 0.017007228071930572,
      "grad_norm": 3.472965717315674,
      "learning_rate": 4.9996033797190524e-05,
      "loss": 0.1014,
      "num_input_tokens_seen": 285952,
      "step": 85
    },
    {
      "epoch": 0.01800765325263237,
      "grad_norm": 1.9901703596115112,
      "learning_rate": 4.999555347544946e-05,
      "loss": 0.1747,
      "num_input_tokens_seen": 302528,
      "step": 90
    },
    {
      "epoch": 0.01900807843333417,
      "grad_norm": 2.2979788780212402,
      "learning_rate": 4.999504571009682e-05,
      "loss": 0.0997,
      "num_input_tokens_seen": 317728,
      "step": 95
    },
    {
      "epoch": 0.020008503614035965,
      "grad_norm": 1.4213311672210693,
      "learning_rate": 4.999451050169011e-05,
      "loss": 0.1037,
      "num_input_tokens_seen": 333152,
      "step": 100
    },
    {
      "epoch": 0.021008928794737765,
      "grad_norm": 1.5144597291946411,
      "learning_rate": 4.999394785081696e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 350592,
      "step": 105
    },
    {
      "epoch": 0.02200935397543956,
      "grad_norm": 1.9958913326263428,
      "learning_rate": 4.9993357758095115e-05,
      "loss": 0.0654,
      "num_input_tokens_seen": 366624,
      "step": 110
    },
    {
      "epoch": 0.02300977915614136,
      "grad_norm": 2.1754813194274902,
      "learning_rate": 4.9992740224172476e-05,
      "loss": 0.081,
      "num_input_tokens_seen": 383296,
      "step": 115
    },
    {
      "epoch": 0.024010204336843158,
      "grad_norm": 1.4228804111480713,
      "learning_rate": 4.999209524972704e-05,
      "loss": 0.068,
      "num_input_tokens_seen": 399008,
      "step": 120
    },
    {
      "epoch": 0.025010629517544958,
      "grad_norm": 1.2373309135437012,
      "learning_rate": 4.9991422835466974e-05,
      "loss": 0.0602,
      "num_input_tokens_seen": 415552,
      "step": 125
    },
    {
      "epoch": 0.026011054698246754,
      "grad_norm": 1.2388474941253662,
      "learning_rate": 4.999072298213053e-05,
      "loss": 0.071,
      "num_input_tokens_seen": 431808,
      "step": 130
    },
    {
      "epoch": 0.027011479878948554,
      "grad_norm": 0.6230462193489075,
      "learning_rate": 4.99899956904861e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 448416,
      "step": 135
    },
    {
      "epoch": 0.02801190505965035,
      "grad_norm": 1.511096477508545,
      "learning_rate": 4.9989240961332226e-05,
      "loss": 0.0745,
      "num_input_tokens_seen": 464512,
      "step": 140
    },
    {
      "epoch": 0.02901233024035215,
      "grad_norm": 1.797107219696045,
      "learning_rate": 4.998845879549754e-05,
      "loss": 0.0943,
      "num_input_tokens_seen": 480448,
      "step": 145
    },
    {
      "epoch": 0.030012755421053947,
      "grad_norm": 1.3896641731262207,
      "learning_rate": 4.998764919384082e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 497600,
      "step": 150
    },
    {
      "epoch": 0.031013180601755747,
      "grad_norm": 1.2604399919509888,
      "learning_rate": 4.9986812157250955e-05,
      "loss": 0.0573,
      "num_input_tokens_seen": 513120,
      "step": 155
    },
    {
      "epoch": 0.032013605782457544,
      "grad_norm": 2.3319883346557617,
      "learning_rate": 4.998594768664697e-05,
      "loss": 0.0862,
      "num_input_tokens_seen": 531648,
      "step": 160
    },
    {
      "epoch": 0.033014030963159344,
      "grad_norm": 1.6746535301208496,
      "learning_rate": 4.9985055782977985e-05,
      "loss": 0.1011,
      "num_input_tokens_seen": 547520,
      "step": 165
    },
    {
      "epoch": 0.034014456143861144,
      "grad_norm": 3.3541808128356934,
      "learning_rate": 4.998413644722327e-05,
      "loss": 0.0811,
      "num_input_tokens_seen": 563328,
      "step": 170
    },
    {
      "epoch": 0.03501488132456294,
      "grad_norm": 3.3083388805389404,
      "learning_rate": 4.998318968039221e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 579872,
      "step": 175
    },
    {
      "epoch": 0.03601530650526474,
      "grad_norm": 1.5279896259307861,
      "learning_rate": 4.9982215483524264e-05,
      "loss": 0.0648,
      "num_input_tokens_seen": 595808,
      "step": 180
    },
    {
      "epoch": 0.03701573168596654,
      "grad_norm": 0.45309773087501526,
      "learning_rate": 4.998121385768907e-05,
      "loss": 0.0563,
      "num_input_tokens_seen": 612288,
      "step": 185
    },
    {
      "epoch": 0.03801615686666834,
      "grad_norm": 1.0328619480133057,
      "learning_rate": 4.998018480398635e-05,
      "loss": 0.0553,
      "num_input_tokens_seen": 628448,
      "step": 190
    },
    {
      "epoch": 0.03901658204737013,
      "grad_norm": 0.4076535701751709,
      "learning_rate": 4.997912832354593e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 645440,
      "step": 195
    },
    {
      "epoch": 0.04001700722807193,
      "grad_norm": 2.606926679611206,
      "learning_rate": 4.997804441752778e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 661056,
      "step": 200
    },
    {
      "epoch": 0.04101743240877373,
      "grad_norm": 3.2455272674560547,
      "learning_rate": 4.997693308712195e-05,
      "loss": 0.0781,
      "num_input_tokens_seen": 677312,
      "step": 205
    },
    {
      "epoch": 0.04201785758947553,
      "grad_norm": 2.260077476501465,
      "learning_rate": 4.997579433354861e-05,
      "loss": 0.0725,
      "num_input_tokens_seen": 694560,
      "step": 210
    },
    {
      "epoch": 0.04301828277017732,
      "grad_norm": 1.78510320186615,
      "learning_rate": 4.997462815805806e-05,
      "loss": 0.0848,
      "num_input_tokens_seen": 711968,
      "step": 215
    },
    {
      "epoch": 0.04401870795087912,
      "grad_norm": 1.5576282739639282,
      "learning_rate": 4.9973434561930674e-05,
      "loss": 0.0829,
      "num_input_tokens_seen": 729664,
      "step": 220
    },
    {
      "epoch": 0.04501913313158092,
      "grad_norm": 0.7414575815200806,
      "learning_rate": 4.997221354647695e-05,
      "loss": 0.0526,
      "num_input_tokens_seen": 746496,
      "step": 225
    },
    {
      "epoch": 0.04601955831228272,
      "grad_norm": 1.7496589422225952,
      "learning_rate": 4.9970965113037506e-05,
      "loss": 0.0671,
      "num_input_tokens_seen": 763200,
      "step": 230
    },
    {
      "epoch": 0.047019983492984516,
      "grad_norm": 1.5983532667160034,
      "learning_rate": 4.9969689262983036e-05,
      "loss": 0.0721,
      "num_input_tokens_seen": 779552,
      "step": 235
    },
    {
      "epoch": 0.048020408673686316,
      "grad_norm": 1.5230525732040405,
      "learning_rate": 4.996838599771433e-05,
      "loss": 0.0822,
      "num_input_tokens_seen": 796096,
      "step": 240
    },
    {
      "epoch": 0.049020833854388116,
      "grad_norm": 2.191336154937744,
      "learning_rate": 4.996705531866232e-05,
      "loss": 0.0525,
      "num_input_tokens_seen": 811904,
      "step": 245
    },
    {
      "epoch": 0.050021259035089916,
      "grad_norm": 0.9807636737823486,
      "learning_rate": 4.9965697227287997e-05,
      "loss": 0.0981,
      "num_input_tokens_seen": 828800,
      "step": 250
    },
    {
      "epoch": 0.05102168421579171,
      "grad_norm": 0.8154246211051941,
      "learning_rate": 4.996431172508248e-05,
      "loss": 0.0715,
      "num_input_tokens_seen": 846656,
      "step": 255
    },
    {
      "epoch": 0.05202210939649351,
      "grad_norm": 0.8523686528205872,
      "learning_rate": 4.996289881356695e-05,
      "loss": 0.0787,
      "num_input_tokens_seen": 862240,
      "step": 260
    },
    {
      "epoch": 0.05302253457719531,
      "grad_norm": 1.0898770093917847,
      "learning_rate": 4.99614584942927e-05,
      "loss": 0.072,
      "num_input_tokens_seen": 878528,
      "step": 265
    },
    {
      "epoch": 0.05402295975789711,
      "grad_norm": 0.9473728537559509,
      "learning_rate": 4.9959990768841115e-05,
      "loss": 0.1188,
      "num_input_tokens_seen": 895456,
      "step": 270
    },
    {
      "epoch": 0.0550233849385989,
      "grad_norm": 0.6116834878921509,
      "learning_rate": 4.9958495638823676e-05,
      "loss": 0.0519,
      "num_input_tokens_seen": 912416,
      "step": 275
    },
    {
      "epoch": 0.0560238101193007,
      "grad_norm": 1.0323799848556519,
      "learning_rate": 4.9956973105881947e-05,
      "loss": 0.0459,
      "num_input_tokens_seen": 929120,
      "step": 280
    },
    {
      "epoch": 0.0570242353000025,
      "grad_norm": 0.6817773580551147,
      "learning_rate": 4.995542317168756e-05,
      "loss": 0.0558,
      "num_input_tokens_seen": 945824,
      "step": 285
    },
    {
      "epoch": 0.0580246604807043,
      "grad_norm": 2.252847909927368,
      "learning_rate": 4.995384583794227e-05,
      "loss": 0.0584,
      "num_input_tokens_seen": 962976,
      "step": 290
    },
    {
      "epoch": 0.059025085661406095,
      "grad_norm": 1.859374761581421,
      "learning_rate": 4.995224110637788e-05,
      "loss": 0.0575,
      "num_input_tokens_seen": 980864,
      "step": 295
    },
    {
      "epoch": 0.060025510842107895,
      "grad_norm": 2.2975878715515137,
      "learning_rate": 4.99506089787563e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 997184,
      "step": 300
    },
    {
      "epoch": 0.061025936022809694,
      "grad_norm": 1.8312894105911255,
      "learning_rate": 4.99489494568695e-05,
      "loss": 0.0619,
      "num_input_tokens_seen": 1014688,
      "step": 305
    },
    {
      "epoch": 0.062026361203511494,
      "grad_norm": 1.4239940643310547,
      "learning_rate": 4.994726254253953e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 1030432,
      "step": 310
    },
    {
      "epoch": 0.06302678638421329,
      "grad_norm": 0.9822717308998108,
      "learning_rate": 4.9945548237618535e-05,
      "loss": 0.0643,
      "num_input_tokens_seen": 1046048,
      "step": 315
    },
    {
      "epoch": 0.06402721156491509,
      "grad_norm": 1.1589330434799194,
      "learning_rate": 4.994380654398871e-05,
      "loss": 0.0598,
      "num_input_tokens_seen": 1062496,
      "step": 320
    },
    {
      "epoch": 0.06502763674561689,
      "grad_norm": 1.6940858364105225,
      "learning_rate": 4.994203746356233e-05,
      "loss": 0.0545,
      "num_input_tokens_seen": 1080000,
      "step": 325
    },
    {
      "epoch": 0.06602806192631869,
      "grad_norm": 1.0039169788360596,
      "learning_rate": 4.994024099828174e-05,
      "loss": 0.0665,
      "num_input_tokens_seen": 1095712,
      "step": 330
    },
    {
      "epoch": 0.06702848710702049,
      "grad_norm": 0.5740472674369812,
      "learning_rate": 4.993841715011934e-05,
      "loss": 0.0706,
      "num_input_tokens_seen": 1113088,
      "step": 335
    },
    {
      "epoch": 0.06802891228772229,
      "grad_norm": 2.5024569034576416,
      "learning_rate": 4.993656592107763e-05,
      "loss": 0.0935,
      "num_input_tokens_seen": 1129952,
      "step": 340
    },
    {
      "epoch": 0.06902933746842407,
      "grad_norm": 1.3382577896118164,
      "learning_rate": 4.9934687313189124e-05,
      "loss": 0.0693,
      "num_input_tokens_seen": 1147104,
      "step": 345
    },
    {
      "epoch": 0.07002976264912587,
      "grad_norm": 1.107261300086975,
      "learning_rate": 4.993278132851643e-05,
      "loss": 0.0647,
      "num_input_tokens_seen": 1163232,
      "step": 350
    },
    {
      "epoch": 0.07103018782982767,
      "grad_norm": 1.4886616468429565,
      "learning_rate": 4.99308479691522e-05,
      "loss": 0.0763,
      "num_input_tokens_seen": 1180160,
      "step": 355
    },
    {
      "epoch": 0.07203061301052947,
      "grad_norm": 1.7975897789001465,
      "learning_rate": 4.992888723721916e-05,
      "loss": 0.0688,
      "num_input_tokens_seen": 1197632,
      "step": 360
    },
    {
      "epoch": 0.07303103819123127,
      "grad_norm": 0.9495922327041626,
      "learning_rate": 4.992689913487006e-05,
      "loss": 0.0537,
      "num_input_tokens_seen": 1214016,
      "step": 365
    },
    {
      "epoch": 0.07403146337193307,
      "grad_norm": 1.2237321138381958,
      "learning_rate": 4.9924883664287724e-05,
      "loss": 0.0775,
      "num_input_tokens_seen": 1230400,
      "step": 370
    },
    {
      "epoch": 0.07503188855263487,
      "grad_norm": 0.574914813041687,
      "learning_rate": 4.992284082768501e-05,
      "loss": 0.0599,
      "num_input_tokens_seen": 1246720,
      "step": 375
    },
    {
      "epoch": 0.07603231373333667,
      "grad_norm": 1.1832525730133057,
      "learning_rate": 4.9920770627304845e-05,
      "loss": 0.0605,
      "num_input_tokens_seen": 1265152,
      "step": 380
    },
    {
      "epoch": 0.07703273891403846,
      "grad_norm": 0.732682466506958,
      "learning_rate": 4.991867306542016e-05,
      "loss": 0.0544,
      "num_input_tokens_seen": 1281952,
      "step": 385
    },
    {
      "epoch": 0.07803316409474026,
      "grad_norm": 1.6287379264831543,
      "learning_rate": 4.9916548144333974e-05,
      "loss": 0.0678,
      "num_input_tokens_seen": 1299456,
      "step": 390
    },
    {
      "epoch": 0.07903358927544206,
      "grad_norm": 1.1713122129440308,
      "learning_rate": 4.991439586637931e-05,
      "loss": 0.0566,
      "num_input_tokens_seen": 1315808,
      "step": 395
    },
    {
      "epoch": 0.08003401445614386,
      "grad_norm": 1.2388888597488403,
      "learning_rate": 4.991221623391925e-05,
      "loss": 0.0779,
      "num_input_tokens_seen": 1332224,
      "step": 400
    },
    {
      "epoch": 0.08103443963684566,
      "grad_norm": 1.4474679231643677,
      "learning_rate": 4.991000924934689e-05,
      "loss": 0.077,
      "num_input_tokens_seen": 1348928,
      "step": 405
    },
    {
      "epoch": 0.08203486481754746,
      "grad_norm": 1.1414148807525635,
      "learning_rate": 4.990777491508536e-05,
      "loss": 0.0774,
      "num_input_tokens_seen": 1365600,
      "step": 410
    },
    {
      "epoch": 0.08303528999824926,
      "grad_norm": 0.7782608270645142,
      "learning_rate": 4.990551323358784e-05,
      "loss": 0.061,
      "num_input_tokens_seen": 1381696,
      "step": 415
    },
    {
      "epoch": 0.08403571517895106,
      "grad_norm": 0.6450750827789307,
      "learning_rate": 4.990322420733751e-05,
      "loss": 0.0484,
      "num_input_tokens_seen": 1398496,
      "step": 420
    },
    {
      "epoch": 0.08503614035965286,
      "grad_norm": 0.9892857670783997,
      "learning_rate": 4.990090783884759e-05,
      "loss": 0.0422,
      "num_input_tokens_seen": 1414368,
      "step": 425
    },
    {
      "epoch": 0.08603656554035465,
      "grad_norm": 1.7115873098373413,
      "learning_rate": 4.989856413066131e-05,
      "loss": 0.0594,
      "num_input_tokens_seen": 1431168,
      "step": 430
    },
    {
      "epoch": 0.08703699072105645,
      "grad_norm": 0.8290610313415527,
      "learning_rate": 4.9896193085351904e-05,
      "loss": 0.0448,
      "num_input_tokens_seen": 1447520,
      "step": 435
    },
    {
      "epoch": 0.08803741590175825,
      "grad_norm": 1.5415170192718506,
      "learning_rate": 4.9893794705522666e-05,
      "loss": 0.0707,
      "num_input_tokens_seen": 1464448,
      "step": 440
    },
    {
      "epoch": 0.08903784108246005,
      "grad_norm": 1.3662270307540894,
      "learning_rate": 4.989136899380686e-05,
      "loss": 0.0374,
      "num_input_tokens_seen": 1481280,
      "step": 445
    },
    {
      "epoch": 0.09003826626316185,
      "grad_norm": 0.9870457649230957,
      "learning_rate": 4.9888915952867766e-05,
      "loss": 0.0422,
      "num_input_tokens_seen": 1497792,
      "step": 450
    },
    {
      "epoch": 0.09103869144386365,
      "grad_norm": 2.193429470062256,
      "learning_rate": 4.988643558539867e-05,
      "loss": 0.0861,
      "num_input_tokens_seen": 1513920,
      "step": 455
    },
    {
      "epoch": 0.09203911662456545,
      "grad_norm": 0.9253290891647339,
      "learning_rate": 4.988392789412288e-05,
      "loss": 0.0626,
      "num_input_tokens_seen": 1530240,
      "step": 460
    },
    {
      "epoch": 0.09303954180526725,
      "grad_norm": 0.678046464920044,
      "learning_rate": 4.98813928817937e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 1547264,
      "step": 465
    },
    {
      "epoch": 0.09403996698596903,
      "grad_norm": 0.8769763708114624,
      "learning_rate": 4.987883055119439e-05,
      "loss": 0.0509,
      "num_input_tokens_seen": 1564288,
      "step": 470
    },
    {
      "epoch": 0.09504039216667083,
      "grad_norm": 0.7749297022819519,
      "learning_rate": 4.987624090513825e-05,
      "loss": 0.0408,
      "num_input_tokens_seen": 1581312,
      "step": 475
    },
    {
      "epoch": 0.09604081734737263,
      "grad_norm": 1.0127102136611938,
      "learning_rate": 4.987362394646856e-05,
      "loss": 0.0818,
      "num_input_tokens_seen": 1598432,
      "step": 480
    },
    {
      "epoch": 0.09704124252807443,
      "grad_norm": 1.330907940864563,
      "learning_rate": 4.987097967805858e-05,
      "loss": 0.0999,
      "num_input_tokens_seen": 1614560,
      "step": 485
    },
    {
      "epoch": 0.09804166770877623,
      "grad_norm": 1.55445396900177,
      "learning_rate": 4.986830810281156e-05,
      "loss": 0.0471,
      "num_input_tokens_seen": 1631136,
      "step": 490
    },
    {
      "epoch": 0.09904209288947803,
      "grad_norm": 0.9867338538169861,
      "learning_rate": 4.9865609223660714e-05,
      "loss": 0.0904,
      "num_input_tokens_seen": 1647872,
      "step": 495
    },
    {
      "epoch": 0.10004251807017983,
      "grad_norm": 1.0111794471740723,
      "learning_rate": 4.986288304356926e-05,
      "loss": 0.061,
      "num_input_tokens_seen": 1664800,
      "step": 500
    },
    {
      "epoch": 0.10104294325088163,
      "grad_norm": 1.2285752296447754,
      "learning_rate": 4.986012956553038e-05,
      "loss": 0.0839,
      "num_input_tokens_seen": 1681056,
      "step": 505
    },
    {
      "epoch": 0.10204336843158342,
      "grad_norm": 0.5017119646072388,
      "learning_rate": 4.985734879256723e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 1696960,
      "step": 510
    },
    {
      "epoch": 0.10304379361228522,
      "grad_norm": 0.6441024541854858,
      "learning_rate": 4.985454072773292e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 1714432,
      "step": 515
    },
    {
      "epoch": 0.10404421879298702,
      "grad_norm": 0.23873034119606018,
      "learning_rate": 4.9851705374110556e-05,
      "loss": 0.0572,
      "num_input_tokens_seen": 1731104,
      "step": 520
    },
    {
      "epoch": 0.10504464397368882,
      "grad_norm": 0.7182571291923523,
      "learning_rate": 4.984884273481316e-05,
      "loss": 0.0638,
      "num_input_tokens_seen": 1747168,
      "step": 525
    },
    {
      "epoch": 0.10604506915439062,
      "grad_norm": 1.063759207725525,
      "learning_rate": 4.984595281298375e-05,
      "loss": 0.0577,
      "num_input_tokens_seen": 1762912,
      "step": 530
    },
    {
      "epoch": 0.10704549433509242,
      "grad_norm": 0.7074857354164124,
      "learning_rate": 4.9843035611795295e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 1779904,
      "step": 535
    },
    {
      "epoch": 0.10804591951579422,
      "grad_norm": 1.392200231552124,
      "learning_rate": 4.98400911344507e-05,
      "loss": 0.0688,
      "num_input_tokens_seen": 1795360,
      "step": 540
    },
    {
      "epoch": 0.10904634469649602,
      "grad_norm": 0.5381610989570618,
      "learning_rate": 4.983711938418281e-05,
      "loss": 0.0209,
      "num_input_tokens_seen": 1812128,
      "step": 545
    },
    {
      "epoch": 0.1100467698771978,
      "grad_norm": 2.0378992557525635,
      "learning_rate": 4.9834120364254436e-05,
      "loss": 0.0653,
      "num_input_tokens_seen": 1828960,
      "step": 550
    },
    {
      "epoch": 0.1110471950578996,
      "grad_norm": 1.9617708921432495,
      "learning_rate": 4.9831094077958326e-05,
      "loss": 0.0744,
      "num_input_tokens_seen": 1845344,
      "step": 555
    },
    {
      "epoch": 0.1120476202386014,
      "grad_norm": 0.4756849706172943,
      "learning_rate": 4.9828040528617156e-05,
      "loss": 0.0814,
      "num_input_tokens_seen": 1861344,
      "step": 560
    },
    {
      "epoch": 0.1130480454193032,
      "grad_norm": 0.37431013584136963,
      "learning_rate": 4.9824959719583545e-05,
      "loss": 0.045,
      "num_input_tokens_seen": 1877696,
      "step": 565
    },
    {
      "epoch": 0.114048470600005,
      "grad_norm": 0.8085149526596069,
      "learning_rate": 4.9821851654240026e-05,
      "loss": 0.0442,
      "num_input_tokens_seen": 1894912,
      "step": 570
    },
    {
      "epoch": 0.1150488957807068,
      "grad_norm": 0.6764988303184509,
      "learning_rate": 4.981871633599908e-05,
      "loss": 0.0417,
      "num_input_tokens_seen": 1911520,
      "step": 575
    },
    {
      "epoch": 0.1160493209614086,
      "grad_norm": 0.9866617918014526,
      "learning_rate": 4.981555376830309e-05,
      "loss": 0.0564,
      "num_input_tokens_seen": 1928160,
      "step": 580
    },
    {
      "epoch": 0.1170497461421104,
      "grad_norm": 0.6540876626968384,
      "learning_rate": 4.9812363954624356e-05,
      "loss": 0.0401,
      "num_input_tokens_seen": 1944096,
      "step": 585
    },
    {
      "epoch": 0.11805017132281219,
      "grad_norm": 0.5941640138626099,
      "learning_rate": 4.980914689846512e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 1959904,
      "step": 590
    },
    {
      "epoch": 0.11905059650351399,
      "grad_norm": 1.2908644676208496,
      "learning_rate": 4.98059026033575e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 1976160,
      "step": 595
    },
    {
      "epoch": 0.12005102168421579,
      "grad_norm": 1.6685086488723755,
      "learning_rate": 4.980263107286355e-05,
      "loss": 0.104,
      "num_input_tokens_seen": 1991648,
      "step": 600
    },
    {
      "epoch": 0.12105144686491759,
      "grad_norm": 1.4214237928390503,
      "learning_rate": 4.97993323105752e-05,
      "loss": 0.0701,
      "num_input_tokens_seen": 2006848,
      "step": 605
    },
    {
      "epoch": 0.12205187204561939,
      "grad_norm": 1.3490134477615356,
      "learning_rate": 4.9796006320114295e-05,
      "loss": 0.0475,
      "num_input_tokens_seen": 2022304,
      "step": 610
    },
    {
      "epoch": 0.12305229722632119,
      "grad_norm": 0.7132759094238281,
      "learning_rate": 4.979265310513258e-05,
      "loss": 0.0437,
      "num_input_tokens_seen": 2038624,
      "step": 615
    },
    {
      "epoch": 0.12405272240702299,
      "grad_norm": 1.148813009262085,
      "learning_rate": 4.978927266931167e-05,
      "loss": 0.0514,
      "num_input_tokens_seen": 2054656,
      "step": 620
    },
    {
      "epoch": 0.1250531475877248,
      "grad_norm": 1.119343876838684,
      "learning_rate": 4.9785865016363096e-05,
      "loss": 0.0478,
      "num_input_tokens_seen": 2070912,
      "step": 625
    },
    {
      "epoch": 0.12605357276842657,
      "grad_norm": 1.2424360513687134,
      "learning_rate": 4.9782430150028246e-05,
      "loss": 0.0646,
      "num_input_tokens_seen": 2086272,
      "step": 630
    },
    {
      "epoch": 0.1270539979491284,
      "grad_norm": 0.9024567008018494,
      "learning_rate": 4.97789680740784e-05,
      "loss": 0.0561,
      "num_input_tokens_seen": 2102464,
      "step": 635
    },
    {
      "epoch": 0.12805442312983017,
      "grad_norm": 0.31885960698127747,
      "learning_rate": 4.977547879231471e-05,
      "loss": 0.0416,
      "num_input_tokens_seen": 2119200,
      "step": 640
    },
    {
      "epoch": 0.129054848310532,
      "grad_norm": 0.7536543011665344,
      "learning_rate": 4.977196230856819e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 2134560,
      "step": 645
    },
    {
      "epoch": 0.13005527349123377,
      "grad_norm": 0.8086920976638794,
      "learning_rate": 4.9768418626699734e-05,
      "loss": 0.0586,
      "num_input_tokens_seen": 2150464,
      "step": 650
    },
    {
      "epoch": 0.13105569867193556,
      "grad_norm": 0.47185018658638,
      "learning_rate": 4.9764847750600086e-05,
      "loss": 0.0331,
      "num_input_tokens_seen": 2167776,
      "step": 655
    },
    {
      "epoch": 0.13205612385263737,
      "grad_norm": 0.4128211438655853,
      "learning_rate": 4.976124968418985e-05,
      "loss": 0.0486,
      "num_input_tokens_seen": 2185792,
      "step": 660
    },
    {
      "epoch": 0.13305654903333916,
      "grad_norm": 1.0512120723724365,
      "learning_rate": 4.975762443141949e-05,
      "loss": 0.0673,
      "num_input_tokens_seen": 2202848,
      "step": 665
    },
    {
      "epoch": 0.13405697421404097,
      "grad_norm": 1.3533042669296265,
      "learning_rate": 4.975397199626933e-05,
      "loss": 0.0771,
      "num_input_tokens_seen": 2219904,
      "step": 670
    },
    {
      "epoch": 0.13505739939474276,
      "grad_norm": 0.9410749673843384,
      "learning_rate": 4.97502923827495e-05,
      "loss": 0.0692,
      "num_input_tokens_seen": 2236128,
      "step": 675
    },
    {
      "epoch": 0.13605782457544457,
      "grad_norm": 0.7121180891990662,
      "learning_rate": 4.9746585594900005e-05,
      "loss": 0.0424,
      "num_input_tokens_seen": 2253216,
      "step": 680
    },
    {
      "epoch": 0.13705824975614636,
      "grad_norm": 0.4436166286468506,
      "learning_rate": 4.9742851636790676e-05,
      "loss": 0.0518,
      "num_input_tokens_seen": 2269120,
      "step": 685
    },
    {
      "epoch": 0.13805867493684815,
      "grad_norm": 0.9069826006889343,
      "learning_rate": 4.973909051252117e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 2286560,
      "step": 690
    },
    {
      "epoch": 0.13905910011754996,
      "grad_norm": 0.47360703349113464,
      "learning_rate": 4.973530222622097e-05,
      "loss": 0.057,
      "num_input_tokens_seen": 2302880,
      "step": 695
    },
    {
      "epoch": 0.14005952529825175,
      "grad_norm": 0.3753754794597626,
      "learning_rate": 4.9731486782049404e-05,
      "loss": 0.0747,
      "num_input_tokens_seen": 2320384,
      "step": 700
    },
    {
      "epoch": 0.14105995047895356,
      "grad_norm": 0.3111303150653839,
      "learning_rate": 4.972764418419558e-05,
      "loss": 0.0506,
      "num_input_tokens_seen": 2336576,
      "step": 705
    },
    {
      "epoch": 0.14206037565965535,
      "grad_norm": 0.8782441020011902,
      "learning_rate": 4.972377443687845e-05,
      "loss": 0.0489,
      "num_input_tokens_seen": 2353152,
      "step": 710
    },
    {
      "epoch": 0.14306080084035716,
      "grad_norm": 0.26745501160621643,
      "learning_rate": 4.971987754434676e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 2370336,
      "step": 715
    },
    {
      "epoch": 0.14406122602105895,
      "grad_norm": 0.4768848121166229,
      "learning_rate": 4.971595351087907e-05,
      "loss": 0.0382,
      "num_input_tokens_seen": 2386496,
      "step": 720
    },
    {
      "epoch": 0.14506165120176076,
      "grad_norm": 1.0148855447769165,
      "learning_rate": 4.971200234078372e-05,
      "loss": 0.0584,
      "num_input_tokens_seen": 2402496,
      "step": 725
    },
    {
      "epoch": 0.14606207638246255,
      "grad_norm": 0.7883719801902771,
      "learning_rate": 4.970802403839886e-05,
      "loss": 0.032,
      "num_input_tokens_seen": 2419008,
      "step": 730
    },
    {
      "epoch": 0.14706250156316433,
      "grad_norm": 1.0178841352462769,
      "learning_rate": 4.970401860809244e-05,
      "loss": 0.0398,
      "num_input_tokens_seen": 2435360,
      "step": 735
    },
    {
      "epoch": 0.14806292674386615,
      "grad_norm": 1.6463992595672607,
      "learning_rate": 4.969998605426216e-05,
      "loss": 0.0649,
      "num_input_tokens_seen": 2451360,
      "step": 740
    },
    {
      "epoch": 0.14906335192456793,
      "grad_norm": 1.079568862915039,
      "learning_rate": 4.969592638133553e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 2467360,
      "step": 745
    },
    {
      "epoch": 0.15006377710526975,
      "grad_norm": 0.28534814715385437,
      "learning_rate": 4.9691839593769836e-05,
      "loss": 0.0699,
      "num_input_tokens_seen": 2483840,
      "step": 750
    },
    {
      "epoch": 0.15106420228597153,
      "grad_norm": 0.7496660351753235,
      "learning_rate": 4.968772569605211e-05,
      "loss": 0.0264,
      "num_input_tokens_seen": 2500640,
      "step": 755
    },
    {
      "epoch": 0.15206462746667335,
      "grad_norm": 1.5654454231262207,
      "learning_rate": 4.968358469269917e-05,
      "loss": 0.0303,
      "num_input_tokens_seen": 2516960,
      "step": 760
    },
    {
      "epoch": 0.15306505264737513,
      "grad_norm": 0.4835239350795746,
      "learning_rate": 4.967941658825759e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 2533152,
      "step": 765
    },
    {
      "epoch": 0.15406547782807692,
      "grad_norm": 0.7468734383583069,
      "learning_rate": 4.96752213873037e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 2549792,
      "step": 770
    },
    {
      "epoch": 0.15506590300877873,
      "grad_norm": 1.111474871635437,
      "learning_rate": 4.9670999094443574e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 2566272,
      "step": 775
    },
    {
      "epoch": 0.15606632818948052,
      "grad_norm": 0.8435876369476318,
      "learning_rate": 4.966674971431302e-05,
      "loss": 0.0585,
      "num_input_tokens_seen": 2583552,
      "step": 780
    },
    {
      "epoch": 0.15706675337018233,
      "grad_norm": 0.9625574350357056,
      "learning_rate": 4.966247325157763e-05,
      "loss": 0.0463,
      "num_input_tokens_seen": 2599648,
      "step": 785
    },
    {
      "epoch": 0.15806717855088412,
      "grad_norm": 0.7700160145759583,
      "learning_rate": 4.9658169710932676e-05,
      "loss": 0.0816,
      "num_input_tokens_seen": 2615712,
      "step": 790
    },
    {
      "epoch": 0.15906760373158593,
      "grad_norm": 0.42034628987312317,
      "learning_rate": 4.965383909710321e-05,
      "loss": 0.0543,
      "num_input_tokens_seen": 2633312,
      "step": 795
    },
    {
      "epoch": 0.16006802891228772,
      "grad_norm": 0.4920070469379425,
      "learning_rate": 4.964948141484397e-05,
      "loss": 0.0604,
      "num_input_tokens_seen": 2650496,
      "step": 800
    },
    {
      "epoch": 0.16106845409298953,
      "grad_norm": 0.43571746349334717,
      "learning_rate": 4.964509666893943e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 2666432,
      "step": 805
    },
    {
      "epoch": 0.16206887927369132,
      "grad_norm": 1.0017355680465698,
      "learning_rate": 4.9640684864203786e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 2683168,
      "step": 810
    },
    {
      "epoch": 0.1630693044543931,
      "grad_norm": 0.2180410772562027,
      "learning_rate": 4.9636246005480916e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 2698784,
      "step": 815
    },
    {
      "epoch": 0.16406972963509492,
      "grad_norm": 0.5853586792945862,
      "learning_rate": 4.963178009764444e-05,
      "loss": 0.0676,
      "num_input_tokens_seen": 2715104,
      "step": 820
    },
    {
      "epoch": 0.1650701548157967,
      "grad_norm": 0.7395941019058228,
      "learning_rate": 4.962728714559765e-05,
      "loss": 0.0539,
      "num_input_tokens_seen": 2731616,
      "step": 825
    },
    {
      "epoch": 0.16607057999649852,
      "grad_norm": 0.8851535320281982,
      "learning_rate": 4.962276715427351e-05,
      "loss": 0.0689,
      "num_input_tokens_seen": 2748448,
      "step": 830
    },
    {
      "epoch": 0.1670710051772003,
      "grad_norm": 0.2044210433959961,
      "learning_rate": 4.961822012863474e-05,
      "loss": 0.0397,
      "num_input_tokens_seen": 2766528,
      "step": 835
    },
    {
      "epoch": 0.16807143035790212,
      "grad_norm": 0.7691633105278015,
      "learning_rate": 4.9613646073673673e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 2783488,
      "step": 840
    },
    {
      "epoch": 0.1690718555386039,
      "grad_norm": 0.8963852524757385,
      "learning_rate": 4.9609044994412356e-05,
      "loss": 0.0359,
      "num_input_tokens_seen": 2799904,
      "step": 845
    },
    {
      "epoch": 0.17007228071930572,
      "grad_norm": 1.6576282978057861,
      "learning_rate": 4.960441689590249e-05,
      "loss": 0.0608,
      "num_input_tokens_seen": 2817376,
      "step": 850
    },
    {
      "epoch": 0.1710727059000075,
      "grad_norm": 0.40670549869537354,
      "learning_rate": 4.959976178322546e-05,
      "loss": 0.0646,
      "num_input_tokens_seen": 2834432,
      "step": 855
    },
    {
      "epoch": 0.1720731310807093,
      "grad_norm": 0.5909866690635681,
      "learning_rate": 4.959507966149229e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 2851040,
      "step": 860
    },
    {
      "epoch": 0.1730735562614111,
      "grad_norm": 1.1720274686813354,
      "learning_rate": 4.9590370535843664e-05,
      "loss": 0.0504,
      "num_input_tokens_seen": 2867072,
      "step": 865
    },
    {
      "epoch": 0.1740739814421129,
      "grad_norm": 0.6624909043312073,
      "learning_rate": 4.9585634411449935e-05,
      "loss": 0.0558,
      "num_input_tokens_seen": 2882944,
      "step": 870
    },
    {
      "epoch": 0.1750744066228147,
      "grad_norm": 0.8819348216056824,
      "learning_rate": 4.958087129351106e-05,
      "loss": 0.0659,
      "num_input_tokens_seen": 2900832,
      "step": 875
    },
    {
      "epoch": 0.1760748318035165,
      "grad_norm": 0.5899621844291687,
      "learning_rate": 4.9576081187256666e-05,
      "loss": 0.0465,
      "num_input_tokens_seen": 2917056,
      "step": 880
    },
    {
      "epoch": 0.1770752569842183,
      "grad_norm": 0.7371861934661865,
      "learning_rate": 4.9571264097946e-05,
      "loss": 0.0401,
      "num_input_tokens_seen": 2934144,
      "step": 885
    },
    {
      "epoch": 0.1780756821649201,
      "grad_norm": 0.2233559936285019,
      "learning_rate": 4.956642003086794e-05,
      "loss": 0.0296,
      "num_input_tokens_seen": 2949984,
      "step": 890
    },
    {
      "epoch": 0.17907610734562188,
      "grad_norm": 1.363047480583191,
      "learning_rate": 4.956154899134098e-05,
      "loss": 0.0571,
      "num_input_tokens_seen": 2966176,
      "step": 895
    },
    {
      "epoch": 0.1800765325263237,
      "grad_norm": 0.9176114201545715,
      "learning_rate": 4.955665098471323e-05,
      "loss": 0.0433,
      "num_input_tokens_seen": 2982144,
      "step": 900
    },
    {
      "epoch": 0.18107695770702548,
      "grad_norm": 0.9925788640975952,
      "learning_rate": 4.9551726016362386e-05,
      "loss": 0.0587,
      "num_input_tokens_seen": 2998688,
      "step": 905
    },
    {
      "epoch": 0.1820773828877273,
      "grad_norm": 0.8520196676254272,
      "learning_rate": 4.954677409169579e-05,
      "loss": 0.044,
      "num_input_tokens_seen": 3015360,
      "step": 910
    },
    {
      "epoch": 0.18307780806842908,
      "grad_norm": 0.9778660535812378,
      "learning_rate": 4.954179521615035e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 3032416,
      "step": 915
    },
    {
      "epoch": 0.1840782332491309,
      "grad_norm": 0.682713508605957,
      "learning_rate": 4.953678939519256e-05,
      "loss": 0.0689,
      "num_input_tokens_seen": 3048608,
      "step": 920
    },
    {
      "epoch": 0.18507865842983268,
      "grad_norm": 0.3935362994670868,
      "learning_rate": 4.9531756634318524e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 3065568,
      "step": 925
    },
    {
      "epoch": 0.1860790836105345,
      "grad_norm": 0.8297384977340698,
      "learning_rate": 4.952669693905391e-05,
      "loss": 0.07,
      "num_input_tokens_seen": 3083264,
      "step": 930
    },
    {
      "epoch": 0.18707950879123628,
      "grad_norm": 0.9590235948562622,
      "learning_rate": 4.952161031495395e-05,
      "loss": 0.0725,
      "num_input_tokens_seen": 3099296,
      "step": 935
    },
    {
      "epoch": 0.18807993397193806,
      "grad_norm": 0.3106594383716583,
      "learning_rate": 4.951649676760346e-05,
      "loss": 0.032,
      "num_input_tokens_seen": 3116000,
      "step": 940
    },
    {
      "epoch": 0.18908035915263988,
      "grad_norm": 0.5347669124603271,
      "learning_rate": 4.951135630261679e-05,
      "loss": 0.0584,
      "num_input_tokens_seen": 3132544,
      "step": 945
    },
    {
      "epoch": 0.19008078433334166,
      "grad_norm": 0.9589521288871765,
      "learning_rate": 4.950618892563789e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 3149472,
      "step": 950
    },
    {
      "epoch": 0.19108120951404348,
      "grad_norm": 1.2461116313934326,
      "learning_rate": 4.9500994642340195e-05,
      "loss": 0.0605,
      "num_input_tokens_seen": 3165152,
      "step": 955
    },
    {
      "epoch": 0.19208163469474526,
      "grad_norm": 0.48353829979896545,
      "learning_rate": 4.949577345842674e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 3181568,
      "step": 960
    },
    {
      "epoch": 0.19308205987544708,
      "grad_norm": 0.8187334537506104,
      "learning_rate": 4.949052537963006e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 3198336,
      "step": 965
    },
    {
      "epoch": 0.19408248505614886,
      "grad_norm": 0.49673548340797424,
      "learning_rate": 4.9485250411712224e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 3215392,
      "step": 970
    },
    {
      "epoch": 0.19508291023685065,
      "grad_norm": 0.5972616076469421,
      "learning_rate": 4.947994856046484e-05,
      "loss": 0.07,
      "num_input_tokens_seen": 3232576,
      "step": 975
    },
    {
      "epoch": 0.19608333541755246,
      "grad_norm": 0.7993170619010925,
      "learning_rate": 4.947461983170902e-05,
      "loss": 0.0381,
      "num_input_tokens_seen": 3248320,
      "step": 980
    },
    {
      "epoch": 0.19708376059825425,
      "grad_norm": 0.7358092665672302,
      "learning_rate": 4.9469264231295376e-05,
      "loss": 0.0422,
      "num_input_tokens_seen": 3266432,
      "step": 985
    },
    {
      "epoch": 0.19808418577895606,
      "grad_norm": 0.4396766722202301,
      "learning_rate": 4.9463881765104045e-05,
      "loss": 0.0448,
      "num_input_tokens_seen": 3283168,
      "step": 990
    },
    {
      "epoch": 0.19908461095965785,
      "grad_norm": 0.2774893343448639,
      "learning_rate": 4.945847243904466e-05,
      "loss": 0.0579,
      "num_input_tokens_seen": 3301280,
      "step": 995
    },
    {
      "epoch": 0.20008503614035966,
      "grad_norm": 0.6029003858566284,
      "learning_rate": 4.945303625905631e-05,
      "loss": 0.0329,
      "num_input_tokens_seen": 3317536,
      "step": 1000
    },
    {
      "epoch": 0.20108546132106145,
      "grad_norm": 1.284956932067871,
      "learning_rate": 4.944757323110761e-05,
      "loss": 0.06,
      "num_input_tokens_seen": 3334336,
      "step": 1005
    },
    {
      "epoch": 0.20208588650176326,
      "grad_norm": 0.3467988073825836,
      "learning_rate": 4.9442083361196627e-05,
      "loss": 0.0432,
      "num_input_tokens_seen": 3351616,
      "step": 1010
    },
    {
      "epoch": 0.20308631168246505,
      "grad_norm": 0.5883722901344299,
      "learning_rate": 4.9436566655350917e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 3367840,
      "step": 1015
    },
    {
      "epoch": 0.20408673686316683,
      "grad_norm": 0.6289622783660889,
      "learning_rate": 4.9431023119627484e-05,
      "loss": 0.0511,
      "num_input_tokens_seen": 3384128,
      "step": 1020
    },
    {
      "epoch": 0.20508716204386865,
      "grad_norm": 0.5500460267066956,
      "learning_rate": 4.94254527601128e-05,
      "loss": 0.0319,
      "num_input_tokens_seen": 3400000,
      "step": 1025
    },
    {
      "epoch": 0.20608758722457043,
      "grad_norm": 0.6038098335266113,
      "learning_rate": 4.941985558292278e-05,
      "loss": 0.0427,
      "num_input_tokens_seen": 3417504,
      "step": 1030
    },
    {
      "epoch": 0.20708801240527225,
      "grad_norm": 0.3944091498851776,
      "learning_rate": 4.9414231594202784e-05,
      "loss": 0.0407,
      "num_input_tokens_seen": 3434080,
      "step": 1035
    },
    {
      "epoch": 0.20808843758597403,
      "grad_norm": 0.7011579275131226,
      "learning_rate": 4.940858080012763e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 3450336,
      "step": 1040
    },
    {
      "epoch": 0.20908886276667585,
      "grad_norm": 0.8541505336761475,
      "learning_rate": 4.940290320690153e-05,
      "loss": 0.0663,
      "num_input_tokens_seen": 3468064,
      "step": 1045
    },
    {
      "epoch": 0.21008928794737763,
      "grad_norm": 0.6061210036277771,
      "learning_rate": 4.9397198820758144e-05,
      "loss": 0.0371,
      "num_input_tokens_seen": 3484384,
      "step": 1050
    },
    {
      "epoch": 0.21108971312807942,
      "grad_norm": 0.5961213707923889,
      "learning_rate": 4.939146764796055e-05,
      "loss": 0.022,
      "num_input_tokens_seen": 3502240,
      "step": 1055
    },
    {
      "epoch": 0.21209013830878123,
      "grad_norm": 0.48619797825813293,
      "learning_rate": 4.9385709694801227e-05,
      "loss": 0.0513,
      "num_input_tokens_seen": 3518784,
      "step": 1060
    },
    {
      "epoch": 0.21309056348948302,
      "grad_norm": 0.4226716160774231,
      "learning_rate": 4.937992496760207e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 3535360,
      "step": 1065
    },
    {
      "epoch": 0.21409098867018483,
      "grad_norm": 0.11164263635873795,
      "learning_rate": 4.937411347271434e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 3550816,
      "step": 1070
    },
    {
      "epoch": 0.21509141385088662,
      "grad_norm": 0.3861413300037384,
      "learning_rate": 4.9368275216518716e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 3568128,
      "step": 1075
    },
    {
      "epoch": 0.21609183903158843,
      "grad_norm": 0.9424998760223389,
      "learning_rate": 4.936241020542526e-05,
      "loss": 0.0394,
      "num_input_tokens_seen": 3585344,
      "step": 1080
    },
    {
      "epoch": 0.21709226421229022,
      "grad_norm": 0.4098639488220215,
      "learning_rate": 4.9356518445873404e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 3601472,
      "step": 1085
    },
    {
      "epoch": 0.21809268939299203,
      "grad_norm": 0.5707319378852844,
      "learning_rate": 4.935059994433192e-05,
      "loss": 0.0547,
      "num_input_tokens_seen": 3619584,
      "step": 1090
    },
    {
      "epoch": 0.21909311457369382,
      "grad_norm": 0.36102646589279175,
      "learning_rate": 4.9344654707298996e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 3636576,
      "step": 1095
    },
    {
      "epoch": 0.2200935397543956,
      "grad_norm": 0.8916822075843811,
      "learning_rate": 4.9338682741302134e-05,
      "loss": 0.0506,
      "num_input_tokens_seen": 3653440,
      "step": 1100
    },
    {
      "epoch": 0.22109396493509742,
      "grad_norm": 0.6885157823562622,
      "learning_rate": 4.933268405289818e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 3671072,
      "step": 1105
    },
    {
      "epoch": 0.2220943901157992,
      "grad_norm": 0.5270304083824158,
      "learning_rate": 4.932665864867335e-05,
      "loss": 0.0615,
      "num_input_tokens_seen": 3687968,
      "step": 1110
    },
    {
      "epoch": 0.22309481529650102,
      "grad_norm": 0.16785143315792084,
      "learning_rate": 4.932060653524316e-05,
      "loss": 0.0527,
      "num_input_tokens_seen": 3704768,
      "step": 1115
    },
    {
      "epoch": 0.2240952404772028,
      "grad_norm": 0.6057844758033752,
      "learning_rate": 4.931452771925248e-05,
      "loss": 0.0554,
      "num_input_tokens_seen": 3720512,
      "step": 1120
    },
    {
      "epoch": 0.22509566565790462,
      "grad_norm": 0.23519811034202576,
      "learning_rate": 4.9308422207375475e-05,
      "loss": 0.0406,
      "num_input_tokens_seen": 3738176,
      "step": 1125
    },
    {
      "epoch": 0.2260960908386064,
      "grad_norm": 1.2040951251983643,
      "learning_rate": 4.930229000631563e-05,
      "loss": 0.0585,
      "num_input_tokens_seen": 3754592,
      "step": 1130
    },
    {
      "epoch": 0.2270965160193082,
      "grad_norm": 0.15998463332653046,
      "learning_rate": 4.929613112280573e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 3770976,
      "step": 1135
    },
    {
      "epoch": 0.22809694120001,
      "grad_norm": 0.4066956341266632,
      "learning_rate": 4.928994556360787e-05,
      "loss": 0.0552,
      "num_input_tokens_seen": 3787872,
      "step": 1140
    },
    {
      "epoch": 0.2290973663807118,
      "grad_norm": 0.967108964920044,
      "learning_rate": 4.9283733335513404e-05,
      "loss": 0.0601,
      "num_input_tokens_seen": 3804416,
      "step": 1145
    },
    {
      "epoch": 0.2300977915614136,
      "grad_norm": 0.3622339963912964,
      "learning_rate": 4.9277494445342995e-05,
      "loss": 0.0451,
      "num_input_tokens_seen": 3820672,
      "step": 1150
    },
    {
      "epoch": 0.2310982167421154,
      "grad_norm": 0.39043447375297546,
      "learning_rate": 4.927122889994657e-05,
      "loss": 0.0348,
      "num_input_tokens_seen": 3836960,
      "step": 1155
    },
    {
      "epoch": 0.2320986419228172,
      "grad_norm": 0.3957688510417938,
      "learning_rate": 4.926493670620331e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 3854112,
      "step": 1160
    },
    {
      "epoch": 0.233099067103519,
      "grad_norm": 0.8970000147819519,
      "learning_rate": 4.9258617871021675e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 3870752,
      "step": 1165
    },
    {
      "epoch": 0.2340994922842208,
      "grad_norm": 0.2305895835161209,
      "learning_rate": 4.925227240133936e-05,
      "loss": 0.0372,
      "num_input_tokens_seen": 3887168,
      "step": 1170
    },
    {
      "epoch": 0.2350999174649226,
      "grad_norm": 0.4444639980792999,
      "learning_rate": 4.924590030412331e-05,
      "loss": 0.034,
      "num_input_tokens_seen": 3903104,
      "step": 1175
    },
    {
      "epoch": 0.23610034264562438,
      "grad_norm": 0.38624247908592224,
      "learning_rate": 4.92395015863697e-05,
      "loss": 0.0674,
      "num_input_tokens_seen": 3919232,
      "step": 1180
    },
    {
      "epoch": 0.2371007678263262,
      "grad_norm": 0.9376976490020752,
      "learning_rate": 4.923307625510395e-05,
      "loss": 0.069,
      "num_input_tokens_seen": 3935872,
      "step": 1185
    },
    {
      "epoch": 0.23810119300702798,
      "grad_norm": 0.168256938457489,
      "learning_rate": 4.922662431738066e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 3952896,
      "step": 1190
    },
    {
      "epoch": 0.2391016181877298,
      "grad_norm": 0.3504430651664734,
      "learning_rate": 4.92201457802837e-05,
      "loss": 0.0462,
      "num_input_tokens_seen": 3970400,
      "step": 1195
    },
    {
      "epoch": 0.24010204336843158,
      "grad_norm": 0.051882412284612656,
      "learning_rate": 4.921364065092609e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 3986880,
      "step": 1200
    },
    {
      "epoch": 0.2411024685491334,
      "grad_norm": 0.6538947820663452,
      "learning_rate": 4.920710893645008e-05,
      "loss": 0.032,
      "num_input_tokens_seen": 4002752,
      "step": 1205
    },
    {
      "epoch": 0.24210289372983518,
      "grad_norm": 0.5337681174278259,
      "learning_rate": 4.9200550644027096e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 4019744,
      "step": 1210
    },
    {
      "epoch": 0.243103318910537,
      "grad_norm": 0.4791374206542969,
      "learning_rate": 4.9193965780857754e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 4035968,
      "step": 1215
    },
    {
      "epoch": 0.24410374409123878,
      "grad_norm": 1.150654911994934,
      "learning_rate": 4.9187354354171835e-05,
      "loss": 0.0907,
      "num_input_tokens_seen": 4053472,
      "step": 1220
    },
    {
      "epoch": 0.24510416927194056,
      "grad_norm": 0.6459364891052246,
      "learning_rate": 4.9180716371228275e-05,
      "loss": 0.0526,
      "num_input_tokens_seen": 4070208,
      "step": 1225
    },
    {
      "epoch": 0.24610459445264238,
      "grad_norm": 0.7645431160926819,
      "learning_rate": 4.9174051839315196e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 4085728,
      "step": 1230
    },
    {
      "epoch": 0.24710501963334416,
      "grad_norm": 0.3979432284832001,
      "learning_rate": 4.916736076574984e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 4102368,
      "step": 1235
    },
    {
      "epoch": 0.24810544481404598,
      "grad_norm": 0.6675825715065002,
      "learning_rate": 4.9160643157878616e-05,
      "loss": 0.0395,
      "num_input_tokens_seen": 4118912,
      "step": 1240
    },
    {
      "epoch": 0.24910586999474776,
      "grad_norm": 0.6228586435317993,
      "learning_rate": 4.915389902307704e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 4136192,
      "step": 1245
    },
    {
      "epoch": 0.2501062951754496,
      "grad_norm": 0.48848792910575867,
      "learning_rate": 4.914712836874977e-05,
      "loss": 0.0375,
      "num_input_tokens_seen": 4153024,
      "step": 1250
    },
    {
      "epoch": 0.2511067203561514,
      "grad_norm": 0.7998842597007751,
      "learning_rate": 4.914033120233057e-05,
      "loss": 0.0433,
      "num_input_tokens_seen": 4169984,
      "step": 1255
    },
    {
      "epoch": 0.25210714553685315,
      "grad_norm": 0.8730126023292542,
      "learning_rate": 4.913350753128233e-05,
      "loss": 0.0702,
      "num_input_tokens_seen": 4187168,
      "step": 1260
    },
    {
      "epoch": 0.25310757071755496,
      "grad_norm": 0.9048335552215576,
      "learning_rate": 4.912665736309703e-05,
      "loss": 0.0475,
      "num_input_tokens_seen": 4205120,
      "step": 1265
    },
    {
      "epoch": 0.2541079958982568,
      "grad_norm": 0.4183575212955475,
      "learning_rate": 4.9119780705295735e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 4221184,
      "step": 1270
    },
    {
      "epoch": 0.25510842107895854,
      "grad_norm": 0.2847769260406494,
      "learning_rate": 4.91128775654286e-05,
      "loss": 0.0421,
      "num_input_tokens_seen": 4236928,
      "step": 1275
    },
    {
      "epoch": 0.25610884625966035,
      "grad_norm": 0.45041540265083313,
      "learning_rate": 4.9105947951074874e-05,
      "loss": 0.0422,
      "num_input_tokens_seen": 4254752,
      "step": 1280
    },
    {
      "epoch": 0.25710927144036216,
      "grad_norm": 0.6893653273582458,
      "learning_rate": 4.909899186984283e-05,
      "loss": 0.0594,
      "num_input_tokens_seen": 4271648,
      "step": 1285
    },
    {
      "epoch": 0.258109696621064,
      "grad_norm": 0.876491904258728,
      "learning_rate": 4.9092009329369834e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 4288224,
      "step": 1290
    },
    {
      "epoch": 0.25911012180176574,
      "grad_norm": 0.7342787981033325,
      "learning_rate": 4.908500033732231e-05,
      "loss": 0.0511,
      "num_input_tokens_seen": 4304512,
      "step": 1295
    },
    {
      "epoch": 0.26011054698246755,
      "grad_norm": 0.6910397410392761,
      "learning_rate": 4.9077964901395686e-05,
      "loss": 0.0735,
      "num_input_tokens_seen": 4320992,
      "step": 1300
    },
    {
      "epoch": 0.26111097216316936,
      "grad_norm": 0.5218247771263123,
      "learning_rate": 4.907090302931446e-05,
      "loss": 0.0532,
      "num_input_tokens_seen": 4336160,
      "step": 1305
    },
    {
      "epoch": 0.2621113973438711,
      "grad_norm": 0.5280895233154297,
      "learning_rate": 4.906381472883215e-05,
      "loss": 0.0516,
      "num_input_tokens_seen": 4352640,
      "step": 1310
    },
    {
      "epoch": 0.26311182252457294,
      "grad_norm": 0.47305095195770264,
      "learning_rate": 4.905670000773126e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 4369312,
      "step": 1315
    },
    {
      "epoch": 0.26411224770527475,
      "grad_norm": 0.8043111562728882,
      "learning_rate": 4.904955887382335e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 4384512,
      "step": 1320
    },
    {
      "epoch": 0.26511267288597656,
      "grad_norm": 0.705629289150238,
      "learning_rate": 4.904239133494893e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 4401056,
      "step": 1325
    },
    {
      "epoch": 0.2661130980666783,
      "grad_norm": 0.7742530107498169,
      "learning_rate": 4.903519739897755e-05,
      "loss": 0.0341,
      "num_input_tokens_seen": 4417440,
      "step": 1330
    },
    {
      "epoch": 0.26711352324738014,
      "grad_norm": 0.48411840200424194,
      "learning_rate": 4.902797707380771e-05,
      "loss": 0.033,
      "num_input_tokens_seen": 4434080,
      "step": 1335
    },
    {
      "epoch": 0.26811394842808195,
      "grad_norm": 1.2817881107330322,
      "learning_rate": 4.9020730367366875e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 4450592,
      "step": 1340
    },
    {
      "epoch": 0.2691143736087837,
      "grad_norm": 0.5104709267616272,
      "learning_rate": 4.901345728761151e-05,
      "loss": 0.0342,
      "num_input_tokens_seen": 4466496,
      "step": 1345
    },
    {
      "epoch": 0.2701147987894855,
      "grad_norm": 0.7885850071907043,
      "learning_rate": 4.9006157842527024e-05,
      "loss": 0.0485,
      "num_input_tokens_seen": 4483584,
      "step": 1350
    },
    {
      "epoch": 0.27111522397018734,
      "grad_norm": 0.8310132026672363,
      "learning_rate": 4.899883204012775e-05,
      "loss": 0.0605,
      "num_input_tokens_seen": 4498976,
      "step": 1355
    },
    {
      "epoch": 0.27211564915088915,
      "grad_norm": 0.43888920545578003,
      "learning_rate": 4.899147988845698e-05,
      "loss": 0.034,
      "num_input_tokens_seen": 4515072,
      "step": 1360
    },
    {
      "epoch": 0.2731160743315909,
      "grad_norm": 0.47633183002471924,
      "learning_rate": 4.898410139558695e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 4532160,
      "step": 1365
    },
    {
      "epoch": 0.2741164995122927,
      "grad_norm": 0.5134224891662598,
      "learning_rate": 4.8976696569618785e-05,
      "loss": 0.0312,
      "num_input_tokens_seen": 4548672,
      "step": 1370
    },
    {
      "epoch": 0.27511692469299454,
      "grad_norm": 0.9848766326904297,
      "learning_rate": 4.896926541868254e-05,
      "loss": 0.0432,
      "num_input_tokens_seen": 4565728,
      "step": 1375
    },
    {
      "epoch": 0.2761173498736963,
      "grad_norm": 0.6101632118225098,
      "learning_rate": 4.896180795093718e-05,
      "loss": 0.0403,
      "num_input_tokens_seen": 4583680,
      "step": 1380
    },
    {
      "epoch": 0.2771177750543981,
      "grad_norm": 0.3657839596271515,
      "learning_rate": 4.895432417457054e-05,
      "loss": 0.0331,
      "num_input_tokens_seen": 4600096,
      "step": 1385
    },
    {
      "epoch": 0.2781182002350999,
      "grad_norm": 0.634440541267395,
      "learning_rate": 4.8946814097799375e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 4618208,
      "step": 1390
    },
    {
      "epoch": 0.27911862541580174,
      "grad_norm": 0.4174501299858093,
      "learning_rate": 4.893927772886928e-05,
      "loss": 0.0529,
      "num_input_tokens_seen": 4634720,
      "step": 1395
    },
    {
      "epoch": 0.2801190505965035,
      "grad_norm": 1.0442984104156494,
      "learning_rate": 4.893171507605474e-05,
      "loss": 0.0493,
      "num_input_tokens_seen": 4650688,
      "step": 1400
    },
    {
      "epoch": 0.2811194757772053,
      "grad_norm": 0.3786543309688568,
      "learning_rate": 4.8924126147659085e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 4666912,
      "step": 1405
    },
    {
      "epoch": 0.2821199009579071,
      "grad_norm": 0.8404293656349182,
      "learning_rate": 4.8916510952014515e-05,
      "loss": 0.0476,
      "num_input_tokens_seen": 4682688,
      "step": 1410
    },
    {
      "epoch": 0.28312032613860894,
      "grad_norm": 1.00434410572052,
      "learning_rate": 4.890886949748205e-05,
      "loss": 0.0612,
      "num_input_tokens_seen": 4699168,
      "step": 1415
    },
    {
      "epoch": 0.2841207513193107,
      "grad_norm": 0.49716895818710327,
      "learning_rate": 4.890120179245155e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 4715488,
      "step": 1420
    },
    {
      "epoch": 0.2851211765000125,
      "grad_norm": 0.8475269079208374,
      "learning_rate": 4.889350784534168e-05,
      "loss": 0.0558,
      "num_input_tokens_seen": 4731456,
      "step": 1425
    },
    {
      "epoch": 0.2861216016807143,
      "grad_norm": 1.1672545671463013,
      "learning_rate": 4.888578766459994e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 4748224,
      "step": 1430
    },
    {
      "epoch": 0.2871220268614161,
      "grad_norm": 0.5872476100921631,
      "learning_rate": 4.887804125870262e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 4765280,
      "step": 1435
    },
    {
      "epoch": 0.2881224520421179,
      "grad_norm": 0.16453994810581207,
      "learning_rate": 4.887026863615482e-05,
      "loss": 0.0446,
      "num_input_tokens_seen": 4781760,
      "step": 1440
    },
    {
      "epoch": 0.2891228772228197,
      "grad_norm": 0.47434720396995544,
      "learning_rate": 4.8862469805490396e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 4798432,
      "step": 1445
    },
    {
      "epoch": 0.2901233024035215,
      "grad_norm": 0.9123254418373108,
      "learning_rate": 4.8854644775271996e-05,
      "loss": 0.026,
      "num_input_tokens_seen": 4815808,
      "step": 1450
    },
    {
      "epoch": 0.2911237275842233,
      "grad_norm": 0.26198312640190125,
      "learning_rate": 4.884679355409103e-05,
      "loss": 0.0598,
      "num_input_tokens_seen": 4831328,
      "step": 1455
    },
    {
      "epoch": 0.2921241527649251,
      "grad_norm": 0.6132438778877258,
      "learning_rate": 4.883891615056768e-05,
      "loss": 0.0673,
      "num_input_tokens_seen": 4847616,
      "step": 1460
    },
    {
      "epoch": 0.2931245779456269,
      "grad_norm": 1.4866887331008911,
      "learning_rate": 4.883101257335084e-05,
      "loss": 0.0445,
      "num_input_tokens_seen": 4863872,
      "step": 1465
    },
    {
      "epoch": 0.29412500312632867,
      "grad_norm": 0.7224521040916443,
      "learning_rate": 4.882308283111818e-05,
      "loss": 0.0547,
      "num_input_tokens_seen": 4881280,
      "step": 1470
    },
    {
      "epoch": 0.2951254283070305,
      "grad_norm": 1.2841455936431885,
      "learning_rate": 4.881512693257607e-05,
      "loss": 0.0731,
      "num_input_tokens_seen": 4898432,
      "step": 1475
    },
    {
      "epoch": 0.2961258534877323,
      "grad_norm": 0.356650710105896,
      "learning_rate": 4.88071448864596e-05,
      "loss": 0.0431,
      "num_input_tokens_seen": 4915744,
      "step": 1480
    },
    {
      "epoch": 0.2971262786684341,
      "grad_norm": 0.39753177762031555,
      "learning_rate": 4.879913670153259e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 4932800,
      "step": 1485
    },
    {
      "epoch": 0.29812670384913587,
      "grad_norm": 1.4896060228347778,
      "learning_rate": 4.879110238658754e-05,
      "loss": 0.0437,
      "num_input_tokens_seen": 4949664,
      "step": 1490
    },
    {
      "epoch": 0.2991271290298377,
      "grad_norm": 0.6999287605285645,
      "learning_rate": 4.8783041950445644e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 4966432,
      "step": 1495
    },
    {
      "epoch": 0.3001275542105395,
      "grad_norm": 0.699177086353302,
      "learning_rate": 4.877495540195677e-05,
      "loss": 0.0673,
      "num_input_tokens_seen": 4983392,
      "step": 1500
    },
    {
      "epoch": 0.30112797939124125,
      "grad_norm": 0.6445897221565247,
      "learning_rate": 4.8766842749999465e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 5000736,
      "step": 1505
    },
    {
      "epoch": 0.30212840457194307,
      "grad_norm": 0.2621614634990692,
      "learning_rate": 4.875870400348093e-05,
      "loss": 0.0631,
      "num_input_tokens_seen": 5017696,
      "step": 1510
    },
    {
      "epoch": 0.3031288297526449,
      "grad_norm": 0.7124674320220947,
      "learning_rate": 4.875053917133702e-05,
      "loss": 0.0576,
      "num_input_tokens_seen": 5034496,
      "step": 1515
    },
    {
      "epoch": 0.3041292549333467,
      "grad_norm": 0.4209171533584595,
      "learning_rate": 4.874234826253223e-05,
      "loss": 0.0398,
      "num_input_tokens_seen": 5051968,
      "step": 1520
    },
    {
      "epoch": 0.30512968011404845,
      "grad_norm": 0.6975089311599731,
      "learning_rate": 4.873413128605968e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 5067776,
      "step": 1525
    },
    {
      "epoch": 0.30613010529475027,
      "grad_norm": 0.41621458530426025,
      "learning_rate": 4.872588825094112e-05,
      "loss": 0.0465,
      "num_input_tokens_seen": 5083680,
      "step": 1530
    },
    {
      "epoch": 0.3071305304754521,
      "grad_norm": 1.3077008724212646,
      "learning_rate": 4.8717619166226904e-05,
      "loss": 0.0611,
      "num_input_tokens_seen": 5099744,
      "step": 1535
    },
    {
      "epoch": 0.30813095565615384,
      "grad_norm": 0.12024356424808502,
      "learning_rate": 4.8709324040995985e-05,
      "loss": 0.0527,
      "num_input_tokens_seen": 5117632,
      "step": 1540
    },
    {
      "epoch": 0.30913138083685565,
      "grad_norm": 0.7145063281059265,
      "learning_rate": 4.8701002884355915e-05,
      "loss": 0.046,
      "num_input_tokens_seen": 5134592,
      "step": 1545
    },
    {
      "epoch": 0.31013180601755747,
      "grad_norm": 0.8199822902679443,
      "learning_rate": 4.869265570544282e-05,
      "loss": 0.0561,
      "num_input_tokens_seen": 5151648,
      "step": 1550
    },
    {
      "epoch": 0.3111322311982593,
      "grad_norm": 0.5546838641166687,
      "learning_rate": 4.86842825134214e-05,
      "loss": 0.0398,
      "num_input_tokens_seen": 5169440,
      "step": 1555
    },
    {
      "epoch": 0.31213265637896104,
      "grad_norm": 1.3934036493301392,
      "learning_rate": 4.8675883317484916e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 5185984,
      "step": 1560
    },
    {
      "epoch": 0.31313308155966285,
      "grad_norm": 0.3828260004520416,
      "learning_rate": 4.8667458126855184e-05,
      "loss": 0.0371,
      "num_input_tokens_seen": 5202784,
      "step": 1565
    },
    {
      "epoch": 0.31413350674036467,
      "grad_norm": 0.6127660274505615,
      "learning_rate": 4.865900695078255e-05,
      "loss": 0.0428,
      "num_input_tokens_seen": 5219776,
      "step": 1570
    },
    {
      "epoch": 0.3151339319210665,
      "grad_norm": 0.2966005504131317,
      "learning_rate": 4.86505297985459e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 5235904,
      "step": 1575
    },
    {
      "epoch": 0.31613435710176824,
      "grad_norm": 0.5187402963638306,
      "learning_rate": 4.8642026679452626e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 5253216,
      "step": 1580
    },
    {
      "epoch": 0.31713478228247005,
      "grad_norm": 1.12813401222229,
      "learning_rate": 4.8633497602838654e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 5269632,
      "step": 1585
    },
    {
      "epoch": 0.31813520746317187,
      "grad_norm": 0.7938510775566101,
      "learning_rate": 4.8624942578068386e-05,
      "loss": 0.0525,
      "num_input_tokens_seen": 5287296,
      "step": 1590
    },
    {
      "epoch": 0.3191356326438736,
      "grad_norm": 0.5008676648139954,
      "learning_rate": 4.861636161453473e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 5304928,
      "step": 1595
    },
    {
      "epoch": 0.32013605782457544,
      "grad_norm": 0.5503880381584167,
      "learning_rate": 4.860775472165906e-05,
      "loss": 0.0429,
      "num_input_tokens_seen": 5320960,
      "step": 1600
    },
    {
      "epoch": 0.32113648300527725,
      "grad_norm": 0.8359606862068176,
      "learning_rate": 4.859912190889123e-05,
      "loss": 0.0564,
      "num_input_tokens_seen": 5337536,
      "step": 1605
    },
    {
      "epoch": 0.32213690818597907,
      "grad_norm": 0.3948289752006531,
      "learning_rate": 4.8590463185709555e-05,
      "loss": 0.0284,
      "num_input_tokens_seen": 5354272,
      "step": 1610
    },
    {
      "epoch": 0.3231373333666808,
      "grad_norm": 0.5302601456642151,
      "learning_rate": 4.858177856162078e-05,
      "loss": 0.0538,
      "num_input_tokens_seen": 5371968,
      "step": 1615
    },
    {
      "epoch": 0.32413775854738264,
      "grad_norm": 1.3221774101257324,
      "learning_rate": 4.857306804616011e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 5387936,
      "step": 1620
    },
    {
      "epoch": 0.32513818372808445,
      "grad_norm": 1.5478042364120483,
      "learning_rate": 4.856433164889116e-05,
      "loss": 0.0556,
      "num_input_tokens_seen": 5404800,
      "step": 1625
    },
    {
      "epoch": 0.3261386089087862,
      "grad_norm": 0.3014490604400635,
      "learning_rate": 4.855556937940596e-05,
      "loss": 0.0394,
      "num_input_tokens_seen": 5420992,
      "step": 1630
    },
    {
      "epoch": 0.327139034089488,
      "grad_norm": 0.424380362033844,
      "learning_rate": 4.854678124732496e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 5436704,
      "step": 1635
    },
    {
      "epoch": 0.32813945927018984,
      "grad_norm": 0.38589712977409363,
      "learning_rate": 4.853796726229701e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 5453472,
      "step": 1640
    },
    {
      "epoch": 0.32913988445089165,
      "grad_norm": 0.35621973872184753,
      "learning_rate": 4.852912743399932e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 5470656,
      "step": 1645
    },
    {
      "epoch": 0.3301403096315934,
      "grad_norm": 0.6708896160125732,
      "learning_rate": 4.85202617721375e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 5486752,
      "step": 1650
    },
    {
      "epoch": 0.3311407348122952,
      "grad_norm": 1.6848129034042358,
      "learning_rate": 4.85113702864455e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 5502464,
      "step": 1655
    },
    {
      "epoch": 0.33214115999299704,
      "grad_norm": 0.3120448887348175,
      "learning_rate": 4.850245298668564e-05,
      "loss": 0.0341,
      "num_input_tokens_seen": 5519680,
      "step": 1660
    },
    {
      "epoch": 0.3331415851736988,
      "grad_norm": 1.8794240951538086,
      "learning_rate": 4.849350988264859e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 5535200,
      "step": 1665
    },
    {
      "epoch": 0.3341420103544006,
      "grad_norm": 0.48631739616394043,
      "learning_rate": 4.848454098415333e-05,
      "loss": 0.0297,
      "num_input_tokens_seen": 5553664,
      "step": 1670
    },
    {
      "epoch": 0.3351424355351024,
      "grad_norm": 0.2804921567440033,
      "learning_rate": 4.847554630104716e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 5570304,
      "step": 1675
    },
    {
      "epoch": 0.33614286071580424,
      "grad_norm": 0.3838098347187042,
      "learning_rate": 4.846652584320571e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 5586624,
      "step": 1680
    },
    {
      "epoch": 0.337143285896506,
      "grad_norm": 0.7481986284255981,
      "learning_rate": 4.8457479620532906e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 5603424,
      "step": 1685
    },
    {
      "epoch": 0.3381437110772078,
      "grad_norm": 0.2822953760623932,
      "learning_rate": 4.8448407642960946e-05,
      "loss": 0.03,
      "num_input_tokens_seen": 5619040,
      "step": 1690
    },
    {
      "epoch": 0.3391441362579096,
      "grad_norm": 0.313073992729187,
      "learning_rate": 4.8439309920450314e-05,
      "loss": 0.0632,
      "num_input_tokens_seen": 5636512,
      "step": 1695
    },
    {
      "epoch": 0.34014456143861144,
      "grad_norm": 0.4042295217514038,
      "learning_rate": 4.8430186462989765e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 5653408,
      "step": 1700
    },
    {
      "epoch": 0.3411449866193132,
      "grad_norm": 0.6126199960708618,
      "learning_rate": 4.8421037280596304e-05,
      "loss": 0.0275,
      "num_input_tokens_seen": 5670752,
      "step": 1705
    },
    {
      "epoch": 0.342145411800015,
      "grad_norm": 0.6158226728439331,
      "learning_rate": 4.8411862383315185e-05,
      "loss": 0.0315,
      "num_input_tokens_seen": 5686752,
      "step": 1710
    },
    {
      "epoch": 0.3431458369807168,
      "grad_norm": 0.19262298941612244,
      "learning_rate": 4.840266178121989e-05,
      "loss": 0.0419,
      "num_input_tokens_seen": 5702880,
      "step": 1715
    },
    {
      "epoch": 0.3441462621614186,
      "grad_norm": 0.6169284582138062,
      "learning_rate": 4.839343548441213e-05,
      "loss": 0.0423,
      "num_input_tokens_seen": 5720384,
      "step": 1720
    },
    {
      "epoch": 0.3451466873421204,
      "grad_norm": 0.5302843451499939,
      "learning_rate": 4.8384183503021815e-05,
      "loss": 0.0549,
      "num_input_tokens_seen": 5735808,
      "step": 1725
    },
    {
      "epoch": 0.3461471125228222,
      "grad_norm": 0.3886967897415161,
      "learning_rate": 4.837490584720707e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 5752544,
      "step": 1730
    },
    {
      "epoch": 0.347147537703524,
      "grad_norm": 0.5098423957824707,
      "learning_rate": 4.836560252715419e-05,
      "loss": 0.0405,
      "num_input_tokens_seen": 5769312,
      "step": 1735
    },
    {
      "epoch": 0.3481479628842258,
      "grad_norm": 0.8524506688117981,
      "learning_rate": 4.835627355307767e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 5785152,
      "step": 1740
    },
    {
      "epoch": 0.3491483880649276,
      "grad_norm": 1.0980675220489502,
      "learning_rate": 4.834691893522016e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 5802848,
      "step": 1745
    },
    {
      "epoch": 0.3501488132456294,
      "grad_norm": 0.59861159324646,
      "learning_rate": 4.833753868385247e-05,
      "loss": 0.0317,
      "num_input_tokens_seen": 5818816,
      "step": 1750
    },
    {
      "epoch": 0.35114923842633117,
      "grad_norm": 0.36550024151802063,
      "learning_rate": 4.832813280927355e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 5835648,
      "step": 1755
    },
    {
      "epoch": 0.352149663607033,
      "grad_norm": 0.2625948488712311,
      "learning_rate": 4.8318701321810475e-05,
      "loss": 0.0443,
      "num_input_tokens_seen": 5852192,
      "step": 1760
    },
    {
      "epoch": 0.3531500887877348,
      "grad_norm": 0.7051465511322021,
      "learning_rate": 4.830924423181846e-05,
      "loss": 0.0431,
      "num_input_tokens_seen": 5869184,
      "step": 1765
    },
    {
      "epoch": 0.3541505139684366,
      "grad_norm": 0.3694879412651062,
      "learning_rate": 4.829976154968081e-05,
      "loss": 0.0525,
      "num_input_tokens_seen": 5885856,
      "step": 1770
    },
    {
      "epoch": 0.35515093914913837,
      "grad_norm": 0.4274010956287384,
      "learning_rate": 4.829025328580895e-05,
      "loss": 0.045,
      "num_input_tokens_seen": 5902560,
      "step": 1775
    },
    {
      "epoch": 0.3561513643298402,
      "grad_norm": 0.6589645147323608,
      "learning_rate": 4.828071945064238e-05,
      "loss": 0.0434,
      "num_input_tokens_seen": 5920576,
      "step": 1780
    },
    {
      "epoch": 0.357151789510542,
      "grad_norm": 0.632914662361145,
      "learning_rate": 4.8271160054648665e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 5939456,
      "step": 1785
    },
    {
      "epoch": 0.35815221469124375,
      "grad_norm": 0.7634440660476685,
      "learning_rate": 4.8261575108323456e-05,
      "loss": 0.0541,
      "num_input_tokens_seen": 5955744,
      "step": 1790
    },
    {
      "epoch": 0.35915263987194557,
      "grad_norm": 0.7146453857421875,
      "learning_rate": 4.8251964622190436e-05,
      "loss": 0.0549,
      "num_input_tokens_seen": 5971488,
      "step": 1795
    },
    {
      "epoch": 0.3601530650526474,
      "grad_norm": 0.6739281415939331,
      "learning_rate": 4.824232860680136e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 5988160,
      "step": 1800
    },
    {
      "epoch": 0.3611534902333492,
      "grad_norm": 0.37205588817596436,
      "learning_rate": 4.8232667072735966e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 6004576,
      "step": 1805
    },
    {
      "epoch": 0.36215391541405095,
      "grad_norm": 0.5811402797698975,
      "learning_rate": 4.8222980030602054e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 6020992,
      "step": 1810
    },
    {
      "epoch": 0.36315434059475277,
      "grad_norm": 0.19677260518074036,
      "learning_rate": 4.8213267491035406e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 6039168,
      "step": 1815
    },
    {
      "epoch": 0.3641547657754546,
      "grad_norm": 0.8598901033401489,
      "learning_rate": 4.820352946469982e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 6055488,
      "step": 1820
    },
    {
      "epoch": 0.36515519095615634,
      "grad_norm": 0.5264309644699097,
      "learning_rate": 4.819376596228703e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 6072544,
      "step": 1825
    },
    {
      "epoch": 0.36615561613685815,
      "grad_norm": 0.5197092294692993,
      "learning_rate": 4.81839769945168e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 6087808,
      "step": 1830
    },
    {
      "epoch": 0.36715604131755997,
      "grad_norm": 0.36588457226753235,
      "learning_rate": 4.817416257213681e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 6104320,
      "step": 1835
    },
    {
      "epoch": 0.3681564664982618,
      "grad_norm": 0.8020911812782288,
      "learning_rate": 4.8164322705922736e-05,
      "loss": 0.0289,
      "num_input_tokens_seen": 6120960,
      "step": 1840
    },
    {
      "epoch": 0.36915689167896354,
      "grad_norm": 0.32438233494758606,
      "learning_rate": 4.815445740667812e-05,
      "loss": 0.0443,
      "num_input_tokens_seen": 6138432,
      "step": 1845
    },
    {
      "epoch": 0.37015731685966535,
      "grad_norm": 0.5802657008171082,
      "learning_rate": 4.814456668523448e-05,
      "loss": 0.0549,
      "num_input_tokens_seen": 6154496,
      "step": 1850
    },
    {
      "epoch": 0.37115774204036717,
      "grad_norm": 0.11963453888893127,
      "learning_rate": 4.813465055245124e-05,
      "loss": 0.0433,
      "num_input_tokens_seen": 6172352,
      "step": 1855
    },
    {
      "epoch": 0.372158167221069,
      "grad_norm": 0.34996163845062256,
      "learning_rate": 4.8124709019215716e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 6188000,
      "step": 1860
    },
    {
      "epoch": 0.37315859240177074,
      "grad_norm": 0.24354368448257446,
      "learning_rate": 4.81147420964431e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 6203712,
      "step": 1865
    },
    {
      "epoch": 0.37415901758247255,
      "grad_norm": 0.5403783321380615,
      "learning_rate": 4.810474979507648e-05,
      "loss": 0.0487,
      "num_input_tokens_seen": 6219968,
      "step": 1870
    },
    {
      "epoch": 0.37515944276317437,
      "grad_norm": 0.46348753571510315,
      "learning_rate": 4.80947321260868e-05,
      "loss": 0.0342,
      "num_input_tokens_seen": 6237184,
      "step": 1875
    },
    {
      "epoch": 0.3761598679438761,
      "grad_norm": 1.2938811779022217,
      "learning_rate": 4.808468910047286e-05,
      "loss": 0.0432,
      "num_input_tokens_seen": 6252800,
      "step": 1880
    },
    {
      "epoch": 0.37716029312457794,
      "grad_norm": 0.25433051586151123,
      "learning_rate": 4.8074620729261287e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 6269344,
      "step": 1885
    },
    {
      "epoch": 0.37816071830527975,
      "grad_norm": 0.9118472933769226,
      "learning_rate": 4.806452702350656e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 6286592,
      "step": 1890
    },
    {
      "epoch": 0.37916114348598157,
      "grad_norm": 0.7628657817840576,
      "learning_rate": 4.805440799429095e-05,
      "loss": 0.0428,
      "num_input_tokens_seen": 6302336,
      "step": 1895
    },
    {
      "epoch": 0.3801615686666833,
      "grad_norm": 0.45024359226226807,
      "learning_rate": 4.804426365272455e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 6318816,
      "step": 1900
    },
    {
      "epoch": 0.38116199384738514,
      "grad_norm": 1.1999708414077759,
      "learning_rate": 4.8034094009945215e-05,
      "loss": 0.0451,
      "num_input_tokens_seen": 6335008,
      "step": 1905
    },
    {
      "epoch": 0.38216241902808695,
      "grad_norm": 0.4844655394554138,
      "learning_rate": 4.802389907711863e-05,
      "loss": 0.0685,
      "num_input_tokens_seen": 6351904,
      "step": 1910
    },
    {
      "epoch": 0.3831628442087887,
      "grad_norm": 0.32273247838020325,
      "learning_rate": 4.801367886543819e-05,
      "loss": 0.0402,
      "num_input_tokens_seen": 6367680,
      "step": 1915
    },
    {
      "epoch": 0.3841632693894905,
      "grad_norm": 0.2264421135187149,
      "learning_rate": 4.80034333861251e-05,
      "loss": 0.0514,
      "num_input_tokens_seen": 6384768,
      "step": 1920
    },
    {
      "epoch": 0.38516369457019234,
      "grad_norm": 0.3693816661834717,
      "learning_rate": 4.799316265042825e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 6400928,
      "step": 1925
    },
    {
      "epoch": 0.38616411975089415,
      "grad_norm": 0.19210654497146606,
      "learning_rate": 4.798286666962431e-05,
      "loss": 0.0415,
      "num_input_tokens_seen": 6417984,
      "step": 1930
    },
    {
      "epoch": 0.3871645449315959,
      "grad_norm": 0.9294410943984985,
      "learning_rate": 4.797254545501763e-05,
      "loss": 0.0506,
      "num_input_tokens_seen": 6434432,
      "step": 1935
    },
    {
      "epoch": 0.3881649701122977,
      "grad_norm": 0.18243177235126495,
      "learning_rate": 4.7962199017940283e-05,
      "loss": 0.0712,
      "num_input_tokens_seen": 6450624,
      "step": 1940
    },
    {
      "epoch": 0.38916539529299954,
      "grad_norm": 0.40890705585479736,
      "learning_rate": 4.795182736975205e-05,
      "loss": 0.0394,
      "num_input_tokens_seen": 6466272,
      "step": 1945
    },
    {
      "epoch": 0.3901658204737013,
      "grad_norm": 0.22906385362148285,
      "learning_rate": 4.794143052184037e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 6482784,
      "step": 1950
    },
    {
      "epoch": 0.3911662456544031,
      "grad_norm": 0.27830296754837036,
      "learning_rate": 4.793100848562034e-05,
      "loss": 0.0245,
      "num_input_tokens_seen": 6500160,
      "step": 1955
    },
    {
      "epoch": 0.3921666708351049,
      "grad_norm": 0.32560041546821594,
      "learning_rate": 4.7920561272534745e-05,
      "loss": 0.0445,
      "num_input_tokens_seen": 6517312,
      "step": 1960
    },
    {
      "epoch": 0.39316709601580674,
      "grad_norm": 0.34138888120651245,
      "learning_rate": 4.7910088894053983e-05,
      "loss": 0.0602,
      "num_input_tokens_seen": 6533760,
      "step": 1965
    },
    {
      "epoch": 0.3941675211965085,
      "grad_norm": 0.45417314767837524,
      "learning_rate": 4.789959136167611e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 6550752,
      "step": 1970
    },
    {
      "epoch": 0.3951679463772103,
      "grad_norm": 0.6495351791381836,
      "learning_rate": 4.788906868692677e-05,
      "loss": 0.0194,
      "num_input_tokens_seen": 6567328,
      "step": 1975
    },
    {
      "epoch": 0.3961683715579121,
      "grad_norm": 0.8971573114395142,
      "learning_rate": 4.787852088135923e-05,
      "loss": 0.039,
      "num_input_tokens_seen": 6584096,
      "step": 1980
    },
    {
      "epoch": 0.39716879673861394,
      "grad_norm": 0.6822695732116699,
      "learning_rate": 4.7867947956554346e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 6600736,
      "step": 1985
    },
    {
      "epoch": 0.3981692219193157,
      "grad_norm": 0.8708227872848511,
      "learning_rate": 4.7857349924120556e-05,
      "loss": 0.0649,
      "num_input_tokens_seen": 6616896,
      "step": 1990
    },
    {
      "epoch": 0.3991696471000175,
      "grad_norm": 0.4138895273208618,
      "learning_rate": 4.7846726795693855e-05,
      "loss": 0.0444,
      "num_input_tokens_seen": 6633472,
      "step": 1995
    },
    {
      "epoch": 0.4001700722807193,
      "grad_norm": 0.2843296229839325,
      "learning_rate": 4.78360785829378e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 6650496,
      "step": 2000
    },
    {
      "epoch": 0.4011704974614211,
      "grad_norm": 0.4477090537548065,
      "learning_rate": 4.782540529754348e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 6666528,
      "step": 2005
    },
    {
      "epoch": 0.4021709226421229,
      "grad_norm": 0.31738659739494324,
      "learning_rate": 4.781470695122954e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 6683136,
      "step": 2010
    },
    {
      "epoch": 0.4031713478228247,
      "grad_norm": 0.3024028241634369,
      "learning_rate": 4.7803983555742096e-05,
      "loss": 0.0272,
      "num_input_tokens_seen": 6700128,
      "step": 2015
    },
    {
      "epoch": 0.4041717730035265,
      "grad_norm": 0.5160415172576904,
      "learning_rate": 4.77932351228548e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 6716608,
      "step": 2020
    },
    {
      "epoch": 0.4051721981842283,
      "grad_norm": 0.6938416361808777,
      "learning_rate": 4.7782461664368786e-05,
      "loss": 0.0464,
      "num_input_tokens_seen": 6733312,
      "step": 2025
    },
    {
      "epoch": 0.4061726233649301,
      "grad_norm": 0.5055555701255798,
      "learning_rate": 4.777166319211266e-05,
      "loss": 0.0489,
      "num_input_tokens_seen": 6748160,
      "step": 2030
    },
    {
      "epoch": 0.4071730485456319,
      "grad_norm": 0.6181210279464722,
      "learning_rate": 4.7760839717942495e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 6765088,
      "step": 2035
    },
    {
      "epoch": 0.40817347372633367,
      "grad_norm": 0.4572533667087555,
      "learning_rate": 4.774999125374181e-05,
      "loss": 0.0659,
      "num_input_tokens_seen": 6781760,
      "step": 2040
    },
    {
      "epoch": 0.4091738989070355,
      "grad_norm": 0.6048961877822876,
      "learning_rate": 4.7739117811421566e-05,
      "loss": 0.0367,
      "num_input_tokens_seen": 6798848,
      "step": 2045
    },
    {
      "epoch": 0.4101743240877373,
      "grad_norm": 0.15084992349147797,
      "learning_rate": 4.7728219402920156e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 6815264,
      "step": 2050
    },
    {
      "epoch": 0.4111747492684391,
      "grad_norm": 0.36724817752838135,
      "learning_rate": 4.771729604020336e-05,
      "loss": 0.0495,
      "num_input_tokens_seen": 6830816,
      "step": 2055
    },
    {
      "epoch": 0.41217517444914087,
      "grad_norm": 0.23971237242221832,
      "learning_rate": 4.7706347735264385e-05,
      "loss": 0.0271,
      "num_input_tokens_seen": 6847072,
      "step": 2060
    },
    {
      "epoch": 0.4131755996298427,
      "grad_norm": 0.7865434885025024,
      "learning_rate": 4.76953745001238e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 6863072,
      "step": 2065
    },
    {
      "epoch": 0.4141760248105445,
      "grad_norm": 0.28162121772766113,
      "learning_rate": 4.768437634682957e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 6880288,
      "step": 2070
    },
    {
      "epoch": 0.41517644999124625,
      "grad_norm": 0.6597485542297363,
      "learning_rate": 4.767335328745699e-05,
      "loss": 0.0639,
      "num_input_tokens_seen": 6896928,
      "step": 2075
    },
    {
      "epoch": 0.41617687517194807,
      "grad_norm": 0.0439864918589592,
      "learning_rate": 4.766230533410872e-05,
      "loss": 0.0387,
      "num_input_tokens_seen": 6912736,
      "step": 2080
    },
    {
      "epoch": 0.4171773003526499,
      "grad_norm": 0.547297477722168,
      "learning_rate": 4.7651232498914755e-05,
      "loss": 0.0766,
      "num_input_tokens_seen": 6929312,
      "step": 2085
    },
    {
      "epoch": 0.4181777255333517,
      "grad_norm": 0.5888490676879883,
      "learning_rate": 4.7640134794032386e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 6945888,
      "step": 2090
    },
    {
      "epoch": 0.41917815071405345,
      "grad_norm": 0.6623382568359375,
      "learning_rate": 4.762901223164624e-05,
      "loss": 0.0408,
      "num_input_tokens_seen": 6964000,
      "step": 2095
    },
    {
      "epoch": 0.42017857589475527,
      "grad_norm": 0.33597245812416077,
      "learning_rate": 4.7617864823968205e-05,
      "loss": 0.0318,
      "num_input_tokens_seen": 6980672,
      "step": 2100
    },
    {
      "epoch": 0.4211790010754571,
      "grad_norm": 0.5683799982070923,
      "learning_rate": 4.7606692583237476e-05,
      "loss": 0.035,
      "num_input_tokens_seen": 6998688,
      "step": 2105
    },
    {
      "epoch": 0.42217942625615884,
      "grad_norm": 1.111972689628601,
      "learning_rate": 4.759549552172049e-05,
      "loss": 0.0634,
      "num_input_tokens_seen": 7015104,
      "step": 2110
    },
    {
      "epoch": 0.42317985143686065,
      "grad_norm": 0.43278875946998596,
      "learning_rate": 4.7584273651710955e-05,
      "loss": 0.0396,
      "num_input_tokens_seen": 7032224,
      "step": 2115
    },
    {
      "epoch": 0.42418027661756247,
      "grad_norm": 0.49375009536743164,
      "learning_rate": 4.75730269855298e-05,
      "loss": 0.0433,
      "num_input_tokens_seen": 7048192,
      "step": 2120
    },
    {
      "epoch": 0.4251807017982643,
      "grad_norm": 0.6873200535774231,
      "learning_rate": 4.7561755535525196e-05,
      "loss": 0.043,
      "num_input_tokens_seen": 7064416,
      "step": 2125
    },
    {
      "epoch": 0.42618112697896604,
      "grad_norm": 0.6794641613960266,
      "learning_rate": 4.7550459314072505e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 7080320,
      "step": 2130
    },
    {
      "epoch": 0.42718155215966785,
      "grad_norm": 0.6741289496421814,
      "learning_rate": 4.75391383335743e-05,
      "loss": 0.0339,
      "num_input_tokens_seen": 7097120,
      "step": 2135
    },
    {
      "epoch": 0.42818197734036967,
      "grad_norm": 1.288682460784912,
      "learning_rate": 4.7527792606460344e-05,
      "loss": 0.0538,
      "num_input_tokens_seen": 7114848,
      "step": 2140
    },
    {
      "epoch": 0.4291824025210715,
      "grad_norm": 0.36019641160964966,
      "learning_rate": 4.751642214518756e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 7131008,
      "step": 2145
    },
    {
      "epoch": 0.43018282770177324,
      "grad_norm": 0.5206799507141113,
      "learning_rate": 4.750502696224002e-05,
      "loss": 0.0383,
      "num_input_tokens_seen": 7148160,
      "step": 2150
    },
    {
      "epoch": 0.43118325288247505,
      "grad_norm": 0.5463341474533081,
      "learning_rate": 4.749360707012895e-05,
      "loss": 0.0585,
      "num_input_tokens_seen": 7164288,
      "step": 2155
    },
    {
      "epoch": 0.43218367806317687,
      "grad_norm": 0.43360671401023865,
      "learning_rate": 4.7482162481392704e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 7181632,
      "step": 2160
    },
    {
      "epoch": 0.4331841032438786,
      "grad_norm": 0.2629452049732208,
      "learning_rate": 4.7470693208596764e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 7198112,
      "step": 2165
    },
    {
      "epoch": 0.43418452842458044,
      "grad_norm": 0.801326334476471,
      "learning_rate": 4.745919926433368e-05,
      "loss": 0.0458,
      "num_input_tokens_seen": 7213696,
      "step": 2170
    },
    {
      "epoch": 0.43518495360528225,
      "grad_norm": 0.30609095096588135,
      "learning_rate": 4.7447680661223126e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 7229920,
      "step": 2175
    },
    {
      "epoch": 0.43618537878598407,
      "grad_norm": 0.5648137331008911,
      "learning_rate": 4.743613741191183e-05,
      "loss": 0.0497,
      "num_input_tokens_seen": 7245472,
      "step": 2180
    },
    {
      "epoch": 0.4371858039666858,
      "grad_norm": 0.9186530709266663,
      "learning_rate": 4.742456952907358e-05,
      "loss": 0.0508,
      "num_input_tokens_seen": 7261088,
      "step": 2185
    },
    {
      "epoch": 0.43818622914738764,
      "grad_norm": 0.7290270328521729,
      "learning_rate": 4.7412977025409225e-05,
      "loss": 0.0526,
      "num_input_tokens_seen": 7276160,
      "step": 2190
    },
    {
      "epoch": 0.43918665432808945,
      "grad_norm": 0.6139854788780212,
      "learning_rate": 4.740135991364662e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 7293600,
      "step": 2195
    },
    {
      "epoch": 0.4401870795087912,
      "grad_norm": 0.17772167921066284,
      "learning_rate": 4.738971820654066e-05,
      "loss": 0.0341,
      "num_input_tokens_seen": 7310304,
      "step": 2200
    },
    {
      "epoch": 0.441187504689493,
      "grad_norm": 0.6025674343109131,
      "learning_rate": 4.737805191687323e-05,
      "loss": 0.0473,
      "num_input_tokens_seen": 7327072,
      "step": 2205
    },
    {
      "epoch": 0.44218792987019484,
      "grad_norm": 0.23449702560901642,
      "learning_rate": 4.7366361057453236e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 7344064,
      "step": 2210
    },
    {
      "epoch": 0.44318835505089665,
      "grad_norm": 0.6478351950645447,
      "learning_rate": 4.7354645641116526e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 7360096,
      "step": 2215
    },
    {
      "epoch": 0.4441887802315984,
      "grad_norm": 0.8951166272163391,
      "learning_rate": 4.7342905680725906e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 7376704,
      "step": 2220
    },
    {
      "epoch": 0.4451892054123002,
      "grad_norm": 0.7574548721313477,
      "learning_rate": 4.733114118917117e-05,
      "loss": 0.0548,
      "num_input_tokens_seen": 7394304,
      "step": 2225
    },
    {
      "epoch": 0.44618963059300204,
      "grad_norm": 0.2258099615573883,
      "learning_rate": 4.7319352179369e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 7411456,
      "step": 2230
    },
    {
      "epoch": 0.4471900557737038,
      "grad_norm": 0.49726584553718567,
      "learning_rate": 4.730753866426303e-05,
      "loss": 0.0257,
      "num_input_tokens_seen": 7428512,
      "step": 2235
    },
    {
      "epoch": 0.4481904809544056,
      "grad_norm": 0.6473379135131836,
      "learning_rate": 4.729570065682378e-05,
      "loss": 0.0392,
      "num_input_tokens_seen": 7445408,
      "step": 2240
    },
    {
      "epoch": 0.4491909061351074,
      "grad_norm": 0.7364275455474854,
      "learning_rate": 4.7283838170048674e-05,
      "loss": 0.033,
      "num_input_tokens_seen": 7462464,
      "step": 2245
    },
    {
      "epoch": 0.45019133131580924,
      "grad_norm": 0.2830129861831665,
      "learning_rate": 4.727195121696202e-05,
      "loss": 0.046,
      "num_input_tokens_seen": 7479392,
      "step": 2250
    },
    {
      "epoch": 0.451191756496511,
      "grad_norm": 0.3266063332557678,
      "learning_rate": 4.7260039810614954e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 7495424,
      "step": 2255
    },
    {
      "epoch": 0.4521921816772128,
      "grad_norm": 0.38082149624824524,
      "learning_rate": 4.724810396408549e-05,
      "loss": 0.0245,
      "num_input_tokens_seen": 7512448,
      "step": 2260
    },
    {
      "epoch": 0.4531926068579146,
      "grad_norm": 1.0159626007080078,
      "learning_rate": 4.723614369047847e-05,
      "loss": 0.0455,
      "num_input_tokens_seen": 7528640,
      "step": 2265
    },
    {
      "epoch": 0.4541930320386164,
      "grad_norm": 0.5355516076087952,
      "learning_rate": 4.722415900292555e-05,
      "loss": 0.042,
      "num_input_tokens_seen": 7544576,
      "step": 2270
    },
    {
      "epoch": 0.4551934572193182,
      "grad_norm": 0.17202123999595642,
      "learning_rate": 4.721214991458521e-05,
      "loss": 0.0529,
      "num_input_tokens_seen": 7561664,
      "step": 2275
    },
    {
      "epoch": 0.45619388240002,
      "grad_norm": 0.4654875099658966,
      "learning_rate": 4.7200116438642686e-05,
      "loss": 0.0495,
      "num_input_tokens_seen": 7579136,
      "step": 2280
    },
    {
      "epoch": 0.4571943075807218,
      "grad_norm": 1.3036562204360962,
      "learning_rate": 4.718805858831002e-05,
      "loss": 0.039,
      "num_input_tokens_seen": 7597056,
      "step": 2285
    },
    {
      "epoch": 0.4581947327614236,
      "grad_norm": 0.6863433718681335,
      "learning_rate": 4.7175976376826e-05,
      "loss": 0.0385,
      "num_input_tokens_seen": 7615296,
      "step": 2290
    },
    {
      "epoch": 0.4591951579421254,
      "grad_norm": 0.6734871864318848,
      "learning_rate": 4.7163869817456176e-05,
      "loss": 0.0464,
      "num_input_tokens_seen": 7632064,
      "step": 2295
    },
    {
      "epoch": 0.4601955831228272,
      "grad_norm": 0.3591286242008209,
      "learning_rate": 4.7151738923492804e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 7648224,
      "step": 2300
    },
    {
      "epoch": 0.461196008303529,
      "grad_norm": 0.5754177570343018,
      "learning_rate": 4.713958370825489e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 7664576,
      "step": 2305
    },
    {
      "epoch": 0.4621964334842308,
      "grad_norm": 0.21992503106594086,
      "learning_rate": 4.712740418508812e-05,
      "loss": 0.0412,
      "num_input_tokens_seen": 7681504,
      "step": 2310
    },
    {
      "epoch": 0.4631968586649326,
      "grad_norm": 0.712186336517334,
      "learning_rate": 4.711520036736488e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 7698656,
      "step": 2315
    },
    {
      "epoch": 0.4641972838456344,
      "grad_norm": 0.35100919008255005,
      "learning_rate": 4.7102972268484226e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 7716448,
      "step": 2320
    },
    {
      "epoch": 0.46519770902633617,
      "grad_norm": 0.6114473938941956,
      "learning_rate": 4.709071990187187e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 7732192,
      "step": 2325
    },
    {
      "epoch": 0.466198134207038,
      "grad_norm": 0.28752997517585754,
      "learning_rate": 4.707844328098018e-05,
      "loss": 0.0311,
      "num_input_tokens_seen": 7747904,
      "step": 2330
    },
    {
      "epoch": 0.4671985593877398,
      "grad_norm": 0.757440447807312,
      "learning_rate": 4.706614241928813e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 7764928,
      "step": 2335
    },
    {
      "epoch": 0.4681989845684416,
      "grad_norm": 1.7856647968292236,
      "learning_rate": 4.7053817330301345e-05,
      "loss": 0.0484,
      "num_input_tokens_seen": 7781248,
      "step": 2340
    },
    {
      "epoch": 0.46919940974914337,
      "grad_norm": 0.4046674966812134,
      "learning_rate": 4.7041468027552016e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 7796864,
      "step": 2345
    },
    {
      "epoch": 0.4701998349298452,
      "grad_norm": 0.8157446384429932,
      "learning_rate": 4.7029094524598936e-05,
      "loss": 0.0429,
      "num_input_tokens_seen": 7813888,
      "step": 2350
    },
    {
      "epoch": 0.471200260110547,
      "grad_norm": 0.7077159881591797,
      "learning_rate": 4.701669683502747e-05,
      "loss": 0.0748,
      "num_input_tokens_seen": 7829440,
      "step": 2355
    },
    {
      "epoch": 0.47220068529124876,
      "grad_norm": 0.5294864177703857,
      "learning_rate": 4.700427497244954e-05,
      "loss": 0.0427,
      "num_input_tokens_seen": 7847584,
      "step": 2360
    },
    {
      "epoch": 0.47320111047195057,
      "grad_norm": 0.44839006662368774,
      "learning_rate": 4.699182895050359e-05,
      "loss": 0.0316,
      "num_input_tokens_seen": 7864384,
      "step": 2365
    },
    {
      "epoch": 0.4742015356526524,
      "grad_norm": 0.2019323706626892,
      "learning_rate": 4.697935878285461e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 7881056,
      "step": 2370
    },
    {
      "epoch": 0.4752019608333542,
      "grad_norm": 0.6341044306755066,
      "learning_rate": 4.696686448319408e-05,
      "loss": 0.0392,
      "num_input_tokens_seen": 7897184,
      "step": 2375
    },
    {
      "epoch": 0.47620238601405596,
      "grad_norm": 0.020971478894352913,
      "learning_rate": 4.695434606524e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 7913824,
      "step": 2380
    },
    {
      "epoch": 0.47720281119475777,
      "grad_norm": 0.5249700546264648,
      "learning_rate": 4.694180354273684e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 7930560,
      "step": 2385
    },
    {
      "epoch": 0.4782032363754596,
      "grad_norm": 1.1561530828475952,
      "learning_rate": 4.692923692945553e-05,
      "loss": 0.0347,
      "num_input_tokens_seen": 7946976,
      "step": 2390
    },
    {
      "epoch": 0.47920366155616134,
      "grad_norm": 2.0148797035217285,
      "learning_rate": 4.6916646239193454e-05,
      "loss": 0.0359,
      "num_input_tokens_seen": 7963328,
      "step": 2395
    },
    {
      "epoch": 0.48020408673686316,
      "grad_norm": 0.909793496131897,
      "learning_rate": 4.690403148577443e-05,
      "loss": 0.0366,
      "num_input_tokens_seen": 7980576,
      "step": 2400
    },
    {
      "epoch": 0.48120451191756497,
      "grad_norm": 0.5779248476028442,
      "learning_rate": 4.68913926830487e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 7996224,
      "step": 2405
    },
    {
      "epoch": 0.4822049370982668,
      "grad_norm": 0.21549813449382782,
      "learning_rate": 4.6878729844892896e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 8012736,
      "step": 2410
    },
    {
      "epoch": 0.48320536227896854,
      "grad_norm": 0.44577720761299133,
      "learning_rate": 4.686604298521008e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 8029472,
      "step": 2415
    },
    {
      "epoch": 0.48420578745967036,
      "grad_norm": 0.7483829855918884,
      "learning_rate": 4.685333211792963e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 8045376,
      "step": 2420
    },
    {
      "epoch": 0.48520621264037217,
      "grad_norm": 0.357555627822876,
      "learning_rate": 4.6840597257007325e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 8061504,
      "step": 2425
    },
    {
      "epoch": 0.486206637821074,
      "grad_norm": 0.308903306722641,
      "learning_rate": 4.682783841642527e-05,
      "loss": 0.0392,
      "num_input_tokens_seen": 8077824,
      "step": 2430
    },
    {
      "epoch": 0.48720706300177574,
      "grad_norm": 0.27499833703041077,
      "learning_rate": 4.681505561019192e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 8094400,
      "step": 2435
    },
    {
      "epoch": 0.48820748818247756,
      "grad_norm": 0.9674980640411377,
      "learning_rate": 4.6802248852342e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 8110880,
      "step": 2440
    },
    {
      "epoch": 0.48920791336317937,
      "grad_norm": 0.9127982258796692,
      "learning_rate": 4.678941815693657e-05,
      "loss": 0.0543,
      "num_input_tokens_seen": 8127040,
      "step": 2445
    },
    {
      "epoch": 0.49020833854388113,
      "grad_norm": 0.6197338104248047,
      "learning_rate": 4.677656353806297e-05,
      "loss": 0.0293,
      "num_input_tokens_seen": 8143712,
      "step": 2450
    },
    {
      "epoch": 0.49120876372458294,
      "grad_norm": 0.7034648656845093,
      "learning_rate": 4.6763685009834784e-05,
      "loss": 0.0416,
      "num_input_tokens_seen": 8159424,
      "step": 2455
    },
    {
      "epoch": 0.49220918890528476,
      "grad_norm": 0.5745025277137756,
      "learning_rate": 4.675078258639187e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 8176768,
      "step": 2460
    },
    {
      "epoch": 0.49320961408598657,
      "grad_norm": 0.3975605368614197,
      "learning_rate": 4.6737856281900295e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 8193248,
      "step": 2465
    },
    {
      "epoch": 0.49421003926668833,
      "grad_norm": 0.2682710587978363,
      "learning_rate": 4.672490611055238e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 8209824,
      "step": 2470
    },
    {
      "epoch": 0.49521046444739014,
      "grad_norm": 0.23972515761852264,
      "learning_rate": 4.671193208656662e-05,
      "loss": 0.0211,
      "num_input_tokens_seen": 8227232,
      "step": 2475
    },
    {
      "epoch": 0.49621088962809196,
      "grad_norm": 0.36315664649009705,
      "learning_rate": 4.669893422418773e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 8243520,
      "step": 2480
    },
    {
      "epoch": 0.4972113148087937,
      "grad_norm": 0.44635969400405884,
      "learning_rate": 4.668591253768656e-05,
      "loss": 0.0396,
      "num_input_tokens_seen": 8260384,
      "step": 2485
    },
    {
      "epoch": 0.49821173998949553,
      "grad_norm": 0.4753713011741638,
      "learning_rate": 4.667286704136014e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 8276192,
      "step": 2490
    },
    {
      "epoch": 0.49921216517019734,
      "grad_norm": 0.20319467782974243,
      "learning_rate": 4.665979774953165e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 8292288,
      "step": 2495
    },
    {
      "epoch": 0.5002125903508992,
      "grad_norm": 0.5693745613098145,
      "learning_rate": 4.664670467655038e-05,
      "loss": 0.0336,
      "num_input_tokens_seen": 8308960,
      "step": 2500
    },
    {
      "epoch": 0.5012130155316009,
      "grad_norm": 0.30443742871284485,
      "learning_rate": 4.663358783679173e-05,
      "loss": 0.0371,
      "num_input_tokens_seen": 8327200,
      "step": 2505
    },
    {
      "epoch": 0.5022134407123028,
      "grad_norm": 0.6215730309486389,
      "learning_rate": 4.662044724465719e-05,
      "loss": 0.0523,
      "num_input_tokens_seen": 8344864,
      "step": 2510
    },
    {
      "epoch": 0.5032138658930045,
      "grad_norm": 0.48964619636535645,
      "learning_rate": 4.660728291457437e-05,
      "loss": 0.0298,
      "num_input_tokens_seen": 8361696,
      "step": 2515
    },
    {
      "epoch": 0.5042142910737063,
      "grad_norm": 0.5137848854064941,
      "learning_rate": 4.659409486099689e-05,
      "loss": 0.0186,
      "num_input_tokens_seen": 8378048,
      "step": 2520
    },
    {
      "epoch": 0.5052147162544082,
      "grad_norm": 0.3688870370388031,
      "learning_rate": 4.658088309840444e-05,
      "loss": 0.034,
      "num_input_tokens_seen": 8395104,
      "step": 2525
    },
    {
      "epoch": 0.5062151414351099,
      "grad_norm": 0.5594642162322998,
      "learning_rate": 4.656764764130274e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 8411360,
      "step": 2530
    },
    {
      "epoch": 0.5072155666158117,
      "grad_norm": 0.21476143598556519,
      "learning_rate": 4.655438850422352e-05,
      "loss": 0.0466,
      "num_input_tokens_seen": 8428480,
      "step": 2535
    },
    {
      "epoch": 0.5082159917965136,
      "grad_norm": 0.4187186360359192,
      "learning_rate": 4.654110570172452e-05,
      "loss": 0.0319,
      "num_input_tokens_seen": 8445312,
      "step": 2540
    },
    {
      "epoch": 0.5092164169772153,
      "grad_norm": 0.7518160939216614,
      "learning_rate": 4.652779924838945e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 8462528,
      "step": 2545
    },
    {
      "epoch": 0.5102168421579171,
      "grad_norm": 0.717582106590271,
      "learning_rate": 4.6514469158828014e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 8479360,
      "step": 2550
    },
    {
      "epoch": 0.5112172673386189,
      "grad_norm": 0.16009634733200073,
      "learning_rate": 4.6501115447675814e-05,
      "loss": 0.0367,
      "num_input_tokens_seen": 8496672,
      "step": 2555
    },
    {
      "epoch": 0.5122176925193207,
      "grad_norm": 0.5227007865905762,
      "learning_rate": 4.648773812959445e-05,
      "loss": 0.0696,
      "num_input_tokens_seen": 8514176,
      "step": 2560
    },
    {
      "epoch": 0.5132181177000225,
      "grad_norm": 0.6089304685592651,
      "learning_rate": 4.647433721927139e-05,
      "loss": 0.0266,
      "num_input_tokens_seen": 8529824,
      "step": 2565
    },
    {
      "epoch": 0.5142185428807243,
      "grad_norm": 0.5312008857727051,
      "learning_rate": 4.6460912731420045e-05,
      "loss": 0.0512,
      "num_input_tokens_seen": 8546912,
      "step": 2570
    },
    {
      "epoch": 0.5152189680614261,
      "grad_norm": 0.2808869779109955,
      "learning_rate": 4.644746468077968e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 8563264,
      "step": 2575
    },
    {
      "epoch": 0.516219393242128,
      "grad_norm": 0.32890409231185913,
      "learning_rate": 4.6433993082115444e-05,
      "loss": 0.0659,
      "num_input_tokens_seen": 8580608,
      "step": 2580
    },
    {
      "epoch": 0.5172198184228297,
      "grad_norm": 0.6253911852836609,
      "learning_rate": 4.642049795021836e-05,
      "loss": 0.0375,
      "num_input_tokens_seen": 8596736,
      "step": 2585
    },
    {
      "epoch": 0.5182202436035315,
      "grad_norm": 0.7056679129600525,
      "learning_rate": 4.6406979299905236e-05,
      "loss": 0.0472,
      "num_input_tokens_seen": 8613184,
      "step": 2590
    },
    {
      "epoch": 0.5192206687842333,
      "grad_norm": 0.6428877115249634,
      "learning_rate": 4.639343714601876e-05,
      "loss": 0.0387,
      "num_input_tokens_seen": 8629984,
      "step": 2595
    },
    {
      "epoch": 0.5202210939649351,
      "grad_norm": 0.7325150370597839,
      "learning_rate": 4.63798715034274e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 8646688,
      "step": 2600
    },
    {
      "epoch": 0.5212215191456369,
      "grad_norm": 1.1029064655303955,
      "learning_rate": 4.636628238702541e-05,
      "loss": 0.0601,
      "num_input_tokens_seen": 8662240,
      "step": 2605
    },
    {
      "epoch": 0.5222219443263387,
      "grad_norm": 0.7259523868560791,
      "learning_rate": 4.6352669811732805e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 8679488,
      "step": 2610
    },
    {
      "epoch": 0.5232223695070405,
      "grad_norm": 0.5248921513557434,
      "learning_rate": 4.63390337924954e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 8695744,
      "step": 2615
    },
    {
      "epoch": 0.5242227946877422,
      "grad_norm": 0.9337000846862793,
      "learning_rate": 4.6325374344284704e-05,
      "loss": 0.0512,
      "num_input_tokens_seen": 8712544,
      "step": 2620
    },
    {
      "epoch": 0.5252232198684441,
      "grad_norm": 0.2723657488822937,
      "learning_rate": 4.6311691482097985e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 8729504,
      "step": 2625
    },
    {
      "epoch": 0.5262236450491459,
      "grad_norm": 0.29864421486854553,
      "learning_rate": 4.629798522095818e-05,
      "loss": 0.0361,
      "num_input_tokens_seen": 8745696,
      "step": 2630
    },
    {
      "epoch": 0.5272240702298476,
      "grad_norm": 0.750019371509552,
      "learning_rate": 4.628425557591395e-05,
      "loss": 0.0341,
      "num_input_tokens_seen": 8762016,
      "step": 2635
    },
    {
      "epoch": 0.5282244954105495,
      "grad_norm": 0.2977132499217987,
      "learning_rate": 4.627050256203962e-05,
      "loss": 0.0425,
      "num_input_tokens_seen": 8778400,
      "step": 2640
    },
    {
      "epoch": 0.5292249205912513,
      "grad_norm": 0.9307669401168823,
      "learning_rate": 4.6256726194435164e-05,
      "loss": 0.0541,
      "num_input_tokens_seen": 8795616,
      "step": 2645
    },
    {
      "epoch": 0.5302253457719531,
      "grad_norm": 0.29842454195022583,
      "learning_rate": 4.62429264882262e-05,
      "loss": 0.033,
      "num_input_tokens_seen": 8811584,
      "step": 2650
    },
    {
      "epoch": 0.5312257709526549,
      "grad_norm": 0.880999743938446,
      "learning_rate": 4.622910345856399e-05,
      "loss": 0.034,
      "num_input_tokens_seen": 8828096,
      "step": 2655
    },
    {
      "epoch": 0.5322261961333566,
      "grad_norm": 0.32106754183769226,
      "learning_rate": 4.621525712062537e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 8845216,
      "step": 2660
    },
    {
      "epoch": 0.5332266213140585,
      "grad_norm": 0.3914763331413269,
      "learning_rate": 4.62013874896128e-05,
      "loss": 0.0383,
      "num_input_tokens_seen": 8861216,
      "step": 2665
    },
    {
      "epoch": 0.5342270464947603,
      "grad_norm": 0.38581305742263794,
      "learning_rate": 4.6187494580754284e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 8877696,
      "step": 2670
    },
    {
      "epoch": 0.535227471675462,
      "grad_norm": 0.8075183033943176,
      "learning_rate": 4.617357840930341e-05,
      "loss": 0.0371,
      "num_input_tokens_seen": 8894688,
      "step": 2675
    },
    {
      "epoch": 0.5362278968561639,
      "grad_norm": 0.766015350818634,
      "learning_rate": 4.615963899053929e-05,
      "loss": 0.0526,
      "num_input_tokens_seen": 8910144,
      "step": 2680
    },
    {
      "epoch": 0.5372283220368657,
      "grad_norm": 1.2620230913162231,
      "learning_rate": 4.614567633976656e-05,
      "loss": 0.0485,
      "num_input_tokens_seen": 8925632,
      "step": 2685
    },
    {
      "epoch": 0.5382287472175674,
      "grad_norm": 0.4963071942329407,
      "learning_rate": 4.6131690472315394e-05,
      "loss": 0.027,
      "num_input_tokens_seen": 8942688,
      "step": 2690
    },
    {
      "epoch": 0.5392291723982693,
      "grad_norm": 0.4999218285083771,
      "learning_rate": 4.6117681403541406e-05,
      "loss": 0.0288,
      "num_input_tokens_seen": 8959744,
      "step": 2695
    },
    {
      "epoch": 0.540229597578971,
      "grad_norm": 0.45712122321128845,
      "learning_rate": 4.610364914882572e-05,
      "loss": 0.0353,
      "num_input_tokens_seen": 8976256,
      "step": 2700
    },
    {
      "epoch": 0.5412300227596729,
      "grad_norm": 0.7429212331771851,
      "learning_rate": 4.60895937235749e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 8992288,
      "step": 2705
    },
    {
      "epoch": 0.5422304479403747,
      "grad_norm": 0.8552815914154053,
      "learning_rate": 4.607551514322096e-05,
      "loss": 0.0519,
      "num_input_tokens_seen": 9008960,
      "step": 2710
    },
    {
      "epoch": 0.5432308731210764,
      "grad_norm": 0.2171797752380371,
      "learning_rate": 4.606141342322134e-05,
      "loss": 0.0296,
      "num_input_tokens_seen": 9024800,
      "step": 2715
    },
    {
      "epoch": 0.5442312983017783,
      "grad_norm": 0.6828743815422058,
      "learning_rate": 4.6047288579058864e-05,
      "loss": 0.0401,
      "num_input_tokens_seen": 9041152,
      "step": 2720
    },
    {
      "epoch": 0.5452317234824801,
      "grad_norm": 0.501909613609314,
      "learning_rate": 4.6033140626241776e-05,
      "loss": 0.0597,
      "num_input_tokens_seen": 9059040,
      "step": 2725
    },
    {
      "epoch": 0.5462321486631818,
      "grad_norm": 0.2294810712337494,
      "learning_rate": 4.6018969580303664e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 9076256,
      "step": 2730
    },
    {
      "epoch": 0.5472325738438837,
      "grad_norm": 0.20600394904613495,
      "learning_rate": 4.600477545680348e-05,
      "loss": 0.0271,
      "num_input_tokens_seen": 9093792,
      "step": 2735
    },
    {
      "epoch": 0.5482329990245854,
      "grad_norm": 0.3571174144744873,
      "learning_rate": 4.599055827132553e-05,
      "loss": 0.0293,
      "num_input_tokens_seen": 9110016,
      "step": 2740
    },
    {
      "epoch": 0.5492334242052872,
      "grad_norm": 0.4264056980609894,
      "learning_rate": 4.597631803947942e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 9127648,
      "step": 2745
    },
    {
      "epoch": 0.5502338493859891,
      "grad_norm": 0.6972314119338989,
      "learning_rate": 4.5962054776900055e-05,
      "loss": 0.0247,
      "num_input_tokens_seen": 9143840,
      "step": 2750
    },
    {
      "epoch": 0.5512342745666908,
      "grad_norm": 1.5259716510772705,
      "learning_rate": 4.5947768499247656e-05,
      "loss": 0.0358,
      "num_input_tokens_seen": 9159648,
      "step": 2755
    },
    {
      "epoch": 0.5522346997473926,
      "grad_norm": 0.6529611349105835,
      "learning_rate": 4.5933459222207694e-05,
      "loss": 0.0183,
      "num_input_tokens_seen": 9177728,
      "step": 2760
    },
    {
      "epoch": 0.5532351249280945,
      "grad_norm": 1.142457365989685,
      "learning_rate": 4.591912696149088e-05,
      "loss": 0.0493,
      "num_input_tokens_seen": 9193536,
      "step": 2765
    },
    {
      "epoch": 0.5542355501087962,
      "grad_norm": 0.47713854908943176,
      "learning_rate": 4.5904771732833176e-05,
      "loss": 0.0504,
      "num_input_tokens_seen": 9209920,
      "step": 2770
    },
    {
      "epoch": 0.5552359752894981,
      "grad_norm": 0.8752053380012512,
      "learning_rate": 4.5890393551995764e-05,
      "loss": 0.0376,
      "num_input_tokens_seen": 9225216,
      "step": 2775
    },
    {
      "epoch": 0.5562364004701998,
      "grad_norm": 0.5204348564147949,
      "learning_rate": 4.5875992434765014e-05,
      "loss": 0.0512,
      "num_input_tokens_seen": 9241440,
      "step": 2780
    },
    {
      "epoch": 0.5572368256509016,
      "grad_norm": 0.43067991733551025,
      "learning_rate": 4.586156839695249e-05,
      "loss": 0.0286,
      "num_input_tokens_seen": 9257984,
      "step": 2785
    },
    {
      "epoch": 0.5582372508316035,
      "grad_norm": 0.9396999478340149,
      "learning_rate": 4.5847121454394925e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 9275264,
      "step": 2790
    },
    {
      "epoch": 0.5592376760123052,
      "grad_norm": 0.22550955414772034,
      "learning_rate": 4.583265162295417e-05,
      "loss": 0.0528,
      "num_input_tokens_seen": 9292608,
      "step": 2795
    },
    {
      "epoch": 0.560238101193007,
      "grad_norm": 1.111317753791809,
      "learning_rate": 4.581815891851724e-05,
      "loss": 0.0521,
      "num_input_tokens_seen": 9308640,
      "step": 2800
    },
    {
      "epoch": 0.5612385263737089,
      "grad_norm": 0.24286577105522156,
      "learning_rate": 4.580364335699626e-05,
      "loss": 0.0315,
      "num_input_tokens_seen": 9326048,
      "step": 2805
    },
    {
      "epoch": 0.5622389515544106,
      "grad_norm": 0.6611106991767883,
      "learning_rate": 4.578910495432842e-05,
      "loss": 0.0456,
      "num_input_tokens_seen": 9341984,
      "step": 2810
    },
    {
      "epoch": 0.5632393767351124,
      "grad_norm": 0.38972610235214233,
      "learning_rate": 4.5774543726476026e-05,
      "loss": 0.0256,
      "num_input_tokens_seen": 9358016,
      "step": 2815
    },
    {
      "epoch": 0.5642398019158142,
      "grad_norm": 0.46140381693840027,
      "learning_rate": 4.5759959689426425e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 9374880,
      "step": 2820
    },
    {
      "epoch": 0.565240227096516,
      "grad_norm": 0.8493245244026184,
      "learning_rate": 4.5745352859192017e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 9391840,
      "step": 2825
    },
    {
      "epoch": 0.5662406522772179,
      "grad_norm": 0.1267855316400528,
      "learning_rate": 4.573072325181021e-05,
      "loss": 0.0227,
      "num_input_tokens_seen": 9407360,
      "step": 2830
    },
    {
      "epoch": 0.5672410774579196,
      "grad_norm": 0.7152971029281616,
      "learning_rate": 4.571607088334344e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 9423200,
      "step": 2835
    },
    {
      "epoch": 0.5682415026386214,
      "grad_norm": 0.5422306060791016,
      "learning_rate": 4.570139576987912e-05,
      "loss": 0.045,
      "num_input_tokens_seen": 9438976,
      "step": 2840
    },
    {
      "epoch": 0.5692419278193233,
      "grad_norm": 0.2335476279258728,
      "learning_rate": 4.568669792752964e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 9455328,
      "step": 2845
    },
    {
      "epoch": 0.570242353000025,
      "grad_norm": 0.24949710071086884,
      "learning_rate": 4.567197737243235e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 9471520,
      "step": 2850
    },
    {
      "epoch": 0.5712427781807268,
      "grad_norm": 0.38113167881965637,
      "learning_rate": 4.565723412074953e-05,
      "loss": 0.0454,
      "num_input_tokens_seen": 9487904,
      "step": 2855
    },
    {
      "epoch": 0.5722432033614286,
      "grad_norm": 1.0451862812042236,
      "learning_rate": 4.564246818866838e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 9503872,
      "step": 2860
    },
    {
      "epoch": 0.5732436285421304,
      "grad_norm": 0.9802286028862,
      "learning_rate": 4.5627679592401e-05,
      "loss": 0.0406,
      "num_input_tokens_seen": 9519936,
      "step": 2865
    },
    {
      "epoch": 0.5742440537228322,
      "grad_norm": 0.46139222383499146,
      "learning_rate": 4.561286834818439e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 9536544,
      "step": 2870
    },
    {
      "epoch": 0.575244478903534,
      "grad_norm": 0.85197913646698,
      "learning_rate": 4.559803447228039e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 9552320,
      "step": 2875
    },
    {
      "epoch": 0.5762449040842358,
      "grad_norm": 0.7129600048065186,
      "learning_rate": 4.558317798097571e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 9568192,
      "step": 2880
    },
    {
      "epoch": 0.5772453292649375,
      "grad_norm": 0.34167492389678955,
      "learning_rate": 4.556829889058188e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 9584640,
      "step": 2885
    },
    {
      "epoch": 0.5782457544456394,
      "grad_norm": 0.6730559468269348,
      "learning_rate": 4.555339721743526e-05,
      "loss": 0.0478,
      "num_input_tokens_seen": 9602656,
      "step": 2890
    },
    {
      "epoch": 0.5792461796263412,
      "grad_norm": 0.5515514612197876,
      "learning_rate": 4.553847297789696e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 9620384,
      "step": 2895
    },
    {
      "epoch": 0.580246604807043,
      "grad_norm": 0.8256182670593262,
      "learning_rate": 4.552352618835291e-05,
      "loss": 0.0416,
      "num_input_tokens_seen": 9636640,
      "step": 2900
    },
    {
      "epoch": 0.5812470299877448,
      "grad_norm": 0.439656525850296,
      "learning_rate": 4.5508556865213784e-05,
      "loss": 0.0344,
      "num_input_tokens_seen": 9653248,
      "step": 2905
    },
    {
      "epoch": 0.5822474551684466,
      "grad_norm": 0.7512446641921997,
      "learning_rate": 4.5493565024915005e-05,
      "loss": 0.0607,
      "num_input_tokens_seen": 9670432,
      "step": 2910
    },
    {
      "epoch": 0.5832478803491484,
      "grad_norm": 0.5968084335327148,
      "learning_rate": 4.54785506839167e-05,
      "loss": 0.0401,
      "num_input_tokens_seen": 9685632,
      "step": 2915
    },
    {
      "epoch": 0.5842483055298502,
      "grad_norm": 0.41218385100364685,
      "learning_rate": 4.546351385870371e-05,
      "loss": 0.0838,
      "num_input_tokens_seen": 9702272,
      "step": 2920
    },
    {
      "epoch": 0.585248730710552,
      "grad_norm": 0.23477399349212646,
      "learning_rate": 4.544845456578557e-05,
      "loss": 0.0402,
      "num_input_tokens_seen": 9718432,
      "step": 2925
    },
    {
      "epoch": 0.5862491558912538,
      "grad_norm": 0.34075719118118286,
      "learning_rate": 4.5433372821696465e-05,
      "loss": 0.0376,
      "num_input_tokens_seen": 9735392,
      "step": 2930
    },
    {
      "epoch": 0.5872495810719556,
      "grad_norm": 0.4260369539260864,
      "learning_rate": 4.541826864299526e-05,
      "loss": 0.0264,
      "num_input_tokens_seen": 9751712,
      "step": 2935
    },
    {
      "epoch": 0.5882500062526573,
      "grad_norm": 0.2027159035205841,
      "learning_rate": 4.5403142046265415e-05,
      "loss": 0.0318,
      "num_input_tokens_seen": 9768864,
      "step": 2940
    },
    {
      "epoch": 0.5892504314333592,
      "grad_norm": 0.6176860332489014,
      "learning_rate": 4.538799304811503e-05,
      "loss": 0.0337,
      "num_input_tokens_seen": 9784768,
      "step": 2945
    },
    {
      "epoch": 0.590250856614061,
      "grad_norm": 1.2780542373657227,
      "learning_rate": 4.537282166517679e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 9801888,
      "step": 2950
    },
    {
      "epoch": 0.5912512817947628,
      "grad_norm": 1.051655650138855,
      "learning_rate": 4.5357627914107956e-05,
      "loss": 0.0387,
      "num_input_tokens_seen": 9818048,
      "step": 2955
    },
    {
      "epoch": 0.5922517069754646,
      "grad_norm": 0.1395687311887741,
      "learning_rate": 4.534241181159037e-05,
      "loss": 0.0465,
      "num_input_tokens_seen": 9834208,
      "step": 2960
    },
    {
      "epoch": 0.5932521321561663,
      "grad_norm": 0.737740159034729,
      "learning_rate": 4.532717337433037e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 9850848,
      "step": 2965
    },
    {
      "epoch": 0.5942525573368682,
      "grad_norm": 0.392226904630661,
      "learning_rate": 4.531191261905885e-05,
      "loss": 0.0542,
      "num_input_tokens_seen": 9866432,
      "step": 2970
    },
    {
      "epoch": 0.59525298251757,
      "grad_norm": 0.33025363087654114,
      "learning_rate": 4.5296629562531204e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 9883872,
      "step": 2975
    },
    {
      "epoch": 0.5962534076982717,
      "grad_norm": 0.18435664474964142,
      "learning_rate": 4.5281324221527314e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 9900512,
      "step": 2980
    },
    {
      "epoch": 0.5972538328789736,
      "grad_norm": 0.26673799753189087,
      "learning_rate": 4.5265996612851504e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 9917600,
      "step": 2985
    },
    {
      "epoch": 0.5982542580596754,
      "grad_norm": 0.2888360917568207,
      "learning_rate": 4.525064675333258e-05,
      "loss": 0.0256,
      "num_input_tokens_seen": 9933728,
      "step": 2990
    },
    {
      "epoch": 0.5992546832403771,
      "grad_norm": 0.37292373180389404,
      "learning_rate": 4.523527465982375e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 9950688,
      "step": 2995
    },
    {
      "epoch": 0.600255108421079,
      "grad_norm": 0.5753964185714722,
      "learning_rate": 4.521988034920266e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 9967072,
      "step": 3000
    },
    {
      "epoch": 0.6012555336017807,
      "grad_norm": 0.7338746190071106,
      "learning_rate": 4.5204463838371333e-05,
      "loss": 0.0381,
      "num_input_tokens_seen": 9983424,
      "step": 3005
    },
    {
      "epoch": 0.6022559587824825,
      "grad_norm": 0.1053917333483696,
      "learning_rate": 4.518902514425616e-05,
      "loss": 0.0168,
      "num_input_tokens_seen": 10000608,
      "step": 3010
    },
    {
      "epoch": 0.6032563839631844,
      "grad_norm": 0.3549809455871582,
      "learning_rate": 4.51735642838079e-05,
      "loss": 0.0602,
      "num_input_tokens_seen": 10017376,
      "step": 3015
    },
    {
      "epoch": 0.6042568091438861,
      "grad_norm": 0.2778543531894684,
      "learning_rate": 4.515808127400165e-05,
      "loss": 0.0358,
      "num_input_tokens_seen": 10033504,
      "step": 3020
    },
    {
      "epoch": 0.605257234324588,
      "grad_norm": 0.5423609614372253,
      "learning_rate": 4.514257613183682e-05,
      "loss": 0.0649,
      "num_input_tokens_seen": 10050816,
      "step": 3025
    },
    {
      "epoch": 0.6062576595052898,
      "grad_norm": 0.5223437547683716,
      "learning_rate": 4.5127048874337123e-05,
      "loss": 0.0475,
      "num_input_tokens_seen": 10066080,
      "step": 3030
    },
    {
      "epoch": 0.6072580846859915,
      "grad_norm": 0.2898062765598297,
      "learning_rate": 4.511149951855055e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 10083200,
      "step": 3035
    },
    {
      "epoch": 0.6082585098666934,
      "grad_norm": 0.3376716077327728,
      "learning_rate": 4.509592808154936e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 10099616,
      "step": 3040
    },
    {
      "epoch": 0.6092589350473951,
      "grad_norm": 0.6940332651138306,
      "learning_rate": 4.508033458043005e-05,
      "loss": 0.0412,
      "num_input_tokens_seen": 10115744,
      "step": 3045
    },
    {
      "epoch": 0.6102593602280969,
      "grad_norm": 0.4612104892730713,
      "learning_rate": 4.506471903231334e-05,
      "loss": 0.0272,
      "num_input_tokens_seen": 10132224,
      "step": 3050
    },
    {
      "epoch": 0.6112597854087988,
      "grad_norm": 0.33771753311157227,
      "learning_rate": 4.5049081454344173e-05,
      "loss": 0.039,
      "num_input_tokens_seen": 10151008,
      "step": 3055
    },
    {
      "epoch": 0.6122602105895005,
      "grad_norm": 0.5704123377799988,
      "learning_rate": 4.503342186369165e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 10168096,
      "step": 3060
    },
    {
      "epoch": 0.6132606357702023,
      "grad_norm": 0.1791711002588272,
      "learning_rate": 4.501774027754908e-05,
      "loss": 0.0151,
      "num_input_tokens_seen": 10183616,
      "step": 3065
    },
    {
      "epoch": 0.6142610609509042,
      "grad_norm": 0.5915095806121826,
      "learning_rate": 4.5002036713133876e-05,
      "loss": 0.0321,
      "num_input_tokens_seen": 10200288,
      "step": 3070
    },
    {
      "epoch": 0.6152614861316059,
      "grad_norm": 0.5338975787162781,
      "learning_rate": 4.4986311187687624e-05,
      "loss": 0.0217,
      "num_input_tokens_seen": 10217248,
      "step": 3075
    },
    {
      "epoch": 0.6162619113123077,
      "grad_norm": 1.0453840494155884,
      "learning_rate": 4.4970563718476e-05,
      "loss": 0.0589,
      "num_input_tokens_seen": 10234688,
      "step": 3080
    },
    {
      "epoch": 0.6172623364930095,
      "grad_norm": 0.9612706899642944,
      "learning_rate": 4.495479432278876e-05,
      "loss": 0.0498,
      "num_input_tokens_seen": 10252096,
      "step": 3085
    },
    {
      "epoch": 0.6182627616737113,
      "grad_norm": 0.31238457560539246,
      "learning_rate": 4.493900301793977e-05,
      "loss": 0.0497,
      "num_input_tokens_seen": 10268512,
      "step": 3090
    },
    {
      "epoch": 0.6192631868544132,
      "grad_norm": 0.5907064080238342,
      "learning_rate": 4.492318982126693e-05,
      "loss": 0.0504,
      "num_input_tokens_seen": 10284736,
      "step": 3095
    },
    {
      "epoch": 0.6202636120351149,
      "grad_norm": 0.2940639853477478,
      "learning_rate": 4.490735475013217e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 10301632,
      "step": 3100
    },
    {
      "epoch": 0.6212640372158167,
      "grad_norm": 0.20434249937534332,
      "learning_rate": 4.4891497821921436e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 10319680,
      "step": 3105
    },
    {
      "epoch": 0.6222644623965186,
      "grad_norm": 0.44756656885147095,
      "learning_rate": 4.487561905404469e-05,
      "loss": 0.0591,
      "num_input_tokens_seen": 10336960,
      "step": 3110
    },
    {
      "epoch": 0.6232648875772203,
      "grad_norm": 0.3095456659793854,
      "learning_rate": 4.485971846393587e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 10352928,
      "step": 3115
    },
    {
      "epoch": 0.6242653127579221,
      "grad_norm": 0.6086257100105286,
      "learning_rate": 4.484379606905286e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 10369632,
      "step": 3120
    },
    {
      "epoch": 0.6252657379386239,
      "grad_norm": 0.8644992709159851,
      "learning_rate": 4.482785188687747e-05,
      "loss": 0.0522,
      "num_input_tokens_seen": 10384992,
      "step": 3125
    },
    {
      "epoch": 0.6262661631193257,
      "grad_norm": 0.22823557257652283,
      "learning_rate": 4.4811885934915485e-05,
      "loss": 0.0193,
      "num_input_tokens_seen": 10401536,
      "step": 3130
    },
    {
      "epoch": 0.6272665883000275,
      "grad_norm": 0.35614490509033203,
      "learning_rate": 4.479589823069653e-05,
      "loss": 0.0364,
      "num_input_tokens_seen": 10418144,
      "step": 3135
    },
    {
      "epoch": 0.6282670134807293,
      "grad_norm": 0.395463228225708,
      "learning_rate": 4.477988879177416e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 10434176,
      "step": 3140
    },
    {
      "epoch": 0.6292674386614311,
      "grad_norm": 0.5990577936172485,
      "learning_rate": 4.476385763572576e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 10450432,
      "step": 3145
    },
    {
      "epoch": 0.630267863842133,
      "grad_norm": 0.7763248085975647,
      "learning_rate": 4.474780478015259e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 10466464,
      "step": 3150
    },
    {
      "epoch": 0.6312682890228347,
      "grad_norm": 0.3586103618144989,
      "learning_rate": 4.47317302426797e-05,
      "loss": 0.0341,
      "num_input_tokens_seen": 10483200,
      "step": 3155
    },
    {
      "epoch": 0.6322687142035365,
      "grad_norm": 0.2397426962852478,
      "learning_rate": 4.4715634040955977e-05,
      "loss": 0.0341,
      "num_input_tokens_seen": 10499008,
      "step": 3160
    },
    {
      "epoch": 0.6332691393842383,
      "grad_norm": 0.14368750154972076,
      "learning_rate": 4.469951619265408e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 10516928,
      "step": 3165
    },
    {
      "epoch": 0.6342695645649401,
      "grad_norm": 0.5184279680252075,
      "learning_rate": 4.468337671547042e-05,
      "loss": 0.0322,
      "num_input_tokens_seen": 10532800,
      "step": 3170
    },
    {
      "epoch": 0.6352699897456419,
      "grad_norm": 0.610159158706665,
      "learning_rate": 4.466721562712518e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 10550112,
      "step": 3175
    },
    {
      "epoch": 0.6362704149263437,
      "grad_norm": 0.4582042396068573,
      "learning_rate": 4.465103294536227e-05,
      "loss": 0.0614,
      "num_input_tokens_seen": 10565376,
      "step": 3180
    },
    {
      "epoch": 0.6372708401070455,
      "grad_norm": 0.4239751696586609,
      "learning_rate": 4.4634828687949296e-05,
      "loss": 0.0217,
      "num_input_tokens_seen": 10582112,
      "step": 3185
    },
    {
      "epoch": 0.6382712652877472,
      "grad_norm": 0.6495357155799866,
      "learning_rate": 4.461860287267754e-05,
      "loss": 0.0331,
      "num_input_tokens_seen": 10598592,
      "step": 3190
    },
    {
      "epoch": 0.6392716904684491,
      "grad_norm": 0.39509135484695435,
      "learning_rate": 4.460235551736199e-05,
      "loss": 0.0351,
      "num_input_tokens_seen": 10615392,
      "step": 3195
    },
    {
      "epoch": 0.6402721156491509,
      "grad_norm": 0.40606188774108887,
      "learning_rate": 4.458608663984125e-05,
      "loss": 0.0383,
      "num_input_tokens_seen": 10632352,
      "step": 3200
    },
    {
      "epoch": 0.6412725408298526,
      "grad_norm": 0.1430123746395111,
      "learning_rate": 4.4569796257977576e-05,
      "loss": 0.013,
      "num_input_tokens_seen": 10649504,
      "step": 3205
    },
    {
      "epoch": 0.6422729660105545,
      "grad_norm": 0.1357680708169937,
      "learning_rate": 4.455348438965682e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 10667008,
      "step": 3210
    },
    {
      "epoch": 0.6432733911912563,
      "grad_norm": 0.4743283689022064,
      "learning_rate": 4.4537151052788425e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 10683744,
      "step": 3215
    },
    {
      "epoch": 0.6442738163719581,
      "grad_norm": 1.1936310529708862,
      "learning_rate": 4.452079626530543e-05,
      "loss": 0.0565,
      "num_input_tokens_seen": 10699840,
      "step": 3220
    },
    {
      "epoch": 0.6452742415526599,
      "grad_norm": 0.24838638305664062,
      "learning_rate": 4.450442004516439e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 10716416,
      "step": 3225
    },
    {
      "epoch": 0.6462746667333616,
      "grad_norm": 0.38544249534606934,
      "learning_rate": 4.44880224103454e-05,
      "loss": 0.0364,
      "num_input_tokens_seen": 10732960,
      "step": 3230
    },
    {
      "epoch": 0.6472750919140635,
      "grad_norm": 0.5896895527839661,
      "learning_rate": 4.44716033788521e-05,
      "loss": 0.0392,
      "num_input_tokens_seen": 10749632,
      "step": 3235
    },
    {
      "epoch": 0.6482755170947653,
      "grad_norm": 0.49441149830818176,
      "learning_rate": 4.4455162968711585e-05,
      "loss": 0.019,
      "num_input_tokens_seen": 10767808,
      "step": 3240
    },
    {
      "epoch": 0.649275942275467,
      "grad_norm": 0.14837044477462769,
      "learning_rate": 4.4438701197974424e-05,
      "loss": 0.0485,
      "num_input_tokens_seen": 10784512,
      "step": 3245
    },
    {
      "epoch": 0.6502763674561689,
      "grad_norm": 0.5606879591941833,
      "learning_rate": 4.4422218084714664e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 10800896,
      "step": 3250
    },
    {
      "epoch": 0.6512767926368707,
      "grad_norm": 0.6518362164497375,
      "learning_rate": 4.440571364702977e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 10817568,
      "step": 3255
    },
    {
      "epoch": 0.6522772178175724,
      "grad_norm": 0.6149171590805054,
      "learning_rate": 4.4389187903040605e-05,
      "loss": 0.0726,
      "num_input_tokens_seen": 10833824,
      "step": 3260
    },
    {
      "epoch": 0.6532776429982743,
      "grad_norm": 0.4839974343776703,
      "learning_rate": 4.437264087089146e-05,
      "loss": 0.0421,
      "num_input_tokens_seen": 10850432,
      "step": 3265
    },
    {
      "epoch": 0.654278068178976,
      "grad_norm": 0.5093735456466675,
      "learning_rate": 4.435607256874996e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 10867296,
      "step": 3270
    },
    {
      "epoch": 0.6552784933596779,
      "grad_norm": 0.5983829498291016,
      "learning_rate": 4.433948301480712e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 10883136,
      "step": 3275
    },
    {
      "epoch": 0.6562789185403797,
      "grad_norm": 0.37272733449935913,
      "learning_rate": 4.4322872227277255e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 10899904,
      "step": 3280
    },
    {
      "epoch": 0.6572793437210814,
      "grad_norm": 0.5240197777748108,
      "learning_rate": 4.4306240224398024e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 10915808,
      "step": 3285
    },
    {
      "epoch": 0.6582797689017833,
      "grad_norm": 0.26496824622154236,
      "learning_rate": 4.428958702443035e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 10932608,
      "step": 3290
    },
    {
      "epoch": 0.6592801940824851,
      "grad_norm": 0.16277429461479187,
      "learning_rate": 4.4272912645658454e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 10949152,
      "step": 3295
    },
    {
      "epoch": 0.6602806192631868,
      "grad_norm": 0.9645645618438721,
      "learning_rate": 4.4256217106389796e-05,
      "loss": 0.0397,
      "num_input_tokens_seen": 10965792,
      "step": 3300
    },
    {
      "epoch": 0.6612810444438887,
      "grad_norm": 0.878473699092865,
      "learning_rate": 4.423950042495506e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 10981184,
      "step": 3305
    },
    {
      "epoch": 0.6622814696245904,
      "grad_norm": 0.39659085869789124,
      "learning_rate": 4.422276261970818e-05,
      "loss": 0.0518,
      "num_input_tokens_seen": 10997088,
      "step": 3310
    },
    {
      "epoch": 0.6632818948052922,
      "grad_norm": 0.41067275404930115,
      "learning_rate": 4.420600370902623e-05,
      "loss": 0.0386,
      "num_input_tokens_seen": 11014784,
      "step": 3315
    },
    {
      "epoch": 0.6642823199859941,
      "grad_norm": 0.33324503898620605,
      "learning_rate": 4.4189223711309514e-05,
      "loss": 0.0388,
      "num_input_tokens_seen": 11030464,
      "step": 3320
    },
    {
      "epoch": 0.6652827451666958,
      "grad_norm": 0.4756176769733429,
      "learning_rate": 4.4172422644981435e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 11046464,
      "step": 3325
    },
    {
      "epoch": 0.6662831703473976,
      "grad_norm": 0.32805490493774414,
      "learning_rate": 4.415560052848855e-05,
      "loss": 0.0431,
      "num_input_tokens_seen": 11063744,
      "step": 3330
    },
    {
      "epoch": 0.6672835955280995,
      "grad_norm": 0.776425838470459,
      "learning_rate": 4.413875738030054e-05,
      "loss": 0.0389,
      "num_input_tokens_seen": 11080832,
      "step": 3335
    },
    {
      "epoch": 0.6682840207088012,
      "grad_norm": 0.3049832284450531,
      "learning_rate": 4.412189321891017e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 11098624,
      "step": 3340
    },
    {
      "epoch": 0.6692844458895031,
      "grad_norm": 0.5459462404251099,
      "learning_rate": 4.4105008062833264e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 11116064,
      "step": 3345
    },
    {
      "epoch": 0.6702848710702048,
      "grad_norm": 0.6240223050117493,
      "learning_rate": 4.408810193060871e-05,
      "loss": 0.0471,
      "num_input_tokens_seen": 11134208,
      "step": 3350
    },
    {
      "epoch": 0.6712852962509066,
      "grad_norm": 0.5435937643051147,
      "learning_rate": 4.407117484079843e-05,
      "loss": 0.0446,
      "num_input_tokens_seen": 11151168,
      "step": 3355
    },
    {
      "epoch": 0.6722857214316085,
      "grad_norm": 0.27000948786735535,
      "learning_rate": 4.405422681198732e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 11167296,
      "step": 3360
    },
    {
      "epoch": 0.6732861466123102,
      "grad_norm": 0.4941251873970032,
      "learning_rate": 4.403725786278333e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 11184128,
      "step": 3365
    },
    {
      "epoch": 0.674286571793012,
      "grad_norm": 0.551177442073822,
      "learning_rate": 4.4020268011817324e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 11201888,
      "step": 3370
    },
    {
      "epoch": 0.6752869969737139,
      "grad_norm": 0.43434232473373413,
      "learning_rate": 4.4003257277743135e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 11218912,
      "step": 3375
    },
    {
      "epoch": 0.6762874221544156,
      "grad_norm": 0.8790416717529297,
      "learning_rate": 4.3986225679237534e-05,
      "loss": 0.0464,
      "num_input_tokens_seen": 11234464,
      "step": 3380
    },
    {
      "epoch": 0.6772878473351174,
      "grad_norm": 0.6582038402557373,
      "learning_rate": 4.396917323500018e-05,
      "loss": 0.0226,
      "num_input_tokens_seen": 11251136,
      "step": 3385
    },
    {
      "epoch": 0.6782882725158192,
      "grad_norm": 0.34015730023384094,
      "learning_rate": 4.395209996375363e-05,
      "loss": 0.0342,
      "num_input_tokens_seen": 11268000,
      "step": 3390
    },
    {
      "epoch": 0.679288697696521,
      "grad_norm": 0.15952935814857483,
      "learning_rate": 4.393500588424333e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 11284256,
      "step": 3395
    },
    {
      "epoch": 0.6802891228772229,
      "grad_norm": 0.21244794130325317,
      "learning_rate": 4.391789101523751e-05,
      "loss": 0.0266,
      "num_input_tokens_seen": 11300896,
      "step": 3400
    },
    {
      "epoch": 0.6812895480579246,
      "grad_norm": 0.2520093023777008,
      "learning_rate": 4.390075537552729e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 11316768,
      "step": 3405
    },
    {
      "epoch": 0.6822899732386264,
      "grad_norm": 0.6954443454742432,
      "learning_rate": 4.388359898392656e-05,
      "loss": 0.074,
      "num_input_tokens_seen": 11333312,
      "step": 3410
    },
    {
      "epoch": 0.6832903984193283,
      "grad_norm": 0.2341863512992859,
      "learning_rate": 4.386642185927201e-05,
      "loss": 0.0247,
      "num_input_tokens_seen": 11350144,
      "step": 3415
    },
    {
      "epoch": 0.68429082360003,
      "grad_norm": 0.4664306044578552,
      "learning_rate": 4.3849224020423095e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 11365952,
      "step": 3420
    },
    {
      "epoch": 0.6852912487807318,
      "grad_norm": 0.22717294096946716,
      "learning_rate": 4.383200548626199e-05,
      "loss": 0.0527,
      "num_input_tokens_seen": 11382944,
      "step": 3425
    },
    {
      "epoch": 0.6862916739614336,
      "grad_norm": 0.8043995499610901,
      "learning_rate": 4.381476627569361e-05,
      "loss": 0.0347,
      "num_input_tokens_seen": 11398720,
      "step": 3430
    },
    {
      "epoch": 0.6872920991421354,
      "grad_norm": 0.6292429566383362,
      "learning_rate": 4.379750640764558e-05,
      "loss": 0.037,
      "num_input_tokens_seen": 11415168,
      "step": 3435
    },
    {
      "epoch": 0.6882925243228372,
      "grad_norm": 0.342271625995636,
      "learning_rate": 4.3780225901068194e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 11430912,
      "step": 3440
    },
    {
      "epoch": 0.689292949503539,
      "grad_norm": 0.6785433292388916,
      "learning_rate": 4.37629247749344e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 11447904,
      "step": 3445
    },
    {
      "epoch": 0.6902933746842408,
      "grad_norm": 0.21225985884666443,
      "learning_rate": 4.374560304823979e-05,
      "loss": 0.0374,
      "num_input_tokens_seen": 11465152,
      "step": 3450
    },
    {
      "epoch": 0.6912937998649425,
      "grad_norm": 0.32921814918518066,
      "learning_rate": 4.372826074000258e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 11481760,
      "step": 3455
    },
    {
      "epoch": 0.6922942250456444,
      "grad_norm": 0.07804366946220398,
      "learning_rate": 4.371089786926359e-05,
      "loss": 0.0168,
      "num_input_tokens_seen": 11497888,
      "step": 3460
    },
    {
      "epoch": 0.6932946502263462,
      "grad_norm": 0.16711276769638062,
      "learning_rate": 4.369351445508618e-05,
      "loss": 0.0356,
      "num_input_tokens_seen": 11514592,
      "step": 3465
    },
    {
      "epoch": 0.694295075407048,
      "grad_norm": 0.49618369340896606,
      "learning_rate": 4.367611051655632e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 11532352,
      "step": 3470
    },
    {
      "epoch": 0.6952955005877498,
      "grad_norm": 0.2727205157279968,
      "learning_rate": 4.3658686072782476e-05,
      "loss": 0.0588,
      "num_input_tokens_seen": 11549408,
      "step": 3475
    },
    {
      "epoch": 0.6962959257684516,
      "grad_norm": 0.38875845074653625,
      "learning_rate": 4.364124114289562e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 11566912,
      "step": 3480
    },
    {
      "epoch": 0.6972963509491534,
      "grad_norm": 0.2872340679168701,
      "learning_rate": 4.362377574604926e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 11584960,
      "step": 3485
    },
    {
      "epoch": 0.6982967761298552,
      "grad_norm": 0.41954946517944336,
      "learning_rate": 4.3606289901419326e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 11601312,
      "step": 3490
    },
    {
      "epoch": 0.699297201310557,
      "grad_norm": 0.5533486604690552,
      "learning_rate": 4.358878362820425e-05,
      "loss": 0.0394,
      "num_input_tokens_seen": 11617440,
      "step": 3495
    },
    {
      "epoch": 0.7002976264912588,
      "grad_norm": 0.19497890770435333,
      "learning_rate": 4.357125694562484e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 11633888,
      "step": 3500
    },
    {
      "epoch": 0.7012980516719606,
      "grad_norm": 0.34632545709609985,
      "learning_rate": 4.355370987292434e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 11651136,
      "step": 3505
    },
    {
      "epoch": 0.7022984768526623,
      "grad_norm": 0.6666107773780823,
      "learning_rate": 4.353614242936839e-05,
      "loss": 0.0492,
      "num_input_tokens_seen": 11666016,
      "step": 3510
    },
    {
      "epoch": 0.7032989020333642,
      "grad_norm": 0.43019816279411316,
      "learning_rate": 4.351855463424498e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 11682688,
      "step": 3515
    },
    {
      "epoch": 0.704299327214066,
      "grad_norm": 1.1388988494873047,
      "learning_rate": 4.350094650686445e-05,
      "loss": 0.0467,
      "num_input_tokens_seen": 11698688,
      "step": 3520
    },
    {
      "epoch": 0.7052997523947678,
      "grad_norm": 0.2856043875217438,
      "learning_rate": 4.3483318066559456e-05,
      "loss": 0.0315,
      "num_input_tokens_seen": 11716416,
      "step": 3525
    },
    {
      "epoch": 0.7063001775754696,
      "grad_norm": 0.1407749503850937,
      "learning_rate": 4.3465669332684965e-05,
      "loss": 0.0342,
      "num_input_tokens_seen": 11732960,
      "step": 3530
    },
    {
      "epoch": 0.7073006027561713,
      "grad_norm": 0.32876652479171753,
      "learning_rate": 4.344800032461823e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 11748480,
      "step": 3535
    },
    {
      "epoch": 0.7083010279368732,
      "grad_norm": 0.5137848258018494,
      "learning_rate": 4.343031106175876e-05,
      "loss": 0.0258,
      "num_input_tokens_seen": 11766048,
      "step": 3540
    },
    {
      "epoch": 0.709301453117575,
      "grad_norm": 0.499584436416626,
      "learning_rate": 4.341260156352828e-05,
      "loss": 0.0293,
      "num_input_tokens_seen": 11783232,
      "step": 3545
    },
    {
      "epoch": 0.7103018782982767,
      "grad_norm": 0.5204300284385681,
      "learning_rate": 4.339487184937078e-05,
      "loss": 0.0415,
      "num_input_tokens_seen": 11798976,
      "step": 3550
    },
    {
      "epoch": 0.7113023034789786,
      "grad_norm": 1.127689242362976,
      "learning_rate": 4.337712193875239e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 11815680,
      "step": 3555
    },
    {
      "epoch": 0.7123027286596804,
      "grad_norm": 0.36804962158203125,
      "learning_rate": 4.335935185116147e-05,
      "loss": 0.0303,
      "num_input_tokens_seen": 11832864,
      "step": 3560
    },
    {
      "epoch": 0.7133031538403821,
      "grad_norm": 0.9016924500465393,
      "learning_rate": 4.334156160610847e-05,
      "loss": 0.0586,
      "num_input_tokens_seen": 11850272,
      "step": 3565
    },
    {
      "epoch": 0.714303579021084,
      "grad_norm": 0.5556852221488953,
      "learning_rate": 4.332375122312605e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 11866624,
      "step": 3570
    },
    {
      "epoch": 0.7153040042017857,
      "grad_norm": 0.5024580955505371,
      "learning_rate": 4.33059207217689e-05,
      "loss": 0.0317,
      "num_input_tokens_seen": 11882752,
      "step": 3575
    },
    {
      "epoch": 0.7163044293824875,
      "grad_norm": 0.4733322858810425,
      "learning_rate": 4.3288070121613854e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 11898720,
      "step": 3580
    },
    {
      "epoch": 0.7173048545631894,
      "grad_norm": 0.7094988822937012,
      "learning_rate": 4.32701994422598e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 11915264,
      "step": 3585
    },
    {
      "epoch": 0.7183052797438911,
      "grad_norm": 0.4604605436325073,
      "learning_rate": 4.3252308703327656e-05,
      "loss": 0.0415,
      "num_input_tokens_seen": 11932544,
      "step": 3590
    },
    {
      "epoch": 0.719305704924593,
      "grad_norm": 0.4017398953437805,
      "learning_rate": 4.323439792446038e-05,
      "loss": 0.0361,
      "num_input_tokens_seen": 11949856,
      "step": 3595
    },
    {
      "epoch": 0.7203061301052948,
      "grad_norm": 0.09843380004167557,
      "learning_rate": 4.321646712532292e-05,
      "loss": 0.0506,
      "num_input_tokens_seen": 11966144,
      "step": 3600
    },
    {
      "epoch": 0.7213065552859965,
      "grad_norm": 0.33211007714271545,
      "learning_rate": 4.319851632560222e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 11984736,
      "step": 3605
    },
    {
      "epoch": 0.7223069804666984,
      "grad_norm": 0.42315033078193665,
      "learning_rate": 4.318054554500719e-05,
      "loss": 0.0391,
      "num_input_tokens_seen": 12000384,
      "step": 3610
    },
    {
      "epoch": 0.7233074056474001,
      "grad_norm": 0.5389741063117981,
      "learning_rate": 4.316255480326864e-05,
      "loss": 0.0388,
      "num_input_tokens_seen": 12017056,
      "step": 3615
    },
    {
      "epoch": 0.7243078308281019,
      "grad_norm": 0.49584582448005676,
      "learning_rate": 4.314454412013934e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 12033408,
      "step": 3620
    },
    {
      "epoch": 0.7253082560088038,
      "grad_norm": 0.6536705493927002,
      "learning_rate": 4.312651351539392e-05,
      "loss": 0.0361,
      "num_input_tokens_seen": 12049856,
      "step": 3625
    },
    {
      "epoch": 0.7263086811895055,
      "grad_norm": 0.4414742887020111,
      "learning_rate": 4.31084630088289e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 12067008,
      "step": 3630
    },
    {
      "epoch": 0.7273091063702073,
      "grad_norm": 0.4404211640357971,
      "learning_rate": 4.309039262026265e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 12084160,
      "step": 3635
    },
    {
      "epoch": 0.7283095315509092,
      "grad_norm": 0.20715558528900146,
      "learning_rate": 4.307230236953538e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 12100960,
      "step": 3640
    },
    {
      "epoch": 0.7293099567316109,
      "grad_norm": 0.30401870608329773,
      "learning_rate": 4.305419227650908e-05,
      "loss": 0.0337,
      "num_input_tokens_seen": 12116352,
      "step": 3645
    },
    {
      "epoch": 0.7303103819123127,
      "grad_norm": 0.3494117558002472,
      "learning_rate": 4.303606236106754e-05,
      "loss": 0.0366,
      "num_input_tokens_seen": 12132832,
      "step": 3650
    },
    {
      "epoch": 0.7313108070930145,
      "grad_norm": 0.653393030166626,
      "learning_rate": 4.3017912643116324e-05,
      "loss": 0.028,
      "num_input_tokens_seen": 12148256,
      "step": 3655
    },
    {
      "epoch": 0.7323112322737163,
      "grad_norm": 0.4546215534210205,
      "learning_rate": 4.2999743142582735e-05,
      "loss": 0.0266,
      "num_input_tokens_seen": 12164768,
      "step": 3660
    },
    {
      "epoch": 0.7333116574544182,
      "grad_norm": 0.6860993504524231,
      "learning_rate": 4.298155387941577e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 12181696,
      "step": 3665
    },
    {
      "epoch": 0.7343120826351199,
      "grad_norm": 0.6686315536499023,
      "learning_rate": 4.296334487358615e-05,
      "loss": 0.0417,
      "num_input_tokens_seen": 12199712,
      "step": 3670
    },
    {
      "epoch": 0.7353125078158217,
      "grad_norm": 0.9289959073066711,
      "learning_rate": 4.294511614508628e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 12216000,
      "step": 3675
    },
    {
      "epoch": 0.7363129329965236,
      "grad_norm": 1.0001676082611084,
      "learning_rate": 4.29268677139302e-05,
      "loss": 0.0626,
      "num_input_tokens_seen": 12232768,
      "step": 3680
    },
    {
      "epoch": 0.7373133581772253,
      "grad_norm": 0.10140807181596756,
      "learning_rate": 4.290859960015357e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 12249440,
      "step": 3685
    },
    {
      "epoch": 0.7383137833579271,
      "grad_norm": 0.5375101566314697,
      "learning_rate": 4.2890311823813696e-05,
      "loss": 0.0407,
      "num_input_tokens_seen": 12265024,
      "step": 3690
    },
    {
      "epoch": 0.739314208538629,
      "grad_norm": 0.38093164563179016,
      "learning_rate": 4.2872004404989454e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 12282144,
      "step": 3695
    },
    {
      "epoch": 0.7403146337193307,
      "grad_norm": 0.461363822221756,
      "learning_rate": 4.2853677363781274e-05,
      "loss": 0.0271,
      "num_input_tokens_seen": 12298080,
      "step": 3700
    },
    {
      "epoch": 0.7413150589000325,
      "grad_norm": 1.1793761253356934,
      "learning_rate": 4.283533072031116e-05,
      "loss": 0.043,
      "num_input_tokens_seen": 12314080,
      "step": 3705
    },
    {
      "epoch": 0.7423154840807343,
      "grad_norm": 0.13559503853321075,
      "learning_rate": 4.281696449472261e-05,
      "loss": 0.0193,
      "num_input_tokens_seen": 12331680,
      "step": 3710
    },
    {
      "epoch": 0.7433159092614361,
      "grad_norm": 0.27876919507980347,
      "learning_rate": 4.279857870718064e-05,
      "loss": 0.0371,
      "num_input_tokens_seen": 12347424,
      "step": 3715
    },
    {
      "epoch": 0.744316334442138,
      "grad_norm": 0.5252572894096375,
      "learning_rate": 4.2780173377871725e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 12363392,
      "step": 3720
    },
    {
      "epoch": 0.7453167596228397,
      "grad_norm": 0.5381074547767639,
      "learning_rate": 4.276174852700382e-05,
      "loss": 0.0668,
      "num_input_tokens_seen": 12379616,
      "step": 3725
    },
    {
      "epoch": 0.7463171848035415,
      "grad_norm": 1.9420994520187378,
      "learning_rate": 4.2743304174806295e-05,
      "loss": 0.0542,
      "num_input_tokens_seen": 12396896,
      "step": 3730
    },
    {
      "epoch": 0.7473176099842433,
      "grad_norm": 0.14084391295909882,
      "learning_rate": 4.2724840341529936e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 12413472,
      "step": 3735
    },
    {
      "epoch": 0.7483180351649451,
      "grad_norm": 0.19236116111278534,
      "learning_rate": 4.270635704744693e-05,
      "loss": 0.0285,
      "num_input_tokens_seen": 12429760,
      "step": 3740
    },
    {
      "epoch": 0.7493184603456469,
      "grad_norm": 0.2585514485836029,
      "learning_rate": 4.268785431285081e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 12446816,
      "step": 3745
    },
    {
      "epoch": 0.7503188855263487,
      "grad_norm": 0.4992278218269348,
      "learning_rate": 4.266933215805645e-05,
      "loss": 0.0154,
      "num_input_tokens_seen": 12462464,
      "step": 3750
    },
    {
      "epoch": 0.7513193107070505,
      "grad_norm": 0.556191086769104,
      "learning_rate": 4.2650790603400096e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 12478592,
      "step": 3755
    },
    {
      "epoch": 0.7523197358877523,
      "grad_norm": 0.7319716215133667,
      "learning_rate": 4.263222966923923e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 12495936,
      "step": 3760
    },
    {
      "epoch": 0.7533201610684541,
      "grad_norm": 0.29031893610954285,
      "learning_rate": 4.261364937595265e-05,
      "loss": 0.0574,
      "num_input_tokens_seen": 12512992,
      "step": 3765
    },
    {
      "epoch": 0.7543205862491559,
      "grad_norm": 0.34786590933799744,
      "learning_rate": 4.2595049743940386e-05,
      "loss": 0.0243,
      "num_input_tokens_seen": 12529504,
      "step": 3770
    },
    {
      "epoch": 0.7553210114298576,
      "grad_norm": 0.21003863215446472,
      "learning_rate": 4.257643079362372e-05,
      "loss": 0.0462,
      "num_input_tokens_seen": 12547072,
      "step": 3775
    },
    {
      "epoch": 0.7563214366105595,
      "grad_norm": 0.4604493975639343,
      "learning_rate": 4.255779254544515e-05,
      "loss": 0.0226,
      "num_input_tokens_seen": 12563712,
      "step": 3780
    },
    {
      "epoch": 0.7573218617912613,
      "grad_norm": 0.5216447114944458,
      "learning_rate": 4.253913501986834e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 12580544,
      "step": 3785
    },
    {
      "epoch": 0.7583222869719631,
      "grad_norm": 0.39751800894737244,
      "learning_rate": 4.252045823737812e-05,
      "loss": 0.0335,
      "num_input_tokens_seen": 12596832,
      "step": 3790
    },
    {
      "epoch": 0.7593227121526649,
      "grad_norm": 0.2205200493335724,
      "learning_rate": 4.250176221848049e-05,
      "loss": 0.0424,
      "num_input_tokens_seen": 12614656,
      "step": 3795
    },
    {
      "epoch": 0.7603231373333666,
      "grad_norm": 0.404427170753479,
      "learning_rate": 4.248304698370253e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 12631392,
      "step": 3800
    },
    {
      "epoch": 0.7613235625140685,
      "grad_norm": 0.8761758804321289,
      "learning_rate": 4.2464312553592456e-05,
      "loss": 0.0381,
      "num_input_tokens_seen": 12648768,
      "step": 3805
    },
    {
      "epoch": 0.7623239876947703,
      "grad_norm": 0.7704747319221497,
      "learning_rate": 4.244555894871954e-05,
      "loss": 0.0398,
      "num_input_tokens_seen": 12664736,
      "step": 3810
    },
    {
      "epoch": 0.763324412875472,
      "grad_norm": 0.2651312053203583,
      "learning_rate": 4.2426786189674116e-05,
      "loss": 0.0293,
      "num_input_tokens_seen": 12679712,
      "step": 3815
    },
    {
      "epoch": 0.7643248380561739,
      "grad_norm": 0.5004244446754456,
      "learning_rate": 4.2407994297067534e-05,
      "loss": 0.0598,
      "num_input_tokens_seen": 12697792,
      "step": 3820
    },
    {
      "epoch": 0.7653252632368757,
      "grad_norm": 0.3987187147140503,
      "learning_rate": 4.238918329153215e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 12714080,
      "step": 3825
    },
    {
      "epoch": 0.7663256884175774,
      "grad_norm": 0.35544225573539734,
      "learning_rate": 4.2370353193721336e-05,
      "loss": 0.02,
      "num_input_tokens_seen": 12729504,
      "step": 3830
    },
    {
      "epoch": 0.7673261135982793,
      "grad_norm": 0.6649848222732544,
      "learning_rate": 4.23515040243094e-05,
      "loss": 0.0341,
      "num_input_tokens_seen": 12746048,
      "step": 3835
    },
    {
      "epoch": 0.768326538778981,
      "grad_norm": 1.0058033466339111,
      "learning_rate": 4.2332635803991575e-05,
      "loss": 0.0267,
      "num_input_tokens_seen": 12763040,
      "step": 3840
    },
    {
      "epoch": 0.7693269639596829,
      "grad_norm": 0.18642783164978027,
      "learning_rate": 4.2313748553484056e-05,
      "loss": 0.0268,
      "num_input_tokens_seen": 12778784,
      "step": 3845
    },
    {
      "epoch": 0.7703273891403847,
      "grad_norm": 0.3343425393104553,
      "learning_rate": 4.229484229352388e-05,
      "loss": 0.0348,
      "num_input_tokens_seen": 12795264,
      "step": 3850
    },
    {
      "epoch": 0.7713278143210864,
      "grad_norm": 0.22427132725715637,
      "learning_rate": 4.2275917044869015e-05,
      "loss": 0.0389,
      "num_input_tokens_seen": 12812032,
      "step": 3855
    },
    {
      "epoch": 0.7723282395017883,
      "grad_norm": 0.2122434675693512,
      "learning_rate": 4.2256972828298214e-05,
      "loss": 0.0137,
      "num_input_tokens_seen": 12829440,
      "step": 3860
    },
    {
      "epoch": 0.7733286646824901,
      "grad_norm": 0.7551167607307434,
      "learning_rate": 4.2238009664611096e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 12845952,
      "step": 3865
    },
    {
      "epoch": 0.7743290898631918,
      "grad_norm": 0.5768938064575195,
      "learning_rate": 4.221902757462807e-05,
      "loss": 0.0353,
      "num_input_tokens_seen": 12862848,
      "step": 3870
    },
    {
      "epoch": 0.7753295150438937,
      "grad_norm": 0.44510433077812195,
      "learning_rate": 4.220002657919032e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 12879424,
      "step": 3875
    },
    {
      "epoch": 0.7763299402245954,
      "grad_norm": 0.992270827293396,
      "learning_rate": 4.218100669915982e-05,
      "loss": 0.0444,
      "num_input_tokens_seen": 12895968,
      "step": 3880
    },
    {
      "epoch": 0.7773303654052972,
      "grad_norm": 0.4635940492153168,
      "learning_rate": 4.2161967955419226e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 12912704,
      "step": 3885
    },
    {
      "epoch": 0.7783307905859991,
      "grad_norm": 0.28763484954833984,
      "learning_rate": 4.214291036887194e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 12929600,
      "step": 3890
    },
    {
      "epoch": 0.7793312157667008,
      "grad_norm": 0.44379696249961853,
      "learning_rate": 4.212383396044204e-05,
      "loss": 0.0319,
      "num_input_tokens_seen": 12945280,
      "step": 3895
    },
    {
      "epoch": 0.7803316409474026,
      "grad_norm": 0.8486232757568359,
      "learning_rate": 4.210473875107428e-05,
      "loss": 0.0622,
      "num_input_tokens_seen": 12962048,
      "step": 3900
    },
    {
      "epoch": 0.7813320661281045,
      "grad_norm": 0.3712434470653534,
      "learning_rate": 4.208562476173404e-05,
      "loss": 0.0239,
      "num_input_tokens_seen": 12977888,
      "step": 3905
    },
    {
      "epoch": 0.7823324913088062,
      "grad_norm": 0.13711759448051453,
      "learning_rate": 4.206649201340735e-05,
      "loss": 0.0243,
      "num_input_tokens_seen": 12994752,
      "step": 3910
    },
    {
      "epoch": 0.7833329164895081,
      "grad_norm": 0.23532859981060028,
      "learning_rate": 4.2047340527100786e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 13011776,
      "step": 3915
    },
    {
      "epoch": 0.7843333416702098,
      "grad_norm": 0.6952040195465088,
      "learning_rate": 4.202817032384155e-05,
      "loss": 0.0422,
      "num_input_tokens_seen": 13028352,
      "step": 3920
    },
    {
      "epoch": 0.7853337668509116,
      "grad_norm": 0.4595640003681183,
      "learning_rate": 4.2008981424677354e-05,
      "loss": 0.0513,
      "num_input_tokens_seen": 13044704,
      "step": 3925
    },
    {
      "epoch": 0.7863341920316135,
      "grad_norm": 0.35172584652900696,
      "learning_rate": 4.198977385067645e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 13061952,
      "step": 3930
    },
    {
      "epoch": 0.7873346172123152,
      "grad_norm": 0.1835983693599701,
      "learning_rate": 4.1970547622927615e-05,
      "loss": 0.035,
      "num_input_tokens_seen": 13077760,
      "step": 3935
    },
    {
      "epoch": 0.788335042393017,
      "grad_norm": 0.37188035249710083,
      "learning_rate": 4.1951302762540074e-05,
      "loss": 0.0202,
      "num_input_tokens_seen": 13094912,
      "step": 3940
    },
    {
      "epoch": 0.7893354675737189,
      "grad_norm": 0.17795595526695251,
      "learning_rate": 4.193203929064353e-05,
      "loss": 0.0132,
      "num_input_tokens_seen": 13111520,
      "step": 3945
    },
    {
      "epoch": 0.7903358927544206,
      "grad_norm": 0.6609641909599304,
      "learning_rate": 4.191275722838811e-05,
      "loss": 0.0494,
      "num_input_tokens_seen": 13128960,
      "step": 3950
    },
    {
      "epoch": 0.7913363179351224,
      "grad_norm": 0.22620953619480133,
      "learning_rate": 4.189345659694436e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 13146656,
      "step": 3955
    },
    {
      "epoch": 0.7923367431158242,
      "grad_norm": 0.7801216244697571,
      "learning_rate": 4.187413741750322e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 13163104,
      "step": 3960
    },
    {
      "epoch": 0.793337168296526,
      "grad_norm": 0.884833037853241,
      "learning_rate": 4.185479971127596e-05,
      "loss": 0.0286,
      "num_input_tokens_seen": 13179744,
      "step": 3965
    },
    {
      "epoch": 0.7943375934772279,
      "grad_norm": 0.5032994151115417,
      "learning_rate": 4.1835443499494245e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 13197504,
      "step": 3970
    },
    {
      "epoch": 0.7953380186579296,
      "grad_norm": 1.011844277381897,
      "learning_rate": 4.181606880341001e-05,
      "loss": 0.042,
      "num_input_tokens_seen": 13213824,
      "step": 3975
    },
    {
      "epoch": 0.7963384438386314,
      "grad_norm": 0.6795077323913574,
      "learning_rate": 4.179667564429551e-05,
      "loss": 0.028,
      "num_input_tokens_seen": 13230656,
      "step": 3980
    },
    {
      "epoch": 0.7973388690193333,
      "grad_norm": 0.35327115654945374,
      "learning_rate": 4.177726404344326e-05,
      "loss": 0.0344,
      "num_input_tokens_seen": 13246944,
      "step": 3985
    },
    {
      "epoch": 0.798339294200035,
      "grad_norm": 0.2503950893878937,
      "learning_rate": 4.1757834022166034e-05,
      "loss": 0.0305,
      "num_input_tokens_seen": 13263872,
      "step": 3990
    },
    {
      "epoch": 0.7993397193807368,
      "grad_norm": 0.27673956751823425,
      "learning_rate": 4.173838560179682e-05,
      "loss": 0.051,
      "num_input_tokens_seen": 13280032,
      "step": 3995
    },
    {
      "epoch": 0.8003401445614386,
      "grad_norm": 0.47994551062583923,
      "learning_rate": 4.171891880368882e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 13296128,
      "step": 4000
    },
    {
      "epoch": 0.8013405697421404,
      "grad_norm": 0.5603805184364319,
      "learning_rate": 4.16994336492154e-05,
      "loss": 0.0375,
      "num_input_tokens_seen": 13312704,
      "step": 4005
    },
    {
      "epoch": 0.8023409949228422,
      "grad_norm": 0.2629021406173706,
      "learning_rate": 4.167993015977009e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 13329920,
      "step": 4010
    },
    {
      "epoch": 0.803341420103544,
      "grad_norm": 0.29546400904655457,
      "learning_rate": 4.166040835676653e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 13346976,
      "step": 4015
    },
    {
      "epoch": 0.8043418452842458,
      "grad_norm": 0.349707692861557,
      "learning_rate": 4.1640868261638505e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 13364512,
      "step": 4020
    },
    {
      "epoch": 0.8053422704649476,
      "grad_norm": 0.8134015798568726,
      "learning_rate": 4.162130989583985e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 13381696,
      "step": 4025
    },
    {
      "epoch": 0.8063426956456494,
      "grad_norm": 0.6181686520576477,
      "learning_rate": 4.160173328084447e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 13398976,
      "step": 4030
    },
    {
      "epoch": 0.8073431208263512,
      "grad_norm": 0.3832380473613739,
      "learning_rate": 4.158213843814631e-05,
      "loss": 0.0215,
      "num_input_tokens_seen": 13416032,
      "step": 4035
    },
    {
      "epoch": 0.808343546007053,
      "grad_norm": 0.44530972838401794,
      "learning_rate": 4.156252538925933e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 13432352,
      "step": 4040
    },
    {
      "epoch": 0.8093439711877548,
      "grad_norm": 0.5037532448768616,
      "learning_rate": 4.154289415571749e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 13449472,
      "step": 4045
    },
    {
      "epoch": 0.8103443963684566,
      "grad_norm": 0.2083626538515091,
      "learning_rate": 4.152324475907468e-05,
      "loss": 0.0348,
      "num_input_tokens_seen": 13465184,
      "step": 4050
    },
    {
      "epoch": 0.8113448215491584,
      "grad_norm": 0.7969433665275574,
      "learning_rate": 4.1503577220904747e-05,
      "loss": 0.0309,
      "num_input_tokens_seen": 13481312,
      "step": 4055
    },
    {
      "epoch": 0.8123452467298602,
      "grad_norm": 0.5527857542037964,
      "learning_rate": 4.148389156280149e-05,
      "loss": 0.0498,
      "num_input_tokens_seen": 13498944,
      "step": 4060
    },
    {
      "epoch": 0.813345671910562,
      "grad_norm": 0.28512829542160034,
      "learning_rate": 4.146418780637855e-05,
      "loss": 0.0264,
      "num_input_tokens_seen": 13514528,
      "step": 4065
    },
    {
      "epoch": 0.8143460970912638,
      "grad_norm": 0.4216727018356323,
      "learning_rate": 4.144446597326948e-05,
      "loss": 0.0388,
      "num_input_tokens_seen": 13532128,
      "step": 4070
    },
    {
      "epoch": 0.8153465222719656,
      "grad_norm": 0.38498228788375854,
      "learning_rate": 4.1424726085127666e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 13548320,
      "step": 4075
    },
    {
      "epoch": 0.8163469474526673,
      "grad_norm": 0.8461599349975586,
      "learning_rate": 4.140496816362631e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 13564320,
      "step": 4080
    },
    {
      "epoch": 0.8173473726333692,
      "grad_norm": 1.0074468851089478,
      "learning_rate": 4.1385192230458417e-05,
      "loss": 0.0312,
      "num_input_tokens_seen": 13579904,
      "step": 4085
    },
    {
      "epoch": 0.818347797814071,
      "grad_norm": 0.4712158739566803,
      "learning_rate": 4.136539830733677e-05,
      "loss": 0.0453,
      "num_input_tokens_seen": 13596832,
      "step": 4090
    },
    {
      "epoch": 0.8193482229947727,
      "grad_norm": 0.4534381628036499,
      "learning_rate": 4.13455864159939e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 13613920,
      "step": 4095
    },
    {
      "epoch": 0.8203486481754746,
      "grad_norm": 0.1941436380147934,
      "learning_rate": 4.1325756578182084e-05,
      "loss": 0.0408,
      "num_input_tokens_seen": 13631232,
      "step": 4100
    },
    {
      "epoch": 0.8213490733561764,
      "grad_norm": 0.18202431499958038,
      "learning_rate": 4.130590881567328e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 13647392,
      "step": 4105
    },
    {
      "epoch": 0.8223494985368782,
      "grad_norm": 0.4083399176597595,
      "learning_rate": 4.1286043150259134e-05,
      "loss": 0.0398,
      "num_input_tokens_seen": 13663872,
      "step": 4110
    },
    {
      "epoch": 0.82334992371758,
      "grad_norm": 0.29242146015167236,
      "learning_rate": 4.126615960375095e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 13679616,
      "step": 4115
    },
    {
      "epoch": 0.8243503488982817,
      "grad_norm": 0.19398359954357147,
      "learning_rate": 4.124625819797967e-05,
      "loss": 0.0431,
      "num_input_tokens_seen": 13696416,
      "step": 4120
    },
    {
      "epoch": 0.8253507740789836,
      "grad_norm": 0.20560583472251892,
      "learning_rate": 4.122633895479584e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 13713088,
      "step": 4125
    },
    {
      "epoch": 0.8263511992596854,
      "grad_norm": 0.551224410533905,
      "learning_rate": 4.120640189606958e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 13730080,
      "step": 4130
    },
    {
      "epoch": 0.8273516244403871,
      "grad_norm": 0.5998563170433044,
      "learning_rate": 4.118644704369057e-05,
      "loss": 0.0246,
      "num_input_tokens_seen": 13746272,
      "step": 4135
    },
    {
      "epoch": 0.828352049621089,
      "grad_norm": 0.410586953163147,
      "learning_rate": 4.1166474419568055e-05,
      "loss": 0.0608,
      "num_input_tokens_seen": 13762656,
      "step": 4140
    },
    {
      "epoch": 0.8293524748017908,
      "grad_norm": 1.009325623512268,
      "learning_rate": 4.114648404563076e-05,
      "loss": 0.0401,
      "num_input_tokens_seen": 13779584,
      "step": 4145
    },
    {
      "epoch": 0.8303528999824925,
      "grad_norm": 0.4814990758895874,
      "learning_rate": 4.1126475943826924e-05,
      "loss": 0.043,
      "num_input_tokens_seen": 13795648,
      "step": 4150
    },
    {
      "epoch": 0.8313533251631944,
      "grad_norm": 0.48686766624450684,
      "learning_rate": 4.110645013612423e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 13812416,
      "step": 4155
    },
    {
      "epoch": 0.8323537503438961,
      "grad_norm": 0.3141302168369293,
      "learning_rate": 4.10864066445098e-05,
      "loss": 0.0188,
      "num_input_tokens_seen": 13828864,
      "step": 4160
    },
    {
      "epoch": 0.833354175524598,
      "grad_norm": 0.38863256573677063,
      "learning_rate": 4.1066345490990197e-05,
      "loss": 0.0436,
      "num_input_tokens_seen": 13844768,
      "step": 4165
    },
    {
      "epoch": 0.8343546007052998,
      "grad_norm": 0.7164223194122314,
      "learning_rate": 4.104626669759134e-05,
      "loss": 0.0319,
      "num_input_tokens_seen": 13861216,
      "step": 4170
    },
    {
      "epoch": 0.8353550258860015,
      "grad_norm": 0.37101924419403076,
      "learning_rate": 4.1026170286358544e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 13877888,
      "step": 4175
    },
    {
      "epoch": 0.8363554510667034,
      "grad_norm": 0.3636489510536194,
      "learning_rate": 4.100605627935647e-05,
      "loss": 0.0268,
      "num_input_tokens_seen": 13894336,
      "step": 4180
    },
    {
      "epoch": 0.8373558762474052,
      "grad_norm": 0.8379486203193665,
      "learning_rate": 4.098592469866906e-05,
      "loss": 0.0376,
      "num_input_tokens_seen": 13911360,
      "step": 4185
    },
    {
      "epoch": 0.8383563014281069,
      "grad_norm": 0.40897300839424133,
      "learning_rate": 4.0965775566399593e-05,
      "loss": 0.0222,
      "num_input_tokens_seen": 13927936,
      "step": 4190
    },
    {
      "epoch": 0.8393567266088088,
      "grad_norm": 0.8142618536949158,
      "learning_rate": 4.09456089046706e-05,
      "loss": 0.0574,
      "num_input_tokens_seen": 13945536,
      "step": 4195
    },
    {
      "epoch": 0.8403571517895105,
      "grad_norm": 0.18752454221248627,
      "learning_rate": 4.092542473562386e-05,
      "loss": 0.0598,
      "num_input_tokens_seen": 13962592,
      "step": 4200
    },
    {
      "epoch": 0.8413575769702123,
      "grad_norm": 0.6369066834449768,
      "learning_rate": 4.090522308142038e-05,
      "loss": 0.0315,
      "num_input_tokens_seen": 13979776,
      "step": 4205
    },
    {
      "epoch": 0.8423580021509142,
      "grad_norm": 0.27474310994148254,
      "learning_rate": 4.088500396424033e-05,
      "loss": 0.0249,
      "num_input_tokens_seen": 13996192,
      "step": 4210
    },
    {
      "epoch": 0.8433584273316159,
      "grad_norm": 0.548443615436554,
      "learning_rate": 4.086476740628312e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 14012096,
      "step": 4215
    },
    {
      "epoch": 0.8443588525123177,
      "grad_norm": 0.49320584535598755,
      "learning_rate": 4.084451342976726e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 14028832,
      "step": 4220
    },
    {
      "epoch": 0.8453592776930196,
      "grad_norm": 0.5967304110527039,
      "learning_rate": 4.082424205693038e-05,
      "loss": 0.0366,
      "num_input_tokens_seen": 14046336,
      "step": 4225
    },
    {
      "epoch": 0.8463597028737213,
      "grad_norm": 0.21384429931640625,
      "learning_rate": 4.080395331002925e-05,
      "loss": 0.0308,
      "num_input_tokens_seen": 14062304,
      "step": 4230
    },
    {
      "epoch": 0.8473601280544232,
      "grad_norm": 0.6950636506080627,
      "learning_rate": 4.078364721133967e-05,
      "loss": 0.0457,
      "num_input_tokens_seen": 14078368,
      "step": 4235
    },
    {
      "epoch": 0.8483605532351249,
      "grad_norm": 0.4137358069419861,
      "learning_rate": 4.0763323783156525e-05,
      "loss": 0.026,
      "num_input_tokens_seen": 14094944,
      "step": 4240
    },
    {
      "epoch": 0.8493609784158267,
      "grad_norm": 0.5785394906997681,
      "learning_rate": 4.074298304779371e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 14112384,
      "step": 4245
    },
    {
      "epoch": 0.8503614035965286,
      "grad_norm": 0.6416853070259094,
      "learning_rate": 4.0722625027584126e-05,
      "loss": 0.0321,
      "num_input_tokens_seen": 14129536,
      "step": 4250
    },
    {
      "epoch": 0.8513618287772303,
      "grad_norm": 0.499644011259079,
      "learning_rate": 4.070224974487966e-05,
      "loss": 0.0286,
      "num_input_tokens_seen": 14146976,
      "step": 4255
    },
    {
      "epoch": 0.8523622539579321,
      "grad_norm": 0.7566441893577576,
      "learning_rate": 4.0681857222051134e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 14163168,
      "step": 4260
    },
    {
      "epoch": 0.853362679138634,
      "grad_norm": 0.3304438292980194,
      "learning_rate": 4.066144748148832e-05,
      "loss": 0.037,
      "num_input_tokens_seen": 14180416,
      "step": 4265
    },
    {
      "epoch": 0.8543631043193357,
      "grad_norm": 0.24730849266052246,
      "learning_rate": 4.0641020545599875e-05,
      "loss": 0.0272,
      "num_input_tokens_seen": 14197472,
      "step": 4270
    },
    {
      "epoch": 0.8553635295000375,
      "grad_norm": 1.0695536136627197,
      "learning_rate": 4.062057643681335e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 14214336,
      "step": 4275
    },
    {
      "epoch": 0.8563639546807393,
      "grad_norm": 0.5582004189491272,
      "learning_rate": 4.0600115177575144e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 14230400,
      "step": 4280
    },
    {
      "epoch": 0.8573643798614411,
      "grad_norm": 0.5465623140335083,
      "learning_rate": 4.057963679035048e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 14247136,
      "step": 4285
    },
    {
      "epoch": 0.858364805042143,
      "grad_norm": 0.5054659843444824,
      "learning_rate": 4.055914129762339e-05,
      "loss": 0.044,
      "num_input_tokens_seen": 14265056,
      "step": 4290
    },
    {
      "epoch": 0.8593652302228447,
      "grad_norm": 0.44049715995788574,
      "learning_rate": 4.053862872189671e-05,
      "loss": 0.0246,
      "num_input_tokens_seen": 14281696,
      "step": 4295
    },
    {
      "epoch": 0.8603656554035465,
      "grad_norm": 1.1087952852249146,
      "learning_rate": 4.0518099085692e-05,
      "loss": 0.0559,
      "num_input_tokens_seen": 14299200,
      "step": 4300
    },
    {
      "epoch": 0.8613660805842484,
      "grad_norm": 0.21384133398532867,
      "learning_rate": 4.049755241154955e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 14316608,
      "step": 4305
    },
    {
      "epoch": 0.8623665057649501,
      "grad_norm": 0.3457808792591095,
      "learning_rate": 4.047698872202839e-05,
      "loss": 0.0212,
      "num_input_tokens_seen": 14332992,
      "step": 4310
    },
    {
      "epoch": 0.8633669309456519,
      "grad_norm": 0.172733336687088,
      "learning_rate": 4.0456408039706204e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 14349216,
      "step": 4315
    },
    {
      "epoch": 0.8643673561263537,
      "grad_norm": 0.5303016304969788,
      "learning_rate": 4.043581038717934e-05,
      "loss": 0.038,
      "num_input_tokens_seen": 14367488,
      "step": 4320
    },
    {
      "epoch": 0.8653677813070555,
      "grad_norm": 0.3045515716075897,
      "learning_rate": 4.041519578706279e-05,
      "loss": 0.0265,
      "num_input_tokens_seen": 14385280,
      "step": 4325
    },
    {
      "epoch": 0.8663682064877573,
      "grad_norm": 0.18472611904144287,
      "learning_rate": 4.039456426199012e-05,
      "loss": 0.0412,
      "num_input_tokens_seen": 14402528,
      "step": 4330
    },
    {
      "epoch": 0.8673686316684591,
      "grad_norm": 0.5062572956085205,
      "learning_rate": 4.037391583461353e-05,
      "loss": 0.0209,
      "num_input_tokens_seen": 14420160,
      "step": 4335
    },
    {
      "epoch": 0.8683690568491609,
      "grad_norm": 0.6826553344726562,
      "learning_rate": 4.0353250527603726e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 14437824,
      "step": 4340
    },
    {
      "epoch": 0.8693694820298626,
      "grad_norm": 0.11310242116451263,
      "learning_rate": 4.0332568363649974e-05,
      "loss": 0.0559,
      "num_input_tokens_seen": 14455264,
      "step": 4345
    },
    {
      "epoch": 0.8703699072105645,
      "grad_norm": 0.291314035654068,
      "learning_rate": 4.0311869365460054e-05,
      "loss": 0.028,
      "num_input_tokens_seen": 14470944,
      "step": 4350
    },
    {
      "epoch": 0.8713703323912663,
      "grad_norm": 0.6921257972717285,
      "learning_rate": 4.029115355576022e-05,
      "loss": 0.0204,
      "num_input_tokens_seen": 14486784,
      "step": 4355
    },
    {
      "epoch": 0.8723707575719681,
      "grad_norm": 0.3011789321899414,
      "learning_rate": 4.027042095729517e-05,
      "loss": 0.0439,
      "num_input_tokens_seen": 14503392,
      "step": 4360
    },
    {
      "epoch": 0.8733711827526699,
      "grad_norm": 0.09817075729370117,
      "learning_rate": 4.024967159282805e-05,
      "loss": 0.0226,
      "num_input_tokens_seen": 14520000,
      "step": 4365
    },
    {
      "epoch": 0.8743716079333717,
      "grad_norm": 0.28903499245643616,
      "learning_rate": 4.0228905485140415e-05,
      "loss": 0.0106,
      "num_input_tokens_seen": 14535648,
      "step": 4370
    },
    {
      "epoch": 0.8753720331140735,
      "grad_norm": 0.3278878927230835,
      "learning_rate": 4.020812265703221e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 14552096,
      "step": 4375
    },
    {
      "epoch": 0.8763724582947753,
      "grad_norm": 0.8860204219818115,
      "learning_rate": 4.018732313132171e-05,
      "loss": 0.04,
      "num_input_tokens_seen": 14568736,
      "step": 4380
    },
    {
      "epoch": 0.877372883475477,
      "grad_norm": 0.23948296904563904,
      "learning_rate": 4.016650693084555e-05,
      "loss": 0.0361,
      "num_input_tokens_seen": 14586208,
      "step": 4385
    },
    {
      "epoch": 0.8783733086561789,
      "grad_norm": 0.1991083174943924,
      "learning_rate": 4.014567407845866e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 14602752,
      "step": 4390
    },
    {
      "epoch": 0.8793737338368807,
      "grad_norm": 0.6444497108459473,
      "learning_rate": 4.0124824597034274e-05,
      "loss": 0.0399,
      "num_input_tokens_seen": 14619360,
      "step": 4395
    },
    {
      "epoch": 0.8803741590175824,
      "grad_norm": 0.44131720066070557,
      "learning_rate": 4.0103958509463835e-05,
      "loss": 0.0316,
      "num_input_tokens_seen": 14636256,
      "step": 4400
    },
    {
      "epoch": 0.8813745841982843,
      "grad_norm": 0.23221592605113983,
      "learning_rate": 4.008307583865708e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 14652992,
      "step": 4405
    },
    {
      "epoch": 0.882375009378986,
      "grad_norm": 0.5025535225868225,
      "learning_rate": 4.00621766075419e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 14669632,
      "step": 4410
    },
    {
      "epoch": 0.8833754345596879,
      "grad_norm": 0.27033570408821106,
      "learning_rate": 4.004126083906441e-05,
      "loss": 0.0296,
      "num_input_tokens_seen": 14685440,
      "step": 4415
    },
    {
      "epoch": 0.8843758597403897,
      "grad_norm": 0.0698275938630104,
      "learning_rate": 4.002032855618885e-05,
      "loss": 0.0109,
      "num_input_tokens_seen": 14702240,
      "step": 4420
    },
    {
      "epoch": 0.8853762849210914,
      "grad_norm": 1.5025676488876343,
      "learning_rate": 3.999937978189761e-05,
      "loss": 0.0513,
      "num_input_tokens_seen": 14719072,
      "step": 4425
    },
    {
      "epoch": 0.8863767101017933,
      "grad_norm": 0.9077886343002319,
      "learning_rate": 3.9978414539191184e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 14737344,
      "step": 4430
    },
    {
      "epoch": 0.8873771352824951,
      "grad_norm": 0.20032626390457153,
      "learning_rate": 3.995743285108814e-05,
      "loss": 0.0327,
      "num_input_tokens_seen": 14754432,
      "step": 4435
    },
    {
      "epoch": 0.8883775604631968,
      "grad_norm": 0.7127216458320618,
      "learning_rate": 3.993643474062512e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 14771008,
      "step": 4440
    },
    {
      "epoch": 0.8893779856438987,
      "grad_norm": 0.3725959360599518,
      "learning_rate": 3.991542023085676e-05,
      "loss": 0.0155,
      "num_input_tokens_seen": 14788352,
      "step": 4445
    },
    {
      "epoch": 0.8903784108246005,
      "grad_norm": 0.587356448173523,
      "learning_rate": 3.9894389344855755e-05,
      "loss": 0.0648,
      "num_input_tokens_seen": 14804928,
      "step": 4450
    },
    {
      "epoch": 0.8913788360053022,
      "grad_norm": 0.41772809624671936,
      "learning_rate": 3.9873342105712725e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 14821312,
      "step": 4455
    },
    {
      "epoch": 0.8923792611860041,
      "grad_norm": 0.6791223287582397,
      "learning_rate": 3.985227853653628e-05,
      "loss": 0.0475,
      "num_input_tokens_seen": 14837504,
      "step": 4460
    },
    {
      "epoch": 0.8933796863667058,
      "grad_norm": 0.6203396320343018,
      "learning_rate": 3.983119866045296e-05,
      "loss": 0.0505,
      "num_input_tokens_seen": 14855584,
      "step": 4465
    },
    {
      "epoch": 0.8943801115474076,
      "grad_norm": 1.6683917045593262,
      "learning_rate": 3.9810102500607184e-05,
      "loss": 0.0485,
      "num_input_tokens_seen": 14872640,
      "step": 4470
    },
    {
      "epoch": 0.8953805367281095,
      "grad_norm": 0.2774893343448639,
      "learning_rate": 3.978899008016127e-05,
      "loss": 0.044,
      "num_input_tokens_seen": 14890848,
      "step": 4475
    },
    {
      "epoch": 0.8963809619088112,
      "grad_norm": 0.5847212672233582,
      "learning_rate": 3.976786142229538e-05,
      "loss": 0.043,
      "num_input_tokens_seen": 14906144,
      "step": 4480
    },
    {
      "epoch": 0.8973813870895131,
      "grad_norm": 0.14882434904575348,
      "learning_rate": 3.974671655020752e-05,
      "loss": 0.0396,
      "num_input_tokens_seen": 14922400,
      "step": 4485
    },
    {
      "epoch": 0.8983818122702149,
      "grad_norm": 0.16097332537174225,
      "learning_rate": 3.972555548711347e-05,
      "loss": 0.0185,
      "num_input_tokens_seen": 14939008,
      "step": 4490
    },
    {
      "epoch": 0.8993822374509166,
      "grad_norm": 0.20941989123821259,
      "learning_rate": 3.9704378256246805e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 14956256,
      "step": 4495
    },
    {
      "epoch": 0.9003826626316185,
      "grad_norm": 0.7076755166053772,
      "learning_rate": 3.968318488085886e-05,
      "loss": 0.0432,
      "num_input_tokens_seen": 14971968,
      "step": 4500
    },
    {
      "epoch": 0.9013830878123202,
      "grad_norm": 0.6078043580055237,
      "learning_rate": 3.9661975384218664e-05,
      "loss": 0.0355,
      "num_input_tokens_seen": 14988800,
      "step": 4505
    },
    {
      "epoch": 0.902383512993022,
      "grad_norm": 0.481460303068161,
      "learning_rate": 3.9640749789612986e-05,
      "loss": 0.029,
      "num_input_tokens_seen": 15005120,
      "step": 4510
    },
    {
      "epoch": 0.9033839381737239,
      "grad_norm": 0.4094735085964203,
      "learning_rate": 3.9619508120346235e-05,
      "loss": 0.0356,
      "num_input_tokens_seen": 15022176,
      "step": 4515
    },
    {
      "epoch": 0.9043843633544256,
      "grad_norm": 0.18756358325481415,
      "learning_rate": 3.95982503997405e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 15038400,
      "step": 4520
    },
    {
      "epoch": 0.9053847885351274,
      "grad_norm": 0.2987261116504669,
      "learning_rate": 3.957697665113547e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 15054624,
      "step": 4525
    },
    {
      "epoch": 0.9063852137158293,
      "grad_norm": 0.13014473021030426,
      "learning_rate": 3.9555686897888426e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 15070848,
      "step": 4530
    },
    {
      "epoch": 0.907385638896531,
      "grad_norm": 0.504302442073822,
      "learning_rate": 3.953438116337425e-05,
      "loss": 0.0464,
      "num_input_tokens_seen": 15087936,
      "step": 4535
    },
    {
      "epoch": 0.9083860640772328,
      "grad_norm": 0.47447165846824646,
      "learning_rate": 3.951305947098535e-05,
      "loss": 0.0302,
      "num_input_tokens_seen": 15104512,
      "step": 4540
    },
    {
      "epoch": 0.9093864892579346,
      "grad_norm": 0.5051746368408203,
      "learning_rate": 3.949172184413166e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 15120064,
      "step": 4545
    },
    {
      "epoch": 0.9103869144386364,
      "grad_norm": 0.5719951391220093,
      "learning_rate": 3.9470368306240615e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 15136000,
      "step": 4550
    },
    {
      "epoch": 0.9113873396193383,
      "grad_norm": 0.32244595885276794,
      "learning_rate": 3.944899888075709e-05,
      "loss": 0.0392,
      "num_input_tokens_seen": 15152704,
      "step": 4555
    },
    {
      "epoch": 0.91238776480004,
      "grad_norm": 0.2153882533311844,
      "learning_rate": 3.942761359114345e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 15170176,
      "step": 4560
    },
    {
      "epoch": 0.9133881899807418,
      "grad_norm": 0.40789932012557983,
      "learning_rate": 3.940621246087946e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 15186656,
      "step": 4565
    },
    {
      "epoch": 0.9143886151614437,
      "grad_norm": 0.38933348655700684,
      "learning_rate": 3.9384795513462234e-05,
      "loss": 0.0465,
      "num_input_tokens_seen": 15202592,
      "step": 4570
    },
    {
      "epoch": 0.9153890403421454,
      "grad_norm": 0.19240064918994904,
      "learning_rate": 3.936336277240633e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 15219904,
      "step": 4575
    },
    {
      "epoch": 0.9163894655228472,
      "grad_norm": 0.5518398880958557,
      "learning_rate": 3.9341914261243584e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 15236608,
      "step": 4580
    },
    {
      "epoch": 0.917389890703549,
      "grad_norm": 0.6835047602653503,
      "learning_rate": 3.932045000352318e-05,
      "loss": 0.0397,
      "num_input_tokens_seen": 15254016,
      "step": 4585
    },
    {
      "epoch": 0.9183903158842508,
      "grad_norm": 0.8669648170471191,
      "learning_rate": 3.9298970022811575e-05,
      "loss": 0.0398,
      "num_input_tokens_seen": 15270624,
      "step": 4590
    },
    {
      "epoch": 0.9193907410649526,
      "grad_norm": 0.37144604325294495,
      "learning_rate": 3.927747434269249e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 15287392,
      "step": 4595
    },
    {
      "epoch": 0.9203911662456544,
      "grad_norm": 1.0355428457260132,
      "learning_rate": 3.925596298676689e-05,
      "loss": 0.0466,
      "num_input_tokens_seen": 15304928,
      "step": 4600
    },
    {
      "epoch": 0.9213915914263562,
      "grad_norm": 0.20743988454341888,
      "learning_rate": 3.923443597865295e-05,
      "loss": 0.0258,
      "num_input_tokens_seen": 15322208,
      "step": 4605
    },
    {
      "epoch": 0.922392016607058,
      "grad_norm": 0.384706050157547,
      "learning_rate": 3.921289334198602e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 15338560,
      "step": 4610
    },
    {
      "epoch": 0.9233924417877598,
      "grad_norm": 0.7780063152313232,
      "learning_rate": 3.919133510041863e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 15356608,
      "step": 4615
    },
    {
      "epoch": 0.9243928669684616,
      "grad_norm": 1.501051425933838,
      "learning_rate": 3.916976127762042e-05,
      "loss": 0.0684,
      "num_input_tokens_seen": 15373184,
      "step": 4620
    },
    {
      "epoch": 0.9253932921491634,
      "grad_norm": 0.08556798845529556,
      "learning_rate": 3.914817189727815e-05,
      "loss": 0.0202,
      "num_input_tokens_seen": 15390112,
      "step": 4625
    },
    {
      "epoch": 0.9263937173298652,
      "grad_norm": 0.6085869073867798,
      "learning_rate": 3.912656698309565e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 15407680,
      "step": 4630
    },
    {
      "epoch": 0.927394142510567,
      "grad_norm": 0.22436808049678802,
      "learning_rate": 3.9104946558793834e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 15424384,
      "step": 4635
    },
    {
      "epoch": 0.9283945676912688,
      "grad_norm": 0.4365352392196655,
      "learning_rate": 3.908331064811061e-05,
      "loss": 0.0386,
      "num_input_tokens_seen": 15441152,
      "step": 4640
    },
    {
      "epoch": 0.9293949928719706,
      "grad_norm": 0.2901744544506073,
      "learning_rate": 3.90616592748009e-05,
      "loss": 0.0372,
      "num_input_tokens_seen": 15457344,
      "step": 4645
    },
    {
      "epoch": 0.9303954180526723,
      "grad_norm": 0.3928404152393341,
      "learning_rate": 3.903999246263662e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 15474400,
      "step": 4650
    },
    {
      "epoch": 0.9313958432333742,
      "grad_norm": 0.5478469729423523,
      "learning_rate": 3.901831023540662e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 15490272,
      "step": 4655
    },
    {
      "epoch": 0.932396268414076,
      "grad_norm": 0.5972186923027039,
      "learning_rate": 3.8996612616916686e-05,
      "loss": 0.0457,
      "num_input_tokens_seen": 15507392,
      "step": 4660
    },
    {
      "epoch": 0.9333966935947777,
      "grad_norm": 0.42526888847351074,
      "learning_rate": 3.897489963098948e-05,
      "loss": 0.0367,
      "num_input_tokens_seen": 15523712,
      "step": 4665
    },
    {
      "epoch": 0.9343971187754796,
      "grad_norm": 0.7272816300392151,
      "learning_rate": 3.8953171301464554e-05,
      "loss": 0.0367,
      "num_input_tokens_seen": 15539360,
      "step": 4670
    },
    {
      "epoch": 0.9353975439561814,
      "grad_norm": 0.4032025933265686,
      "learning_rate": 3.893142765219832e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 15556384,
      "step": 4675
    },
    {
      "epoch": 0.9363979691368832,
      "grad_norm": 0.1461431235074997,
      "learning_rate": 3.890966870706398e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 15572672,
      "step": 4680
    },
    {
      "epoch": 0.937398394317585,
      "grad_norm": 0.31067830324172974,
      "learning_rate": 3.8887894489951546e-05,
      "loss": 0.0396,
      "num_input_tokens_seen": 15589248,
      "step": 4685
    },
    {
      "epoch": 0.9383988194982867,
      "grad_norm": 0.2968614101409912,
      "learning_rate": 3.886610502476781e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 15604608,
      "step": 4690
    },
    {
      "epoch": 0.9393992446789886,
      "grad_norm": 0.48350194096565247,
      "learning_rate": 3.8844300335436286e-05,
      "loss": 0.043,
      "num_input_tokens_seen": 15621696,
      "step": 4695
    },
    {
      "epoch": 0.9403996698596904,
      "grad_norm": 0.6612088084220886,
      "learning_rate": 3.882248044589719e-05,
      "loss": 0.0405,
      "num_input_tokens_seen": 15638368,
      "step": 4700
    },
    {
      "epoch": 0.9414000950403921,
      "grad_norm": 0.32954537868499756,
      "learning_rate": 3.880064538010747e-05,
      "loss": 0.0104,
      "num_input_tokens_seen": 15655616,
      "step": 4705
    },
    {
      "epoch": 0.942400520221094,
      "grad_norm": 0.3914512097835541,
      "learning_rate": 3.8778795162040705e-05,
      "loss": 0.0312,
      "num_input_tokens_seen": 15672224,
      "step": 4710
    },
    {
      "epoch": 0.9434009454017958,
      "grad_norm": 0.1047367975115776,
      "learning_rate": 3.875692981568712e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 15689248,
      "step": 4715
    },
    {
      "epoch": 0.9444013705824975,
      "grad_norm": 0.23204562067985535,
      "learning_rate": 3.8735049365053536e-05,
      "loss": 0.0241,
      "num_input_tokens_seen": 15706144,
      "step": 4720
    },
    {
      "epoch": 0.9454017957631994,
      "grad_norm": 0.9369616508483887,
      "learning_rate": 3.871315383416337e-05,
      "loss": 0.0239,
      "num_input_tokens_seen": 15724032,
      "step": 4725
    },
    {
      "epoch": 0.9464022209439011,
      "grad_norm": 0.4923385679721832,
      "learning_rate": 3.869124324705662e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 15741184,
      "step": 4730
    },
    {
      "epoch": 0.947402646124603,
      "grad_norm": 0.5696051120758057,
      "learning_rate": 3.866931762778976e-05,
      "loss": 0.0722,
      "num_input_tokens_seen": 15757536,
      "step": 4735
    },
    {
      "epoch": 0.9484030713053048,
      "grad_norm": 1.1380066871643066,
      "learning_rate": 3.8647377000435825e-05,
      "loss": 0.0327,
      "num_input_tokens_seen": 15774208,
      "step": 4740
    },
    {
      "epoch": 0.9494034964860065,
      "grad_norm": 0.8190211057662964,
      "learning_rate": 3.862542138908428e-05,
      "loss": 0.0543,
      "num_input_tokens_seen": 15790752,
      "step": 4745
    },
    {
      "epoch": 0.9504039216667084,
      "grad_norm": 0.7366694211959839,
      "learning_rate": 3.860345081784107e-05,
      "loss": 0.0494,
      "num_input_tokens_seen": 15808608,
      "step": 4750
    },
    {
      "epoch": 0.9514043468474102,
      "grad_norm": 0.44672852754592896,
      "learning_rate": 3.8581465310828566e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 15824544,
      "step": 4755
    },
    {
      "epoch": 0.9524047720281119,
      "grad_norm": 0.23382925987243652,
      "learning_rate": 3.8559464892185526e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 15840672,
      "step": 4760
    },
    {
      "epoch": 0.9534051972088138,
      "grad_norm": 0.20152555406093597,
      "learning_rate": 3.853744958606708e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 15857824,
      "step": 4765
    },
    {
      "epoch": 0.9544056223895155,
      "grad_norm": 0.15209810435771942,
      "learning_rate": 3.851541941664471e-05,
      "loss": 0.0288,
      "num_input_tokens_seen": 15873472,
      "step": 4770
    },
    {
      "epoch": 0.9554060475702173,
      "grad_norm": 0.9689774513244629,
      "learning_rate": 3.849337440810622e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 15889248,
      "step": 4775
    },
    {
      "epoch": 0.9564064727509192,
      "grad_norm": 0.7377488017082214,
      "learning_rate": 3.847131458465568e-05,
      "loss": 0.0498,
      "num_input_tokens_seen": 15907136,
      "step": 4780
    },
    {
      "epoch": 0.9574068979316209,
      "grad_norm": 0.4979168772697449,
      "learning_rate": 3.8449239970513476e-05,
      "loss": 0.0604,
      "num_input_tokens_seen": 15923456,
      "step": 4785
    },
    {
      "epoch": 0.9584073231123227,
      "grad_norm": 0.21843913197517395,
      "learning_rate": 3.842715058991619e-05,
      "loss": 0.0398,
      "num_input_tokens_seen": 15938880,
      "step": 4790
    },
    {
      "epoch": 0.9594077482930246,
      "grad_norm": 0.2384534627199173,
      "learning_rate": 3.8405046467116636e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 15956320,
      "step": 4795
    },
    {
      "epoch": 0.9604081734737263,
      "grad_norm": 0.26655861735343933,
      "learning_rate": 3.8382927626383794e-05,
      "loss": 0.0161,
      "num_input_tokens_seen": 15973088,
      "step": 4800
    },
    {
      "epoch": 0.9614085986544282,
      "grad_norm": 0.07932254672050476,
      "learning_rate": 3.836079409200283e-05,
      "loss": 0.0228,
      "num_input_tokens_seen": 15989696,
      "step": 4805
    },
    {
      "epoch": 0.9624090238351299,
      "grad_norm": 0.4905978739261627,
      "learning_rate": 3.833864588827504e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 16006784,
      "step": 4810
    },
    {
      "epoch": 0.9634094490158317,
      "grad_norm": 0.2714848220348358,
      "learning_rate": 3.83164830395178e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 16022688,
      "step": 4815
    },
    {
      "epoch": 0.9644098741965336,
      "grad_norm": 0.7874173521995544,
      "learning_rate": 3.8294305570064596e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 16040064,
      "step": 4820
    },
    {
      "epoch": 0.9654102993772353,
      "grad_norm": 0.5085760354995728,
      "learning_rate": 3.8272113504264936e-05,
      "loss": 0.0271,
      "num_input_tokens_seen": 16055456,
      "step": 4825
    },
    {
      "epoch": 0.9664107245579371,
      "grad_norm": 0.5401554107666016,
      "learning_rate": 3.8249906866484386e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 16072352,
      "step": 4830
    },
    {
      "epoch": 0.967411149738639,
      "grad_norm": 0.45530328154563904,
      "learning_rate": 3.8227685681104484e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 16088928,
      "step": 4835
    },
    {
      "epoch": 0.9684115749193407,
      "grad_norm": 0.1726125329732895,
      "learning_rate": 3.8205449972522775e-05,
      "loss": 0.0349,
      "num_input_tokens_seen": 16105440,
      "step": 4840
    },
    {
      "epoch": 0.9694120001000425,
      "grad_norm": 0.5761016011238098,
      "learning_rate": 3.8183199765152706e-05,
      "loss": 0.021,
      "num_input_tokens_seen": 16121344,
      "step": 4845
    },
    {
      "epoch": 0.9704124252807443,
      "grad_norm": 0.6431677341461182,
      "learning_rate": 3.816093508342366e-05,
      "loss": 0.0484,
      "num_input_tokens_seen": 16137152,
      "step": 4850
    },
    {
      "epoch": 0.9714128504614461,
      "grad_norm": 0.3383553624153137,
      "learning_rate": 3.813865595178094e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 16153728,
      "step": 4855
    },
    {
      "epoch": 0.972413275642148,
      "grad_norm": 0.38180163502693176,
      "learning_rate": 3.811636239468568e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 16171584,
      "step": 4860
    },
    {
      "epoch": 0.9734137008228497,
      "grad_norm": 0.22540175914764404,
      "learning_rate": 3.8094054436614855e-05,
      "loss": 0.0188,
      "num_input_tokens_seen": 16188864,
      "step": 4865
    },
    {
      "epoch": 0.9744141260035515,
      "grad_norm": 0.22031109035015106,
      "learning_rate": 3.8071732102061254e-05,
      "loss": 0.0158,
      "num_input_tokens_seen": 16206048,
      "step": 4870
    },
    {
      "epoch": 0.9754145511842534,
      "grad_norm": 0.7438973784446716,
      "learning_rate": 3.8049395415533474e-05,
      "loss": 0.0556,
      "num_input_tokens_seen": 16222016,
      "step": 4875
    },
    {
      "epoch": 0.9764149763649551,
      "grad_norm": 0.841015100479126,
      "learning_rate": 3.802704440155583e-05,
      "loss": 0.0667,
      "num_input_tokens_seen": 16238272,
      "step": 4880
    },
    {
      "epoch": 0.9774154015456569,
      "grad_norm": 0.34464555978775024,
      "learning_rate": 3.800467908466841e-05,
      "loss": 0.0339,
      "num_input_tokens_seen": 16254304,
      "step": 4885
    },
    {
      "epoch": 0.9784158267263587,
      "grad_norm": 0.39327627420425415,
      "learning_rate": 3.7982299489426975e-05,
      "loss": 0.0432,
      "num_input_tokens_seen": 16270944,
      "step": 4890
    },
    {
      "epoch": 0.9794162519070605,
      "grad_norm": 0.5456660389900208,
      "learning_rate": 3.795990564040296e-05,
      "loss": 0.0416,
      "num_input_tokens_seen": 16287744,
      "step": 4895
    },
    {
      "epoch": 0.9804166770877623,
      "grad_norm": 0.5775066018104553,
      "learning_rate": 3.793749756218349e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 16303680,
      "step": 4900
    },
    {
      "epoch": 0.9814171022684641,
      "grad_norm": 0.5876214504241943,
      "learning_rate": 3.791507527937125e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 16321152,
      "step": 4905
    },
    {
      "epoch": 0.9824175274491659,
      "grad_norm": 0.42364928126335144,
      "learning_rate": 3.789263881658458e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 16338368,
      "step": 4910
    },
    {
      "epoch": 0.9834179526298676,
      "grad_norm": 1.0304877758026123,
      "learning_rate": 3.7870188198457356e-05,
      "loss": 0.0555,
      "num_input_tokens_seen": 16355712,
      "step": 4915
    },
    {
      "epoch": 0.9844183778105695,
      "grad_norm": 0.11261558532714844,
      "learning_rate": 3.784772344963901e-05,
      "loss": 0.0547,
      "num_input_tokens_seen": 16372480,
      "step": 4920
    },
    {
      "epoch": 0.9854188029912713,
      "grad_norm": 0.1952620893716812,
      "learning_rate": 3.782524459479446e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 16389088,
      "step": 4925
    },
    {
      "epoch": 0.9864192281719731,
      "grad_norm": 0.42790642380714417,
      "learning_rate": 3.780275165860416e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 16405536,
      "step": 4930
    },
    {
      "epoch": 0.9874196533526749,
      "grad_norm": 0.08422848582267761,
      "learning_rate": 3.7780244665763995e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 16421504,
      "step": 4935
    },
    {
      "epoch": 0.9884200785333767,
      "grad_norm": 0.544394314289093,
      "learning_rate": 3.775772364098529e-05,
      "loss": 0.0247,
      "num_input_tokens_seen": 16437952,
      "step": 4940
    },
    {
      "epoch": 0.9894205037140785,
      "grad_norm": 0.4700765609741211,
      "learning_rate": 3.7735188608994745e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 16455168,
      "step": 4945
    },
    {
      "epoch": 0.9904209288947803,
      "grad_norm": 0.3246154189109802,
      "learning_rate": 3.77126395945345e-05,
      "loss": 0.0246,
      "num_input_tokens_seen": 16472448,
      "step": 4950
    },
    {
      "epoch": 0.991421354075482,
      "grad_norm": 0.35189178586006165,
      "learning_rate": 3.7690076622362006e-05,
      "loss": 0.0154,
      "num_input_tokens_seen": 16488608,
      "step": 4955
    },
    {
      "epoch": 0.9924217792561839,
      "grad_norm": 0.09933869540691376,
      "learning_rate": 3.7667499717250026e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 16504640,
      "step": 4960
    },
    {
      "epoch": 0.9934222044368857,
      "grad_norm": 0.8066585659980774,
      "learning_rate": 3.764490890398667e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 16521760,
      "step": 4965
    },
    {
      "epoch": 0.9944226296175874,
      "grad_norm": 0.07991819828748703,
      "learning_rate": 3.762230420737528e-05,
      "loss": 0.0271,
      "num_input_tokens_seen": 16537888,
      "step": 4970
    },
    {
      "epoch": 0.9954230547982893,
      "grad_norm": 0.7777450680732727,
      "learning_rate": 3.759968565223444e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 16555168,
      "step": 4975
    },
    {
      "epoch": 0.9964234799789911,
      "grad_norm": 0.35846781730651855,
      "learning_rate": 3.7577053263397974e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 16573280,
      "step": 4980
    },
    {
      "epoch": 0.9974239051596928,
      "grad_norm": 0.32245972752571106,
      "learning_rate": 3.755440706571487e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 16590048,
      "step": 4985
    },
    {
      "epoch": 0.9984243303403947,
      "grad_norm": 0.7595393657684326,
      "learning_rate": 3.7531747084049304e-05,
      "loss": 0.029,
      "num_input_tokens_seen": 16606080,
      "step": 4990
    },
    {
      "epoch": 0.9994247555210964,
      "grad_norm": 0.27887552976608276,
      "learning_rate": 3.7509073343280555e-05,
      "loss": 0.0375,
      "num_input_tokens_seen": 16622848,
      "step": 4995
    },
    {
      "epoch": 1.0004251807017983,
      "grad_norm": 0.27957552671432495,
      "learning_rate": 3.748638586830303e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 16639872,
      "step": 5000
    },
    {
      "epoch": 1.0014256058825,
      "grad_norm": 0.5249574184417725,
      "learning_rate": 3.7463684684026215e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 16656160,
      "step": 5005
    },
    {
      "epoch": 1.0024260310632018,
      "grad_norm": 0.2980789542198181,
      "learning_rate": 3.744096981537463e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 16671392,
      "step": 5010
    },
    {
      "epoch": 1.0034264562439037,
      "grad_norm": 0.18490222096443176,
      "learning_rate": 3.741824128728784e-05,
      "loss": 0.0208,
      "num_input_tokens_seen": 16688672,
      "step": 5015
    },
    {
      "epoch": 1.0044268814246056,
      "grad_norm": 0.2901025414466858,
      "learning_rate": 3.73954991247204e-05,
      "loss": 0.0164,
      "num_input_tokens_seen": 16704704,
      "step": 5020
    },
    {
      "epoch": 1.0054273066053072,
      "grad_norm": 0.8393234610557556,
      "learning_rate": 3.737274335264184e-05,
      "loss": 0.0342,
      "num_input_tokens_seen": 16721664,
      "step": 5025
    },
    {
      "epoch": 1.006427731786009,
      "grad_norm": 0.24501626193523407,
      "learning_rate": 3.7349973996036606e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 16739200,
      "step": 5030
    },
    {
      "epoch": 1.007428156966711,
      "grad_norm": 0.625282347202301,
      "learning_rate": 3.7327191079904094e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 16755488,
      "step": 5035
    },
    {
      "epoch": 1.0084285821474126,
      "grad_norm": 0.390674352645874,
      "learning_rate": 3.730439462925858e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 16771712,
      "step": 5040
    },
    {
      "epoch": 1.0094290073281145,
      "grad_norm": 0.1857205480337143,
      "learning_rate": 3.728158466912918e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 16788256,
      "step": 5045
    },
    {
      "epoch": 1.0104294325088163,
      "grad_norm": 0.08821652084589005,
      "learning_rate": 3.7258761224559855e-05,
      "loss": 0.027,
      "num_input_tokens_seen": 16805056,
      "step": 5050
    },
    {
      "epoch": 1.011429857689518,
      "grad_norm": 0.3787899613380432,
      "learning_rate": 3.72359243206094e-05,
      "loss": 0.044,
      "num_input_tokens_seen": 16822848,
      "step": 5055
    },
    {
      "epoch": 1.0124302828702199,
      "grad_norm": 0.2202107459306717,
      "learning_rate": 3.721307398235131e-05,
      "loss": 0.0186,
      "num_input_tokens_seen": 16839360,
      "step": 5060
    },
    {
      "epoch": 1.0134307080509217,
      "grad_norm": 0.793282687664032,
      "learning_rate": 3.7190210234873924e-05,
      "loss": 0.0421,
      "num_input_tokens_seen": 16856896,
      "step": 5065
    },
    {
      "epoch": 1.0144311332316234,
      "grad_norm": 0.602552056312561,
      "learning_rate": 3.7167333103280234e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 16873568,
      "step": 5070
    },
    {
      "epoch": 1.0154315584123252,
      "grad_norm": 0.3753560185432434,
      "learning_rate": 3.714444261268796e-05,
      "loss": 0.0124,
      "num_input_tokens_seen": 16890880,
      "step": 5075
    },
    {
      "epoch": 1.016431983593027,
      "grad_norm": 0.4732748866081238,
      "learning_rate": 3.7121538788229485e-05,
      "loss": 0.0163,
      "num_input_tokens_seen": 16906816,
      "step": 5080
    },
    {
      "epoch": 1.0174324087737288,
      "grad_norm": 0.6718013286590576,
      "learning_rate": 3.709862165505182e-05,
      "loss": 0.0284,
      "num_input_tokens_seen": 16922368,
      "step": 5085
    },
    {
      "epoch": 1.0184328339544306,
      "grad_norm": 0.340224027633667,
      "learning_rate": 3.707569123831659e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 16939200,
      "step": 5090
    },
    {
      "epoch": 1.0194332591351325,
      "grad_norm": 0.8260764479637146,
      "learning_rate": 3.705274756320004e-05,
      "loss": 0.0347,
      "num_input_tokens_seen": 16956608,
      "step": 5095
    },
    {
      "epoch": 1.0204336843158341,
      "grad_norm": 0.41199642419815063,
      "learning_rate": 3.7029790654892924e-05,
      "loss": 0.0319,
      "num_input_tokens_seen": 16974112,
      "step": 5100
    },
    {
      "epoch": 1.021434109496536,
      "grad_norm": 0.15774880349636078,
      "learning_rate": 3.700682053860053e-05,
      "loss": 0.0108,
      "num_input_tokens_seen": 16990592,
      "step": 5105
    },
    {
      "epoch": 1.0224345346772379,
      "grad_norm": 0.8791263103485107,
      "learning_rate": 3.698383723954268e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 17005824,
      "step": 5110
    },
    {
      "epoch": 1.0234349598579395,
      "grad_norm": 0.4210505783557892,
      "learning_rate": 3.6960840782953654e-05,
      "loss": 0.0302,
      "num_input_tokens_seen": 17021824,
      "step": 5115
    },
    {
      "epoch": 1.0244353850386414,
      "grad_norm": 0.305221289396286,
      "learning_rate": 3.6937831194082166e-05,
      "loss": 0.0214,
      "num_input_tokens_seen": 17038208,
      "step": 5120
    },
    {
      "epoch": 1.0254358102193433,
      "grad_norm": 0.25967296957969666,
      "learning_rate": 3.6914808498191355e-05,
      "loss": 0.0197,
      "num_input_tokens_seen": 17056096,
      "step": 5125
    },
    {
      "epoch": 1.026436235400045,
      "grad_norm": 0.5764554142951965,
      "learning_rate": 3.689177272055877e-05,
      "loss": 0.0222,
      "num_input_tokens_seen": 17072960,
      "step": 5130
    },
    {
      "epoch": 1.0274366605807468,
      "grad_norm": 0.4865210950374603,
      "learning_rate": 3.6868723886476295e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 17090176,
      "step": 5135
    },
    {
      "epoch": 1.0284370857614487,
      "grad_norm": 0.23601755499839783,
      "learning_rate": 3.684566202125015e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 17106528,
      "step": 5140
    },
    {
      "epoch": 1.0294375109421505,
      "grad_norm": 0.6067863702774048,
      "learning_rate": 3.6822587150200897e-05,
      "loss": 0.0243,
      "num_input_tokens_seen": 17123008,
      "step": 5145
    },
    {
      "epoch": 1.0304379361228522,
      "grad_norm": 0.5614882111549377,
      "learning_rate": 3.679949929866334e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 17139200,
      "step": 5150
    },
    {
      "epoch": 1.031438361303554,
      "grad_norm": 0.16848532855510712,
      "learning_rate": 3.6776398491986555e-05,
      "loss": 0.033,
      "num_input_tokens_seen": 17155648,
      "step": 5155
    },
    {
      "epoch": 1.032438786484256,
      "grad_norm": 0.20804527401924133,
      "learning_rate": 3.675328475553382e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 17172352,
      "step": 5160
    },
    {
      "epoch": 1.0334392116649576,
      "grad_norm": 0.3758269250392914,
      "learning_rate": 3.6730158114682645e-05,
      "loss": 0.0119,
      "num_input_tokens_seen": 17188640,
      "step": 5165
    },
    {
      "epoch": 1.0344396368456594,
      "grad_norm": 0.06102520599961281,
      "learning_rate": 3.670701859482468e-05,
      "loss": 0.039,
      "num_input_tokens_seen": 17204128,
      "step": 5170
    },
    {
      "epoch": 1.0354400620263613,
      "grad_norm": 0.3474910855293274,
      "learning_rate": 3.668386622136573e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 17219232,
      "step": 5175
    },
    {
      "epoch": 1.036440487207063,
      "grad_norm": 0.6429199576377869,
      "learning_rate": 3.666070101972569e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 17235328,
      "step": 5180
    },
    {
      "epoch": 1.0374409123877648,
      "grad_norm": 0.5511621236801147,
      "learning_rate": 3.663752301533857e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 17252128,
      "step": 5185
    },
    {
      "epoch": 1.0384413375684667,
      "grad_norm": 0.99336177110672,
      "learning_rate": 3.661433223365241e-05,
      "loss": 0.0424,
      "num_input_tokens_seen": 17268224,
      "step": 5190
    },
    {
      "epoch": 1.0394417627491683,
      "grad_norm": 0.26638805866241455,
      "learning_rate": 3.659112870012932e-05,
      "loss": 0.0117,
      "num_input_tokens_seen": 17284960,
      "step": 5195
    },
    {
      "epoch": 1.0404421879298702,
      "grad_norm": 0.3554804027080536,
      "learning_rate": 3.656791244024534e-05,
      "loss": 0.0205,
      "num_input_tokens_seen": 17302272,
      "step": 5200
    },
    {
      "epoch": 1.041442613110572,
      "grad_norm": 1.1736732721328735,
      "learning_rate": 3.654468347949055e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 17318976,
      "step": 5205
    },
    {
      "epoch": 1.0424430382912737,
      "grad_norm": 0.4835314154624939,
      "learning_rate": 3.6521441843368945e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 17335840,
      "step": 5210
    },
    {
      "epoch": 1.0434434634719756,
      "grad_norm": 0.3860490620136261,
      "learning_rate": 3.649818755739844e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 17352704,
      "step": 5215
    },
    {
      "epoch": 1.0444438886526775,
      "grad_norm": 0.6804181337356567,
      "learning_rate": 3.647492064711083e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 17369472,
      "step": 5220
    },
    {
      "epoch": 1.045444313833379,
      "grad_norm": 0.349334716796875,
      "learning_rate": 3.645164113805181e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 17387008,
      "step": 5225
    },
    {
      "epoch": 1.046444739014081,
      "grad_norm": 0.3037664592266083,
      "learning_rate": 3.642834905578084e-05,
      "loss": 0.0212,
      "num_input_tokens_seen": 17403552,
      "step": 5230
    },
    {
      "epoch": 1.0474451641947828,
      "grad_norm": 0.2728258967399597,
      "learning_rate": 3.640504442587125e-05,
      "loss": 0.012,
      "num_input_tokens_seen": 17421408,
      "step": 5235
    },
    {
      "epoch": 1.0484455893754845,
      "grad_norm": 0.8211174011230469,
      "learning_rate": 3.6381727273910104e-05,
      "loss": 0.0286,
      "num_input_tokens_seen": 17438240,
      "step": 5240
    },
    {
      "epoch": 1.0494460145561864,
      "grad_norm": 0.2891646921634674,
      "learning_rate": 3.635839762549824e-05,
      "loss": 0.027,
      "num_input_tokens_seen": 17455488,
      "step": 5245
    },
    {
      "epoch": 1.0504464397368882,
      "grad_norm": 0.75777268409729,
      "learning_rate": 3.633505550625021e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 17471872,
      "step": 5250
    },
    {
      "epoch": 1.0514468649175899,
      "grad_norm": 0.709787130355835,
      "learning_rate": 3.6311700941794236e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 17489024,
      "step": 5255
    },
    {
      "epoch": 1.0524472900982917,
      "grad_norm": 0.4019218385219574,
      "learning_rate": 3.628833395777224e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 17506912,
      "step": 5260
    },
    {
      "epoch": 1.0534477152789936,
      "grad_norm": 0.47778013348579407,
      "learning_rate": 3.626495457983975e-05,
      "loss": 0.0211,
      "num_input_tokens_seen": 17524320,
      "step": 5265
    },
    {
      "epoch": 1.0544481404596953,
      "grad_norm": 0.5299893617630005,
      "learning_rate": 3.624156283366592e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 17539904,
      "step": 5270
    },
    {
      "epoch": 1.0554485656403971,
      "grad_norm": 0.16689853370189667,
      "learning_rate": 3.6218158744933476e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 17557504,
      "step": 5275
    },
    {
      "epoch": 1.056448990821099,
      "grad_norm": 0.5492450594902039,
      "learning_rate": 3.619474233933871e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 17574112,
      "step": 5280
    },
    {
      "epoch": 1.0574494160018009,
      "grad_norm": 0.7357029318809509,
      "learning_rate": 3.617131364259141e-05,
      "loss": 0.0196,
      "num_input_tokens_seen": 17590176,
      "step": 5285
    },
    {
      "epoch": 1.0584498411825025,
      "grad_norm": 0.4019847810268402,
      "learning_rate": 3.614787268041487e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 17606400,
      "step": 5290
    },
    {
      "epoch": 1.0594502663632044,
      "grad_norm": 0.6422977447509766,
      "learning_rate": 3.6124419478545854e-05,
      "loss": 0.0256,
      "num_input_tokens_seen": 17623584,
      "step": 5295
    },
    {
      "epoch": 1.0604506915439063,
      "grad_norm": 0.3176085948944092,
      "learning_rate": 3.610095406273459e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 17641504,
      "step": 5300
    },
    {
      "epoch": 1.061451116724608,
      "grad_norm": 0.6598035097122192,
      "learning_rate": 3.6077476458744664e-05,
      "loss": 0.0237,
      "num_input_tokens_seen": 17659232,
      "step": 5305
    },
    {
      "epoch": 1.0624515419053098,
      "grad_norm": 0.17913129925727844,
      "learning_rate": 3.605398669235307e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 17675744,
      "step": 5310
    },
    {
      "epoch": 1.0634519670860116,
      "grad_norm": 0.6988956928253174,
      "learning_rate": 3.6030484789350176e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 17693056,
      "step": 5315
    },
    {
      "epoch": 1.0644523922667133,
      "grad_norm": 0.41953983902931213,
      "learning_rate": 3.600697077553964e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 17710336,
      "step": 5320
    },
    {
      "epoch": 1.0654528174474152,
      "grad_norm": 0.4209190309047699,
      "learning_rate": 3.598344467673843e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 17726176,
      "step": 5325
    },
    {
      "epoch": 1.066453242628117,
      "grad_norm": 0.17477113008499146,
      "learning_rate": 3.595990651877679e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 17741664,
      "step": 5330
    },
    {
      "epoch": 1.0674536678088187,
      "grad_norm": 0.30309292674064636,
      "learning_rate": 3.593635632749821e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 17757888,
      "step": 5335
    },
    {
      "epoch": 1.0684540929895205,
      "grad_norm": 0.3024982810020447,
      "learning_rate": 3.591279412875937e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 17775584,
      "step": 5340
    },
    {
      "epoch": 1.0694545181702224,
      "grad_norm": 0.7211211323738098,
      "learning_rate": 3.588921994843013e-05,
      "loss": 0.0188,
      "num_input_tokens_seen": 17791040,
      "step": 5345
    },
    {
      "epoch": 1.070454943350924,
      "grad_norm": 0.4666048586368561,
      "learning_rate": 3.586563381239355e-05,
      "loss": 0.0329,
      "num_input_tokens_seen": 17807264,
      "step": 5350
    },
    {
      "epoch": 1.071455368531626,
      "grad_norm": 0.24549226462841034,
      "learning_rate": 3.5842035746545765e-05,
      "loss": 0.0155,
      "num_input_tokens_seen": 17824096,
      "step": 5355
    },
    {
      "epoch": 1.0724557937123278,
      "grad_norm": 0.3099384009838104,
      "learning_rate": 3.5818425776796044e-05,
      "loss": 0.016,
      "num_input_tokens_seen": 17840480,
      "step": 5360
    },
    {
      "epoch": 1.0734562188930294,
      "grad_norm": 0.2598489224910736,
      "learning_rate": 3.579480392906669e-05,
      "loss": 0.0387,
      "num_input_tokens_seen": 17857504,
      "step": 5365
    },
    {
      "epoch": 1.0744566440737313,
      "grad_norm": 1.099205732345581,
      "learning_rate": 3.5771170229293085e-05,
      "loss": 0.0285,
      "num_input_tokens_seen": 17873600,
      "step": 5370
    },
    {
      "epoch": 1.0754570692544332,
      "grad_norm": 0.8734391331672668,
      "learning_rate": 3.574752470342361e-05,
      "loss": 0.0526,
      "num_input_tokens_seen": 17890400,
      "step": 5375
    },
    {
      "epoch": 1.0764574944351348,
      "grad_norm": 0.34526267647743225,
      "learning_rate": 3.5723867377419614e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 17907104,
      "step": 5380
    },
    {
      "epoch": 1.0774579196158367,
      "grad_norm": 0.3123202621936798,
      "learning_rate": 3.570019827725542e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 17923200,
      "step": 5385
    },
    {
      "epoch": 1.0784583447965386,
      "grad_norm": 0.6335588693618774,
      "learning_rate": 3.567651742891828e-05,
      "loss": 0.0212,
      "num_input_tokens_seen": 17940448,
      "step": 5390
    },
    {
      "epoch": 1.0794587699772404,
      "grad_norm": 0.10668666660785675,
      "learning_rate": 3.5652824858408326e-05,
      "loss": 0.0146,
      "num_input_tokens_seen": 17956544,
      "step": 5395
    },
    {
      "epoch": 1.080459195157942,
      "grad_norm": 0.7952919602394104,
      "learning_rate": 3.562912059173859e-05,
      "loss": 0.0317,
      "num_input_tokens_seen": 17973312,
      "step": 5400
    },
    {
      "epoch": 1.081459620338644,
      "grad_norm": 0.8085919618606567,
      "learning_rate": 3.560540465493492e-05,
      "loss": 0.0285,
      "num_input_tokens_seen": 17989376,
      "step": 5405
    },
    {
      "epoch": 1.0824600455193458,
      "grad_norm": 0.3211195468902588,
      "learning_rate": 3.558167707403599e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 18005536,
      "step": 5410
    },
    {
      "epoch": 1.0834604707000475,
      "grad_norm": 0.7521841526031494,
      "learning_rate": 3.555793787509324e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 18021888,
      "step": 5415
    },
    {
      "epoch": 1.0844608958807493,
      "grad_norm": 0.7724508047103882,
      "learning_rate": 3.5534187084170884e-05,
      "loss": 0.0276,
      "num_input_tokens_seen": 18037664,
      "step": 5420
    },
    {
      "epoch": 1.0854613210614512,
      "grad_norm": 0.40795546770095825,
      "learning_rate": 3.5510424727345874e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 18054400,
      "step": 5425
    },
    {
      "epoch": 1.0864617462421529,
      "grad_norm": 0.40630871057510376,
      "learning_rate": 3.548665083070784e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 18072672,
      "step": 5430
    },
    {
      "epoch": 1.0874621714228547,
      "grad_norm": 0.7899190783500671,
      "learning_rate": 3.5462865420359073e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 18089280,
      "step": 5435
    },
    {
      "epoch": 1.0884625966035566,
      "grad_norm": 0.5796297788619995,
      "learning_rate": 3.543906852241454e-05,
      "loss": 0.0321,
      "num_input_tokens_seen": 18105600,
      "step": 5440
    },
    {
      "epoch": 1.0894630217842582,
      "grad_norm": 0.7630406022071838,
      "learning_rate": 3.54152601630018e-05,
      "loss": 0.0297,
      "num_input_tokens_seen": 18121152,
      "step": 5445
    },
    {
      "epoch": 1.0904634469649601,
      "grad_norm": 0.6281930208206177,
      "learning_rate": 3.5391440368260975e-05,
      "loss": 0.0432,
      "num_input_tokens_seen": 18138976,
      "step": 5450
    },
    {
      "epoch": 1.091463872145662,
      "grad_norm": 0.6421923041343689,
      "learning_rate": 3.536760916434478e-05,
      "loss": 0.0412,
      "num_input_tokens_seen": 18156800,
      "step": 5455
    },
    {
      "epoch": 1.0924642973263636,
      "grad_norm": 0.4233951270580292,
      "learning_rate": 3.534376657741845e-05,
      "loss": 0.0372,
      "num_input_tokens_seen": 18172512,
      "step": 5460
    },
    {
      "epoch": 1.0934647225070655,
      "grad_norm": 0.30023810267448425,
      "learning_rate": 3.53199126336597e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 18188352,
      "step": 5465
    },
    {
      "epoch": 1.0944651476877674,
      "grad_norm": 0.8775084614753723,
      "learning_rate": 3.529604735925871e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 18204960,
      "step": 5470
    },
    {
      "epoch": 1.095465572868469,
      "grad_norm": 0.2205784022808075,
      "learning_rate": 3.527217078041813e-05,
      "loss": 0.0205,
      "num_input_tokens_seen": 18220000,
      "step": 5475
    },
    {
      "epoch": 1.0964659980491709,
      "grad_norm": 0.7994363307952881,
      "learning_rate": 3.5248282923353004e-05,
      "loss": 0.032,
      "num_input_tokens_seen": 18235712,
      "step": 5480
    },
    {
      "epoch": 1.0974664232298728,
      "grad_norm": 1.386659026145935,
      "learning_rate": 3.5224383814290754e-05,
      "loss": 0.0451,
      "num_input_tokens_seen": 18252384,
      "step": 5485
    },
    {
      "epoch": 1.0984668484105744,
      "grad_norm": 0.6417826414108276,
      "learning_rate": 3.520047347947116e-05,
      "loss": 0.037,
      "num_input_tokens_seen": 18268256,
      "step": 5490
    },
    {
      "epoch": 1.0994672735912763,
      "grad_norm": 0.3438206613063812,
      "learning_rate": 3.517655194514634e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 18284128,
      "step": 5495
    },
    {
      "epoch": 1.1004676987719781,
      "grad_norm": 0.7664891481399536,
      "learning_rate": 3.515261923758068e-05,
      "loss": 0.0237,
      "num_input_tokens_seen": 18300480,
      "step": 5500
    },
    {
      "epoch": 1.1014681239526798,
      "grad_norm": 0.6775107383728027,
      "learning_rate": 3.512867538305086e-05,
      "loss": 0.0164,
      "num_input_tokens_seen": 18318144,
      "step": 5505
    },
    {
      "epoch": 1.1024685491333817,
      "grad_norm": 0.3344127833843231,
      "learning_rate": 3.510472040784579e-05,
      "loss": 0.0171,
      "num_input_tokens_seen": 18333760,
      "step": 5510
    },
    {
      "epoch": 1.1034689743140835,
      "grad_norm": 0.2144256830215454,
      "learning_rate": 3.50807543382666e-05,
      "loss": 0.0102,
      "num_input_tokens_seen": 18349056,
      "step": 5515
    },
    {
      "epoch": 1.1044693994947852,
      "grad_norm": 0.3108498156070709,
      "learning_rate": 3.5056777200626574e-05,
      "loss": 0.018,
      "num_input_tokens_seen": 18366240,
      "step": 5520
    },
    {
      "epoch": 1.105469824675487,
      "grad_norm": 0.23828475177288055,
      "learning_rate": 3.5032789021251167e-05,
      "loss": 0.0204,
      "num_input_tokens_seen": 18382816,
      "step": 5525
    },
    {
      "epoch": 1.106470249856189,
      "grad_norm": 0.316902220249176,
      "learning_rate": 3.500878982647795e-05,
      "loss": 0.0144,
      "num_input_tokens_seen": 18399808,
      "step": 5530
    },
    {
      "epoch": 1.1074706750368906,
      "grad_norm": 1.227034330368042,
      "learning_rate": 3.498477964265661e-05,
      "loss": 0.0239,
      "num_input_tokens_seen": 18417088,
      "step": 5535
    },
    {
      "epoch": 1.1084711002175924,
      "grad_norm": 1.0346033573150635,
      "learning_rate": 3.496075849614887e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 18432448,
      "step": 5540
    },
    {
      "epoch": 1.1094715253982943,
      "grad_norm": 0.13602112233638763,
      "learning_rate": 3.49367264133285e-05,
      "loss": 0.0256,
      "num_input_tokens_seen": 18449440,
      "step": 5545
    },
    {
      "epoch": 1.1104719505789962,
      "grad_norm": 0.771540105342865,
      "learning_rate": 3.491268342058128e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 18465312,
      "step": 5550
    },
    {
      "epoch": 1.1114723757596978,
      "grad_norm": 0.5797554850578308,
      "learning_rate": 3.488862954430499e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 18481824,
      "step": 5555
    },
    {
      "epoch": 1.1124728009403997,
      "grad_norm": 0.7463229298591614,
      "learning_rate": 3.4864564810909296e-05,
      "loss": 0.0164,
      "num_input_tokens_seen": 18498624,
      "step": 5560
    },
    {
      "epoch": 1.1134732261211016,
      "grad_norm": 0.5763323903083801,
      "learning_rate": 3.484048924681584e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 18513824,
      "step": 5565
    },
    {
      "epoch": 1.1144736513018032,
      "grad_norm": 0.17134538292884827,
      "learning_rate": 3.481640287845817e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 18530592,
      "step": 5570
    },
    {
      "epoch": 1.115474076482505,
      "grad_norm": 0.802047610282898,
      "learning_rate": 3.479230573228162e-05,
      "loss": 0.0349,
      "num_input_tokens_seen": 18547104,
      "step": 5575
    },
    {
      "epoch": 1.116474501663207,
      "grad_norm": 0.6504844427108765,
      "learning_rate": 3.476819783474344e-05,
      "loss": 0.0289,
      "num_input_tokens_seen": 18562560,
      "step": 5580
    },
    {
      "epoch": 1.1174749268439086,
      "grad_norm": 0.61082923412323,
      "learning_rate": 3.474407921231263e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 18578752,
      "step": 5585
    },
    {
      "epoch": 1.1184753520246105,
      "grad_norm": 0.09099405258893967,
      "learning_rate": 3.471994989147e-05,
      "loss": 0.0114,
      "num_input_tokens_seen": 18596128,
      "step": 5590
    },
    {
      "epoch": 1.1194757772053123,
      "grad_norm": 0.28908923268318176,
      "learning_rate": 3.4695809898708075e-05,
      "loss": 0.0127,
      "num_input_tokens_seen": 18611872,
      "step": 5595
    },
    {
      "epoch": 1.120476202386014,
      "grad_norm": 0.3132735788822174,
      "learning_rate": 3.4671659260531116e-05,
      "loss": 0.0168,
      "num_input_tokens_seen": 18628416,
      "step": 5600
    },
    {
      "epoch": 1.1214766275667158,
      "grad_norm": 0.7768466472625732,
      "learning_rate": 3.464749800345507e-05,
      "loss": 0.0145,
      "num_input_tokens_seen": 18645376,
      "step": 5605
    },
    {
      "epoch": 1.1224770527474177,
      "grad_norm": 0.35140541195869446,
      "learning_rate": 3.462332615400755e-05,
      "loss": 0.0288,
      "num_input_tokens_seen": 18662048,
      "step": 5610
    },
    {
      "epoch": 1.1234774779281194,
      "grad_norm": 0.6198104023933411,
      "learning_rate": 3.459914373872778e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 18678432,
      "step": 5615
    },
    {
      "epoch": 1.1244779031088212,
      "grad_norm": 0.14505791664123535,
      "learning_rate": 3.457495078416659e-05,
      "loss": 0.0202,
      "num_input_tokens_seen": 18694304,
      "step": 5620
    },
    {
      "epoch": 1.125478328289523,
      "grad_norm": 0.19856490194797516,
      "learning_rate": 3.45507473168864e-05,
      "loss": 0.0171,
      "num_input_tokens_seen": 18710080,
      "step": 5625
    },
    {
      "epoch": 1.126478753470225,
      "grad_norm": 0.11967971920967102,
      "learning_rate": 3.452653336346115e-05,
      "loss": 0.0109,
      "num_input_tokens_seen": 18726720,
      "step": 5630
    },
    {
      "epoch": 1.1274791786509266,
      "grad_norm": 0.2559490203857422,
      "learning_rate": 3.450230895047631e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 18743680,
      "step": 5635
    },
    {
      "epoch": 1.1284796038316285,
      "grad_norm": 0.4048309028148651,
      "learning_rate": 3.447807410452881e-05,
      "loss": 0.0364,
      "num_input_tokens_seen": 18760928,
      "step": 5640
    },
    {
      "epoch": 1.1294800290123304,
      "grad_norm": 0.05227162688970566,
      "learning_rate": 3.445382885222707e-05,
      "loss": 0.0132,
      "num_input_tokens_seen": 18777504,
      "step": 5645
    },
    {
      "epoch": 1.130480454193032,
      "grad_norm": 0.2708294093608856,
      "learning_rate": 3.44295732201909e-05,
      "loss": 0.0227,
      "num_input_tokens_seen": 18794752,
      "step": 5650
    },
    {
      "epoch": 1.1314808793737339,
      "grad_norm": 0.8836909532546997,
      "learning_rate": 3.4405307235051534e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 18812352,
      "step": 5655
    },
    {
      "epoch": 1.1324813045544357,
      "grad_norm": 0.8740631937980652,
      "learning_rate": 3.438103092345155e-05,
      "loss": 0.0261,
      "num_input_tokens_seen": 18829472,
      "step": 5660
    },
    {
      "epoch": 1.1334817297351374,
      "grad_norm": 0.13569481670856476,
      "learning_rate": 3.435674431204489e-05,
      "loss": 0.0297,
      "num_input_tokens_seen": 18845664,
      "step": 5665
    },
    {
      "epoch": 1.1344821549158393,
      "grad_norm": 0.2504052519798279,
      "learning_rate": 3.4332447427496775e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 18861632,
      "step": 5670
    },
    {
      "epoch": 1.1354825800965411,
      "grad_norm": 0.8606333136558533,
      "learning_rate": 3.4308140296483724e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 18878240,
      "step": 5675
    },
    {
      "epoch": 1.1364830052772428,
      "grad_norm": 0.5849329233169556,
      "learning_rate": 3.4283822945693514e-05,
      "loss": 0.0286,
      "num_input_tokens_seen": 18894624,
      "step": 5680
    },
    {
      "epoch": 1.1374834304579446,
      "grad_norm": 0.30441370606422424,
      "learning_rate": 3.425949540182512e-05,
      "loss": 0.0366,
      "num_input_tokens_seen": 18911168,
      "step": 5685
    },
    {
      "epoch": 1.1384838556386465,
      "grad_norm": 0.29574379324913025,
      "learning_rate": 3.423515769158872e-05,
      "loss": 0.0133,
      "num_input_tokens_seen": 18927712,
      "step": 5690
    },
    {
      "epoch": 1.1394842808193482,
      "grad_norm": 0.2827134430408478,
      "learning_rate": 3.421080984170565e-05,
      "loss": 0.0249,
      "num_input_tokens_seen": 18944288,
      "step": 5695
    },
    {
      "epoch": 1.14048470600005,
      "grad_norm": 0.09477187693119049,
      "learning_rate": 3.418645187890839e-05,
      "loss": 0.0237,
      "num_input_tokens_seen": 18960864,
      "step": 5700
    },
    {
      "epoch": 1.141485131180752,
      "grad_norm": 0.2097742259502411,
      "learning_rate": 3.416208382994051e-05,
      "loss": 0.0155,
      "num_input_tokens_seen": 18977568,
      "step": 5705
    },
    {
      "epoch": 1.1424855563614535,
      "grad_norm": 0.887037456035614,
      "learning_rate": 3.413770572155667e-05,
      "loss": 0.0303,
      "num_input_tokens_seen": 18993920,
      "step": 5710
    },
    {
      "epoch": 1.1434859815421554,
      "grad_norm": 0.1299106776714325,
      "learning_rate": 3.411331758052254e-05,
      "loss": 0.0171,
      "num_input_tokens_seen": 19010688,
      "step": 5715
    },
    {
      "epoch": 1.1444864067228573,
      "grad_norm": 1.0206340551376343,
      "learning_rate": 3.408891943361486e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 19028992,
      "step": 5720
    },
    {
      "epoch": 1.145486831903559,
      "grad_norm": 0.44007593393325806,
      "learning_rate": 3.406451130762131e-05,
      "loss": 0.019,
      "num_input_tokens_seen": 19045792,
      "step": 5725
    },
    {
      "epoch": 1.1464872570842608,
      "grad_norm": 0.297586590051651,
      "learning_rate": 3.404009322934052e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 19062208,
      "step": 5730
    },
    {
      "epoch": 1.1474876822649627,
      "grad_norm": 0.34563490748405457,
      "learning_rate": 3.4015665225582104e-05,
      "loss": 0.0117,
      "num_input_tokens_seen": 19078304,
      "step": 5735
    },
    {
      "epoch": 1.1484881074456643,
      "grad_norm": 0.38076719641685486,
      "learning_rate": 3.399122732316653e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 19095904,
      "step": 5740
    },
    {
      "epoch": 1.1494885326263662,
      "grad_norm": 0.1036958321928978,
      "learning_rate": 3.396677954892514e-05,
      "loss": 0.0243,
      "num_input_tokens_seen": 19111264,
      "step": 5745
    },
    {
      "epoch": 1.150488957807068,
      "grad_norm": 0.0501471608877182,
      "learning_rate": 3.3942321929700096e-05,
      "loss": 0.0102,
      "num_input_tokens_seen": 19128064,
      "step": 5750
    },
    {
      "epoch": 1.1514893829877697,
      "grad_norm": 0.27944546937942505,
      "learning_rate": 3.3917854492344416e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 19143904,
      "step": 5755
    },
    {
      "epoch": 1.1524898081684716,
      "grad_norm": 0.653773307800293,
      "learning_rate": 3.389337726372187e-05,
      "loss": 0.0128,
      "num_input_tokens_seen": 19159328,
      "step": 5760
    },
    {
      "epoch": 1.1534902333491734,
      "grad_norm": 0.3465997874736786,
      "learning_rate": 3.386889027070696e-05,
      "loss": 0.0109,
      "num_input_tokens_seen": 19175488,
      "step": 5765
    },
    {
      "epoch": 1.154490658529875,
      "grad_norm": 0.22995132207870483,
      "learning_rate": 3.3844393540184954e-05,
      "loss": 0.0209,
      "num_input_tokens_seen": 19192416,
      "step": 5770
    },
    {
      "epoch": 1.155491083710577,
      "grad_norm": 1.1864521503448486,
      "learning_rate": 3.381988709905177e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 19209952,
      "step": 5775
    },
    {
      "epoch": 1.1564915088912788,
      "grad_norm": 0.29596784710884094,
      "learning_rate": 3.379537097421401e-05,
      "loss": 0.0179,
      "num_input_tokens_seen": 19226112,
      "step": 5780
    },
    {
      "epoch": 1.1574919340719805,
      "grad_norm": 0.20320992171764374,
      "learning_rate": 3.37708451925889e-05,
      "loss": 0.0267,
      "num_input_tokens_seen": 19242816,
      "step": 5785
    },
    {
      "epoch": 1.1584923592526823,
      "grad_norm": 0.21257950365543365,
      "learning_rate": 3.374630978110428e-05,
      "loss": 0.0265,
      "num_input_tokens_seen": 19259456,
      "step": 5790
    },
    {
      "epoch": 1.1594927844333842,
      "grad_norm": 0.32195860147476196,
      "learning_rate": 3.372176476669853e-05,
      "loss": 0.024,
      "num_input_tokens_seen": 19275488,
      "step": 5795
    },
    {
      "epoch": 1.1604932096140859,
      "grad_norm": 0.3061094880104065,
      "learning_rate": 3.3697210176320615e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 19291232,
      "step": 5800
    },
    {
      "epoch": 1.1614936347947877,
      "grad_norm": 0.5570156574249268,
      "learning_rate": 3.3672646036929985e-05,
      "loss": 0.024,
      "num_input_tokens_seen": 19308448,
      "step": 5805
    },
    {
      "epoch": 1.1624940599754896,
      "grad_norm": 0.05850246548652649,
      "learning_rate": 3.3648072375496594e-05,
      "loss": 0.024,
      "num_input_tokens_seen": 19324416,
      "step": 5810
    },
    {
      "epoch": 1.1634944851561915,
      "grad_norm": 0.42457568645477295,
      "learning_rate": 3.362348921900085e-05,
      "loss": 0.0221,
      "num_input_tokens_seen": 19342176,
      "step": 5815
    },
    {
      "epoch": 1.1644949103368931,
      "grad_norm": 0.6461185216903687,
      "learning_rate": 3.359889659443356e-05,
      "loss": 0.0171,
      "num_input_tokens_seen": 19359264,
      "step": 5820
    },
    {
      "epoch": 1.165495335517595,
      "grad_norm": 0.24088598787784576,
      "learning_rate": 3.3574294528795945e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 19375808,
      "step": 5825
    },
    {
      "epoch": 1.1664957606982969,
      "grad_norm": 0.24455271661281586,
      "learning_rate": 3.35496830490996e-05,
      "loss": 0.021,
      "num_input_tokens_seen": 19391296,
      "step": 5830
    },
    {
      "epoch": 1.1674961858789985,
      "grad_norm": 0.48601317405700684,
      "learning_rate": 3.352506218236644e-05,
      "loss": 0.0147,
      "num_input_tokens_seen": 19408896,
      "step": 5835
    },
    {
      "epoch": 1.1684966110597004,
      "grad_norm": 0.13113577663898468,
      "learning_rate": 3.35004319556287e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 19425856,
      "step": 5840
    },
    {
      "epoch": 1.1694970362404022,
      "grad_norm": 0.21660153567790985,
      "learning_rate": 3.347579239592888e-05,
      "loss": 0.022,
      "num_input_tokens_seen": 19443200,
      "step": 5845
    },
    {
      "epoch": 1.170497461421104,
      "grad_norm": 0.39085274934768677,
      "learning_rate": 3.3451143530319725e-05,
      "loss": 0.0165,
      "num_input_tokens_seen": 19460800,
      "step": 5850
    },
    {
      "epoch": 1.1714978866018058,
      "grad_norm": 1.0306813716888428,
      "learning_rate": 3.342648538586422e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 19477280,
      "step": 5855
    },
    {
      "epoch": 1.1724983117825076,
      "grad_norm": 0.38740503787994385,
      "learning_rate": 3.340181798963551e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 19495232,
      "step": 5860
    },
    {
      "epoch": 1.1734987369632093,
      "grad_norm": 0.10791005939245224,
      "learning_rate": 3.337714136871691e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 19511232,
      "step": 5865
    },
    {
      "epoch": 1.1744991621439111,
      "grad_norm": 0.49075236916542053,
      "learning_rate": 3.335245555020187e-05,
      "loss": 0.0241,
      "num_input_tokens_seen": 19528256,
      "step": 5870
    },
    {
      "epoch": 1.175499587324613,
      "grad_norm": 0.8124425411224365,
      "learning_rate": 3.332776056119392e-05,
      "loss": 0.0284,
      "num_input_tokens_seen": 19545344,
      "step": 5875
    },
    {
      "epoch": 1.1765000125053147,
      "grad_norm": 0.543897807598114,
      "learning_rate": 3.330305642880669e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 19561792,
      "step": 5880
    },
    {
      "epoch": 1.1775004376860165,
      "grad_norm": 1.1725205183029175,
      "learning_rate": 3.327834318016381e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 19578016,
      "step": 5885
    },
    {
      "epoch": 1.1785008628667184,
      "grad_norm": 0.5985966324806213,
      "learning_rate": 3.325362084239894e-05,
      "loss": 0.0297,
      "num_input_tokens_seen": 19595776,
      "step": 5890
    },
    {
      "epoch": 1.1795012880474203,
      "grad_norm": 0.3250117301940918,
      "learning_rate": 3.322888944265572e-05,
      "loss": 0.0451,
      "num_input_tokens_seen": 19612960,
      "step": 5895
    },
    {
      "epoch": 1.180501713228122,
      "grad_norm": 0.5441887378692627,
      "learning_rate": 3.3204149008087745e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 19629728,
      "step": 5900
    },
    {
      "epoch": 1.1815021384088238,
      "grad_norm": 0.22758100926876068,
      "learning_rate": 3.317939956585851e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 19647360,
      "step": 5905
    },
    {
      "epoch": 1.1825025635895257,
      "grad_norm": 0.4903544783592224,
      "learning_rate": 3.315464114314141e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 19664128,
      "step": 5910
    },
    {
      "epoch": 1.1835029887702273,
      "grad_norm": 0.549884557723999,
      "learning_rate": 3.312987376711971e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 19681440,
      "step": 5915
    },
    {
      "epoch": 1.1845034139509292,
      "grad_norm": 0.3596501052379608,
      "learning_rate": 3.310509746498649e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 19697888,
      "step": 5920
    },
    {
      "epoch": 1.185503839131631,
      "grad_norm": 0.14788438379764557,
      "learning_rate": 3.308031226394464e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 19714400,
      "step": 5925
    },
    {
      "epoch": 1.1865042643123327,
      "grad_norm": 0.21356885135173798,
      "learning_rate": 3.3055518191206824e-05,
      "loss": 0.0247,
      "num_input_tokens_seen": 19730848,
      "step": 5930
    },
    {
      "epoch": 1.1875046894930346,
      "grad_norm": 0.20204849541187286,
      "learning_rate": 3.303071527399543e-05,
      "loss": 0.0174,
      "num_input_tokens_seen": 19747424,
      "step": 5935
    },
    {
      "epoch": 1.1885051146737364,
      "grad_norm": 0.47173598408699036,
      "learning_rate": 3.300590353954256e-05,
      "loss": 0.0239,
      "num_input_tokens_seen": 19763584,
      "step": 5940
    },
    {
      "epoch": 1.189505539854438,
      "grad_norm": 0.3403458893299103,
      "learning_rate": 3.298108301509003e-05,
      "loss": 0.0171,
      "num_input_tokens_seen": 19780000,
      "step": 5945
    },
    {
      "epoch": 1.19050596503514,
      "grad_norm": 0.10040679574012756,
      "learning_rate": 3.295625372788925e-05,
      "loss": 0.0355,
      "num_input_tokens_seen": 19796000,
      "step": 5950
    },
    {
      "epoch": 1.1915063902158418,
      "grad_norm": 0.725883424282074,
      "learning_rate": 3.2931415705201294e-05,
      "loss": 0.0214,
      "num_input_tokens_seen": 19812512,
      "step": 5955
    },
    {
      "epoch": 1.1925068153965435,
      "grad_norm": 0.2431889921426773,
      "learning_rate": 3.290656897429682e-05,
      "loss": 0.0312,
      "num_input_tokens_seen": 19829888,
      "step": 5960
    },
    {
      "epoch": 1.1935072405772453,
      "grad_norm": 0.3370024561882019,
      "learning_rate": 3.2881713562456036e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 19847712,
      "step": 5965
    },
    {
      "epoch": 1.1945076657579472,
      "grad_norm": 0.269488662481308,
      "learning_rate": 3.2856849496968685e-05,
      "loss": 0.0374,
      "num_input_tokens_seen": 19863776,
      "step": 5970
    },
    {
      "epoch": 1.1955080909386488,
      "grad_norm": 0.19755752384662628,
      "learning_rate": 3.283197680513401e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 19880000,
      "step": 5975
    },
    {
      "epoch": 1.1965085161193507,
      "grad_norm": 0.37241336703300476,
      "learning_rate": 3.280709551426074e-05,
      "loss": 0.018,
      "num_input_tokens_seen": 19896896,
      "step": 5980
    },
    {
      "epoch": 1.1975089413000526,
      "grad_norm": 0.15598493814468384,
      "learning_rate": 3.2782205651667013e-05,
      "loss": 0.0124,
      "num_input_tokens_seen": 19912544,
      "step": 5985
    },
    {
      "epoch": 1.1985093664807542,
      "grad_norm": 0.23613229393959045,
      "learning_rate": 3.275730724468041e-05,
      "loss": 0.021,
      "num_input_tokens_seen": 19928608,
      "step": 5990
    },
    {
      "epoch": 1.199509791661456,
      "grad_norm": 0.42579779028892517,
      "learning_rate": 3.27324003206379e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 19944832,
      "step": 5995
    },
    {
      "epoch": 1.200510216842158,
      "grad_norm": 0.6366907954216003,
      "learning_rate": 3.270748490688575e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 19961280,
      "step": 6000
    },
    {
      "epoch": 1.2015106420228596,
      "grad_norm": 0.19301056861877441,
      "learning_rate": 3.268256103077961e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 19977056,
      "step": 6005
    },
    {
      "epoch": 1.2025110672035615,
      "grad_norm": 0.7739194631576538,
      "learning_rate": 3.2657628719684377e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 19993696,
      "step": 6010
    },
    {
      "epoch": 1.2035114923842634,
      "grad_norm": 0.4243657886981964,
      "learning_rate": 3.2632688000974234e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 20010176,
      "step": 6015
    },
    {
      "epoch": 1.204511917564965,
      "grad_norm": 0.7800456881523132,
      "learning_rate": 3.2607738902032586e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 20026752,
      "step": 6020
    },
    {
      "epoch": 1.2055123427456669,
      "grad_norm": 0.4734976887702942,
      "learning_rate": 3.258278145025204e-05,
      "loss": 0.019,
      "num_input_tokens_seen": 20042784,
      "step": 6025
    },
    {
      "epoch": 1.2065127679263687,
      "grad_norm": 0.06025788187980652,
      "learning_rate": 3.255781567303438e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 20059488,
      "step": 6030
    },
    {
      "epoch": 1.2075131931070704,
      "grad_norm": 0.283933162689209,
      "learning_rate": 3.253284159779052e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 20076320,
      "step": 6035
    },
    {
      "epoch": 1.2085136182877723,
      "grad_norm": 0.3365752696990967,
      "learning_rate": 3.250785925194048e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 20093984,
      "step": 6040
    },
    {
      "epoch": 1.2095140434684741,
      "grad_norm": 0.35834792256355286,
      "learning_rate": 3.248286866291338e-05,
      "loss": 0.0196,
      "num_input_tokens_seen": 20109984,
      "step": 6045
    },
    {
      "epoch": 1.2105144686491758,
      "grad_norm": 0.26457342505455017,
      "learning_rate": 3.24578698581474e-05,
      "loss": 0.0292,
      "num_input_tokens_seen": 20125760,
      "step": 6050
    },
    {
      "epoch": 1.2115148938298776,
      "grad_norm": 0.3771316409111023,
      "learning_rate": 3.243286286508968e-05,
      "loss": 0.0289,
      "num_input_tokens_seen": 20141504,
      "step": 6055
    },
    {
      "epoch": 1.2125153190105795,
      "grad_norm": 0.19131837785243988,
      "learning_rate": 3.240784771119642e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 20158752,
      "step": 6060
    },
    {
      "epoch": 1.2135157441912814,
      "grad_norm": 0.7011873722076416,
      "learning_rate": 3.238282442393274e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 20175776,
      "step": 6065
    },
    {
      "epoch": 1.214516169371983,
      "grad_norm": 0.06214110925793648,
      "learning_rate": 3.235779303077272e-05,
      "loss": 0.0193,
      "num_input_tokens_seen": 20192704,
      "step": 6070
    },
    {
      "epoch": 1.215516594552685,
      "grad_norm": 0.32150939106941223,
      "learning_rate": 3.233275355919931e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 20211104,
      "step": 6075
    },
    {
      "epoch": 1.2165170197333868,
      "grad_norm": 0.19849102199077606,
      "learning_rate": 3.230770603670433e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 20228000,
      "step": 6080
    },
    {
      "epoch": 1.2175174449140884,
      "grad_norm": 0.8177050352096558,
      "learning_rate": 3.2282650490788466e-05,
      "loss": 0.022,
      "num_input_tokens_seen": 20244448,
      "step": 6085
    },
    {
      "epoch": 1.2185178700947903,
      "grad_norm": 1.0571411848068237,
      "learning_rate": 3.225758694896118e-05,
      "loss": 0.0463,
      "num_input_tokens_seen": 20261344,
      "step": 6090
    },
    {
      "epoch": 1.2195182952754922,
      "grad_norm": 0.6836508512496948,
      "learning_rate": 3.2232515438740754e-05,
      "loss": 0.0406,
      "num_input_tokens_seen": 20278368,
      "step": 6095
    },
    {
      "epoch": 1.2205187204561938,
      "grad_norm": 0.42014962434768677,
      "learning_rate": 3.220743598765418e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 20295456,
      "step": 6100
    },
    {
      "epoch": 1.2215191456368957,
      "grad_norm": 0.3057481646537781,
      "learning_rate": 3.218234862323718e-05,
      "loss": 0.0307,
      "num_input_tokens_seen": 20311520,
      "step": 6105
    },
    {
      "epoch": 1.2225195708175975,
      "grad_norm": 0.5091171264648438,
      "learning_rate": 3.2157253373034194e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 20328544,
      "step": 6110
    },
    {
      "epoch": 1.2235199959982992,
      "grad_norm": 0.3312574326992035,
      "learning_rate": 3.213215026459826e-05,
      "loss": 0.0137,
      "num_input_tokens_seen": 20344640,
      "step": 6115
    },
    {
      "epoch": 1.224520421179001,
      "grad_norm": 0.09498897194862366,
      "learning_rate": 3.2107039325491095e-05,
      "loss": 0.0186,
      "num_input_tokens_seen": 20359840,
      "step": 6120
    },
    {
      "epoch": 1.225520846359703,
      "grad_norm": 0.7953746914863586,
      "learning_rate": 3.2081920583283005e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 20377056,
      "step": 6125
    },
    {
      "epoch": 1.2265212715404046,
      "grad_norm": 0.34085607528686523,
      "learning_rate": 3.205679406555285e-05,
      "loss": 0.0184,
      "num_input_tokens_seen": 20393152,
      "step": 6130
    },
    {
      "epoch": 1.2275216967211064,
      "grad_norm": 0.37727925181388855,
      "learning_rate": 3.2031659799888016e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 20409888,
      "step": 6135
    },
    {
      "epoch": 1.2285221219018083,
      "grad_norm": 0.23031802475452423,
      "learning_rate": 3.2006517813884434e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 20426528,
      "step": 6140
    },
    {
      "epoch": 1.2295225470825102,
      "grad_norm": 0.8728134632110596,
      "learning_rate": 3.198136813514648e-05,
      "loss": 0.0241,
      "num_input_tokens_seen": 20443488,
      "step": 6145
    },
    {
      "epoch": 1.2305229722632118,
      "grad_norm": 0.5621263980865479,
      "learning_rate": 3.1956210791287e-05,
      "loss": 0.0188,
      "num_input_tokens_seen": 20458944,
      "step": 6150
    },
    {
      "epoch": 1.2315233974439137,
      "grad_norm": 0.6416258811950684,
      "learning_rate": 3.193104580992722e-05,
      "loss": 0.019,
      "num_input_tokens_seen": 20474784,
      "step": 6155
    },
    {
      "epoch": 1.2325238226246156,
      "grad_norm": 0.5702195763587952,
      "learning_rate": 3.190587321869678e-05,
      "loss": 0.0242,
      "num_input_tokens_seen": 20491264,
      "step": 6160
    },
    {
      "epoch": 1.2335242478053172,
      "grad_norm": 0.2383430451154709,
      "learning_rate": 3.188069304523367e-05,
      "loss": 0.022,
      "num_input_tokens_seen": 20507424,
      "step": 6165
    },
    {
      "epoch": 1.234524672986019,
      "grad_norm": 0.33039844036102295,
      "learning_rate": 3.18555053171842e-05,
      "loss": 0.0249,
      "num_input_tokens_seen": 20523360,
      "step": 6170
    },
    {
      "epoch": 1.235525098166721,
      "grad_norm": 0.735704779624939,
      "learning_rate": 3.1830310062202996e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 20539328,
      "step": 6175
    },
    {
      "epoch": 1.2365255233474226,
      "grad_norm": 0.1837497353553772,
      "learning_rate": 3.1805107307952906e-05,
      "loss": 0.0138,
      "num_input_tokens_seen": 20555872,
      "step": 6180
    },
    {
      "epoch": 1.2375259485281245,
      "grad_norm": 0.46224069595336914,
      "learning_rate": 3.1779897082105054e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 20572416,
      "step": 6185
    },
    {
      "epoch": 1.2385263737088263,
      "grad_norm": 0.15932872891426086,
      "learning_rate": 3.175467941233873e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 20589216,
      "step": 6190
    },
    {
      "epoch": 1.239526798889528,
      "grad_norm": 0.43344810605049133,
      "learning_rate": 3.172945432634142e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 20605056,
      "step": 6195
    },
    {
      "epoch": 1.2405272240702299,
      "grad_norm": 0.2447565197944641,
      "learning_rate": 3.170422185180877e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 20622336,
      "step": 6200
    },
    {
      "epoch": 1.2415276492509317,
      "grad_norm": 0.7022611498832703,
      "learning_rate": 3.167898201644449e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 20638528,
      "step": 6205
    },
    {
      "epoch": 1.2425280744316334,
      "grad_norm": 0.3187255859375,
      "learning_rate": 3.165373484796041e-05,
      "loss": 0.014,
      "num_input_tokens_seen": 20654656,
      "step": 6210
    },
    {
      "epoch": 1.2435284996123352,
      "grad_norm": 0.4601322412490845,
      "learning_rate": 3.162848037407641e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 20672128,
      "step": 6215
    },
    {
      "epoch": 1.2445289247930371,
      "grad_norm": 0.6867743730545044,
      "learning_rate": 3.160321862252036e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 20688096,
      "step": 6220
    },
    {
      "epoch": 1.2455293499737388,
      "grad_norm": 0.3737960755825043,
      "learning_rate": 3.157794962102815e-05,
      "loss": 0.0139,
      "num_input_tokens_seen": 20703360,
      "step": 6225
    },
    {
      "epoch": 1.2465297751544406,
      "grad_norm": 0.09392146021127701,
      "learning_rate": 3.155267339734362e-05,
      "loss": 0.0217,
      "num_input_tokens_seen": 20719936,
      "step": 6230
    },
    {
      "epoch": 1.2475302003351425,
      "grad_norm": 0.47164633870124817,
      "learning_rate": 3.1527389979218546e-05,
      "loss": 0.0156,
      "num_input_tokens_seen": 20737440,
      "step": 6235
    },
    {
      "epoch": 1.2485306255158442,
      "grad_norm": 0.5443530082702637,
      "learning_rate": 3.150209939441259e-05,
      "loss": 0.0445,
      "num_input_tokens_seen": 20754176,
      "step": 6240
    },
    {
      "epoch": 1.249531050696546,
      "grad_norm": 0.3592490553855896,
      "learning_rate": 3.14768016706933e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 20771232,
      "step": 6245
    },
    {
      "epoch": 1.2505314758772479,
      "grad_norm": 0.3470192551612854,
      "learning_rate": 3.1451496835836044e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 20789184,
      "step": 6250
    },
    {
      "epoch": 1.2515319010579495,
      "grad_norm": 0.5166939496994019,
      "learning_rate": 3.1426184917624e-05,
      "loss": 0.035,
      "num_input_tokens_seen": 20805152,
      "step": 6255
    },
    {
      "epoch": 1.2525323262386514,
      "grad_norm": 0.2819918692111969,
      "learning_rate": 3.1400865943848146e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 20822432,
      "step": 6260
    },
    {
      "epoch": 1.2535327514193533,
      "grad_norm": 0.39920511841773987,
      "learning_rate": 3.137553994230718e-05,
      "loss": 0.0189,
      "num_input_tokens_seen": 20838656,
      "step": 6265
    },
    {
      "epoch": 1.254533176600055,
      "grad_norm": 0.2822507321834564,
      "learning_rate": 3.135020694080752e-05,
      "loss": 0.0292,
      "num_input_tokens_seen": 20855552,
      "step": 6270
    },
    {
      "epoch": 1.2555336017807568,
      "grad_norm": 1.1919745206832886,
      "learning_rate": 3.132486696716329e-05,
      "loss": 0.0348,
      "num_input_tokens_seen": 20871200,
      "step": 6275
    },
    {
      "epoch": 1.2565340269614587,
      "grad_norm": 0.3402027487754822,
      "learning_rate": 3.129952004919624e-05,
      "loss": 0.0268,
      "num_input_tokens_seen": 20889312,
      "step": 6280
    },
    {
      "epoch": 1.2575344521421603,
      "grad_norm": 0.7842808961868286,
      "learning_rate": 3.127416621473576e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 20905376,
      "step": 6285
    },
    {
      "epoch": 1.2585348773228622,
      "grad_norm": 0.10425746440887451,
      "learning_rate": 3.124880549161885e-05,
      "loss": 0.0185,
      "num_input_tokens_seen": 20923360,
      "step": 6290
    },
    {
      "epoch": 1.259535302503564,
      "grad_norm": 0.6856287717819214,
      "learning_rate": 3.122343790769003e-05,
      "loss": 0.0298,
      "num_input_tokens_seen": 20939840,
      "step": 6295
    },
    {
      "epoch": 1.2605357276842657,
      "grad_norm": 0.2629176080226898,
      "learning_rate": 3.119806349080139e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 20956928,
      "step": 6300
    },
    {
      "epoch": 1.2615361528649676,
      "grad_norm": 0.6491134166717529,
      "learning_rate": 3.117268226881252e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 20973024,
      "step": 6305
    },
    {
      "epoch": 1.2625365780456694,
      "grad_norm": 0.3988141119480133,
      "learning_rate": 3.114729426959046e-05,
      "loss": 0.0158,
      "num_input_tokens_seen": 20988544,
      "step": 6310
    },
    {
      "epoch": 1.263537003226371,
      "grad_norm": 0.30559006333351135,
      "learning_rate": 3.11218995210097e-05,
      "loss": 0.0147,
      "num_input_tokens_seen": 21005920,
      "step": 6315
    },
    {
      "epoch": 1.264537428407073,
      "grad_norm": 0.23855215311050415,
      "learning_rate": 3.109649805095216e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 21022528,
      "step": 6320
    },
    {
      "epoch": 1.2655378535877748,
      "grad_norm": 0.3411101698875427,
      "learning_rate": 3.107108988730712e-05,
      "loss": 0.0375,
      "num_input_tokens_seen": 21038560,
      "step": 6325
    },
    {
      "epoch": 1.2665382787684765,
      "grad_norm": 0.09174685180187225,
      "learning_rate": 3.1045675057971204e-05,
      "loss": 0.0342,
      "num_input_tokens_seen": 21055200,
      "step": 6330
    },
    {
      "epoch": 1.2675387039491783,
      "grad_norm": 0.7968325614929199,
      "learning_rate": 3.102025359084836e-05,
      "loss": 0.0329,
      "num_input_tokens_seen": 21071648,
      "step": 6335
    },
    {
      "epoch": 1.2685391291298802,
      "grad_norm": 0.650685727596283,
      "learning_rate": 3.099482551384984e-05,
      "loss": 0.0205,
      "num_input_tokens_seen": 21088864,
      "step": 6340
    },
    {
      "epoch": 1.269539554310582,
      "grad_norm": 0.7141836285591125,
      "learning_rate": 3.096939085489413e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 21105472,
      "step": 6345
    },
    {
      "epoch": 1.2705399794912837,
      "grad_norm": 0.36796361207962036,
      "learning_rate": 3.094394964190696e-05,
      "loss": 0.0275,
      "num_input_tokens_seen": 21121088,
      "step": 6350
    },
    {
      "epoch": 1.2715404046719856,
      "grad_norm": 1.0574023723602295,
      "learning_rate": 3.091850190282124e-05,
      "loss": 0.0256,
      "num_input_tokens_seen": 21139008,
      "step": 6355
    },
    {
      "epoch": 1.2725408298526875,
      "grad_norm": 0.5442276000976562,
      "learning_rate": 3.089304766557707e-05,
      "loss": 0.0146,
      "num_input_tokens_seen": 21156160,
      "step": 6360
    },
    {
      "epoch": 1.273541255033389,
      "grad_norm": 0.46550431847572327,
      "learning_rate": 3.0867586958121654e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 21172064,
      "step": 6365
    },
    {
      "epoch": 1.274541680214091,
      "grad_norm": 0.11053989082574844,
      "learning_rate": 3.0842119808409315e-05,
      "loss": 0.0173,
      "num_input_tokens_seen": 21188736,
      "step": 6370
    },
    {
      "epoch": 1.2755421053947928,
      "grad_norm": 0.2156449854373932,
      "learning_rate": 3.0816646244401455e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 21204320,
      "step": 6375
    },
    {
      "epoch": 1.2765425305754947,
      "grad_norm": 0.15073303878307343,
      "learning_rate": 3.079116629406651e-05,
      "loss": 0.021,
      "num_input_tokens_seen": 21221120,
      "step": 6380
    },
    {
      "epoch": 1.2775429557561964,
      "grad_norm": 0.4173339307308197,
      "learning_rate": 3.0765679985379935e-05,
      "loss": 0.0205,
      "num_input_tokens_seen": 21237408,
      "step": 6385
    },
    {
      "epoch": 1.2785433809368982,
      "grad_norm": 1.0750230550765991,
      "learning_rate": 3.074018734632416e-05,
      "loss": 0.0493,
      "num_input_tokens_seen": 21253408,
      "step": 6390
    },
    {
      "epoch": 1.2795438061176,
      "grad_norm": 0.5304452776908875,
      "learning_rate": 3.071468840488855e-05,
      "loss": 0.0112,
      "num_input_tokens_seen": 21270432,
      "step": 6395
    },
    {
      "epoch": 1.2805442312983017,
      "grad_norm": 0.32167404890060425,
      "learning_rate": 3.068918318906943e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 21287040,
      "step": 6400
    },
    {
      "epoch": 1.2815446564790036,
      "grad_norm": 0.5210312008857727,
      "learning_rate": 3.066367172686998e-05,
      "loss": 0.014,
      "num_input_tokens_seen": 21302752,
      "step": 6405
    },
    {
      "epoch": 1.2825450816597055,
      "grad_norm": 0.31968334317207336,
      "learning_rate": 3.063815404630024e-05,
      "loss": 0.0105,
      "num_input_tokens_seen": 21318368,
      "step": 6410
    },
    {
      "epoch": 1.2835455068404071,
      "grad_norm": 0.16647179424762726,
      "learning_rate": 3.06126301753771e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 21334464,
      "step": 6415
    },
    {
      "epoch": 1.284545932021109,
      "grad_norm": 0.8157343864440918,
      "learning_rate": 3.0587100142124225e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 21351840,
      "step": 6420
    },
    {
      "epoch": 1.2855463572018109,
      "grad_norm": 0.6097559928894043,
      "learning_rate": 3.056156397457205e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 21368416,
      "step": 6425
    },
    {
      "epoch": 1.2865467823825125,
      "grad_norm": 0.5425021648406982,
      "learning_rate": 3.053602170075775e-05,
      "loss": 0.0437,
      "num_input_tokens_seen": 21385344,
      "step": 6430
    },
    {
      "epoch": 1.2875472075632144,
      "grad_norm": 0.6646090745925903,
      "learning_rate": 3.051047334872521e-05,
      "loss": 0.02,
      "num_input_tokens_seen": 21400288,
      "step": 6435
    },
    {
      "epoch": 1.2885476327439163,
      "grad_norm": 0.26214632391929626,
      "learning_rate": 3.0484918946524983e-05,
      "loss": 0.0205,
      "num_input_tokens_seen": 21416704,
      "step": 6440
    },
    {
      "epoch": 1.289548057924618,
      "grad_norm": 0.2173846811056137,
      "learning_rate": 3.0459358522214244e-05,
      "loss": 0.0151,
      "num_input_tokens_seen": 21432928,
      "step": 6445
    },
    {
      "epoch": 1.2905484831053198,
      "grad_norm": 0.31593844294548035,
      "learning_rate": 3.043379210385681e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 21448928,
      "step": 6450
    },
    {
      "epoch": 1.2915489082860216,
      "grad_norm": 0.6138732433319092,
      "learning_rate": 3.0408219719523077e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 21466240,
      "step": 6455
    },
    {
      "epoch": 1.2925493334667233,
      "grad_norm": 0.15956415235996246,
      "learning_rate": 3.0382641397289967e-05,
      "loss": 0.0154,
      "num_input_tokens_seen": 21482080,
      "step": 6460
    },
    {
      "epoch": 1.2935497586474252,
      "grad_norm": 0.5225535035133362,
      "learning_rate": 3.0357057165240943e-05,
      "loss": 0.0199,
      "num_input_tokens_seen": 21497920,
      "step": 6465
    },
    {
      "epoch": 1.294550183828127,
      "grad_norm": 0.4933944344520569,
      "learning_rate": 3.0331467051465955e-05,
      "loss": 0.03,
      "num_input_tokens_seen": 21514848,
      "step": 6470
    },
    {
      "epoch": 1.2955506090088287,
      "grad_norm": 0.8946163058280945,
      "learning_rate": 3.0305871084061398e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 21531136,
      "step": 6475
    },
    {
      "epoch": 1.2965510341895305,
      "grad_norm": 0.2678624391555786,
      "learning_rate": 3.0280269291130104e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 21546208,
      "step": 6480
    },
    {
      "epoch": 1.2975514593702324,
      "grad_norm": 0.4984210133552551,
      "learning_rate": 3.02546617007813e-05,
      "loss": 0.0194,
      "num_input_tokens_seen": 21561632,
      "step": 6485
    },
    {
      "epoch": 1.298551884550934,
      "grad_norm": 0.8703275918960571,
      "learning_rate": 3.022904834113058e-05,
      "loss": 0.0208,
      "num_input_tokens_seen": 21577760,
      "step": 6490
    },
    {
      "epoch": 1.299552309731636,
      "grad_norm": 0.43906381726264954,
      "learning_rate": 3.0203429240299865e-05,
      "loss": 0.022,
      "num_input_tokens_seen": 21594176,
      "step": 6495
    },
    {
      "epoch": 1.3005527349123378,
      "grad_norm": 0.18665099143981934,
      "learning_rate": 3.0177804426417384e-05,
      "loss": 0.0115,
      "num_input_tokens_seen": 21611552,
      "step": 6500
    },
    {
      "epoch": 1.3015531600930395,
      "grad_norm": 0.3526068925857544,
      "learning_rate": 3.0152173927617634e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 21628864,
      "step": 6505
    },
    {
      "epoch": 1.3025535852737413,
      "grad_norm": 0.19899950921535492,
      "learning_rate": 3.0126537772041375e-05,
      "loss": 0.0621,
      "num_input_tokens_seen": 21645408,
      "step": 6510
    },
    {
      "epoch": 1.3035540104544432,
      "grad_norm": 0.21650636196136475,
      "learning_rate": 3.0100895987835547e-05,
      "loss": 0.0161,
      "num_input_tokens_seen": 21661952,
      "step": 6515
    },
    {
      "epoch": 1.3045544356351448,
      "grad_norm": 0.11611881852149963,
      "learning_rate": 3.0075248603153284e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 21677760,
      "step": 6520
    },
    {
      "epoch": 1.3055548608158467,
      "grad_norm": 0.25818386673927307,
      "learning_rate": 3.0049595646153877e-05,
      "loss": 0.0135,
      "num_input_tokens_seen": 21694560,
      "step": 6525
    },
    {
      "epoch": 1.3065552859965486,
      "grad_norm": 0.281645268201828,
      "learning_rate": 3.0023937145002717e-05,
      "loss": 0.016,
      "num_input_tokens_seen": 21711392,
      "step": 6530
    },
    {
      "epoch": 1.3075557111772502,
      "grad_norm": 0.5636654496192932,
      "learning_rate": 2.9998273127871302e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 21728800,
      "step": 6535
    },
    {
      "epoch": 1.308556136357952,
      "grad_norm": 0.3114864230155945,
      "learning_rate": 2.9972603622937167e-05,
      "loss": 0.0094,
      "num_input_tokens_seen": 21745664,
      "step": 6540
    },
    {
      "epoch": 1.309556561538654,
      "grad_norm": 0.8865411877632141,
      "learning_rate": 2.9946928658383893e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 21761536,
      "step": 6545
    },
    {
      "epoch": 1.3105569867193556,
      "grad_norm": 0.29117724299430847,
      "learning_rate": 2.992124826240104e-05,
      "loss": 0.014,
      "num_input_tokens_seen": 21777888,
      "step": 6550
    },
    {
      "epoch": 1.3115574119000575,
      "grad_norm": 0.3249759376049042,
      "learning_rate": 2.989556246318412e-05,
      "loss": 0.0192,
      "num_input_tokens_seen": 21793952,
      "step": 6555
    },
    {
      "epoch": 1.3125578370807593,
      "grad_norm": 1.2025161981582642,
      "learning_rate": 2.9869871288934615e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 21810208,
      "step": 6560
    },
    {
      "epoch": 1.313558262261461,
      "grad_norm": 0.5750439167022705,
      "learning_rate": 2.9844174767859877e-05,
      "loss": 0.0228,
      "num_input_tokens_seen": 21826368,
      "step": 6565
    },
    {
      "epoch": 1.3145586874421629,
      "grad_norm": 0.24626857042312622,
      "learning_rate": 2.9818472928173136e-05,
      "loss": 0.0097,
      "num_input_tokens_seen": 21844192,
      "step": 6570
    },
    {
      "epoch": 1.3155591126228647,
      "grad_norm": 1.0716522932052612,
      "learning_rate": 2.9792765798093465e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 21861312,
      "step": 6575
    },
    {
      "epoch": 1.3165595378035664,
      "grad_norm": 0.2427041381597519,
      "learning_rate": 2.976705340584574e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 21878080,
      "step": 6580
    },
    {
      "epoch": 1.3175599629842683,
      "grad_norm": 0.1594732105731964,
      "learning_rate": 2.974133577966063e-05,
      "loss": 0.013,
      "num_input_tokens_seen": 21893856,
      "step": 6585
    },
    {
      "epoch": 1.3185603881649701,
      "grad_norm": 0.49334800243377686,
      "learning_rate": 2.9715612947774523e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 21910496,
      "step": 6590
    },
    {
      "epoch": 1.319560813345672,
      "grad_norm": 0.2606200873851776,
      "learning_rate": 2.9689884938429552e-05,
      "loss": 0.0096,
      "num_input_tokens_seen": 21927488,
      "step": 6595
    },
    {
      "epoch": 1.3205612385263736,
      "grad_norm": 0.9450080394744873,
      "learning_rate": 2.9664151779873516e-05,
      "loss": 0.0471,
      "num_input_tokens_seen": 21943360,
      "step": 6600
    },
    {
      "epoch": 1.3215616637070755,
      "grad_norm": 0.2751266360282898,
      "learning_rate": 2.9638413500359868e-05,
      "loss": 0.0349,
      "num_input_tokens_seen": 21960320,
      "step": 6605
    },
    {
      "epoch": 1.3225620888877774,
      "grad_norm": 0.6991470456123352,
      "learning_rate": 2.961267012814769e-05,
      "loss": 0.0194,
      "num_input_tokens_seen": 21977920,
      "step": 6610
    },
    {
      "epoch": 1.323562514068479,
      "grad_norm": 0.5267415642738342,
      "learning_rate": 2.9586921691501662e-05,
      "loss": 0.0303,
      "num_input_tokens_seen": 21996000,
      "step": 6615
    },
    {
      "epoch": 1.324562939249181,
      "grad_norm": 0.19206087291240692,
      "learning_rate": 2.9561168218692008e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 22012064,
      "step": 6620
    },
    {
      "epoch": 1.3255633644298828,
      "grad_norm": 0.3343439996242523,
      "learning_rate": 2.953540973799449e-05,
      "loss": 0.0161,
      "num_input_tokens_seen": 22028832,
      "step": 6625
    },
    {
      "epoch": 1.3265637896105846,
      "grad_norm": 0.031643129885196686,
      "learning_rate": 2.9509646277690362e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 22047168,
      "step": 6630
    },
    {
      "epoch": 1.3275642147912863,
      "grad_norm": 0.3571871221065521,
      "learning_rate": 2.9483877866066368e-05,
      "loss": 0.0201,
      "num_input_tokens_seen": 22063840,
      "step": 6635
    },
    {
      "epoch": 1.3285646399719881,
      "grad_norm": 0.09580302238464355,
      "learning_rate": 2.9458104531414654e-05,
      "loss": 0.0108,
      "num_input_tokens_seen": 22079968,
      "step": 6640
    },
    {
      "epoch": 1.32956506515269,
      "grad_norm": 0.920793354511261,
      "learning_rate": 2.9432326302032793e-05,
      "loss": 0.0355,
      "num_input_tokens_seen": 22096320,
      "step": 6645
    },
    {
      "epoch": 1.3305654903333917,
      "grad_norm": 0.5107559561729431,
      "learning_rate": 2.9406543206223735e-05,
      "loss": 0.0151,
      "num_input_tokens_seen": 22113600,
      "step": 6650
    },
    {
      "epoch": 1.3315659155140935,
      "grad_norm": 0.9166436195373535,
      "learning_rate": 2.9380755272295758e-05,
      "loss": 0.0202,
      "num_input_tokens_seen": 22129664,
      "step": 6655
    },
    {
      "epoch": 1.3325663406947954,
      "grad_norm": 0.2292318493127823,
      "learning_rate": 2.9354962528562463e-05,
      "loss": 0.0141,
      "num_input_tokens_seen": 22145728,
      "step": 6660
    },
    {
      "epoch": 1.333566765875497,
      "grad_norm": 0.16360390186309814,
      "learning_rate": 2.9329165003342722e-05,
      "loss": 0.0226,
      "num_input_tokens_seen": 22162880,
      "step": 6665
    },
    {
      "epoch": 1.334567191056199,
      "grad_norm": 0.2184598594903946,
      "learning_rate": 2.9303362724960677e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 22180672,
      "step": 6670
    },
    {
      "epoch": 1.3355676162369008,
      "grad_norm": 0.5025904774665833,
      "learning_rate": 2.9277555721745664e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 22197568,
      "step": 6675
    },
    {
      "epoch": 1.3365680414176024,
      "grad_norm": 0.45514610409736633,
      "learning_rate": 2.925174402203221e-05,
      "loss": 0.0442,
      "num_input_tokens_seen": 22214368,
      "step": 6680
    },
    {
      "epoch": 1.3375684665983043,
      "grad_norm": 0.19939345121383667,
      "learning_rate": 2.9225927654160023e-05,
      "loss": 0.0143,
      "num_input_tokens_seen": 22230624,
      "step": 6685
    },
    {
      "epoch": 1.3385688917790062,
      "grad_norm": 0.5434321165084839,
      "learning_rate": 2.9200106646473917e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 22246208,
      "step": 6690
    },
    {
      "epoch": 1.3395693169597078,
      "grad_norm": 0.7273253798484802,
      "learning_rate": 2.9174281027323796e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 22263072,
      "step": 6695
    },
    {
      "epoch": 1.3405697421404097,
      "grad_norm": 0.33907535672187805,
      "learning_rate": 2.9148450825064633e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 22279968,
      "step": 6700
    },
    {
      "epoch": 1.3415701673211116,
      "grad_norm": 0.19482874870300293,
      "learning_rate": 2.9122616068056436e-05,
      "loss": 0.0273,
      "num_input_tokens_seen": 22295904,
      "step": 6705
    },
    {
      "epoch": 1.3425705925018132,
      "grad_norm": 0.37009555101394653,
      "learning_rate": 2.9096776784664204e-05,
      "loss": 0.0221,
      "num_input_tokens_seen": 22312704,
      "step": 6710
    },
    {
      "epoch": 1.343571017682515,
      "grad_norm": 0.6175490021705627,
      "learning_rate": 2.9070933003257933e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 22329376,
      "step": 6715
    },
    {
      "epoch": 1.344571442863217,
      "grad_norm": 0.13512542843818665,
      "learning_rate": 2.9045084752212515e-05,
      "loss": 0.0271,
      "num_input_tokens_seen": 22344672,
      "step": 6720
    },
    {
      "epoch": 1.3455718680439186,
      "grad_norm": 0.40606412291526794,
      "learning_rate": 2.901923205990779e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 22361120,
      "step": 6725
    },
    {
      "epoch": 1.3465722932246205,
      "grad_norm": 0.6456787586212158,
      "learning_rate": 2.8993374954728448e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 22378656,
      "step": 6730
    },
    {
      "epoch": 1.3475727184053223,
      "grad_norm": 0.6901668310165405,
      "learning_rate": 2.8967513465064028e-05,
      "loss": 0.0253,
      "num_input_tokens_seen": 22394560,
      "step": 6735
    },
    {
      "epoch": 1.348573143586024,
      "grad_norm": 0.5287696719169617,
      "learning_rate": 2.894164761930889e-05,
      "loss": 0.0214,
      "num_input_tokens_seen": 22410880,
      "step": 6740
    },
    {
      "epoch": 1.3495735687667259,
      "grad_norm": 0.07469778507947922,
      "learning_rate": 2.8915777445862184e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 22427648,
      "step": 6745
    },
    {
      "epoch": 1.3505739939474277,
      "grad_norm": 0.35534486174583435,
      "learning_rate": 2.8889902973127796e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 22442976,
      "step": 6750
    },
    {
      "epoch": 1.3515744191281294,
      "grad_norm": 0.37630361318588257,
      "learning_rate": 2.886402422951433e-05,
      "loss": 0.0297,
      "num_input_tokens_seen": 22458784,
      "step": 6755
    },
    {
      "epoch": 1.3525748443088312,
      "grad_norm": 0.23175548017024994,
      "learning_rate": 2.883814124343509e-05,
      "loss": 0.0152,
      "num_input_tokens_seen": 22475872,
      "step": 6760
    },
    {
      "epoch": 1.353575269489533,
      "grad_norm": 0.23230740427970886,
      "learning_rate": 2.8812254043308052e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 22493344,
      "step": 6765
    },
    {
      "epoch": 1.3545756946702348,
      "grad_norm": 0.1427178680896759,
      "learning_rate": 2.8786362657555782e-05,
      "loss": 0.008,
      "num_input_tokens_seen": 22509344,
      "step": 6770
    },
    {
      "epoch": 1.3555761198509366,
      "grad_norm": 0.9476840496063232,
      "learning_rate": 2.8760467114605462e-05,
      "loss": 0.0318,
      "num_input_tokens_seen": 22525568,
      "step": 6775
    },
    {
      "epoch": 1.3565765450316385,
      "grad_norm": 0.44155165553092957,
      "learning_rate": 2.873456744288885e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 22542560,
      "step": 6780
    },
    {
      "epoch": 1.3575769702123401,
      "grad_norm": 0.45242565870285034,
      "learning_rate": 2.870866367084221e-05,
      "loss": 0.0168,
      "num_input_tokens_seen": 22559424,
      "step": 6785
    },
    {
      "epoch": 1.358577395393042,
      "grad_norm": 1.2894710302352905,
      "learning_rate": 2.868275582690634e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 22576064,
      "step": 6790
    },
    {
      "epoch": 1.3595778205737439,
      "grad_norm": 0.4912334084510803,
      "learning_rate": 2.8656843939526473e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 22591904,
      "step": 6795
    },
    {
      "epoch": 1.3605782457544455,
      "grad_norm": 0.2881946563720703,
      "learning_rate": 2.8630928037152322e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 22609024,
      "step": 6800
    },
    {
      "epoch": 1.3615786709351474,
      "grad_norm": 0.732107400894165,
      "learning_rate": 2.8605008148237965e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 22625088,
      "step": 6805
    },
    {
      "epoch": 1.3625790961158493,
      "grad_norm": 0.23752035200595856,
      "learning_rate": 2.8579084301241886e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 22642144,
      "step": 6810
    },
    {
      "epoch": 1.363579521296551,
      "grad_norm": 0.17034226655960083,
      "learning_rate": 2.855315652462691e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 22658592,
      "step": 6815
    },
    {
      "epoch": 1.3645799464772528,
      "grad_norm": 0.1437387317419052,
      "learning_rate": 2.852722484686018e-05,
      "loss": 0.0087,
      "num_input_tokens_seen": 22674432,
      "step": 6820
    },
    {
      "epoch": 1.3655803716579547,
      "grad_norm": 0.5661661028862,
      "learning_rate": 2.8501289296413103e-05,
      "loss": 0.037,
      "num_input_tokens_seen": 22690624,
      "step": 6825
    },
    {
      "epoch": 1.3665807968386563,
      "grad_norm": 0.27419745922088623,
      "learning_rate": 2.8475349901761344e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 22707424,
      "step": 6830
    },
    {
      "epoch": 1.3675812220193582,
      "grad_norm": 0.33506685495376587,
      "learning_rate": 2.844940669138481e-05,
      "loss": 0.0173,
      "num_input_tokens_seen": 22724576,
      "step": 6835
    },
    {
      "epoch": 1.36858164720006,
      "grad_norm": 0.2337416410446167,
      "learning_rate": 2.8423459693767585e-05,
      "loss": 0.0183,
      "num_input_tokens_seen": 22740992,
      "step": 6840
    },
    {
      "epoch": 1.369582072380762,
      "grad_norm": 0.4036204218864441,
      "learning_rate": 2.8397508937397904e-05,
      "loss": 0.0309,
      "num_input_tokens_seen": 22759104,
      "step": 6845
    },
    {
      "epoch": 1.3705824975614636,
      "grad_norm": 0.2100093513727188,
      "learning_rate": 2.8371554450768124e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 22775008,
      "step": 6850
    },
    {
      "epoch": 1.3715829227421654,
      "grad_norm": 0.11855494976043701,
      "learning_rate": 2.8345596262374718e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 22791424,
      "step": 6855
    },
    {
      "epoch": 1.3725833479228673,
      "grad_norm": 0.7924718856811523,
      "learning_rate": 2.8319634400718215e-05,
      "loss": 0.0223,
      "num_input_tokens_seen": 22807680,
      "step": 6860
    },
    {
      "epoch": 1.373583773103569,
      "grad_norm": 0.4390965402126312,
      "learning_rate": 2.8293668894303165e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 22825568,
      "step": 6865
    },
    {
      "epoch": 1.3745841982842708,
      "grad_norm": 1.074836015701294,
      "learning_rate": 2.8267699771638136e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 22842080,
      "step": 6870
    },
    {
      "epoch": 1.3755846234649727,
      "grad_norm": 0.17005810141563416,
      "learning_rate": 2.8241727061235662e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 22858624,
      "step": 6875
    },
    {
      "epoch": 1.3765850486456745,
      "grad_norm": 0.42438748478889465,
      "learning_rate": 2.8215750791612207e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 22874944,
      "step": 6880
    },
    {
      "epoch": 1.3775854738263762,
      "grad_norm": 0.13305796682834625,
      "learning_rate": 2.818977099128815e-05,
      "loss": 0.0268,
      "num_input_tokens_seen": 22892352,
      "step": 6885
    },
    {
      "epoch": 1.378585899007078,
      "grad_norm": 0.2759329676628113,
      "learning_rate": 2.8163787688787752e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 22908224,
      "step": 6890
    },
    {
      "epoch": 1.37958632418778,
      "grad_norm": 0.8038346171379089,
      "learning_rate": 2.813780091263912e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 22925088,
      "step": 6895
    },
    {
      "epoch": 1.3805867493684816,
      "grad_norm": 0.13557299971580505,
      "learning_rate": 2.811181069137415e-05,
      "loss": 0.0201,
      "num_input_tokens_seen": 22941920,
      "step": 6900
    },
    {
      "epoch": 1.3815871745491834,
      "grad_norm": 0.20478865504264832,
      "learning_rate": 2.8085817053528546e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 22959616,
      "step": 6905
    },
    {
      "epoch": 1.3825875997298853,
      "grad_norm": 0.3666134178638458,
      "learning_rate": 2.8059820027641765e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 22975872,
      "step": 6910
    },
    {
      "epoch": 1.383588024910587,
      "grad_norm": 0.8973509669303894,
      "learning_rate": 2.8033819642256963e-05,
      "loss": 0.0425,
      "num_input_tokens_seen": 22992800,
      "step": 6915
    },
    {
      "epoch": 1.3845884500912888,
      "grad_norm": 0.5391751527786255,
      "learning_rate": 2.8007815925921006e-05,
      "loss": 0.0351,
      "num_input_tokens_seen": 23009856,
      "step": 6920
    },
    {
      "epoch": 1.3855888752719907,
      "grad_norm": 0.22452957928180695,
      "learning_rate": 2.7981808907184398e-05,
      "loss": 0.0241,
      "num_input_tokens_seen": 23026848,
      "step": 6925
    },
    {
      "epoch": 1.3865893004526924,
      "grad_norm": 0.6008860468864441,
      "learning_rate": 2.7955798614601288e-05,
      "loss": 0.0285,
      "num_input_tokens_seen": 23042944,
      "step": 6930
    },
    {
      "epoch": 1.3875897256333942,
      "grad_norm": 0.6881251335144043,
      "learning_rate": 2.792978507672941e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 23060128,
      "step": 6935
    },
    {
      "epoch": 1.388590150814096,
      "grad_norm": 0.07176212221384048,
      "learning_rate": 2.7903768322130043e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 23078208,
      "step": 6940
    },
    {
      "epoch": 1.3895905759947977,
      "grad_norm": 0.32679834961891174,
      "learning_rate": 2.787774837936804e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 23095136,
      "step": 6945
    },
    {
      "epoch": 1.3905910011754996,
      "grad_norm": 0.17960023880004883,
      "learning_rate": 2.785172527701172e-05,
      "loss": 0.0246,
      "num_input_tokens_seen": 23110912,
      "step": 6950
    },
    {
      "epoch": 1.3915914263562015,
      "grad_norm": 0.3369027376174927,
      "learning_rate": 2.782569904363288e-05,
      "loss": 0.0106,
      "num_input_tokens_seen": 23127520,
      "step": 6955
    },
    {
      "epoch": 1.3925918515369031,
      "grad_norm": 0.31298232078552246,
      "learning_rate": 2.7799669707806758e-05,
      "loss": 0.0159,
      "num_input_tokens_seen": 23145792,
      "step": 6960
    },
    {
      "epoch": 1.393592276717605,
      "grad_norm": 1.370901107788086,
      "learning_rate": 2.7773637298111988e-05,
      "loss": 0.0335,
      "num_input_tokens_seen": 23162368,
      "step": 6965
    },
    {
      "epoch": 1.3945927018983069,
      "grad_norm": 0.2301444262266159,
      "learning_rate": 2.7747601843130607e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 23178048,
      "step": 6970
    },
    {
      "epoch": 1.3955931270790085,
      "grad_norm": 0.2517456114292145,
      "learning_rate": 2.772156337144795e-05,
      "loss": 0.0097,
      "num_input_tokens_seen": 23196352,
      "step": 6975
    },
    {
      "epoch": 1.3965935522597104,
      "grad_norm": 0.6408939957618713,
      "learning_rate": 2.7695521911652715e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 23212096,
      "step": 6980
    },
    {
      "epoch": 1.3975939774404122,
      "grad_norm": 0.6506691575050354,
      "learning_rate": 2.7669477492336844e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 23229216,
      "step": 6985
    },
    {
      "epoch": 1.398594402621114,
      "grad_norm": 0.1905229240655899,
      "learning_rate": 2.7643430142095532e-05,
      "loss": 0.0199,
      "num_input_tokens_seen": 23245312,
      "step": 6990
    },
    {
      "epoch": 1.3995948278018158,
      "grad_norm": 0.7158108949661255,
      "learning_rate": 2.761737988952721e-05,
      "loss": 0.0247,
      "num_input_tokens_seen": 23260896,
      "step": 6995
    },
    {
      "epoch": 1.4005952529825176,
      "grad_norm": 0.3425590395927429,
      "learning_rate": 2.7591326763233493e-05,
      "loss": 0.0109,
      "num_input_tokens_seen": 23277216,
      "step": 7000
    },
    {
      "epoch": 1.4015956781632193,
      "grad_norm": 1.0016491413116455,
      "learning_rate": 2.756527079181913e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 23294368,
      "step": 7005
    },
    {
      "epoch": 1.4025961033439212,
      "grad_norm": 0.25987598299980164,
      "learning_rate": 2.7539212003892013e-05,
      "loss": 0.0144,
      "num_input_tokens_seen": 23310816,
      "step": 7010
    },
    {
      "epoch": 1.403596528524623,
      "grad_norm": 0.5691525936126709,
      "learning_rate": 2.7513150428063112e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 23327968,
      "step": 7015
    },
    {
      "epoch": 1.4045969537053247,
      "grad_norm": 0.5938063263893127,
      "learning_rate": 2.7487086092946485e-05,
      "loss": 0.0288,
      "num_input_tokens_seen": 23344896,
      "step": 7020
    },
    {
      "epoch": 1.4055973788860265,
      "grad_norm": 0.743880569934845,
      "learning_rate": 2.7461019027159185e-05,
      "loss": 0.021,
      "num_input_tokens_seen": 23362912,
      "step": 7025
    },
    {
      "epoch": 1.4065978040667284,
      "grad_norm": 0.8439948558807373,
      "learning_rate": 2.743494925932129e-05,
      "loss": 0.045,
      "num_input_tokens_seen": 23378880,
      "step": 7030
    },
    {
      "epoch": 1.40759822924743,
      "grad_norm": 1.1395623683929443,
      "learning_rate": 2.740887681805583e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 23395040,
      "step": 7035
    },
    {
      "epoch": 1.408598654428132,
      "grad_norm": 0.6801261305809021,
      "learning_rate": 2.738280173198877e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 23411424,
      "step": 7040
    },
    {
      "epoch": 1.4095990796088338,
      "grad_norm": 0.37221503257751465,
      "learning_rate": 2.7356724029748994e-05,
      "loss": 0.0158,
      "num_input_tokens_seen": 23429184,
      "step": 7045
    },
    {
      "epoch": 1.4105995047895354,
      "grad_norm": 0.4096348285675049,
      "learning_rate": 2.7330643739968237e-05,
      "loss": 0.0174,
      "num_input_tokens_seen": 23445760,
      "step": 7050
    },
    {
      "epoch": 1.4115999299702373,
      "grad_norm": 0.6790040135383606,
      "learning_rate": 2.7304560891281095e-05,
      "loss": 0.0309,
      "num_input_tokens_seen": 23461120,
      "step": 7055
    },
    {
      "epoch": 1.4126003551509392,
      "grad_norm": 0.37077081203460693,
      "learning_rate": 2.727847551232496e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 23477024,
      "step": 7060
    },
    {
      "epoch": 1.4136007803316408,
      "grad_norm": 0.364713579416275,
      "learning_rate": 2.725238763174e-05,
      "loss": 0.0092,
      "num_input_tokens_seen": 23493856,
      "step": 7065
    },
    {
      "epoch": 1.4146012055123427,
      "grad_norm": 0.5751842856407166,
      "learning_rate": 2.722629727816914e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 23510976,
      "step": 7070
    },
    {
      "epoch": 1.4156016306930446,
      "grad_norm": 0.21734090149402618,
      "learning_rate": 2.7200204480258023e-05,
      "loss": 0.0175,
      "num_input_tokens_seen": 23528864,
      "step": 7075
    },
    {
      "epoch": 1.4166020558737462,
      "grad_norm": 0.5161658525466919,
      "learning_rate": 2.7174109266654966e-05,
      "loss": 0.0222,
      "num_input_tokens_seen": 23544896,
      "step": 7080
    },
    {
      "epoch": 1.417602481054448,
      "grad_norm": 0.2955359220504761,
      "learning_rate": 2.7148011666010936e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 23561440,
      "step": 7085
    },
    {
      "epoch": 1.41860290623515,
      "grad_norm": 0.1447141021490097,
      "learning_rate": 2.712191170697953e-05,
      "loss": 0.0205,
      "num_input_tokens_seen": 23577664,
      "step": 7090
    },
    {
      "epoch": 1.4196033314158516,
      "grad_norm": 0.23074093461036682,
      "learning_rate": 2.709580941821692e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 23594336,
      "step": 7095
    },
    {
      "epoch": 1.4206037565965535,
      "grad_norm": 0.747652530670166,
      "learning_rate": 2.706970482838187e-05,
      "loss": 0.0253,
      "num_input_tokens_seen": 23612000,
      "step": 7100
    },
    {
      "epoch": 1.4216041817772553,
      "grad_norm": 0.6384264826774597,
      "learning_rate": 2.704359796613562e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 23628576,
      "step": 7105
    },
    {
      "epoch": 1.4226046069579572,
      "grad_norm": 1.05685555934906,
      "learning_rate": 2.7017488860141955e-05,
      "loss": 0.0222,
      "num_input_tokens_seen": 23646112,
      "step": 7110
    },
    {
      "epoch": 1.4236050321386589,
      "grad_norm": 0.3217991292476654,
      "learning_rate": 2.699137753906708e-05,
      "loss": 0.0158,
      "num_input_tokens_seen": 23662560,
      "step": 7115
    },
    {
      "epoch": 1.4246054573193607,
      "grad_norm": 1.1819077730178833,
      "learning_rate": 2.696526403157966e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 23679264,
      "step": 7120
    },
    {
      "epoch": 1.4256058825000626,
      "grad_norm": 0.6659606099128723,
      "learning_rate": 2.6939148366350758e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 23695648,
      "step": 7125
    },
    {
      "epoch": 1.4266063076807645,
      "grad_norm": 0.5607231259346008,
      "learning_rate": 2.6913030572053798e-05,
      "loss": 0.0257,
      "num_input_tokens_seen": 23711616,
      "step": 7130
    },
    {
      "epoch": 1.427606732861466,
      "grad_norm": 0.20791594684123993,
      "learning_rate": 2.6886910677364542e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 23728544,
      "step": 7135
    },
    {
      "epoch": 1.428607158042168,
      "grad_norm": 1.3680839538574219,
      "learning_rate": 2.6860788710961054e-05,
      "loss": 0.0464,
      "num_input_tokens_seen": 23745536,
      "step": 7140
    },
    {
      "epoch": 1.4296075832228698,
      "grad_norm": 0.24108925461769104,
      "learning_rate": 2.683466470152369e-05,
      "loss": 0.0127,
      "num_input_tokens_seen": 23761856,
      "step": 7145
    },
    {
      "epoch": 1.4306080084035715,
      "grad_norm": 0.32991474866867065,
      "learning_rate": 2.680853867773504e-05,
      "loss": 0.0298,
      "num_input_tokens_seen": 23777728,
      "step": 7150
    },
    {
      "epoch": 1.4316084335842734,
      "grad_norm": 0.19740597903728485,
      "learning_rate": 2.6782410668279895e-05,
      "loss": 0.0396,
      "num_input_tokens_seen": 23794560,
      "step": 7155
    },
    {
      "epoch": 1.4326088587649752,
      "grad_norm": 0.8721890449523926,
      "learning_rate": 2.675628070184523e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 23812032,
      "step": 7160
    },
    {
      "epoch": 1.4336092839456769,
      "grad_norm": 0.4150925874710083,
      "learning_rate": 2.6730148807120197e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 23828608,
      "step": 7165
    },
    {
      "epoch": 1.4346097091263788,
      "grad_norm": 0.7514519095420837,
      "learning_rate": 2.6704015012796013e-05,
      "loss": 0.03,
      "num_input_tokens_seen": 23844224,
      "step": 7170
    },
    {
      "epoch": 1.4356101343070806,
      "grad_norm": 0.13385280966758728,
      "learning_rate": 2.6677879347566038e-05,
      "loss": 0.0137,
      "num_input_tokens_seen": 23861088,
      "step": 7175
    },
    {
      "epoch": 1.4366105594877823,
      "grad_norm": 0.2709481716156006,
      "learning_rate": 2.6651741840125627e-05,
      "loss": 0.01,
      "num_input_tokens_seen": 23877952,
      "step": 7180
    },
    {
      "epoch": 1.4376109846684841,
      "grad_norm": 0.05890822038054466,
      "learning_rate": 2.6625602519172216e-05,
      "loss": 0.0174,
      "num_input_tokens_seen": 23893568,
      "step": 7185
    },
    {
      "epoch": 1.438611409849186,
      "grad_norm": 0.053118109703063965,
      "learning_rate": 2.6599461413405192e-05,
      "loss": 0.02,
      "num_input_tokens_seen": 23910592,
      "step": 7190
    },
    {
      "epoch": 1.4396118350298877,
      "grad_norm": 0.2782624065876007,
      "learning_rate": 2.6573318551525904e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 23926400,
      "step": 7195
    },
    {
      "epoch": 1.4406122602105895,
      "grad_norm": 0.9280529022216797,
      "learning_rate": 2.6547173962237653e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 23942080,
      "step": 7200
    },
    {
      "epoch": 1.4416126853912914,
      "grad_norm": 0.22394904494285583,
      "learning_rate": 2.6521027674245625e-05,
      "loss": 0.0128,
      "num_input_tokens_seen": 23959136,
      "step": 7205
    },
    {
      "epoch": 1.442613110571993,
      "grad_norm": 0.5346195697784424,
      "learning_rate": 2.649487971625686e-05,
      "loss": 0.024,
      "num_input_tokens_seen": 23976224,
      "step": 7210
    },
    {
      "epoch": 1.443613535752695,
      "grad_norm": 0.18652726709842682,
      "learning_rate": 2.646873011698024e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 23993312,
      "step": 7215
    },
    {
      "epoch": 1.4446139609333968,
      "grad_norm": 0.5600389838218689,
      "learning_rate": 2.6442578905126458e-05,
      "loss": 0.0098,
      "num_input_tokens_seen": 24008896,
      "step": 7220
    },
    {
      "epoch": 1.4456143861140984,
      "grad_norm": 0.423225075006485,
      "learning_rate": 2.6416426109407954e-05,
      "loss": 0.0568,
      "num_input_tokens_seen": 24025632,
      "step": 7225
    },
    {
      "epoch": 1.4466148112948003,
      "grad_norm": 0.2751964330673218,
      "learning_rate": 2.6390271758538944e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 24041792,
      "step": 7230
    },
    {
      "epoch": 1.4476152364755022,
      "grad_norm": 0.10048820078372955,
      "learning_rate": 2.6364115881235302e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 24058656,
      "step": 7235
    },
    {
      "epoch": 1.4486156616562038,
      "grad_norm": 0.5874961614608765,
      "learning_rate": 2.6337958506214627e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 24074048,
      "step": 7240
    },
    {
      "epoch": 1.4496160868369057,
      "grad_norm": 0.6436035633087158,
      "learning_rate": 2.6311799662196136e-05,
      "loss": 0.0228,
      "num_input_tokens_seen": 24090400,
      "step": 7245
    },
    {
      "epoch": 1.4506165120176076,
      "grad_norm": 0.2383117377758026,
      "learning_rate": 2.6285639377900657e-05,
      "loss": 0.0315,
      "num_input_tokens_seen": 24106432,
      "step": 7250
    },
    {
      "epoch": 1.4516169371983092,
      "grad_norm": 0.3349505066871643,
      "learning_rate": 2.6259477682050603e-05,
      "loss": 0.0175,
      "num_input_tokens_seen": 24123840,
      "step": 7255
    },
    {
      "epoch": 1.452617362379011,
      "grad_norm": 0.9861046671867371,
      "learning_rate": 2.6233314603369957e-05,
      "loss": 0.011,
      "num_input_tokens_seen": 24139360,
      "step": 7260
    },
    {
      "epoch": 1.453617787559713,
      "grad_norm": 0.1726267784833908,
      "learning_rate": 2.6207150170584187e-05,
      "loss": 0.0146,
      "num_input_tokens_seen": 24155520,
      "step": 7265
    },
    {
      "epoch": 1.4546182127404146,
      "grad_norm": 0.29923322796821594,
      "learning_rate": 2.6180984412420266e-05,
      "loss": 0.0227,
      "num_input_tokens_seen": 24172672,
      "step": 7270
    },
    {
      "epoch": 1.4556186379211165,
      "grad_norm": 0.3386686444282532,
      "learning_rate": 2.6154817357606626e-05,
      "loss": 0.0265,
      "num_input_tokens_seen": 24189536,
      "step": 7275
    },
    {
      "epoch": 1.4566190631018183,
      "grad_norm": 0.1898878961801529,
      "learning_rate": 2.6128649034873115e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 24204352,
      "step": 7280
    },
    {
      "epoch": 1.45761948828252,
      "grad_norm": 0.1861916184425354,
      "learning_rate": 2.610247947295097e-05,
      "loss": 0.0156,
      "num_input_tokens_seen": 24220832,
      "step": 7285
    },
    {
      "epoch": 1.4586199134632218,
      "grad_norm": 0.7051615715026855,
      "learning_rate": 2.6076308700572794e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 24238016,
      "step": 7290
    },
    {
      "epoch": 1.4596203386439237,
      "grad_norm": 0.3806222379207611,
      "learning_rate": 2.605013674647252e-05,
      "loss": 0.0309,
      "num_input_tokens_seen": 24253888,
      "step": 7295
    },
    {
      "epoch": 1.4606207638246254,
      "grad_norm": 0.6028633117675781,
      "learning_rate": 2.6023963639385374e-05,
      "loss": 0.021,
      "num_input_tokens_seen": 24270880,
      "step": 7300
    },
    {
      "epoch": 1.4616211890053272,
      "grad_norm": 0.2088484764099121,
      "learning_rate": 2.5997789408047863e-05,
      "loss": 0.0192,
      "num_input_tokens_seen": 24288256,
      "step": 7305
    },
    {
      "epoch": 1.462621614186029,
      "grad_norm": 0.35228636860847473,
      "learning_rate": 2.5971614081197692e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 24304704,
      "step": 7310
    },
    {
      "epoch": 1.4636220393667307,
      "grad_norm": 0.8489367961883545,
      "learning_rate": 2.5945437687573816e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 24321952,
      "step": 7315
    },
    {
      "epoch": 1.4646224645474326,
      "grad_norm": 1.9019261598587036,
      "learning_rate": 2.5919260255916322e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 24337696,
      "step": 7320
    },
    {
      "epoch": 1.4656228897281345,
      "grad_norm": 0.20332832634449005,
      "learning_rate": 2.589308181496645e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 24353696,
      "step": 7325
    },
    {
      "epoch": 1.4666233149088361,
      "grad_norm": 0.30129843950271606,
      "learning_rate": 2.5866902393466574e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 24370240,
      "step": 7330
    },
    {
      "epoch": 1.467623740089538,
      "grad_norm": 0.7240298986434937,
      "learning_rate": 2.5840722020160107e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 24387520,
      "step": 7335
    },
    {
      "epoch": 1.4686241652702399,
      "grad_norm": 1.0181207656860352,
      "learning_rate": 2.5814540723791518e-05,
      "loss": 0.0366,
      "num_input_tokens_seen": 24403840,
      "step": 7340
    },
    {
      "epoch": 1.4696245904509415,
      "grad_norm": 1.1026760339736938,
      "learning_rate": 2.5788358533106295e-05,
      "loss": 0.0377,
      "num_input_tokens_seen": 24419936,
      "step": 7345
    },
    {
      "epoch": 1.4706250156316434,
      "grad_norm": 0.7127203941345215,
      "learning_rate": 2.5762175476850896e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 24436736,
      "step": 7350
    },
    {
      "epoch": 1.4716254408123453,
      "grad_norm": 0.2946961224079132,
      "learning_rate": 2.573599158377276e-05,
      "loss": 0.0202,
      "num_input_tokens_seen": 24453120,
      "step": 7355
    },
    {
      "epoch": 1.4726258659930471,
      "grad_norm": 0.17568668723106384,
      "learning_rate": 2.570980688262022e-05,
      "loss": 0.0081,
      "num_input_tokens_seen": 24470368,
      "step": 7360
    },
    {
      "epoch": 1.4736262911737488,
      "grad_norm": 0.2808990478515625,
      "learning_rate": 2.568362140214248e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 24488224,
      "step": 7365
    },
    {
      "epoch": 1.4746267163544506,
      "grad_norm": 0.12591205537319183,
      "learning_rate": 2.5657435171089646e-05,
      "loss": 0.0135,
      "num_input_tokens_seen": 24504896,
      "step": 7370
    },
    {
      "epoch": 1.4756271415351525,
      "grad_norm": 0.5561877489089966,
      "learning_rate": 2.5631248218212595e-05,
      "loss": 0.0185,
      "num_input_tokens_seen": 24521888,
      "step": 7375
    },
    {
      "epoch": 1.4766275667158542,
      "grad_norm": 0.07315061241388321,
      "learning_rate": 2.5605060572263046e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 24538496,
      "step": 7380
    },
    {
      "epoch": 1.477627991896556,
      "grad_norm": 0.4523506760597229,
      "learning_rate": 2.5578872261993447e-05,
      "loss": 0.0196,
      "num_input_tokens_seen": 24554848,
      "step": 7385
    },
    {
      "epoch": 1.478628417077258,
      "grad_norm": 0.07306066155433655,
      "learning_rate": 2.5552683316156983e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 24571840,
      "step": 7390
    },
    {
      "epoch": 1.4796288422579598,
      "grad_norm": 0.3900500535964966,
      "learning_rate": 2.5526493763507548e-05,
      "loss": 0.0221,
      "num_input_tokens_seen": 24589280,
      "step": 7395
    },
    {
      "epoch": 1.4806292674386614,
      "grad_norm": 0.5769615173339844,
      "learning_rate": 2.5500303632799675e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 24605568,
      "step": 7400
    },
    {
      "epoch": 1.4816296926193633,
      "grad_norm": 0.06523500382900238,
      "learning_rate": 2.5474112952788565e-05,
      "loss": 0.0238,
      "num_input_tokens_seen": 24622304,
      "step": 7405
    },
    {
      "epoch": 1.4826301178000652,
      "grad_norm": 0.3345092535018921,
      "learning_rate": 2.5447921752230003e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 24639328,
      "step": 7410
    },
    {
      "epoch": 1.4836305429807668,
      "grad_norm": 0.5575299263000488,
      "learning_rate": 2.542173005988034e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 24657216,
      "step": 7415
    },
    {
      "epoch": 1.4846309681614687,
      "grad_norm": 0.6276900768280029,
      "learning_rate": 2.539553790449649e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 24673664,
      "step": 7420
    },
    {
      "epoch": 1.4856313933421705,
      "grad_norm": 0.5556184649467468,
      "learning_rate": 2.5369345314835856e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 24691264,
      "step": 7425
    },
    {
      "epoch": 1.4866318185228722,
      "grad_norm": 0.6906352043151855,
      "learning_rate": 2.5343152319656316e-05,
      "loss": 0.0217,
      "num_input_tokens_seen": 24708512,
      "step": 7430
    },
    {
      "epoch": 1.487632243703574,
      "grad_norm": 0.267528235912323,
      "learning_rate": 2.5316958947716206e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 24725152,
      "step": 7435
    },
    {
      "epoch": 1.488632668884276,
      "grad_norm": 0.313608318567276,
      "learning_rate": 2.5290765227774277e-05,
      "loss": 0.0158,
      "num_input_tokens_seen": 24741696,
      "step": 7440
    },
    {
      "epoch": 1.4896330940649776,
      "grad_norm": 0.8933184146881104,
      "learning_rate": 2.5264571188589642e-05,
      "loss": 0.0331,
      "num_input_tokens_seen": 24758432,
      "step": 7445
    },
    {
      "epoch": 1.4906335192456794,
      "grad_norm": 0.48685455322265625,
      "learning_rate": 2.5238376858921786e-05,
      "loss": 0.0248,
      "num_input_tokens_seen": 24774400,
      "step": 7450
    },
    {
      "epoch": 1.4916339444263813,
      "grad_norm": 0.806394636631012,
      "learning_rate": 2.5212182267530498e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 24790848,
      "step": 7455
    },
    {
      "epoch": 1.492634369607083,
      "grad_norm": 0.7549716234207153,
      "learning_rate": 2.5185987443175878e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 24808608,
      "step": 7460
    },
    {
      "epoch": 1.4936347947877848,
      "grad_norm": 0.014884378761053085,
      "learning_rate": 2.5159792414618244e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 24824928,
      "step": 7465
    },
    {
      "epoch": 1.4946352199684867,
      "grad_norm": 0.788593053817749,
      "learning_rate": 2.513359721061817e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 24840736,
      "step": 7470
    },
    {
      "epoch": 1.4956356451491883,
      "grad_norm": 0.3795950710773468,
      "learning_rate": 2.510740185993641e-05,
      "loss": 0.0223,
      "num_input_tokens_seen": 24858208,
      "step": 7475
    },
    {
      "epoch": 1.4966360703298902,
      "grad_norm": 1.174605369567871,
      "learning_rate": 2.5081206391333873e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 24874720,
      "step": 7480
    },
    {
      "epoch": 1.497636495510592,
      "grad_norm": 0.6153615713119507,
      "learning_rate": 2.505501083357162e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 24893184,
      "step": 7485
    },
    {
      "epoch": 1.4986369206912937,
      "grad_norm": 0.3919779062271118,
      "learning_rate": 2.502881521541078e-05,
      "loss": 0.0318,
      "num_input_tokens_seen": 24910720,
      "step": 7490
    },
    {
      "epoch": 1.4996373458719956,
      "grad_norm": 0.2815072536468506,
      "learning_rate": 2.500261956561257e-05,
      "loss": 0.007,
      "num_input_tokens_seen": 24927392,
      "step": 7495
    },
    {
      "epoch": 1.5006377710526975,
      "grad_norm": 0.11717823147773743,
      "learning_rate": 2.4976423912938236e-05,
      "loss": 0.0199,
      "num_input_tokens_seen": 24944736,
      "step": 7500
    },
    {
      "epoch": 1.5016381962333991,
      "grad_norm": 1.0984041690826416,
      "learning_rate": 2.4950228286149028e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 24960832,
      "step": 7505
    },
    {
      "epoch": 1.502638621414101,
      "grad_norm": 0.7648619413375854,
      "learning_rate": 2.4924032714006157e-05,
      "loss": 0.0319,
      "num_input_tokens_seen": 24978272,
      "step": 7510
    },
    {
      "epoch": 1.5036390465948029,
      "grad_norm": 0.40908828377723694,
      "learning_rate": 2.4897837225270796e-05,
      "loss": 0.0196,
      "num_input_tokens_seen": 24995904,
      "step": 7515
    },
    {
      "epoch": 1.5046394717755045,
      "grad_norm": 0.14489774405956268,
      "learning_rate": 2.4871641848704004e-05,
      "loss": 0.0096,
      "num_input_tokens_seen": 25013536,
      "step": 7520
    },
    {
      "epoch": 1.5056398969562064,
      "grad_norm": 0.4525126516819,
      "learning_rate": 2.484544661306672e-05,
      "loss": 0.0408,
      "num_input_tokens_seen": 25030560,
      "step": 7525
    },
    {
      "epoch": 1.5066403221369082,
      "grad_norm": 0.21470829844474792,
      "learning_rate": 2.481925154711975e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 25046880,
      "step": 7530
    },
    {
      "epoch": 1.5076407473176099,
      "grad_norm": 0.30370137095451355,
      "learning_rate": 2.479305667962369e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 25064384,
      "step": 7535
    },
    {
      "epoch": 1.5086411724983118,
      "grad_norm": 0.14242781698703766,
      "learning_rate": 2.476686203933892e-05,
      "loss": 0.0163,
      "num_input_tokens_seen": 25079808,
      "step": 7540
    },
    {
      "epoch": 1.5096415976790136,
      "grad_norm": 0.3540135324001312,
      "learning_rate": 2.4740667655025586e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 25096128,
      "step": 7545
    },
    {
      "epoch": 1.5106420228597153,
      "grad_norm": 0.38274118304252625,
      "learning_rate": 2.4714473555443534e-05,
      "loss": 0.0146,
      "num_input_tokens_seen": 25112128,
      "step": 7550
    },
    {
      "epoch": 1.5116424480404171,
      "grad_norm": 0.8952894806861877,
      "learning_rate": 2.4688279769352323e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 25129440,
      "step": 7555
    },
    {
      "epoch": 1.512642873221119,
      "grad_norm": 0.40251901745796204,
      "learning_rate": 2.466208632551114e-05,
      "loss": 0.0231,
      "num_input_tokens_seen": 25144928,
      "step": 7560
    },
    {
      "epoch": 1.5136432984018207,
      "grad_norm": 1.6884913444519043,
      "learning_rate": 2.463589325267881e-05,
      "loss": 0.0253,
      "num_input_tokens_seen": 25161024,
      "step": 7565
    },
    {
      "epoch": 1.5146437235825225,
      "grad_norm": 0.24734361469745636,
      "learning_rate": 2.4609700579613745e-05,
      "loss": 0.024,
      "num_input_tokens_seen": 25177408,
      "step": 7570
    },
    {
      "epoch": 1.5156441487632244,
      "grad_norm": 0.8871567845344543,
      "learning_rate": 2.4583508335073933e-05,
      "loss": 0.0152,
      "num_input_tokens_seen": 25193632,
      "step": 7575
    },
    {
      "epoch": 1.516644573943926,
      "grad_norm": 0.6303677558898926,
      "learning_rate": 2.4557316547816865e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 25211136,
      "step": 7580
    },
    {
      "epoch": 1.517644999124628,
      "grad_norm": 0.1167726218700409,
      "learning_rate": 2.4531125246599556e-05,
      "loss": 0.0211,
      "num_input_tokens_seen": 25228416,
      "step": 7585
    },
    {
      "epoch": 1.5186454243053298,
      "grad_norm": 0.3821775019168854,
      "learning_rate": 2.4504934460178475e-05,
      "loss": 0.0067,
      "num_input_tokens_seen": 25244640,
      "step": 7590
    },
    {
      "epoch": 1.5196458494860314,
      "grad_norm": 0.09346763789653778,
      "learning_rate": 2.4478744217309515e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 25262528,
      "step": 7595
    },
    {
      "epoch": 1.5206462746667335,
      "grad_norm": 1.1715312004089355,
      "learning_rate": 2.445255454674801e-05,
      "loss": 0.0289,
      "num_input_tokens_seen": 25278720,
      "step": 7600
    },
    {
      "epoch": 1.5216466998474352,
      "grad_norm": 0.7903635501861572,
      "learning_rate": 2.4426365477248613e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 25295616,
      "step": 7605
    },
    {
      "epoch": 1.5226471250281368,
      "grad_norm": 0.16973412036895752,
      "learning_rate": 2.4400177037565363e-05,
      "loss": 0.0139,
      "num_input_tokens_seen": 25312384,
      "step": 7610
    },
    {
      "epoch": 1.523647550208839,
      "grad_norm": 0.07989294826984406,
      "learning_rate": 2.437398925645159e-05,
      "loss": 0.0108,
      "num_input_tokens_seen": 25329056,
      "step": 7615
    },
    {
      "epoch": 1.5246479753895406,
      "grad_norm": 0.18625207245349884,
      "learning_rate": 2.4347802162659877e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 25345664,
      "step": 7620
    },
    {
      "epoch": 1.5256484005702422,
      "grad_norm": 0.7733270525932312,
      "learning_rate": 2.43216157849421e-05,
      "loss": 0.0477,
      "num_input_tokens_seen": 25361568,
      "step": 7625
    },
    {
      "epoch": 1.5266488257509443,
      "grad_norm": 0.5695168972015381,
      "learning_rate": 2.4295430152049316e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 25378496,
      "step": 7630
    },
    {
      "epoch": 1.527649250931646,
      "grad_norm": 1.1541272401809692,
      "learning_rate": 2.426924529273176e-05,
      "loss": 0.0317,
      "num_input_tokens_seen": 25394240,
      "step": 7635
    },
    {
      "epoch": 1.5286496761123476,
      "grad_norm": 0.38781797885894775,
      "learning_rate": 2.424306123573885e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 25410656,
      "step": 7640
    },
    {
      "epoch": 1.5296501012930497,
      "grad_norm": 0.1999148726463318,
      "learning_rate": 2.421687800981908e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 25427328,
      "step": 7645
    },
    {
      "epoch": 1.5306505264737513,
      "grad_norm": 0.06788372248411179,
      "learning_rate": 2.4190695643720074e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 25443840,
      "step": 7650
    },
    {
      "epoch": 1.5316509516544532,
      "grad_norm": 0.21277567744255066,
      "learning_rate": 2.416451416618849e-05,
      "loss": 0.0164,
      "num_input_tokens_seen": 25461216,
      "step": 7655
    },
    {
      "epoch": 1.532651376835155,
      "grad_norm": 0.23967178165912628,
      "learning_rate": 2.4138333605970002e-05,
      "loss": 0.0239,
      "num_input_tokens_seen": 25478080,
      "step": 7660
    },
    {
      "epoch": 1.5336518020158567,
      "grad_norm": 0.45840054750442505,
      "learning_rate": 2.4112153991809297e-05,
      "loss": 0.0173,
      "num_input_tokens_seen": 25494528,
      "step": 7665
    },
    {
      "epoch": 1.5346522271965586,
      "grad_norm": 0.1507226526737213,
      "learning_rate": 2.408597535245001e-05,
      "loss": 0.0105,
      "num_input_tokens_seen": 25511904,
      "step": 7670
    },
    {
      "epoch": 1.5356526523772605,
      "grad_norm": 0.5378280878067017,
      "learning_rate": 2.4059797716634707e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 25529120,
      "step": 7675
    },
    {
      "epoch": 1.536653077557962,
      "grad_norm": 0.6452955603599548,
      "learning_rate": 2.403362111310486e-05,
      "loss": 0.0261,
      "num_input_tokens_seen": 25546336,
      "step": 7680
    },
    {
      "epoch": 1.537653502738664,
      "grad_norm": 0.24737097322940826,
      "learning_rate": 2.40074455706008e-05,
      "loss": 0.0114,
      "num_input_tokens_seen": 25562976,
      "step": 7685
    },
    {
      "epoch": 1.5386539279193658,
      "grad_norm": 0.21651674807071686,
      "learning_rate": 2.398127111786169e-05,
      "loss": 0.0212,
      "num_input_tokens_seen": 25579808,
      "step": 7690
    },
    {
      "epoch": 1.5396543531000675,
      "grad_norm": 0.4780050218105316,
      "learning_rate": 2.395509778362552e-05,
      "loss": 0.0358,
      "num_input_tokens_seen": 25595712,
      "step": 7695
    },
    {
      "epoch": 1.5406547782807694,
      "grad_norm": 0.3401283919811249,
      "learning_rate": 2.3928925596629004e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 25612384,
      "step": 7700
    },
    {
      "epoch": 1.5416552034614712,
      "grad_norm": 0.3908470571041107,
      "learning_rate": 2.3902754585607655e-05,
      "loss": 0.0117,
      "num_input_tokens_seen": 25628704,
      "step": 7705
    },
    {
      "epoch": 1.5426556286421729,
      "grad_norm": 0.42244410514831543,
      "learning_rate": 2.387658477929565e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 25645024,
      "step": 7710
    },
    {
      "epoch": 1.5436560538228747,
      "grad_norm": 0.17085978388786316,
      "learning_rate": 2.3850416206425868e-05,
      "loss": 0.0144,
      "num_input_tokens_seen": 25661952,
      "step": 7715
    },
    {
      "epoch": 1.5446564790035766,
      "grad_norm": 0.4565860629081726,
      "learning_rate": 2.3824248895729823e-05,
      "loss": 0.0134,
      "num_input_tokens_seen": 25679520,
      "step": 7720
    },
    {
      "epoch": 1.5456569041842783,
      "grad_norm": 0.8869709968566895,
      "learning_rate": 2.379808287593764e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 25695968,
      "step": 7725
    },
    {
      "epoch": 1.5466573293649801,
      "grad_norm": 0.08044911921024323,
      "learning_rate": 2.3771918175778037e-05,
      "loss": 0.0112,
      "num_input_tokens_seen": 25711168,
      "step": 7730
    },
    {
      "epoch": 1.547657754545682,
      "grad_norm": 0.0679691955447197,
      "learning_rate": 2.374575482397828e-05,
      "loss": 0.0208,
      "num_input_tokens_seen": 25726784,
      "step": 7735
    },
    {
      "epoch": 1.5486581797263836,
      "grad_norm": 0.13976770639419556,
      "learning_rate": 2.3719592849264144e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 25744064,
      "step": 7740
    },
    {
      "epoch": 1.5496586049070855,
      "grad_norm": 0.4790512025356293,
      "learning_rate": 2.3693432280359906e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 25761632,
      "step": 7745
    },
    {
      "epoch": 1.5506590300877874,
      "grad_norm": 0.499796986579895,
      "learning_rate": 2.366727314598829e-05,
      "loss": 0.0483,
      "num_input_tokens_seen": 25777632,
      "step": 7750
    },
    {
      "epoch": 1.551659455268489,
      "grad_norm": 0.5286955237388611,
      "learning_rate": 2.3641115474870462e-05,
      "loss": 0.0174,
      "num_input_tokens_seen": 25795264,
      "step": 7755
    },
    {
      "epoch": 1.552659880449191,
      "grad_norm": 0.3542931377887726,
      "learning_rate": 2.361495929572596e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 25811680,
      "step": 7760
    },
    {
      "epoch": 1.5536603056298928,
      "grad_norm": 0.2267359495162964,
      "learning_rate": 2.358880463727269e-05,
      "loss": 0.0151,
      "num_input_tokens_seen": 25827328,
      "step": 7765
    },
    {
      "epoch": 1.5546607308105944,
      "grad_norm": 0.5924873352050781,
      "learning_rate": 2.35626515282269e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 25843520,
      "step": 7770
    },
    {
      "epoch": 1.5556611559912963,
      "grad_norm": 0.0663023293018341,
      "learning_rate": 2.3536499997303126e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 25859040,
      "step": 7775
    },
    {
      "epoch": 1.5566615811719982,
      "grad_norm": 0.5205705761909485,
      "learning_rate": 2.3510350073214165e-05,
      "loss": 0.0318,
      "num_input_tokens_seen": 25875008,
      "step": 7780
    },
    {
      "epoch": 1.5576620063526998,
      "grad_norm": 0.3358594477176666,
      "learning_rate": 2.3484201784671055e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 25892064,
      "step": 7785
    },
    {
      "epoch": 1.5586624315334017,
      "grad_norm": 0.2526882290840149,
      "learning_rate": 2.3458055160383054e-05,
      "loss": 0.0165,
      "num_input_tokens_seen": 25908896,
      "step": 7790
    },
    {
      "epoch": 1.5596628567141035,
      "grad_norm": 0.286253958940506,
      "learning_rate": 2.343191022905758e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 25926816,
      "step": 7795
    },
    {
      "epoch": 1.5606632818948052,
      "grad_norm": 0.47777873277664185,
      "learning_rate": 2.340576701940017e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 25943296,
      "step": 7800
    },
    {
      "epoch": 1.561663707075507,
      "grad_norm": 0.5363597273826599,
      "learning_rate": 2.3379625560114516e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 25958496,
      "step": 7805
    },
    {
      "epoch": 1.562664132256209,
      "grad_norm": 0.8436539173126221,
      "learning_rate": 2.3353485879902367e-05,
      "loss": 0.0242,
      "num_input_tokens_seen": 25974336,
      "step": 7810
    },
    {
      "epoch": 1.5636645574369106,
      "grad_norm": 0.13085699081420898,
      "learning_rate": 2.3327348007463506e-05,
      "loss": 0.0112,
      "num_input_tokens_seen": 25989792,
      "step": 7815
    },
    {
      "epoch": 1.5646649826176124,
      "grad_norm": 0.12430788576602936,
      "learning_rate": 2.330121197149574e-05,
      "loss": 0.0246,
      "num_input_tokens_seen": 26006080,
      "step": 7820
    },
    {
      "epoch": 1.5656654077983143,
      "grad_norm": 0.13795477151870728,
      "learning_rate": 2.3275077800694872e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 26022496,
      "step": 7825
    },
    {
      "epoch": 1.566665832979016,
      "grad_norm": 0.36709097027778625,
      "learning_rate": 2.324894552375464e-05,
      "loss": 0.0209,
      "num_input_tokens_seen": 26039136,
      "step": 7830
    },
    {
      "epoch": 1.5676662581597178,
      "grad_norm": 0.6450609564781189,
      "learning_rate": 2.3222815169366706e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 26054624,
      "step": 7835
    },
    {
      "epoch": 1.5686666833404197,
      "grad_norm": 0.46975892782211304,
      "learning_rate": 2.319668676622063e-05,
      "loss": 0.0251,
      "num_input_tokens_seen": 26070720,
      "step": 7840
    },
    {
      "epoch": 1.5696671085211213,
      "grad_norm": 0.4092192053794861,
      "learning_rate": 2.3170560343003826e-05,
      "loss": 0.0163,
      "num_input_tokens_seen": 26087552,
      "step": 7845
    },
    {
      "epoch": 1.5706675337018234,
      "grad_norm": 0.4573599398136139,
      "learning_rate": 2.3144435928401528e-05,
      "loss": 0.026,
      "num_input_tokens_seen": 26104224,
      "step": 7850
    },
    {
      "epoch": 1.571667958882525,
      "grad_norm": 0.5140572190284729,
      "learning_rate": 2.3118313551096763e-05,
      "loss": 0.0376,
      "num_input_tokens_seen": 26119712,
      "step": 7855
    },
    {
      "epoch": 1.5726683840632267,
      "grad_norm": 0.6111047863960266,
      "learning_rate": 2.3092193239770338e-05,
      "loss": 0.0205,
      "num_input_tokens_seen": 26136448,
      "step": 7860
    },
    {
      "epoch": 1.5736688092439288,
      "grad_norm": 0.6026287078857422,
      "learning_rate": 2.3066075023100785e-05,
      "loss": 0.0127,
      "num_input_tokens_seen": 26152896,
      "step": 7865
    },
    {
      "epoch": 1.5746692344246305,
      "grad_norm": 0.5394666790962219,
      "learning_rate": 2.303995892976432e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 26169536,
      "step": 7870
    },
    {
      "epoch": 1.5756696596053321,
      "grad_norm": 0.20104095339775085,
      "learning_rate": 2.3013844988434842e-05,
      "loss": 0.0239,
      "num_input_tokens_seen": 26186304,
      "step": 7875
    },
    {
      "epoch": 1.5766700847860342,
      "grad_norm": 0.17194046080112457,
      "learning_rate": 2.298773322778389e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 26203616,
      "step": 7880
    },
    {
      "epoch": 1.5776705099667359,
      "grad_norm": 0.5429868102073669,
      "learning_rate": 2.296162367648061e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 26220736,
      "step": 7885
    },
    {
      "epoch": 1.5786709351474375,
      "grad_norm": 0.4686752259731293,
      "learning_rate": 2.2935516363191693e-05,
      "loss": 0.0352,
      "num_input_tokens_seen": 26237376,
      "step": 7890
    },
    {
      "epoch": 1.5796713603281396,
      "grad_norm": 0.4157911539077759,
      "learning_rate": 2.2909411316581418e-05,
      "loss": 0.0154,
      "num_input_tokens_seen": 26253664,
      "step": 7895
    },
    {
      "epoch": 1.5806717855088412,
      "grad_norm": 0.17762956023216248,
      "learning_rate": 2.288330856531155e-05,
      "loss": 0.0164,
      "num_input_tokens_seen": 26270240,
      "step": 7900
    },
    {
      "epoch": 1.581672210689543,
      "grad_norm": 0.3104545474052429,
      "learning_rate": 2.2857208138041316e-05,
      "loss": 0.0224,
      "num_input_tokens_seen": 26286400,
      "step": 7905
    },
    {
      "epoch": 1.582672635870245,
      "grad_norm": 0.2843524217605591,
      "learning_rate": 2.2831110063427443e-05,
      "loss": 0.0192,
      "num_input_tokens_seen": 26303328,
      "step": 7910
    },
    {
      "epoch": 1.5836730610509466,
      "grad_norm": 0.7966655492782593,
      "learning_rate": 2.2805014370124024e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 26319456,
      "step": 7915
    },
    {
      "epoch": 1.5846734862316485,
      "grad_norm": 0.5430777072906494,
      "learning_rate": 2.277892108678257e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 26335808,
      "step": 7920
    },
    {
      "epoch": 1.5856739114123504,
      "grad_norm": 0.20004427433013916,
      "learning_rate": 2.275283024205193e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 26354496,
      "step": 7925
    },
    {
      "epoch": 1.586674336593052,
      "grad_norm": 0.35423436760902405,
      "learning_rate": 2.2726741864578265e-05,
      "loss": 0.009,
      "num_input_tokens_seen": 26372160,
      "step": 7930
    },
    {
      "epoch": 1.5876747617737539,
      "grad_norm": 0.5483704805374146,
      "learning_rate": 2.2700655983005064e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 26388544,
      "step": 7935
    },
    {
      "epoch": 1.5886751869544558,
      "grad_norm": 0.6132058501243591,
      "learning_rate": 2.2674572625973047e-05,
      "loss": 0.0208,
      "num_input_tokens_seen": 26404192,
      "step": 7940
    },
    {
      "epoch": 1.5896756121351574,
      "grad_norm": 1.8248412609100342,
      "learning_rate": 2.264849182212015e-05,
      "loss": 0.0302,
      "num_input_tokens_seen": 26421152,
      "step": 7945
    },
    {
      "epoch": 1.5906760373158593,
      "grad_norm": 0.5395341515541077,
      "learning_rate": 2.262241360008155e-05,
      "loss": 0.0344,
      "num_input_tokens_seen": 26436992,
      "step": 7950
    },
    {
      "epoch": 1.5916764624965611,
      "grad_norm": 0.19765163958072662,
      "learning_rate": 2.2596337988489537e-05,
      "loss": 0.0122,
      "num_input_tokens_seen": 26453248,
      "step": 7955
    },
    {
      "epoch": 1.5926768876772628,
      "grad_norm": 0.23685020208358765,
      "learning_rate": 2.257026501597359e-05,
      "loss": 0.0223,
      "num_input_tokens_seen": 26469280,
      "step": 7960
    },
    {
      "epoch": 1.5936773128579647,
      "grad_norm": 0.15128207206726074,
      "learning_rate": 2.2544194711160243e-05,
      "loss": 0.0077,
      "num_input_tokens_seen": 26485088,
      "step": 7965
    },
    {
      "epoch": 1.5946777380386665,
      "grad_norm": 0.14374390244483948,
      "learning_rate": 2.2518127102673123e-05,
      "loss": 0.0092,
      "num_input_tokens_seen": 26500992,
      "step": 7970
    },
    {
      "epoch": 1.5956781632193682,
      "grad_norm": 0.05345458537340164,
      "learning_rate": 2.24920622191329e-05,
      "loss": 0.011,
      "num_input_tokens_seen": 26516864,
      "step": 7975
    },
    {
      "epoch": 1.59667858840007,
      "grad_norm": 0.22516244649887085,
      "learning_rate": 2.246600008915724e-05,
      "loss": 0.033,
      "num_input_tokens_seen": 26534144,
      "step": 7980
    },
    {
      "epoch": 1.597679013580772,
      "grad_norm": 0.5458850860595703,
      "learning_rate": 2.2439940741360784e-05,
      "loss": 0.0175,
      "num_input_tokens_seen": 26551328,
      "step": 7985
    },
    {
      "epoch": 1.5986794387614736,
      "grad_norm": 0.33864474296569824,
      "learning_rate": 2.2413884204355144e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 26567616,
      "step": 7990
    },
    {
      "epoch": 1.5996798639421754,
      "grad_norm": 0.9156394004821777,
      "learning_rate": 2.2387830506748812e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 26585824,
      "step": 7995
    },
    {
      "epoch": 1.6006802891228773,
      "grad_norm": 0.8333908915519714,
      "learning_rate": 2.236177967714718e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 26602208,
      "step": 8000
    },
    {
      "epoch": 1.601680714303579,
      "grad_norm": 0.16611388325691223,
      "learning_rate": 2.2335731744152497e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 26618432,
      "step": 8005
    },
    {
      "epoch": 1.6026811394842808,
      "grad_norm": 0.17586727440357208,
      "learning_rate": 2.230968673636381e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 26634368,
      "step": 8010
    },
    {
      "epoch": 1.6036815646649827,
      "grad_norm": 1.160220742225647,
      "learning_rate": 2.2283644682376984e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 26650976,
      "step": 8015
    },
    {
      "epoch": 1.6046819898456843,
      "grad_norm": 0.23919957876205444,
      "learning_rate": 2.2257605610784622e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 26668320,
      "step": 8020
    },
    {
      "epoch": 1.6056824150263862,
      "grad_norm": 0.24601928889751434,
      "learning_rate": 2.2231569550176042e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 26686400,
      "step": 8025
    },
    {
      "epoch": 1.606682840207088,
      "grad_norm": 0.7848149538040161,
      "learning_rate": 2.220553652913729e-05,
      "loss": 0.0239,
      "num_input_tokens_seen": 26702688,
      "step": 8030
    },
    {
      "epoch": 1.6076832653877897,
      "grad_norm": 0.1857006996870041,
      "learning_rate": 2.217950657625104e-05,
      "loss": 0.0397,
      "num_input_tokens_seen": 26719040,
      "step": 8035
    },
    {
      "epoch": 1.6086836905684916,
      "grad_norm": 0.1828911006450653,
      "learning_rate": 2.2153479720096613e-05,
      "loss": 0.0144,
      "num_input_tokens_seen": 26735168,
      "step": 8040
    },
    {
      "epoch": 1.6096841157491935,
      "grad_norm": 0.3723798096179962,
      "learning_rate": 2.2127455989249933e-05,
      "loss": 0.0197,
      "num_input_tokens_seen": 26753280,
      "step": 8045
    },
    {
      "epoch": 1.610684540929895,
      "grad_norm": 0.9142296314239502,
      "learning_rate": 2.2101435412283493e-05,
      "loss": 0.0357,
      "num_input_tokens_seen": 26769472,
      "step": 8050
    },
    {
      "epoch": 1.611684966110597,
      "grad_norm": 0.45003557205200195,
      "learning_rate": 2.20754180177663e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 26785920,
      "step": 8055
    },
    {
      "epoch": 1.6126853912912988,
      "grad_norm": 0.26906803250312805,
      "learning_rate": 2.2049403834263896e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 26802496,
      "step": 8060
    },
    {
      "epoch": 1.6136858164720005,
      "grad_norm": 0.5820974707603455,
      "learning_rate": 2.20233928903383e-05,
      "loss": 0.012,
      "num_input_tokens_seen": 26820320,
      "step": 8065
    },
    {
      "epoch": 1.6146862416527024,
      "grad_norm": 0.3259919285774231,
      "learning_rate": 2.199738521454795e-05,
      "loss": 0.0192,
      "num_input_tokens_seen": 26837664,
      "step": 8070
    },
    {
      "epoch": 1.6156866668334042,
      "grad_norm": 0.9476006031036377,
      "learning_rate": 2.1971380835447708e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 26853248,
      "step": 8075
    },
    {
      "epoch": 1.6166870920141059,
      "grad_norm": 0.7303922772407532,
      "learning_rate": 2.1945379781588808e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 26869440,
      "step": 8080
    },
    {
      "epoch": 1.6176875171948077,
      "grad_norm": 0.12725351750850677,
      "learning_rate": 2.191938208151885e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 26886720,
      "step": 8085
    },
    {
      "epoch": 1.6186879423755096,
      "grad_norm": 0.18985620141029358,
      "learning_rate": 2.1893387763781742e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 26902944,
      "step": 8090
    },
    {
      "epoch": 1.6196883675562113,
      "grad_norm": 1.6756082773208618,
      "learning_rate": 2.1867396856917665e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 26919616,
      "step": 8095
    },
    {
      "epoch": 1.6206887927369134,
      "grad_norm": 0.510668158531189,
      "learning_rate": 2.1841409389463088e-05,
      "loss": 0.0125,
      "num_input_tokens_seen": 26934848,
      "step": 8100
    },
    {
      "epoch": 1.621689217917615,
      "grad_norm": 0.5919009447097778,
      "learning_rate": 2.1815425389950675e-05,
      "loss": 0.013,
      "num_input_tokens_seen": 26951136,
      "step": 8105
    },
    {
      "epoch": 1.6226896430983166,
      "grad_norm": 0.7118249535560608,
      "learning_rate": 2.178944488690928e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 26967424,
      "step": 8110
    },
    {
      "epoch": 1.6236900682790187,
      "grad_norm": 1.4784985780715942,
      "learning_rate": 2.176346790886395e-05,
      "loss": 0.0392,
      "num_input_tokens_seen": 26985984,
      "step": 8115
    },
    {
      "epoch": 1.6246904934597204,
      "grad_norm": 0.253864586353302,
      "learning_rate": 2.1737494484335835e-05,
      "loss": 0.0214,
      "num_input_tokens_seen": 27002016,
      "step": 8120
    },
    {
      "epoch": 1.625690918640422,
      "grad_norm": 0.6842542886734009,
      "learning_rate": 2.171152464184219e-05,
      "loss": 0.0128,
      "num_input_tokens_seen": 27018848,
      "step": 8125
    },
    {
      "epoch": 1.6266913438211241,
      "grad_norm": 0.03511577472090721,
      "learning_rate": 2.1685558409896336e-05,
      "loss": 0.0339,
      "num_input_tokens_seen": 27034848,
      "step": 8130
    },
    {
      "epoch": 1.6276917690018258,
      "grad_norm": 0.21839222311973572,
      "learning_rate": 2.1659595817007623e-05,
      "loss": 0.0134,
      "num_input_tokens_seen": 27052608,
      "step": 8135
    },
    {
      "epoch": 1.6286921941825274,
      "grad_norm": 0.5350809693336487,
      "learning_rate": 2.163363689168143e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 27068512,
      "step": 8140
    },
    {
      "epoch": 1.6296926193632295,
      "grad_norm": 0.44204947352409363,
      "learning_rate": 2.1607681662419085e-05,
      "loss": 0.0143,
      "num_input_tokens_seen": 27084256,
      "step": 8145
    },
    {
      "epoch": 1.6306930445439312,
      "grad_norm": 0.2233804613351822,
      "learning_rate": 2.1581730157717856e-05,
      "loss": 0.0148,
      "num_input_tokens_seen": 27100448,
      "step": 8150
    },
    {
      "epoch": 1.6316934697246328,
      "grad_norm": 0.16001921892166138,
      "learning_rate": 2.1555782406070948e-05,
      "loss": 0.02,
      "num_input_tokens_seen": 27117600,
      "step": 8155
    },
    {
      "epoch": 1.632693894905335,
      "grad_norm": 0.18442663550376892,
      "learning_rate": 2.152983843596741e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 27133248,
      "step": 8160
    },
    {
      "epoch": 1.6336943200860365,
      "grad_norm": 0.1674097180366516,
      "learning_rate": 2.1503898275892177e-05,
      "loss": 0.0084,
      "num_input_tokens_seen": 27150272,
      "step": 8165
    },
    {
      "epoch": 1.6346947452667384,
      "grad_norm": 0.4220287501811981,
      "learning_rate": 2.147796195432597e-05,
      "loss": 0.0106,
      "num_input_tokens_seen": 27167296,
      "step": 8170
    },
    {
      "epoch": 1.6356951704474403,
      "grad_norm": 0.1048525869846344,
      "learning_rate": 2.145202949974532e-05,
      "loss": 0.0117,
      "num_input_tokens_seen": 27184672,
      "step": 8175
    },
    {
      "epoch": 1.636695595628142,
      "grad_norm": 0.7683660387992859,
      "learning_rate": 2.1426100940622483e-05,
      "loss": 0.0221,
      "num_input_tokens_seen": 27200768,
      "step": 8180
    },
    {
      "epoch": 1.6376960208088438,
      "grad_norm": 0.10124363005161285,
      "learning_rate": 2.140017630542546e-05,
      "loss": 0.0246,
      "num_input_tokens_seen": 27216992,
      "step": 8185
    },
    {
      "epoch": 1.6386964459895457,
      "grad_norm": 0.5351779460906982,
      "learning_rate": 2.137425562261795e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 27233184,
      "step": 8190
    },
    {
      "epoch": 1.6396968711702473,
      "grad_norm": 0.08957615494728088,
      "learning_rate": 2.1348338920659285e-05,
      "loss": 0.0223,
      "num_input_tokens_seen": 27248928,
      "step": 8195
    },
    {
      "epoch": 1.6406972963509492,
      "grad_norm": 0.12967830896377563,
      "learning_rate": 2.1322426228004453e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 27265536,
      "step": 8200
    },
    {
      "epoch": 1.641697721531651,
      "grad_norm": 0.0696130022406578,
      "learning_rate": 2.1296517573104015e-05,
      "loss": 0.0145,
      "num_input_tokens_seen": 27282112,
      "step": 8205
    },
    {
      "epoch": 1.6426981467123527,
      "grad_norm": 1.087754249572754,
      "learning_rate": 2.127061298440413e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 27298592,
      "step": 8210
    },
    {
      "epoch": 1.6436985718930546,
      "grad_norm": 0.2898842394351959,
      "learning_rate": 2.1244712490346455e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 27315008,
      "step": 8215
    },
    {
      "epoch": 1.6446989970737564,
      "grad_norm": 0.8593503832817078,
      "learning_rate": 2.1218816119368194e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 27331776,
      "step": 8220
    },
    {
      "epoch": 1.645699422254458,
      "grad_norm": 0.2977852523326874,
      "learning_rate": 2.119292389990199e-05,
      "loss": 0.0149,
      "num_input_tokens_seen": 27347712,
      "step": 8225
    },
    {
      "epoch": 1.64669984743516,
      "grad_norm": 0.6195009350776672,
      "learning_rate": 2.1167035860375947e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 27364992,
      "step": 8230
    },
    {
      "epoch": 1.6477002726158618,
      "grad_norm": 0.445385605096817,
      "learning_rate": 2.114115202921357e-05,
      "loss": 0.0193,
      "num_input_tokens_seen": 27380672,
      "step": 8235
    },
    {
      "epoch": 1.6487006977965635,
      "grad_norm": 0.402042418718338,
      "learning_rate": 2.1115272434833733e-05,
      "loss": 0.0113,
      "num_input_tokens_seen": 27397888,
      "step": 8240
    },
    {
      "epoch": 1.6497011229772653,
      "grad_norm": 0.3347969651222229,
      "learning_rate": 2.108939710565069e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 27415520,
      "step": 8245
    },
    {
      "epoch": 1.6507015481579672,
      "grad_norm": 0.5899851322174072,
      "learning_rate": 2.1063526070073986e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 27432192,
      "step": 8250
    },
    {
      "epoch": 1.6517019733386689,
      "grad_norm": 0.4121209681034088,
      "learning_rate": 2.1037659356508457e-05,
      "loss": 0.0322,
      "num_input_tokens_seen": 27447712,
      "step": 8255
    },
    {
      "epoch": 1.6527023985193707,
      "grad_norm": 0.3264901041984558,
      "learning_rate": 2.1011796993354192e-05,
      "loss": 0.0208,
      "num_input_tokens_seen": 27464288,
      "step": 8260
    },
    {
      "epoch": 1.6537028237000726,
      "grad_norm": 0.5783780813217163,
      "learning_rate": 2.0985939009006507e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 27480480,
      "step": 8265
    },
    {
      "epoch": 1.6547032488807742,
      "grad_norm": 0.177225723862648,
      "learning_rate": 2.0960085431855925e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 27496800,
      "step": 8270
    },
    {
      "epoch": 1.6557036740614761,
      "grad_norm": 0.3714253306388855,
      "learning_rate": 2.093423629028811e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 27513600,
      "step": 8275
    },
    {
      "epoch": 1.656704099242178,
      "grad_norm": 0.38959836959838867,
      "learning_rate": 2.0908391612683854e-05,
      "loss": 0.0197,
      "num_input_tokens_seen": 27529376,
      "step": 8280
    },
    {
      "epoch": 1.6577045244228796,
      "grad_norm": 0.29594042897224426,
      "learning_rate": 2.088255142741906e-05,
      "loss": 0.0079,
      "num_input_tokens_seen": 27545568,
      "step": 8285
    },
    {
      "epoch": 1.6587049496035815,
      "grad_norm": 0.9828856587409973,
      "learning_rate": 2.08567157628647e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 27561536,
      "step": 8290
    },
    {
      "epoch": 1.6597053747842834,
      "grad_norm": 0.7645307779312134,
      "learning_rate": 2.0830884647386767e-05,
      "loss": 0.0311,
      "num_input_tokens_seen": 27578464,
      "step": 8295
    },
    {
      "epoch": 1.660705799964985,
      "grad_norm": 0.29915305972099304,
      "learning_rate": 2.080505810934628e-05,
      "loss": 0.0192,
      "num_input_tokens_seen": 27596192,
      "step": 8300
    },
    {
      "epoch": 1.6617062251456869,
      "grad_norm": 0.6114938259124756,
      "learning_rate": 2.077923617709921e-05,
      "loss": 0.018,
      "num_input_tokens_seen": 27612288,
      "step": 8305
    },
    {
      "epoch": 1.6627066503263888,
      "grad_norm": 0.9057605862617493,
      "learning_rate": 2.075341887899649e-05,
      "loss": 0.0204,
      "num_input_tokens_seen": 27628736,
      "step": 8310
    },
    {
      "epoch": 1.6637070755070904,
      "grad_norm": 0.3348783254623413,
      "learning_rate": 2.072760624338395e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 27644608,
      "step": 8315
    },
    {
      "epoch": 1.6647075006877923,
      "grad_norm": 0.11718225479125977,
      "learning_rate": 2.070179829860232e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 27661344,
      "step": 8320
    },
    {
      "epoch": 1.6657079258684941,
      "grad_norm": 0.1801997274160385,
      "learning_rate": 2.0675995072987165e-05,
      "loss": 0.033,
      "num_input_tokens_seen": 27678368,
      "step": 8325
    },
    {
      "epoch": 1.6667083510491958,
      "grad_norm": 1.089145541191101,
      "learning_rate": 2.065019659486887e-05,
      "loss": 0.0174,
      "num_input_tokens_seen": 27696480,
      "step": 8330
    },
    {
      "epoch": 1.6677087762298977,
      "grad_norm": 0.39637240767478943,
      "learning_rate": 2.0624402892572607e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 27712480,
      "step": 8335
    },
    {
      "epoch": 1.6687092014105995,
      "grad_norm": 1.060081958770752,
      "learning_rate": 2.0598613994418308e-05,
      "loss": 0.0222,
      "num_input_tokens_seen": 27729376,
      "step": 8340
    },
    {
      "epoch": 1.6697096265913012,
      "grad_norm": 0.2965173125267029,
      "learning_rate": 2.0572829928720635e-05,
      "loss": 0.0108,
      "num_input_tokens_seen": 27746560,
      "step": 8345
    },
    {
      "epoch": 1.670710051772003,
      "grad_norm": 0.1402323842048645,
      "learning_rate": 2.054705072378893e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 27764896,
      "step": 8350
    },
    {
      "epoch": 1.671710476952705,
      "grad_norm": 0.20911061763763428,
      "learning_rate": 2.0521276407927212e-05,
      "loss": 0.0231,
      "num_input_tokens_seen": 27782336,
      "step": 8355
    },
    {
      "epoch": 1.6727109021334066,
      "grad_norm": 0.0908847227692604,
      "learning_rate": 2.0495507009434127e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 27798272,
      "step": 8360
    },
    {
      "epoch": 1.6737113273141087,
      "grad_norm": 0.2123504877090454,
      "learning_rate": 2.046974255660291e-05,
      "loss": 0.0115,
      "num_input_tokens_seen": 27814528,
      "step": 8365
    },
    {
      "epoch": 1.6747117524948103,
      "grad_norm": 1.0239791870117188,
      "learning_rate": 2.044398307772139e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 27830080,
      "step": 8370
    },
    {
      "epoch": 1.675712177675512,
      "grad_norm": 0.0846494659781456,
      "learning_rate": 2.0418228601071927e-05,
      "loss": 0.0091,
      "num_input_tokens_seen": 27846432,
      "step": 8375
    },
    {
      "epoch": 1.676712602856214,
      "grad_norm": 0.5295249223709106,
      "learning_rate": 2.0392479154931377e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 27862720,
      "step": 8380
    },
    {
      "epoch": 1.6777130280369157,
      "grad_norm": 0.9511027932167053,
      "learning_rate": 2.036673476757108e-05,
      "loss": 0.0289,
      "num_input_tokens_seen": 27878336,
      "step": 8385
    },
    {
      "epoch": 1.6787134532176173,
      "grad_norm": 0.4604302942752838,
      "learning_rate": 2.034099546725682e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 27894368,
      "step": 8390
    },
    {
      "epoch": 1.6797138783983194,
      "grad_norm": 1.2224334478378296,
      "learning_rate": 2.031526128224881e-05,
      "loss": 0.0183,
      "num_input_tokens_seen": 27910720,
      "step": 8395
    },
    {
      "epoch": 1.680714303579021,
      "grad_norm": 0.639065682888031,
      "learning_rate": 2.0289532240801618e-05,
      "loss": 0.0237,
      "num_input_tokens_seen": 27927584,
      "step": 8400
    },
    {
      "epoch": 1.6817147287597227,
      "grad_norm": 0.16994817554950714,
      "learning_rate": 2.0263808371164185e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 27944064,
      "step": 8405
    },
    {
      "epoch": 1.6827151539404248,
      "grad_norm": 0.16935010254383087,
      "learning_rate": 2.023808970157978e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 27963712,
      "step": 8410
    },
    {
      "epoch": 1.6837155791211265,
      "grad_norm": 0.22500985860824585,
      "learning_rate": 2.0212376260285944e-05,
      "loss": 0.0135,
      "num_input_tokens_seen": 27979360,
      "step": 8415
    },
    {
      "epoch": 1.6847160043018283,
      "grad_norm": 0.639189600944519,
      "learning_rate": 2.018666807551448e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 27997728,
      "step": 8420
    },
    {
      "epoch": 1.6857164294825302,
      "grad_norm": 0.4524305760860443,
      "learning_rate": 2.0160965175491444e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 28014752,
      "step": 8425
    },
    {
      "epoch": 1.6867168546632318,
      "grad_norm": 0.16592879593372345,
      "learning_rate": 2.0135267588437054e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 28030688,
      "step": 8430
    },
    {
      "epoch": 1.6877172798439337,
      "grad_norm": 0.2719612121582031,
      "learning_rate": 2.0109575342565728e-05,
      "loss": 0.0228,
      "num_input_tokens_seen": 28047712,
      "step": 8435
    },
    {
      "epoch": 1.6887177050246356,
      "grad_norm": 0.34056931734085083,
      "learning_rate": 2.008388846608599e-05,
      "loss": 0.0189,
      "num_input_tokens_seen": 28065376,
      "step": 8440
    },
    {
      "epoch": 1.6897181302053372,
      "grad_norm": 0.1926608383655548,
      "learning_rate": 2.0058206987200488e-05,
      "loss": 0.0109,
      "num_input_tokens_seen": 28081760,
      "step": 8445
    },
    {
      "epoch": 1.690718555386039,
      "grad_norm": 1.205339789390564,
      "learning_rate": 2.003253093410594e-05,
      "loss": 0.0221,
      "num_input_tokens_seen": 28097728,
      "step": 8450
    },
    {
      "epoch": 1.691718980566741,
      "grad_norm": 0.37929344177246094,
      "learning_rate": 2.0006860334993105e-05,
      "loss": 0.015,
      "num_input_tokens_seen": 28115488,
      "step": 8455
    },
    {
      "epoch": 1.6927194057474426,
      "grad_norm": 0.446153461933136,
      "learning_rate": 1.9981195218046746e-05,
      "loss": 0.0092,
      "num_input_tokens_seen": 28132128,
      "step": 8460
    },
    {
      "epoch": 1.6937198309281445,
      "grad_norm": 0.4648114740848541,
      "learning_rate": 1.9955535611445625e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 28149056,
      "step": 8465
    },
    {
      "epoch": 1.6947202561088464,
      "grad_norm": 0.5778411626815796,
      "learning_rate": 1.9929881543362432e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 28165472,
      "step": 8470
    },
    {
      "epoch": 1.695720681289548,
      "grad_norm": 0.12256700545549393,
      "learning_rate": 1.9904233041963803e-05,
      "loss": 0.0121,
      "num_input_tokens_seen": 28181376,
      "step": 8475
    },
    {
      "epoch": 1.6967211064702499,
      "grad_norm": 0.1696489453315735,
      "learning_rate": 1.9878590135410244e-05,
      "loss": 0.0144,
      "num_input_tokens_seen": 28196672,
      "step": 8480
    },
    {
      "epoch": 1.6977215316509517,
      "grad_norm": 0.9251601696014404,
      "learning_rate": 1.9852952851856112e-05,
      "loss": 0.0308,
      "num_input_tokens_seen": 28213440,
      "step": 8485
    },
    {
      "epoch": 1.6987219568316534,
      "grad_norm": 1.0078340768814087,
      "learning_rate": 1.9827321219449608e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 28230400,
      "step": 8490
    },
    {
      "epoch": 1.6997223820123553,
      "grad_norm": 0.4716654419898987,
      "learning_rate": 1.9801695266332715e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 28248160,
      "step": 8495
    },
    {
      "epoch": 1.7007228071930571,
      "grad_norm": 0.4144385755062103,
      "learning_rate": 1.9776075020641183e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 28265472,
      "step": 8500
    },
    {
      "epoch": 1.7017232323737588,
      "grad_norm": 0.1860032081604004,
      "learning_rate": 1.9750460510504508e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 28282272,
      "step": 8505
    },
    {
      "epoch": 1.7027236575544606,
      "grad_norm": 0.5012482404708862,
      "learning_rate": 1.972485176404587e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 28300064,
      "step": 8510
    },
    {
      "epoch": 1.7037240827351625,
      "grad_norm": 0.4445567727088928,
      "learning_rate": 1.969924880938213e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 28316384,
      "step": 8515
    },
    {
      "epoch": 1.7047245079158642,
      "grad_norm": 0.2888290584087372,
      "learning_rate": 1.967365167462379e-05,
      "loss": 0.0197,
      "num_input_tokens_seen": 28334368,
      "step": 8520
    },
    {
      "epoch": 1.705724933096566,
      "grad_norm": 0.28883203864097595,
      "learning_rate": 1.9648060387874967e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 28350624,
      "step": 8525
    },
    {
      "epoch": 1.706725358277268,
      "grad_norm": 0.4163239598274231,
      "learning_rate": 1.9622474977233352e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 28367680,
      "step": 8530
    },
    {
      "epoch": 1.7077257834579695,
      "grad_norm": 0.604323148727417,
      "learning_rate": 1.9596895470790174e-05,
      "loss": 0.022,
      "num_input_tokens_seen": 28385120,
      "step": 8535
    },
    {
      "epoch": 1.7087262086386714,
      "grad_norm": 0.5926190614700317,
      "learning_rate": 1.95713218966302e-05,
      "loss": 0.0121,
      "num_input_tokens_seen": 28401568,
      "step": 8540
    },
    {
      "epoch": 1.7097266338193733,
      "grad_norm": 0.17457112669944763,
      "learning_rate": 1.954575428283167e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 28419424,
      "step": 8545
    },
    {
      "epoch": 1.710727059000075,
      "grad_norm": 0.48416203260421753,
      "learning_rate": 1.9520192657466284e-05,
      "loss": 0.0211,
      "num_input_tokens_seen": 28435296,
      "step": 8550
    },
    {
      "epoch": 1.7117274841807768,
      "grad_norm": 0.3750211298465729,
      "learning_rate": 1.9494637048599158e-05,
      "loss": 0.0189,
      "num_input_tokens_seen": 28453120,
      "step": 8555
    },
    {
      "epoch": 1.7127279093614787,
      "grad_norm": 0.8351686000823975,
      "learning_rate": 1.9469087484288825e-05,
      "loss": 0.0353,
      "num_input_tokens_seen": 28469376,
      "step": 8560
    },
    {
      "epoch": 1.7137283345421803,
      "grad_norm": 0.5917026996612549,
      "learning_rate": 1.944354399258716e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 28487520,
      "step": 8565
    },
    {
      "epoch": 1.7147287597228822,
      "grad_norm": 0.6885085105895996,
      "learning_rate": 1.941800660153937e-05,
      "loss": 0.0165,
      "num_input_tokens_seen": 28503936,
      "step": 8570
    },
    {
      "epoch": 1.715729184903584,
      "grad_norm": 0.05982868745923042,
      "learning_rate": 1.9392475339183975e-05,
      "loss": 0.0221,
      "num_input_tokens_seen": 28520224,
      "step": 8575
    },
    {
      "epoch": 1.7167296100842857,
      "grad_norm": 0.817966878414154,
      "learning_rate": 1.9366950233552777e-05,
      "loss": 0.0148,
      "num_input_tokens_seen": 28536416,
      "step": 8580
    },
    {
      "epoch": 1.7177300352649876,
      "grad_norm": 0.600110650062561,
      "learning_rate": 1.9341431312670797e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 28552128,
      "step": 8585
    },
    {
      "epoch": 1.7187304604456894,
      "grad_norm": 0.5472889542579651,
      "learning_rate": 1.931591860455626e-05,
      "loss": 0.0257,
      "num_input_tokens_seen": 28568352,
      "step": 8590
    },
    {
      "epoch": 1.719730885626391,
      "grad_norm": 0.10418272018432617,
      "learning_rate": 1.9290412137220598e-05,
      "loss": 0.0158,
      "num_input_tokens_seen": 28584864,
      "step": 8595
    },
    {
      "epoch": 1.720731310807093,
      "grad_norm": 0.30167078971862793,
      "learning_rate": 1.926491193866837e-05,
      "loss": 0.014,
      "num_input_tokens_seen": 28601920,
      "step": 8600
    },
    {
      "epoch": 1.7217317359877948,
      "grad_norm": 0.22344553470611572,
      "learning_rate": 1.9239418036897253e-05,
      "loss": 0.0215,
      "num_input_tokens_seen": 28618464,
      "step": 8605
    },
    {
      "epoch": 1.7227321611684965,
      "grad_norm": 0.18009479343891144,
      "learning_rate": 1.9213930459898014e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 28634784,
      "step": 8610
    },
    {
      "epoch": 1.7237325863491986,
      "grad_norm": 0.43777209520339966,
      "learning_rate": 1.9188449235654488e-05,
      "loss": 0.0175,
      "num_input_tokens_seen": 28651296,
      "step": 8615
    },
    {
      "epoch": 1.7247330115299002,
      "grad_norm": 0.30056893825531006,
      "learning_rate": 1.9162974392143513e-05,
      "loss": 0.021,
      "num_input_tokens_seen": 28667776,
      "step": 8620
    },
    {
      "epoch": 1.7257334367106019,
      "grad_norm": 0.4077411890029907,
      "learning_rate": 1.9137505957334927e-05,
      "loss": 0.0179,
      "num_input_tokens_seen": 28683648,
      "step": 8625
    },
    {
      "epoch": 1.726733861891304,
      "grad_norm": 0.9050543904304504,
      "learning_rate": 1.911204395919155e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 28699712,
      "step": 8630
    },
    {
      "epoch": 1.7277342870720056,
      "grad_norm": 0.17446036636829376,
      "learning_rate": 1.9086588425669125e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 28715264,
      "step": 8635
    },
    {
      "epoch": 1.7287347122527073,
      "grad_norm": 0.09911100566387177,
      "learning_rate": 1.9061139384716282e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 28734208,
      "step": 8640
    },
    {
      "epoch": 1.7297351374334093,
      "grad_norm": 0.3007764518260956,
      "learning_rate": 1.9035696864274537e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 28752000,
      "step": 8645
    },
    {
      "epoch": 1.730735562614111,
      "grad_norm": 0.3763737976551056,
      "learning_rate": 1.901026089227825e-05,
      "loss": 0.0146,
      "num_input_tokens_seen": 28767776,
      "step": 8650
    },
    {
      "epoch": 1.7317359877948126,
      "grad_norm": 0.645039439201355,
      "learning_rate": 1.898483149665458e-05,
      "loss": 0.0175,
      "num_input_tokens_seen": 28784352,
      "step": 8655
    },
    {
      "epoch": 1.7327364129755147,
      "grad_norm": 0.3146284818649292,
      "learning_rate": 1.8959408705323473e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 28800384,
      "step": 8660
    },
    {
      "epoch": 1.7337368381562164,
      "grad_norm": 0.5576115250587463,
      "learning_rate": 1.8933992546197622e-05,
      "loss": 0.0239,
      "num_input_tokens_seen": 28817248,
      "step": 8665
    },
    {
      "epoch": 1.7347372633369182,
      "grad_norm": 0.5347667932510376,
      "learning_rate": 1.8908583047182436e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 28833344,
      "step": 8670
    },
    {
      "epoch": 1.7357376885176201,
      "grad_norm": 0.22791774570941925,
      "learning_rate": 1.888318023617601e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 28849984,
      "step": 8675
    },
    {
      "epoch": 1.7367381136983218,
      "grad_norm": 0.201163187623024,
      "learning_rate": 1.8857784141069113e-05,
      "loss": 0.029,
      "num_input_tokens_seen": 28865056,
      "step": 8680
    },
    {
      "epoch": 1.7377385388790236,
      "grad_norm": 0.0956418439745903,
      "learning_rate": 1.8832394789745117e-05,
      "loss": 0.0145,
      "num_input_tokens_seen": 28882016,
      "step": 8685
    },
    {
      "epoch": 1.7387389640597255,
      "grad_norm": 0.1477956920862198,
      "learning_rate": 1.8807012210080005e-05,
      "loss": 0.0238,
      "num_input_tokens_seen": 28898624,
      "step": 8690
    },
    {
      "epoch": 1.7397393892404271,
      "grad_norm": 0.6203426122665405,
      "learning_rate": 1.8781636429942318e-05,
      "loss": 0.0249,
      "num_input_tokens_seen": 28915040,
      "step": 8695
    },
    {
      "epoch": 1.740739814421129,
      "grad_norm": 0.15471091866493225,
      "learning_rate": 1.875626747719313e-05,
      "loss": 0.012,
      "num_input_tokens_seen": 28932032,
      "step": 8700
    },
    {
      "epoch": 1.7417402396018309,
      "grad_norm": 0.3013649880886078,
      "learning_rate": 1.8730905379686026e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 28947136,
      "step": 8705
    },
    {
      "epoch": 1.7427406647825325,
      "grad_norm": 0.5449125170707703,
      "learning_rate": 1.8705550165267067e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 28963360,
      "step": 8710
    },
    {
      "epoch": 1.7437410899632344,
      "grad_norm": 0.3873887360095978,
      "learning_rate": 1.8680201861774733e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 28979584,
      "step": 8715
    },
    {
      "epoch": 1.7447415151439363,
      "grad_norm": 0.4926188588142395,
      "learning_rate": 1.8654860497039954e-05,
      "loss": 0.0297,
      "num_input_tokens_seen": 28996288,
      "step": 8720
    },
    {
      "epoch": 1.745741940324638,
      "grad_norm": 0.5046607851982117,
      "learning_rate": 1.8629526098886007e-05,
      "loss": 0.0267,
      "num_input_tokens_seen": 29012384,
      "step": 8725
    },
    {
      "epoch": 1.7467423655053398,
      "grad_norm": 0.5384324193000793,
      "learning_rate": 1.8604198695128534e-05,
      "loss": 0.0204,
      "num_input_tokens_seen": 29028544,
      "step": 8730
    },
    {
      "epoch": 1.7477427906860417,
      "grad_norm": 0.30341100692749023,
      "learning_rate": 1.8578878313575516e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 29045056,
      "step": 8735
    },
    {
      "epoch": 1.7487432158667433,
      "grad_norm": 0.12679381668567657,
      "learning_rate": 1.8553564982027183e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 29062624,
      "step": 8740
    },
    {
      "epoch": 1.7497436410474452,
      "grad_norm": 0.23378930985927582,
      "learning_rate": 1.8528258728276067e-05,
      "loss": 0.0159,
      "num_input_tokens_seen": 29079456,
      "step": 8745
    },
    {
      "epoch": 1.750744066228147,
      "grad_norm": 0.6869626641273499,
      "learning_rate": 1.8502959580106906e-05,
      "loss": 0.0117,
      "num_input_tokens_seen": 29095904,
      "step": 8750
    },
    {
      "epoch": 1.7517444914088487,
      "grad_norm": 1.1471084356307983,
      "learning_rate": 1.8477667565296623e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 29112480,
      "step": 8755
    },
    {
      "epoch": 1.7527449165895506,
      "grad_norm": 0.5438526272773743,
      "learning_rate": 1.845238271161435e-05,
      "loss": 0.02,
      "num_input_tokens_seen": 29128288,
      "step": 8760
    },
    {
      "epoch": 1.7537453417702524,
      "grad_norm": 0.5923305153846741,
      "learning_rate": 1.842710504682132e-05,
      "loss": 0.0242,
      "num_input_tokens_seen": 29145216,
      "step": 8765
    },
    {
      "epoch": 1.754745766950954,
      "grad_norm": 0.9874956011772156,
      "learning_rate": 1.8401834598670887e-05,
      "loss": 0.0268,
      "num_input_tokens_seen": 29161728,
      "step": 8770
    },
    {
      "epoch": 1.755746192131656,
      "grad_norm": 0.6199465990066528,
      "learning_rate": 1.837657139490848e-05,
      "loss": 0.0185,
      "num_input_tokens_seen": 29178048,
      "step": 8775
    },
    {
      "epoch": 1.7567466173123578,
      "grad_norm": 0.3383076786994934,
      "learning_rate": 1.8351315463271568e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 29196032,
      "step": 8780
    },
    {
      "epoch": 1.7577470424930595,
      "grad_norm": 0.3935118317604065,
      "learning_rate": 1.8326066831489663e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 29211840,
      "step": 8785
    },
    {
      "epoch": 1.7587474676737613,
      "grad_norm": 0.1697453409433365,
      "learning_rate": 1.8300825527284225e-05,
      "loss": 0.023,
      "num_input_tokens_seen": 29227392,
      "step": 8790
    },
    {
      "epoch": 1.7597478928544632,
      "grad_norm": 0.48267650604248047,
      "learning_rate": 1.8275591578368683e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 29242528,
      "step": 8795
    },
    {
      "epoch": 1.7607483180351648,
      "grad_norm": 0.3457343876361847,
      "learning_rate": 1.82503650124484e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 29258912,
      "step": 8800
    },
    {
      "epoch": 1.7617487432158667,
      "grad_norm": 0.3773598074913025,
      "learning_rate": 1.822514585722063e-05,
      "loss": 0.009,
      "num_input_tokens_seen": 29275904,
      "step": 8805
    },
    {
      "epoch": 1.7627491683965686,
      "grad_norm": 0.44147181510925293,
      "learning_rate": 1.8199934140374465e-05,
      "loss": 0.0227,
      "num_input_tokens_seen": 29292192,
      "step": 8810
    },
    {
      "epoch": 1.7637495935772702,
      "grad_norm": 0.5126002430915833,
      "learning_rate": 1.8174729889590874e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 29308640,
      "step": 8815
    },
    {
      "epoch": 1.764750018757972,
      "grad_norm": 1.294550895690918,
      "learning_rate": 1.8149533132542594e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 29324672,
      "step": 8820
    },
    {
      "epoch": 1.765750443938674,
      "grad_norm": 0.2473556399345398,
      "learning_rate": 1.8124343896894148e-05,
      "loss": 0.0188,
      "num_input_tokens_seen": 29341504,
      "step": 8825
    },
    {
      "epoch": 1.7667508691193756,
      "grad_norm": 0.36207565665245056,
      "learning_rate": 1.8099162210301795e-05,
      "loss": 0.0115,
      "num_input_tokens_seen": 29357952,
      "step": 8830
    },
    {
      "epoch": 1.7677512943000775,
      "grad_norm": 0.5558183193206787,
      "learning_rate": 1.8073988100413514e-05,
      "loss": 0.0276,
      "num_input_tokens_seen": 29374752,
      "step": 8835
    },
    {
      "epoch": 1.7687517194807794,
      "grad_norm": 0.6651312708854675,
      "learning_rate": 1.804882159486897e-05,
      "loss": 0.0193,
      "num_input_tokens_seen": 29391936,
      "step": 8840
    },
    {
      "epoch": 1.769752144661481,
      "grad_norm": 0.13344928622245789,
      "learning_rate": 1.8023662721299455e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 29408512,
      "step": 8845
    },
    {
      "epoch": 1.7707525698421829,
      "grad_norm": 0.3455791473388672,
      "learning_rate": 1.7998511507327898e-05,
      "loss": 0.0137,
      "num_input_tokens_seen": 29425632,
      "step": 8850
    },
    {
      "epoch": 1.7717529950228847,
      "grad_norm": 0.30371081829071045,
      "learning_rate": 1.7973367980568822e-05,
      "loss": 0.0168,
      "num_input_tokens_seen": 29443840,
      "step": 8855
    },
    {
      "epoch": 1.7727534202035864,
      "grad_norm": 0.3325488269329071,
      "learning_rate": 1.7948232168628305e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 29459584,
      "step": 8860
    },
    {
      "epoch": 1.7737538453842885,
      "grad_norm": 1.1632492542266846,
      "learning_rate": 1.7923104099103945e-05,
      "loss": 0.0253,
      "num_input_tokens_seen": 29476704,
      "step": 8865
    },
    {
      "epoch": 1.7747542705649901,
      "grad_norm": 0.23372527956962585,
      "learning_rate": 1.7897983799584856e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 29493920,
      "step": 8870
    },
    {
      "epoch": 1.7757546957456918,
      "grad_norm": 0.11346451938152313,
      "learning_rate": 1.7872871297651613e-05,
      "loss": 0.0192,
      "num_input_tokens_seen": 29511392,
      "step": 8875
    },
    {
      "epoch": 1.7767551209263939,
      "grad_norm": 0.7949175238609314,
      "learning_rate": 1.7847766620876223e-05,
      "loss": 0.0273,
      "num_input_tokens_seen": 29528096,
      "step": 8880
    },
    {
      "epoch": 1.7777555461070955,
      "grad_norm": 0.3690131902694702,
      "learning_rate": 1.782266979682211e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 29544512,
      "step": 8885
    },
    {
      "epoch": 1.7787559712877972,
      "grad_norm": 0.9416623115539551,
      "learning_rate": 1.779758085304409e-05,
      "loss": 0.0331,
      "num_input_tokens_seen": 29561184,
      "step": 8890
    },
    {
      "epoch": 1.7797563964684993,
      "grad_norm": 0.3709040582180023,
      "learning_rate": 1.7772499817088297e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 29577408,
      "step": 8895
    },
    {
      "epoch": 1.780756821649201,
      "grad_norm": 0.3078094720840454,
      "learning_rate": 1.77474267164922e-05,
      "loss": 0.016,
      "num_input_tokens_seen": 29594944,
      "step": 8900
    },
    {
      "epoch": 1.7817572468299026,
      "grad_norm": 0.5173088908195496,
      "learning_rate": 1.7722361578784547e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 29612416,
      "step": 8905
    },
    {
      "epoch": 1.7827576720106046,
      "grad_norm": 0.11358284950256348,
      "learning_rate": 1.7697304431485358e-05,
      "loss": 0.0117,
      "num_input_tokens_seen": 29628736,
      "step": 8910
    },
    {
      "epoch": 1.7837580971913063,
      "grad_norm": 0.6156454682350159,
      "learning_rate": 1.7672255302105868e-05,
      "loss": 0.0298,
      "num_input_tokens_seen": 29646144,
      "step": 8915
    },
    {
      "epoch": 1.784758522372008,
      "grad_norm": 0.7819226980209351,
      "learning_rate": 1.7647214218148505e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 29662752,
      "step": 8920
    },
    {
      "epoch": 1.78575894755271,
      "grad_norm": 0.25487658381462097,
      "learning_rate": 1.7622181207106884e-05,
      "loss": 0.013,
      "num_input_tokens_seen": 29679296,
      "step": 8925
    },
    {
      "epoch": 1.7867593727334117,
      "grad_norm": 0.5328194499015808,
      "learning_rate": 1.7597156296465734e-05,
      "loss": 0.0133,
      "num_input_tokens_seen": 29695840,
      "step": 8930
    },
    {
      "epoch": 1.7877597979141135,
      "grad_norm": 0.5457685589790344,
      "learning_rate": 1.7572139513700885e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 29713088,
      "step": 8935
    },
    {
      "epoch": 1.7887602230948154,
      "grad_norm": 0.23857742547988892,
      "learning_rate": 1.7547130886279283e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 29729632,
      "step": 8940
    },
    {
      "epoch": 1.789760648275517,
      "grad_norm": 0.4583163857460022,
      "learning_rate": 1.7522130441658884e-05,
      "loss": 0.0196,
      "num_input_tokens_seen": 29745760,
      "step": 8945
    },
    {
      "epoch": 1.790761073456219,
      "grad_norm": 0.3863668143749237,
      "learning_rate": 1.749713820728867e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 29762272,
      "step": 8950
    },
    {
      "epoch": 1.7917614986369208,
      "grad_norm": 0.2928677201271057,
      "learning_rate": 1.747215421060861e-05,
      "loss": 0.0267,
      "num_input_tokens_seen": 29778976,
      "step": 8955
    },
    {
      "epoch": 1.7927619238176224,
      "grad_norm": 0.4981392025947571,
      "learning_rate": 1.7447178479049615e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 29795680,
      "step": 8960
    },
    {
      "epoch": 1.7937623489983243,
      "grad_norm": 0.04591726139187813,
      "learning_rate": 1.742221104003355e-05,
      "loss": 0.011,
      "num_input_tokens_seen": 29813728,
      "step": 8965
    },
    {
      "epoch": 1.7947627741790262,
      "grad_norm": 0.23080898821353912,
      "learning_rate": 1.7397251920973146e-05,
      "loss": 0.0148,
      "num_input_tokens_seen": 29829440,
      "step": 8970
    },
    {
      "epoch": 1.7957631993597278,
      "grad_norm": 0.935471773147583,
      "learning_rate": 1.7372301149272012e-05,
      "loss": 0.019,
      "num_input_tokens_seen": 29848000,
      "step": 8975
    },
    {
      "epoch": 1.7967636245404297,
      "grad_norm": 0.413185715675354,
      "learning_rate": 1.7347358752324605e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 29866240,
      "step": 8980
    },
    {
      "epoch": 1.7977640497211316,
      "grad_norm": 0.6893980503082275,
      "learning_rate": 1.7322424757516155e-05,
      "loss": 0.0212,
      "num_input_tokens_seen": 29883360,
      "step": 8985
    },
    {
      "epoch": 1.7987644749018332,
      "grad_norm": 0.6806471347808838,
      "learning_rate": 1.7297499192222705e-05,
      "loss": 0.0303,
      "num_input_tokens_seen": 29899744,
      "step": 8990
    },
    {
      "epoch": 1.799764900082535,
      "grad_norm": 0.20109929144382477,
      "learning_rate": 1.727258208381101e-05,
      "loss": 0.0311,
      "num_input_tokens_seen": 29917024,
      "step": 8995
    },
    {
      "epoch": 1.800765325263237,
      "grad_norm": 0.954805314540863,
      "learning_rate": 1.7247673459638563e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 29933472,
      "step": 9000
    },
    {
      "epoch": 1.8017657504439386,
      "grad_norm": 0.4293767213821411,
      "learning_rate": 1.722277334705353e-05,
      "loss": 0.0238,
      "num_input_tokens_seen": 29950528,
      "step": 9005
    },
    {
      "epoch": 1.8027661756246405,
      "grad_norm": 0.3819023072719574,
      "learning_rate": 1.7197881773394723e-05,
      "loss": 0.0382,
      "num_input_tokens_seen": 29967008,
      "step": 9010
    },
    {
      "epoch": 1.8037666008053423,
      "grad_norm": 0.42908862233161926,
      "learning_rate": 1.7172998765991604e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 29983808,
      "step": 9015
    },
    {
      "epoch": 1.804767025986044,
      "grad_norm": 0.07979098707437515,
      "learning_rate": 1.7148124352164216e-05,
      "loss": 0.016,
      "num_input_tokens_seen": 30002112,
      "step": 9020
    },
    {
      "epoch": 1.8057674511667459,
      "grad_norm": 0.5107935667037964,
      "learning_rate": 1.712325855922316e-05,
      "loss": 0.0196,
      "num_input_tokens_seen": 30018944,
      "step": 9025
    },
    {
      "epoch": 1.8067678763474477,
      "grad_norm": 0.34811583161354065,
      "learning_rate": 1.7098401414469574e-05,
      "loss": 0.0183,
      "num_input_tokens_seen": 30035616,
      "step": 9030
    },
    {
      "epoch": 1.8077683015281494,
      "grad_norm": 0.20033659040927887,
      "learning_rate": 1.7073552945195105e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 30051072,
      "step": 9035
    },
    {
      "epoch": 1.8087687267088512,
      "grad_norm": 0.5653337836265564,
      "learning_rate": 1.7048713178681892e-05,
      "loss": 0.03,
      "num_input_tokens_seen": 30068864,
      "step": 9040
    },
    {
      "epoch": 1.8097691518895531,
      "grad_norm": 0.3743863105773926,
      "learning_rate": 1.702388214220249e-05,
      "loss": 0.012,
      "num_input_tokens_seen": 30086592,
      "step": 9045
    },
    {
      "epoch": 1.8107695770702548,
      "grad_norm": 0.17334720492362976,
      "learning_rate": 1.6999059863019877e-05,
      "loss": 0.0143,
      "num_input_tokens_seen": 30102752,
      "step": 9050
    },
    {
      "epoch": 1.8117700022509566,
      "grad_norm": 0.31735971570014954,
      "learning_rate": 1.697424636838743e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 30118592,
      "step": 9055
    },
    {
      "epoch": 1.8127704274316585,
      "grad_norm": 0.394786536693573,
      "learning_rate": 1.694944168554886e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 30136320,
      "step": 9060
    },
    {
      "epoch": 1.8137708526123602,
      "grad_norm": 0.0013086630497127771,
      "learning_rate": 1.6924645841738223e-05,
      "loss": 0.0117,
      "num_input_tokens_seen": 30153376,
      "step": 9065
    },
    {
      "epoch": 1.814771277793062,
      "grad_norm": 0.12593862414360046,
      "learning_rate": 1.6899858864179858e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 30169728,
      "step": 9070
    },
    {
      "epoch": 1.815771702973764,
      "grad_norm": 0.44221794605255127,
      "learning_rate": 1.687508078008837e-05,
      "loss": 0.0185,
      "num_input_tokens_seen": 30186976,
      "step": 9075
    },
    {
      "epoch": 1.8167721281544655,
      "grad_norm": 0.30854108929634094,
      "learning_rate": 1.6850311616668606e-05,
      "loss": 0.0211,
      "num_input_tokens_seen": 30204704,
      "step": 9080
    },
    {
      "epoch": 1.8177725533351674,
      "grad_norm": 0.26250120997428894,
      "learning_rate": 1.6825551401115616e-05,
      "loss": 0.0112,
      "num_input_tokens_seen": 30221696,
      "step": 9085
    },
    {
      "epoch": 1.8187729785158693,
      "grad_norm": 0.13034704327583313,
      "learning_rate": 1.680080016061461e-05,
      "loss": 0.0233,
      "num_input_tokens_seen": 30238208,
      "step": 9090
    },
    {
      "epoch": 1.819773403696571,
      "grad_norm": 0.7788635492324829,
      "learning_rate": 1.677605792234099e-05,
      "loss": 0.0147,
      "num_input_tokens_seen": 30255392,
      "step": 9095
    },
    {
      "epoch": 1.8207738288772728,
      "grad_norm": 0.3505137264728546,
      "learning_rate": 1.6751324713460214e-05,
      "loss": 0.006,
      "num_input_tokens_seen": 30272768,
      "step": 9100
    },
    {
      "epoch": 1.8217742540579747,
      "grad_norm": 0.4570329487323761,
      "learning_rate": 1.6726600561127865e-05,
      "loss": 0.0275,
      "num_input_tokens_seen": 30290464,
      "step": 9105
    },
    {
      "epoch": 1.8227746792386763,
      "grad_norm": 0.5529378056526184,
      "learning_rate": 1.6701885492489576e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 30307296,
      "step": 9110
    },
    {
      "epoch": 1.8237751044193784,
      "grad_norm": 0.5718985199928284,
      "learning_rate": 1.6677179534681e-05,
      "loss": 0.0148,
      "num_input_tokens_seen": 30323712,
      "step": 9115
    },
    {
      "epoch": 1.82477552960008,
      "grad_norm": 0.6405457854270935,
      "learning_rate": 1.665248271482778e-05,
      "loss": 0.0151,
      "num_input_tokens_seen": 30341504,
      "step": 9120
    },
    {
      "epoch": 1.8257759547807817,
      "grad_norm": 0.3813199996948242,
      "learning_rate": 1.6627795060045556e-05,
      "loss": 0.0135,
      "num_input_tokens_seen": 30358752,
      "step": 9125
    },
    {
      "epoch": 1.8267763799614838,
      "grad_norm": 0.25917142629623413,
      "learning_rate": 1.660311659743987e-05,
      "loss": 0.0163,
      "num_input_tokens_seen": 30375200,
      "step": 9130
    },
    {
      "epoch": 1.8277768051421854,
      "grad_norm": 0.25971898436546326,
      "learning_rate": 1.6578447354106195e-05,
      "loss": 0.0169,
      "num_input_tokens_seen": 30393312,
      "step": 9135
    },
    {
      "epoch": 1.828777230322887,
      "grad_norm": 0.0786486342549324,
      "learning_rate": 1.655378735712986e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 30410848,
      "step": 9140
    },
    {
      "epoch": 1.8297776555035892,
      "grad_norm": 1.411215901374817,
      "learning_rate": 1.6529136633586075e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 30429664,
      "step": 9145
    },
    {
      "epoch": 1.8307780806842908,
      "grad_norm": 0.7602545619010925,
      "learning_rate": 1.6504495210539845e-05,
      "loss": 0.0066,
      "num_input_tokens_seen": 30446400,
      "step": 9150
    },
    {
      "epoch": 1.8317785058649925,
      "grad_norm": 0.48819780349731445,
      "learning_rate": 1.6479863115045956e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 30462848,
      "step": 9155
    },
    {
      "epoch": 1.8327789310456946,
      "grad_norm": 0.2878517806529999,
      "learning_rate": 1.645524037414897e-05,
      "loss": 0.0188,
      "num_input_tokens_seen": 30480448,
      "step": 9160
    },
    {
      "epoch": 1.8337793562263962,
      "grad_norm": 0.7377524375915527,
      "learning_rate": 1.643062701488318e-05,
      "loss": 0.0225,
      "num_input_tokens_seen": 30496992,
      "step": 9165
    },
    {
      "epoch": 1.8347797814070979,
      "grad_norm": 0.295469731092453,
      "learning_rate": 1.640602306427257e-05,
      "loss": 0.0077,
      "num_input_tokens_seen": 30514016,
      "step": 9170
    },
    {
      "epoch": 1.8357802065878,
      "grad_norm": 0.7386573553085327,
      "learning_rate": 1.638142854933078e-05,
      "loss": 0.0185,
      "num_input_tokens_seen": 30529216,
      "step": 9175
    },
    {
      "epoch": 1.8367806317685016,
      "grad_norm": 0.5637227296829224,
      "learning_rate": 1.6356843497061126e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 30545408,
      "step": 9180
    },
    {
      "epoch": 1.8377810569492035,
      "grad_norm": 0.15489616990089417,
      "learning_rate": 1.63322679344565e-05,
      "loss": 0.0167,
      "num_input_tokens_seen": 30563168,
      "step": 9185
    },
    {
      "epoch": 1.8387814821299053,
      "grad_norm": 0.34152811765670776,
      "learning_rate": 1.6307701888499387e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 30579776,
      "step": 9190
    },
    {
      "epoch": 1.839781907310607,
      "grad_norm": 0.14039745926856995,
      "learning_rate": 1.628314538616184e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 30595744,
      "step": 9195
    },
    {
      "epoch": 1.8407823324913088,
      "grad_norm": 1.1402007341384888,
      "learning_rate": 1.6258598454405402e-05,
      "loss": 0.0237,
      "num_input_tokens_seen": 30610816,
      "step": 9200
    },
    {
      "epoch": 1.8417827576720107,
      "grad_norm": 0.2333715558052063,
      "learning_rate": 1.6234061120181142e-05,
      "loss": 0.0209,
      "num_input_tokens_seen": 30627552,
      "step": 9205
    },
    {
      "epoch": 1.8427831828527124,
      "grad_norm": 0.45906734466552734,
      "learning_rate": 1.6209533410429565e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 30644288,
      "step": 9210
    },
    {
      "epoch": 1.8437836080334142,
      "grad_norm": 0.45235124230384827,
      "learning_rate": 1.618501535208061e-05,
      "loss": 0.0193,
      "num_input_tokens_seen": 30661856,
      "step": 9215
    },
    {
      "epoch": 1.844784033214116,
      "grad_norm": 0.8836572170257568,
      "learning_rate": 1.616050697205365e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 30677568,
      "step": 9220
    },
    {
      "epoch": 1.8457844583948178,
      "grad_norm": 0.4476397633552551,
      "learning_rate": 1.6136008297257395e-05,
      "loss": 0.0089,
      "num_input_tokens_seen": 30693664,
      "step": 9225
    },
    {
      "epoch": 1.8467848835755196,
      "grad_norm": 0.5498594045639038,
      "learning_rate": 1.6111519354589905e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 30710368,
      "step": 9230
    },
    {
      "epoch": 1.8477853087562215,
      "grad_norm": 0.28257307410240173,
      "learning_rate": 1.608704017093858e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 30727584,
      "step": 9235
    },
    {
      "epoch": 1.8487857339369231,
      "grad_norm": 0.4074607491493225,
      "learning_rate": 1.6062570773180083e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 30744224,
      "step": 9240
    },
    {
      "epoch": 1.849786159117625,
      "grad_norm": 0.40445345640182495,
      "learning_rate": 1.6038111188180328e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 30761024,
      "step": 9245
    },
    {
      "epoch": 1.8507865842983269,
      "grad_norm": 0.6417256593704224,
      "learning_rate": 1.601366144279448e-05,
      "loss": 0.0199,
      "num_input_tokens_seen": 30777792,
      "step": 9250
    },
    {
      "epoch": 1.8517870094790285,
      "grad_norm": 0.21276015043258667,
      "learning_rate": 1.5989221563866882e-05,
      "loss": 0.0132,
      "num_input_tokens_seen": 30794880,
      "step": 9255
    },
    {
      "epoch": 1.8527874346597304,
      "grad_norm": 0.8027820587158203,
      "learning_rate": 1.5964791578231052e-05,
      "loss": 0.0174,
      "num_input_tokens_seen": 30811328,
      "step": 9260
    },
    {
      "epoch": 1.8537878598404323,
      "grad_norm": 0.9955580234527588,
      "learning_rate": 1.5940371512709634e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 30826912,
      "step": 9265
    },
    {
      "epoch": 1.854788285021134,
      "grad_norm": 0.3065422475337982,
      "learning_rate": 1.5915961394114386e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 30844224,
      "step": 9270
    },
    {
      "epoch": 1.8557887102018358,
      "grad_norm": 0.17857109010219574,
      "learning_rate": 1.5891561249246162e-05,
      "loss": 0.0105,
      "num_input_tokens_seen": 30860960,
      "step": 9275
    },
    {
      "epoch": 1.8567891353825376,
      "grad_norm": 0.3081003725528717,
      "learning_rate": 1.5867171104894843e-05,
      "loss": 0.0132,
      "num_input_tokens_seen": 30877280,
      "step": 9280
    },
    {
      "epoch": 1.8577895605632393,
      "grad_norm": 0.386561781167984,
      "learning_rate": 1.584279098783933e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 30893088,
      "step": 9285
    },
    {
      "epoch": 1.8587899857439412,
      "grad_norm": 0.22408875823020935,
      "learning_rate": 1.5818420924847535e-05,
      "loss": 0.0305,
      "num_input_tokens_seen": 30910496,
      "step": 9290
    },
    {
      "epoch": 1.859790410924643,
      "grad_norm": 1.0091034173965454,
      "learning_rate": 1.5794060942676302e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 30926432,
      "step": 9295
    },
    {
      "epoch": 1.8607908361053447,
      "grad_norm": 0.43769577145576477,
      "learning_rate": 1.5769711068071446e-05,
      "loss": 0.0366,
      "num_input_tokens_seen": 30942976,
      "step": 9300
    },
    {
      "epoch": 1.8617912612860466,
      "grad_norm": 0.45385193824768066,
      "learning_rate": 1.5745371327767643e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 30961216,
      "step": 9305
    },
    {
      "epoch": 1.8627916864667484,
      "grad_norm": 0.21157459914684296,
      "learning_rate": 1.572104174848848e-05,
      "loss": 0.0315,
      "num_input_tokens_seen": 30976928,
      "step": 9310
    },
    {
      "epoch": 1.86379211164745,
      "grad_norm": 0.009228398092091084,
      "learning_rate": 1.5696722356946354e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 30993440,
      "step": 9315
    },
    {
      "epoch": 1.864792536828152,
      "grad_norm": 0.22807136178016663,
      "learning_rate": 1.5672413179842494e-05,
      "loss": 0.0205,
      "num_input_tokens_seen": 31010304,
      "step": 9320
    },
    {
      "epoch": 1.8657929620088538,
      "grad_norm": 0.08223238587379456,
      "learning_rate": 1.564811424386691e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 31026144,
      "step": 9325
    },
    {
      "epoch": 1.8667933871895555,
      "grad_norm": 0.0866706371307373,
      "learning_rate": 1.5623825575698376e-05,
      "loss": 0.0273,
      "num_input_tokens_seen": 31042368,
      "step": 9330
    },
    {
      "epoch": 1.8677938123702573,
      "grad_norm": 1.7599283456802368,
      "learning_rate": 1.5599547202004385e-05,
      "loss": 0.0312,
      "num_input_tokens_seen": 31059264,
      "step": 9335
    },
    {
      "epoch": 1.8687942375509592,
      "grad_norm": 0.42392227053642273,
      "learning_rate": 1.5575279149441124e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 31076448,
      "step": 9340
    },
    {
      "epoch": 1.8697946627316608,
      "grad_norm": 0.3624468147754669,
      "learning_rate": 1.5551021444653447e-05,
      "loss": 0.0197,
      "num_input_tokens_seen": 31091872,
      "step": 9345
    },
    {
      "epoch": 1.8707950879123627,
      "grad_norm": 0.2185261994600296,
      "learning_rate": 1.5526774114274866e-05,
      "loss": 0.0133,
      "num_input_tokens_seen": 31107968,
      "step": 9350
    },
    {
      "epoch": 1.8717955130930646,
      "grad_norm": 0.10084425657987595,
      "learning_rate": 1.5502537184927486e-05,
      "loss": 0.0194,
      "num_input_tokens_seen": 31124384,
      "step": 9355
    },
    {
      "epoch": 1.8727959382737662,
      "grad_norm": 0.32103344798088074,
      "learning_rate": 1.5478310683221987e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 31140832,
      "step": 9360
    },
    {
      "epoch": 1.873796363454468,
      "grad_norm": 0.2012992799282074,
      "learning_rate": 1.5454094635757616e-05,
      "loss": 0.0147,
      "num_input_tokens_seen": 31157312,
      "step": 9365
    },
    {
      "epoch": 1.87479678863517,
      "grad_norm": 0.29973459243774414,
      "learning_rate": 1.5429889069122133e-05,
      "loss": 0.0149,
      "num_input_tokens_seen": 31173152,
      "step": 9370
    },
    {
      "epoch": 1.8757972138158716,
      "grad_norm": 0.7842708826065063,
      "learning_rate": 1.5405694009891787e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 31191136,
      "step": 9375
    },
    {
      "epoch": 1.8767976389965737,
      "grad_norm": 0.7371029853820801,
      "learning_rate": 1.538150948463129e-05,
      "loss": 0.0169,
      "num_input_tokens_seen": 31208160,
      "step": 9380
    },
    {
      "epoch": 1.8777980641772754,
      "grad_norm": 1.6760576963424683,
      "learning_rate": 1.535733551989381e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 31224224,
      "step": 9385
    },
    {
      "epoch": 1.878798489357977,
      "grad_norm": 0.3584641218185425,
      "learning_rate": 1.5333172142220892e-05,
      "loss": 0.0089,
      "num_input_tokens_seen": 31242272,
      "step": 9390
    },
    {
      "epoch": 1.879798914538679,
      "grad_norm": 0.059135545045137405,
      "learning_rate": 1.5309019378142468e-05,
      "loss": 0.013,
      "num_input_tokens_seen": 31259776,
      "step": 9395
    },
    {
      "epoch": 1.8807993397193807,
      "grad_norm": 0.26289892196655273,
      "learning_rate": 1.5284877254176814e-05,
      "loss": 0.0126,
      "num_input_tokens_seen": 31275808,
      "step": 9400
    },
    {
      "epoch": 1.8817997649000824,
      "grad_norm": 0.49559512734413147,
      "learning_rate": 1.5260745796830545e-05,
      "loss": 0.0194,
      "num_input_tokens_seen": 31292128,
      "step": 9405
    },
    {
      "epoch": 1.8828001900807845,
      "grad_norm": 0.1455199271440506,
      "learning_rate": 1.5236625032598533e-05,
      "loss": 0.0215,
      "num_input_tokens_seen": 31308576,
      "step": 9410
    },
    {
      "epoch": 1.8838006152614861,
      "grad_norm": 0.8640939593315125,
      "learning_rate": 1.5212514987963923e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 31325984,
      "step": 9415
    },
    {
      "epoch": 1.8848010404421878,
      "grad_norm": 0.04954162985086441,
      "learning_rate": 1.51884156893981e-05,
      "loss": 0.0198,
      "num_input_tokens_seen": 31342816,
      "step": 9420
    },
    {
      "epoch": 1.8858014656228899,
      "grad_norm": 0.5669306516647339,
      "learning_rate": 1.516432716336064e-05,
      "loss": 0.0136,
      "num_input_tokens_seen": 31359520,
      "step": 9425
    },
    {
      "epoch": 1.8868018908035915,
      "grad_norm": 0.32257279753685,
      "learning_rate": 1.514024943629928e-05,
      "loss": 0.016,
      "num_input_tokens_seen": 31377184,
      "step": 9430
    },
    {
      "epoch": 1.8878023159842934,
      "grad_norm": 0.2889237701892853,
      "learning_rate": 1.511618253464993e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 31393216,
      "step": 9435
    },
    {
      "epoch": 1.8888027411649952,
      "grad_norm": 0.5731465816497803,
      "learning_rate": 1.5092126484836594e-05,
      "loss": 0.0118,
      "num_input_tokens_seen": 31410272,
      "step": 9440
    },
    {
      "epoch": 1.889803166345697,
      "grad_norm": 0.0007583815604448318,
      "learning_rate": 1.506808131327136e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 31427136,
      "step": 9445
    },
    {
      "epoch": 1.8908035915263988,
      "grad_norm": 0.330671101808548,
      "learning_rate": 1.5044047046354368e-05,
      "loss": 0.0118,
      "num_input_tokens_seen": 31445184,
      "step": 9450
    },
    {
      "epoch": 1.8918040167071006,
      "grad_norm": 0.39024069905281067,
      "learning_rate": 1.5020023710473812e-05,
      "loss": 0.0112,
      "num_input_tokens_seen": 31462016,
      "step": 9455
    },
    {
      "epoch": 1.8928044418878023,
      "grad_norm": 0.10673998296260834,
      "learning_rate": 1.4996011332005867e-05,
      "loss": 0.0333,
      "num_input_tokens_seen": 31479296,
      "step": 9460
    },
    {
      "epoch": 1.8938048670685041,
      "grad_norm": 0.1995709240436554,
      "learning_rate": 1.4972009937314673e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 31497184,
      "step": 9465
    },
    {
      "epoch": 1.894805292249206,
      "grad_norm": 0.5582648515701294,
      "learning_rate": 1.494801955275231e-05,
      "loss": 0.0151,
      "num_input_tokens_seen": 31514784,
      "step": 9470
    },
    {
      "epoch": 1.8958057174299077,
      "grad_norm": 0.05137334764003754,
      "learning_rate": 1.4924040204658785e-05,
      "loss": 0.012,
      "num_input_tokens_seen": 31531488,
      "step": 9475
    },
    {
      "epoch": 1.8968061426106095,
      "grad_norm": 0.010768786072731018,
      "learning_rate": 1.4900071919361969e-05,
      "loss": 0.0155,
      "num_input_tokens_seen": 31549056,
      "step": 9480
    },
    {
      "epoch": 1.8978065677913114,
      "grad_norm": 0.21186643838882446,
      "learning_rate": 1.4876114723177596e-05,
      "loss": 0.013,
      "num_input_tokens_seen": 31566368,
      "step": 9485
    },
    {
      "epoch": 1.898806992972013,
      "grad_norm": 0.3738987445831299,
      "learning_rate": 1.4852168642409226e-05,
      "loss": 0.0105,
      "num_input_tokens_seen": 31582528,
      "step": 9490
    },
    {
      "epoch": 1.899807418152715,
      "grad_norm": 0.7078602910041809,
      "learning_rate": 1.4828233703348216e-05,
      "loss": 0.0289,
      "num_input_tokens_seen": 31599136,
      "step": 9495
    },
    {
      "epoch": 1.9008078433334168,
      "grad_norm": 0.599572479724884,
      "learning_rate": 1.4804309932273669e-05,
      "loss": 0.0364,
      "num_input_tokens_seen": 31615360,
      "step": 9500
    },
    {
      "epoch": 1.9018082685141184,
      "grad_norm": 1.5461905002593994,
      "learning_rate": 1.4780397355452472e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 31633152,
      "step": 9505
    },
    {
      "epoch": 1.9028086936948203,
      "grad_norm": 0.45884576439857483,
      "learning_rate": 1.4756495999139171e-05,
      "loss": 0.0114,
      "num_input_tokens_seen": 31648224,
      "step": 9510
    },
    {
      "epoch": 1.9038091188755222,
      "grad_norm": 0.07703045755624771,
      "learning_rate": 1.4732605889576031e-05,
      "loss": 0.0221,
      "num_input_tokens_seen": 31665632,
      "step": 9515
    },
    {
      "epoch": 1.9048095440562238,
      "grad_norm": 0.08801163733005524,
      "learning_rate": 1.4708727052992943e-05,
      "loss": 0.014,
      "num_input_tokens_seen": 31682944,
      "step": 9520
    },
    {
      "epoch": 1.9058099692369257,
      "grad_norm": 0.22972284257411957,
      "learning_rate": 1.468485951560743e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 31698400,
      "step": 9525
    },
    {
      "epoch": 1.9068103944176276,
      "grad_norm": 0.7997241616249084,
      "learning_rate": 1.4661003303624619e-05,
      "loss": 0.0184,
      "num_input_tokens_seen": 31714784,
      "step": 9530
    },
    {
      "epoch": 1.9078108195983292,
      "grad_norm": 0.12815675139427185,
      "learning_rate": 1.4637158443237187e-05,
      "loss": 0.0082,
      "num_input_tokens_seen": 31731168,
      "step": 9535
    },
    {
      "epoch": 1.908811244779031,
      "grad_norm": 0.04719047620892525,
      "learning_rate": 1.4613324960625346e-05,
      "loss": 0.0084,
      "num_input_tokens_seen": 31748544,
      "step": 9540
    },
    {
      "epoch": 1.909811669959733,
      "grad_norm": 0.7937561273574829,
      "learning_rate": 1.4589502881956838e-05,
      "loss": 0.0273,
      "num_input_tokens_seen": 31764864,
      "step": 9545
    },
    {
      "epoch": 1.9108120951404346,
      "grad_norm": 0.45275047421455383,
      "learning_rate": 1.4565692233386857e-05,
      "loss": 0.0076,
      "num_input_tokens_seen": 31780736,
      "step": 9550
    },
    {
      "epoch": 1.9118125203211365,
      "grad_norm": 0.061869651079177856,
      "learning_rate": 1.4541893041058068e-05,
      "loss": 0.0275,
      "num_input_tokens_seen": 31797856,
      "step": 9555
    },
    {
      "epoch": 1.9128129455018383,
      "grad_norm": 0.6054242253303528,
      "learning_rate": 1.451810533110055e-05,
      "loss": 0.0151,
      "num_input_tokens_seen": 31813440,
      "step": 9560
    },
    {
      "epoch": 1.91381337068254,
      "grad_norm": 0.12409298866987228,
      "learning_rate": 1.4494329129631786e-05,
      "loss": 0.0092,
      "num_input_tokens_seen": 31830048,
      "step": 9565
    },
    {
      "epoch": 1.9148137958632419,
      "grad_norm": 0.36158084869384766,
      "learning_rate": 1.4470564462756591e-05,
      "loss": 0.0245,
      "num_input_tokens_seen": 31846016,
      "step": 9570
    },
    {
      "epoch": 1.9158142210439437,
      "grad_norm": 0.6042336225509644,
      "learning_rate": 1.4446811356567147e-05,
      "loss": 0.0122,
      "num_input_tokens_seen": 31861056,
      "step": 9575
    },
    {
      "epoch": 1.9168146462246454,
      "grad_norm": 0.11119186133146286,
      "learning_rate": 1.442306983714294e-05,
      "loss": 0.0076,
      "num_input_tokens_seen": 31877248,
      "step": 9580
    },
    {
      "epoch": 1.9178150714053472,
      "grad_norm": 0.3187255859375,
      "learning_rate": 1.4399339930550716e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 31892928,
      "step": 9585
    },
    {
      "epoch": 1.918815496586049,
      "grad_norm": 0.45982155203819275,
      "learning_rate": 1.4375621662844485e-05,
      "loss": 0.019,
      "num_input_tokens_seen": 31908576,
      "step": 9590
    },
    {
      "epoch": 1.9198159217667508,
      "grad_norm": 1.2846043109893799,
      "learning_rate": 1.4351915060065488e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 31924160,
      "step": 9595
    },
    {
      "epoch": 1.9208163469474526,
      "grad_norm": 0.534823477268219,
      "learning_rate": 1.4328220148242127e-05,
      "loss": 0.0306,
      "num_input_tokens_seen": 31941920,
      "step": 9600
    },
    {
      "epoch": 1.9218167721281545,
      "grad_norm": 0.16988834738731384,
      "learning_rate": 1.4304536953389996e-05,
      "loss": 0.0084,
      "num_input_tokens_seen": 31959072,
      "step": 9605
    },
    {
      "epoch": 1.9228171973088561,
      "grad_norm": 0.36865609884262085,
      "learning_rate": 1.428086550151182e-05,
      "loss": 0.0119,
      "num_input_tokens_seen": 31975584,
      "step": 9610
    },
    {
      "epoch": 1.923817622489558,
      "grad_norm": 0.47268640995025635,
      "learning_rate": 1.4257205818597424e-05,
      "loss": 0.0138,
      "num_input_tokens_seen": 31991360,
      "step": 9615
    },
    {
      "epoch": 1.9248180476702599,
      "grad_norm": 0.41165751218795776,
      "learning_rate": 1.4233557930623725e-05,
      "loss": 0.0142,
      "num_input_tokens_seen": 32007648,
      "step": 9620
    },
    {
      "epoch": 1.9258184728509615,
      "grad_norm": 0.7218590378761292,
      "learning_rate": 1.4209921863554654e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 32023872,
      "step": 9625
    },
    {
      "epoch": 1.9268188980316636,
      "grad_norm": 0.5501959919929504,
      "learning_rate": 1.4186297643341207e-05,
      "loss": 0.0186,
      "num_input_tokens_seen": 32041280,
      "step": 9630
    },
    {
      "epoch": 1.9278193232123653,
      "grad_norm": 0.3440016806125641,
      "learning_rate": 1.4162685295921358e-05,
      "loss": 0.0127,
      "num_input_tokens_seen": 32057408,
      "step": 9635
    },
    {
      "epoch": 1.928819748393067,
      "grad_norm": 0.5781524181365967,
      "learning_rate": 1.4139084847220024e-05,
      "loss": 0.013,
      "num_input_tokens_seen": 32073344,
      "step": 9640
    },
    {
      "epoch": 1.929820173573769,
      "grad_norm": 1.9640519618988037,
      "learning_rate": 1.4115496323149086e-05,
      "loss": 0.0204,
      "num_input_tokens_seen": 32090464,
      "step": 9645
    },
    {
      "epoch": 1.9308205987544707,
      "grad_norm": 0.5000998973846436,
      "learning_rate": 1.409191974960733e-05,
      "loss": 0.0223,
      "num_input_tokens_seen": 32107936,
      "step": 9650
    },
    {
      "epoch": 1.9318210239351723,
      "grad_norm": 1.0520826578140259,
      "learning_rate": 1.4068355152480384e-05,
      "loss": 0.0345,
      "num_input_tokens_seen": 32125632,
      "step": 9655
    },
    {
      "epoch": 1.9328214491158744,
      "grad_norm": 0.2433328628540039,
      "learning_rate": 1.4044802557640795e-05,
      "loss": 0.0197,
      "num_input_tokens_seen": 32142144,
      "step": 9660
    },
    {
      "epoch": 1.933821874296576,
      "grad_norm": 1.2062532901763916,
      "learning_rate": 1.4021261990947867e-05,
      "loss": 0.0179,
      "num_input_tokens_seen": 32157536,
      "step": 9665
    },
    {
      "epoch": 1.9348222994772777,
      "grad_norm": 0.26744315028190613,
      "learning_rate": 1.399773347824773e-05,
      "loss": 0.0249,
      "num_input_tokens_seen": 32173632,
      "step": 9670
    },
    {
      "epoch": 1.9358227246579798,
      "grad_norm": 0.1802188903093338,
      "learning_rate": 1.3974217045373284e-05,
      "loss": 0.0169,
      "num_input_tokens_seen": 32190816,
      "step": 9675
    },
    {
      "epoch": 1.9368231498386814,
      "grad_norm": 0.3061814606189728,
      "learning_rate": 1.3950712718144133e-05,
      "loss": 0.0166,
      "num_input_tokens_seen": 32206016,
      "step": 9680
    },
    {
      "epoch": 1.9378235750193833,
      "grad_norm": 0.2795793116092682,
      "learning_rate": 1.3927220522366624e-05,
      "loss": 0.0109,
      "num_input_tokens_seen": 32223008,
      "step": 9685
    },
    {
      "epoch": 1.9388240002000852,
      "grad_norm": 0.250918984413147,
      "learning_rate": 1.3903740483833788e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 32239808,
      "step": 9690
    },
    {
      "epoch": 1.9398244253807868,
      "grad_norm": 0.6707315444946289,
      "learning_rate": 1.3880272628325264e-05,
      "loss": 0.0159,
      "num_input_tokens_seen": 32256192,
      "step": 9695
    },
    {
      "epoch": 1.9408248505614887,
      "grad_norm": 0.18448741734027863,
      "learning_rate": 1.3856816981607355e-05,
      "loss": 0.0163,
      "num_input_tokens_seen": 32272736,
      "step": 9700
    },
    {
      "epoch": 1.9418252757421905,
      "grad_norm": 0.5773769617080688,
      "learning_rate": 1.383337356943295e-05,
      "loss": 0.0121,
      "num_input_tokens_seen": 32290144,
      "step": 9705
    },
    {
      "epoch": 1.9428257009228922,
      "grad_norm": 0.10217957198619843,
      "learning_rate": 1.38099424175415e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 32306496,
      "step": 9710
    },
    {
      "epoch": 1.943826126103594,
      "grad_norm": 0.021434210240840912,
      "learning_rate": 1.3786523551659013e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 32324160,
      "step": 9715
    },
    {
      "epoch": 1.944826551284296,
      "grad_norm": 0.20603245496749878,
      "learning_rate": 1.3763116997497968e-05,
      "loss": 0.0249,
      "num_input_tokens_seen": 32340000,
      "step": 9720
    },
    {
      "epoch": 1.9458269764649976,
      "grad_norm": 0.4069787561893463,
      "learning_rate": 1.3739722780757364e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 32356416,
      "step": 9725
    },
    {
      "epoch": 1.9468274016456995,
      "grad_norm": 0.08867352455854416,
      "learning_rate": 1.3716340927122647e-05,
      "loss": 0.0144,
      "num_input_tokens_seen": 32373696,
      "step": 9730
    },
    {
      "epoch": 1.9478278268264013,
      "grad_norm": 0.9749134182929993,
      "learning_rate": 1.369297146226567e-05,
      "loss": 0.0156,
      "num_input_tokens_seen": 32389824,
      "step": 9735
    },
    {
      "epoch": 1.948828252007103,
      "grad_norm": 0.26606303453445435,
      "learning_rate": 1.36696144118447e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 32405568,
      "step": 9740
    },
    {
      "epoch": 1.9498286771878048,
      "grad_norm": 0.22685298323631287,
      "learning_rate": 1.3646269801504388e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 32421344,
      "step": 9745
    },
    {
      "epoch": 1.9508291023685067,
      "grad_norm": 0.25356295704841614,
      "learning_rate": 1.3622937656875685e-05,
      "loss": 0.0138,
      "num_input_tokens_seen": 32437504,
      "step": 9750
    },
    {
      "epoch": 1.9518295275492084,
      "grad_norm": 0.7330453991889954,
      "learning_rate": 1.359961800357589e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 32453152,
      "step": 9755
    },
    {
      "epoch": 1.9528299527299102,
      "grad_norm": 0.6261770725250244,
      "learning_rate": 1.3576310867208577e-05,
      "loss": 0.011,
      "num_input_tokens_seen": 32470912,
      "step": 9760
    },
    {
      "epoch": 1.953830377910612,
      "grad_norm": 0.3458264172077179,
      "learning_rate": 1.3553016273363578e-05,
      "loss": 0.0171,
      "num_input_tokens_seen": 32488000,
      "step": 9765
    },
    {
      "epoch": 1.9548308030913137,
      "grad_norm": 0.7723591327667236,
      "learning_rate": 1.3529734247616965e-05,
      "loss": 0.0124,
      "num_input_tokens_seen": 32503392,
      "step": 9770
    },
    {
      "epoch": 1.9558312282720156,
      "grad_norm": 0.3869456350803375,
      "learning_rate": 1.350646481553098e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 32520160,
      "step": 9775
    },
    {
      "epoch": 1.9568316534527175,
      "grad_norm": 0.3871535360813141,
      "learning_rate": 1.3483208002654063e-05,
      "loss": 0.0171,
      "num_input_tokens_seen": 32537984,
      "step": 9780
    },
    {
      "epoch": 1.9578320786334191,
      "grad_norm": 0.1062915176153183,
      "learning_rate": 1.3459963834520806e-05,
      "loss": 0.0211,
      "num_input_tokens_seen": 32555296,
      "step": 9785
    },
    {
      "epoch": 1.958832503814121,
      "grad_norm": 0.2570017874240875,
      "learning_rate": 1.3436732336651887e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 32572736,
      "step": 9790
    },
    {
      "epoch": 1.9598329289948229,
      "grad_norm": 1.2094992399215698,
      "learning_rate": 1.3413513534554096e-05,
      "loss": 0.0247,
      "num_input_tokens_seen": 32588384,
      "step": 9795
    },
    {
      "epoch": 1.9608333541755245,
      "grad_norm": 0.49232643842697144,
      "learning_rate": 1.3390307453720292e-05,
      "loss": 0.0081,
      "num_input_tokens_seen": 32604768,
      "step": 9800
    },
    {
      "epoch": 1.9618337793562264,
      "grad_norm": 0.025176681578159332,
      "learning_rate": 1.336711411962933e-05,
      "loss": 0.0121,
      "num_input_tokens_seen": 32620384,
      "step": 9805
    },
    {
      "epoch": 1.9628342045369283,
      "grad_norm": 0.6701842546463013,
      "learning_rate": 1.3343933557746108e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 32636128,
      "step": 9810
    },
    {
      "epoch": 1.96383462971763,
      "grad_norm": 0.750227153301239,
      "learning_rate": 1.3320765793521484e-05,
      "loss": 0.0348,
      "num_input_tokens_seen": 32653824,
      "step": 9815
    },
    {
      "epoch": 1.9648350548983318,
      "grad_norm": 0.0017446859274059534,
      "learning_rate": 1.3297610852392264e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 32669536,
      "step": 9820
    },
    {
      "epoch": 1.9658354800790336,
      "grad_norm": 0.49023136496543884,
      "learning_rate": 1.3274468759781195e-05,
      "loss": 0.0238,
      "num_input_tokens_seen": 32687392,
      "step": 9825
    },
    {
      "epoch": 1.9668359052597353,
      "grad_norm": 0.20972096920013428,
      "learning_rate": 1.3251339541096874e-05,
      "loss": 0.0429,
      "num_input_tokens_seen": 32705120,
      "step": 9830
    },
    {
      "epoch": 1.9678363304404372,
      "grad_norm": 0.4810045659542084,
      "learning_rate": 1.32282232217338e-05,
      "loss": 0.0316,
      "num_input_tokens_seen": 32721088,
      "step": 9835
    },
    {
      "epoch": 1.968836755621139,
      "grad_norm": 0.5527705550193787,
      "learning_rate": 1.3205119827072315e-05,
      "loss": 0.0208,
      "num_input_tokens_seen": 32737376,
      "step": 9840
    },
    {
      "epoch": 1.9698371808018407,
      "grad_norm": 1.1657907962799072,
      "learning_rate": 1.3182029382478528e-05,
      "loss": 0.0364,
      "num_input_tokens_seen": 32753312,
      "step": 9845
    },
    {
      "epoch": 1.9708376059825425,
      "grad_norm": 0.5193189382553101,
      "learning_rate": 1.3158951913304366e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 32770880,
      "step": 9850
    },
    {
      "epoch": 1.9718380311632444,
      "grad_norm": 0.48777249455451965,
      "learning_rate": 1.3135887444887513e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 32788544,
      "step": 9855
    },
    {
      "epoch": 1.972838456343946,
      "grad_norm": 0.12154678255319595,
      "learning_rate": 1.3112836002551332e-05,
      "loss": 0.0132,
      "num_input_tokens_seen": 32804384,
      "step": 9860
    },
    {
      "epoch": 1.973838881524648,
      "grad_norm": 0.1488259881734848,
      "learning_rate": 1.3089797611604949e-05,
      "loss": 0.0089,
      "num_input_tokens_seen": 32820864,
      "step": 9865
    },
    {
      "epoch": 1.9748393067053498,
      "grad_norm": 0.40845534205436707,
      "learning_rate": 1.306677229734313e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 32839552,
      "step": 9870
    },
    {
      "epoch": 1.9758397318860514,
      "grad_norm": 0.8280574679374695,
      "learning_rate": 1.3043760085046264e-05,
      "loss": 0.0265,
      "num_input_tokens_seen": 32856800,
      "step": 9875
    },
    {
      "epoch": 1.9768401570667535,
      "grad_norm": 0.20856891572475433,
      "learning_rate": 1.3020760999980385e-05,
      "loss": 0.0125,
      "num_input_tokens_seen": 32873728,
      "step": 9880
    },
    {
      "epoch": 1.9778405822474552,
      "grad_norm": 0.4518989324569702,
      "learning_rate": 1.2997775067397095e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 32888800,
      "step": 9885
    },
    {
      "epoch": 1.9788410074281568,
      "grad_norm": 0.8835201859474182,
      "learning_rate": 1.2974802312533569e-05,
      "loss": 0.0389,
      "num_input_tokens_seen": 32906144,
      "step": 9890
    },
    {
      "epoch": 1.979841432608859,
      "grad_norm": 0.5036636590957642,
      "learning_rate": 1.2951842760612509e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 32921792,
      "step": 9895
    },
    {
      "epoch": 1.9808418577895606,
      "grad_norm": 0.06584909558296204,
      "learning_rate": 1.2928896436842108e-05,
      "loss": 0.0148,
      "num_input_tokens_seen": 32938528,
      "step": 9900
    },
    {
      "epoch": 1.9818422829702622,
      "grad_norm": 0.184793159365654,
      "learning_rate": 1.2905963366416057e-05,
      "loss": 0.0373,
      "num_input_tokens_seen": 32954784,
      "step": 9905
    },
    {
      "epoch": 1.9828427081509643,
      "grad_norm": 0.11269237101078033,
      "learning_rate": 1.288304357451348e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 32971328,
      "step": 9910
    },
    {
      "epoch": 1.983843133331666,
      "grad_norm": 1.0727444887161255,
      "learning_rate": 1.2860137086298934e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 32987552,
      "step": 9915
    },
    {
      "epoch": 1.9848435585123676,
      "grad_norm": 0.5628823637962341,
      "learning_rate": 1.283724392692236e-05,
      "loss": 0.0141,
      "num_input_tokens_seen": 33004672,
      "step": 9920
    },
    {
      "epoch": 1.9858439836930697,
      "grad_norm": 0.7233517169952393,
      "learning_rate": 1.2814364121519074e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 33022400,
      "step": 9925
    },
    {
      "epoch": 1.9868444088737713,
      "grad_norm": 0.36505573987960815,
      "learning_rate": 1.279149769520971e-05,
      "loss": 0.0066,
      "num_input_tokens_seen": 33038528,
      "step": 9930
    },
    {
      "epoch": 1.987844834054473,
      "grad_norm": 0.08851584792137146,
      "learning_rate": 1.2768644673100238e-05,
      "loss": 0.0144,
      "num_input_tokens_seen": 33056864,
      "step": 9935
    },
    {
      "epoch": 1.988845259235175,
      "grad_norm": 0.4341006278991699,
      "learning_rate": 1.2745805080281886e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 33072736,
      "step": 9940
    },
    {
      "epoch": 1.9898456844158767,
      "grad_norm": 0.40328821539878845,
      "learning_rate": 1.2722978941831162e-05,
      "loss": 0.0201,
      "num_input_tokens_seen": 33089344,
      "step": 9945
    },
    {
      "epoch": 1.9908461095965786,
      "grad_norm": 0.24123868346214294,
      "learning_rate": 1.2700166282809794e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 33105408,
      "step": 9950
    },
    {
      "epoch": 1.9918465347772805,
      "grad_norm": 0.962817370891571,
      "learning_rate": 1.2677367128264695e-05,
      "loss": 0.0334,
      "num_input_tokens_seen": 33121856,
      "step": 9955
    },
    {
      "epoch": 1.9928469599579821,
      "grad_norm": 0.2703090012073517,
      "learning_rate": 1.2654581503227963e-05,
      "loss": 0.0077,
      "num_input_tokens_seen": 33140000,
      "step": 9960
    },
    {
      "epoch": 1.993847385138684,
      "grad_norm": 0.5985310077667236,
      "learning_rate": 1.2631809432716852e-05,
      "loss": 0.0256,
      "num_input_tokens_seen": 33156160,
      "step": 9965
    },
    {
      "epoch": 1.9948478103193859,
      "grad_norm": 0.005395437590777874,
      "learning_rate": 1.2609050941733713e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 33172800,
      "step": 9970
    },
    {
      "epoch": 1.9958482355000875,
      "grad_norm": 0.32826924324035645,
      "learning_rate": 1.2586306055266006e-05,
      "loss": 0.0179,
      "num_input_tokens_seen": 33189152,
      "step": 9975
    },
    {
      "epoch": 1.9968486606807894,
      "grad_norm": 0.030345836654305458,
      "learning_rate": 1.2563574798286249e-05,
      "loss": 0.0103,
      "num_input_tokens_seen": 33206208,
      "step": 9980
    },
    {
      "epoch": 1.9978490858614912,
      "grad_norm": 0.15808875858783722,
      "learning_rate": 1.2540857195751978e-05,
      "loss": 0.0246,
      "num_input_tokens_seen": 33222816,
      "step": 9985
    },
    {
      "epoch": 1.9988495110421929,
      "grad_norm": 0.5443884134292603,
      "learning_rate": 1.251815327260576e-05,
      "loss": 0.0197,
      "num_input_tokens_seen": 33239232,
      "step": 9990
    },
    {
      "epoch": 1.9998499362228948,
      "grad_norm": 0.31862539052963257,
      "learning_rate": 1.2495463053775145e-05,
      "loss": 0.0209,
      "num_input_tokens_seen": 33256928,
      "step": 9995
    },
    {
      "epoch": 2.0008503614035966,
      "grad_norm": 0.3561738133430481,
      "learning_rate": 1.2472786564172604e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 33272928,
      "step": 10000
    },
    {
      "epoch": 2.0018507865842983,
      "grad_norm": 0.16990409791469574,
      "learning_rate": 1.2450123828695576e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 33289824,
      "step": 10005
    },
    {
      "epoch": 2.002851211765,
      "grad_norm": 0.13149109482765198,
      "learning_rate": 1.2427474872226363e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 33306080,
      "step": 10010
    },
    {
      "epoch": 2.003851636945702,
      "grad_norm": 0.04130624979734421,
      "learning_rate": 1.240483971963215e-05,
      "loss": 0.0125,
      "num_input_tokens_seen": 33322688,
      "step": 10015
    },
    {
      "epoch": 2.0048520621264037,
      "grad_norm": 0.3268091380596161,
      "learning_rate": 1.2382218395765002e-05,
      "loss": 0.006,
      "num_input_tokens_seen": 33338048,
      "step": 10020
    },
    {
      "epoch": 2.0058524873071053,
      "grad_norm": 0.2938809096813202,
      "learning_rate": 1.2359610925461742e-05,
      "loss": 0.0101,
      "num_input_tokens_seen": 33353952,
      "step": 10025
    },
    {
      "epoch": 2.0068529124878074,
      "grad_norm": 0.4705606997013092,
      "learning_rate": 1.2337017333544018e-05,
      "loss": 0.0095,
      "num_input_tokens_seen": 33370880,
      "step": 10030
    },
    {
      "epoch": 2.007853337668509,
      "grad_norm": 0.09025679528713226,
      "learning_rate": 1.231443764481824e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 33387968,
      "step": 10035
    },
    {
      "epoch": 2.008853762849211,
      "grad_norm": 0.023138917982578278,
      "learning_rate": 1.2291871884075532e-05,
      "loss": 0.0073,
      "num_input_tokens_seen": 33406560,
      "step": 10040
    },
    {
      "epoch": 2.009854188029913,
      "grad_norm": 0.16036993265151978,
      "learning_rate": 1.2269320076091742e-05,
      "loss": 0.0069,
      "num_input_tokens_seen": 33423392,
      "step": 10045
    },
    {
      "epoch": 2.0108546132106144,
      "grad_norm": 0.3855239152908325,
      "learning_rate": 1.2246782245627412e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 33439616,
      "step": 10050
    },
    {
      "epoch": 2.0118550383913165,
      "grad_norm": 0.2660057544708252,
      "learning_rate": 1.22242584174277e-05,
      "loss": 0.009,
      "num_input_tokens_seen": 33455776,
      "step": 10055
    },
    {
      "epoch": 2.012855463572018,
      "grad_norm": 0.12530750036239624,
      "learning_rate": 1.220174861622243e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 33471360,
      "step": 10060
    },
    {
      "epoch": 2.01385588875272,
      "grad_norm": 0.2971198558807373,
      "learning_rate": 1.2179252866725981e-05,
      "loss": 0.007,
      "num_input_tokens_seen": 33487136,
      "step": 10065
    },
    {
      "epoch": 2.014856313933422,
      "grad_norm": 0.12205822020769119,
      "learning_rate": 1.2156771193637359e-05,
      "loss": 0.0077,
      "num_input_tokens_seen": 33502816,
      "step": 10070
    },
    {
      "epoch": 2.0158567391141236,
      "grad_norm": 0.11058386415243149,
      "learning_rate": 1.2134303621640087e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 33518560,
      "step": 10075
    },
    {
      "epoch": 2.016857164294825,
      "grad_norm": 0.11671654880046844,
      "learning_rate": 1.2111850175402183e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 33535328,
      "step": 10080
    },
    {
      "epoch": 2.0178575894755273,
      "grad_norm": 0.25610536336898804,
      "learning_rate": 1.208941087957619e-05,
      "loss": 0.0104,
      "num_input_tokens_seen": 33551520,
      "step": 10085
    },
    {
      "epoch": 2.018858014656229,
      "grad_norm": 0.21770890057086945,
      "learning_rate": 1.2066985758799113e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 33567840,
      "step": 10090
    },
    {
      "epoch": 2.0198584398369306,
      "grad_norm": 0.0675746500492096,
      "learning_rate": 1.204457483769236e-05,
      "loss": 0.0047,
      "num_input_tokens_seen": 33584576,
      "step": 10095
    },
    {
      "epoch": 2.0208588650176327,
      "grad_norm": 0.09338697791099548,
      "learning_rate": 1.2022178140861787e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 33601600,
      "step": 10100
    },
    {
      "epoch": 2.0218592901983343,
      "grad_norm": 0.1736832857131958,
      "learning_rate": 1.1999795692897623e-05,
      "loss": 0.006,
      "num_input_tokens_seen": 33618592,
      "step": 10105
    },
    {
      "epoch": 2.022859715379036,
      "grad_norm": 0.4741303324699402,
      "learning_rate": 1.1977427518374428e-05,
      "loss": 0.0073,
      "num_input_tokens_seen": 33635008,
      "step": 10110
    },
    {
      "epoch": 2.023860140559738,
      "grad_norm": 0.46921661496162415,
      "learning_rate": 1.1955073641851125e-05,
      "loss": 0.0173,
      "num_input_tokens_seen": 33652224,
      "step": 10115
    },
    {
      "epoch": 2.0248605657404397,
      "grad_norm": 0.16761521995067596,
      "learning_rate": 1.1932734087870917e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 33667904,
      "step": 10120
    },
    {
      "epoch": 2.0258609909211414,
      "grad_norm": 0.506805956363678,
      "learning_rate": 1.1910408880961291e-05,
      "loss": 0.0086,
      "num_input_tokens_seen": 33685376,
      "step": 10125
    },
    {
      "epoch": 2.0268614161018434,
      "grad_norm": 0.5889264941215515,
      "learning_rate": 1.1888098045633988e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 33700928,
      "step": 10130
    },
    {
      "epoch": 2.027861841282545,
      "grad_norm": 0.128665953874588,
      "learning_rate": 1.1865801606384944e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 33716384,
      "step": 10135
    },
    {
      "epoch": 2.0288622664632467,
      "grad_norm": 0.0878659337759018,
      "learning_rate": 1.184351958769431e-05,
      "loss": 0.0128,
      "num_input_tokens_seen": 33734272,
      "step": 10140
    },
    {
      "epoch": 2.029862691643949,
      "grad_norm": 0.2913253605365753,
      "learning_rate": 1.1821252014026413e-05,
      "loss": 0.0082,
      "num_input_tokens_seen": 33750752,
      "step": 10145
    },
    {
      "epoch": 2.0308631168246505,
      "grad_norm": 0.05499362573027611,
      "learning_rate": 1.179899890982969e-05,
      "loss": 0.007,
      "num_input_tokens_seen": 33766784,
      "step": 10150
    },
    {
      "epoch": 2.031863542005352,
      "grad_norm": 0.35297057032585144,
      "learning_rate": 1.1776760299536713e-05,
      "loss": 0.006,
      "num_input_tokens_seen": 33784544,
      "step": 10155
    },
    {
      "epoch": 2.032863967186054,
      "grad_norm": 0.06490829586982727,
      "learning_rate": 1.1754536207564149e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 33800960,
      "step": 10160
    },
    {
      "epoch": 2.033864392366756,
      "grad_norm": 0.5872918367385864,
      "learning_rate": 1.1732326658312693e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 33817120,
      "step": 10165
    },
    {
      "epoch": 2.0348648175474575,
      "grad_norm": 0.03993507847189903,
      "learning_rate": 1.17101316761671e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 33833376,
      "step": 10170
    },
    {
      "epoch": 2.0358652427281596,
      "grad_norm": 0.09653317928314209,
      "learning_rate": 1.1687951285496124e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 33850272,
      "step": 10175
    },
    {
      "epoch": 2.0368656679088613,
      "grad_norm": 0.30753350257873535,
      "learning_rate": 1.1665785510652494e-05,
      "loss": 0.0075,
      "num_input_tokens_seen": 33867072,
      "step": 10180
    },
    {
      "epoch": 2.037866093089563,
      "grad_norm": 0.9518786668777466,
      "learning_rate": 1.1643634375972908e-05,
      "loss": 0.009,
      "num_input_tokens_seen": 33883232,
      "step": 10185
    },
    {
      "epoch": 2.038866518270265,
      "grad_norm": 0.5314575433731079,
      "learning_rate": 1.1621497905777962e-05,
      "loss": 0.0088,
      "num_input_tokens_seen": 33899616,
      "step": 10190
    },
    {
      "epoch": 2.0398669434509666,
      "grad_norm": 0.4062546491622925,
      "learning_rate": 1.1599376124372166e-05,
      "loss": 0.0072,
      "num_input_tokens_seen": 33915136,
      "step": 10195
    },
    {
      "epoch": 2.0408673686316683,
      "grad_norm": 0.11941560357809067,
      "learning_rate": 1.1577269056043922e-05,
      "loss": 0.006,
      "num_input_tokens_seen": 33930656,
      "step": 10200
    },
    {
      "epoch": 2.0418677938123704,
      "grad_norm": 0.353001207113266,
      "learning_rate": 1.1555176725065434e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 33948032,
      "step": 10205
    },
    {
      "epoch": 2.042868218993072,
      "grad_norm": 0.20794694125652313,
      "learning_rate": 1.1533099155692762e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 33966816,
      "step": 10210
    },
    {
      "epoch": 2.0438686441737737,
      "grad_norm": 0.7453445792198181,
      "learning_rate": 1.1511036372165754e-05,
      "loss": 0.0133,
      "num_input_tokens_seen": 33982560,
      "step": 10215
    },
    {
      "epoch": 2.0448690693544758,
      "grad_norm": 0.7288349270820618,
      "learning_rate": 1.1488988398707992e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 33999104,
      "step": 10220
    },
    {
      "epoch": 2.0458694945351774,
      "grad_norm": 0.3507794439792633,
      "learning_rate": 1.146695525952686e-05,
      "loss": 0.0109,
      "num_input_tokens_seen": 34015456,
      "step": 10225
    },
    {
      "epoch": 2.046869919715879,
      "grad_norm": 0.008352497592568398,
      "learning_rate": 1.1444936978813386e-05,
      "loss": 0.0114,
      "num_input_tokens_seen": 34032032,
      "step": 10230
    },
    {
      "epoch": 2.047870344896581,
      "grad_norm": 0.28601670265197754,
      "learning_rate": 1.1422933580742327e-05,
      "loss": 0.0056,
      "num_input_tokens_seen": 34048544,
      "step": 10235
    },
    {
      "epoch": 2.048870770077283,
      "grad_norm": 0.3275420367717743,
      "learning_rate": 1.1400945089472095e-05,
      "loss": 0.005,
      "num_input_tokens_seen": 34065024,
      "step": 10240
    },
    {
      "epoch": 2.0498711952579844,
      "grad_norm": 0.2914477586746216,
      "learning_rate": 1.1378971529144714e-05,
      "loss": 0.0087,
      "num_input_tokens_seen": 34080320,
      "step": 10245
    },
    {
      "epoch": 2.0508716204386865,
      "grad_norm": 0.12133269011974335,
      "learning_rate": 1.1357012923885835e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 34096352,
      "step": 10250
    },
    {
      "epoch": 2.051872045619388,
      "grad_norm": 0.3188471496105194,
      "learning_rate": 1.1335069297804693e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 34113696,
      "step": 10255
    },
    {
      "epoch": 2.05287247080009,
      "grad_norm": 0.023096419870853424,
      "learning_rate": 1.1313140674994052e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 34130304,
      "step": 10260
    },
    {
      "epoch": 2.053872895980792,
      "grad_norm": 0.6211907267570496,
      "learning_rate": 1.1291227079530222e-05,
      "loss": 0.0073,
      "num_input_tokens_seen": 34147840,
      "step": 10265
    },
    {
      "epoch": 2.0548733211614936,
      "grad_norm": 0.0840262845158577,
      "learning_rate": 1.1269328535473025e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 34163680,
      "step": 10270
    },
    {
      "epoch": 2.055873746342195,
      "grad_norm": 0.15804554522037506,
      "learning_rate": 1.1247445066865714e-05,
      "loss": 0.0145,
      "num_input_tokens_seen": 34180288,
      "step": 10275
    },
    {
      "epoch": 2.0568741715228973,
      "grad_norm": 0.09390754997730255,
      "learning_rate": 1.1225576697735055e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 34197440,
      "step": 10280
    },
    {
      "epoch": 2.057874596703599,
      "grad_norm": 1.0508612394332886,
      "learning_rate": 1.120372345209117e-05,
      "loss": 0.0104,
      "num_input_tokens_seen": 34213504,
      "step": 10285
    },
    {
      "epoch": 2.058875021884301,
      "grad_norm": 0.0515342615544796,
      "learning_rate": 1.1181885353927616e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 34230784,
      "step": 10290
    },
    {
      "epoch": 2.0598754470650027,
      "grad_norm": 1.2798871994018555,
      "learning_rate": 1.116006242722132e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 34246464,
      "step": 10295
    },
    {
      "epoch": 2.0608758722457043,
      "grad_norm": 0.10574700683355331,
      "learning_rate": 1.1138254695932524e-05,
      "loss": 0.0123,
      "num_input_tokens_seen": 34263424,
      "step": 10300
    },
    {
      "epoch": 2.0618762974264064,
      "grad_norm": 0.6526563763618469,
      "learning_rate": 1.1116462184004806e-05,
      "loss": 0.0127,
      "num_input_tokens_seen": 34279488,
      "step": 10305
    },
    {
      "epoch": 2.062876722607108,
      "grad_norm": 0.6106184720993042,
      "learning_rate": 1.1094684915365045e-05,
      "loss": 0.0089,
      "num_input_tokens_seen": 34295456,
      "step": 10310
    },
    {
      "epoch": 2.0638771477878097,
      "grad_norm": 0.10786963254213333,
      "learning_rate": 1.1072922913923348e-05,
      "loss": 0.0047,
      "num_input_tokens_seen": 34311904,
      "step": 10315
    },
    {
      "epoch": 2.064877572968512,
      "grad_norm": 0.5320205688476562,
      "learning_rate": 1.1051176203573094e-05,
      "loss": 0.0079,
      "num_input_tokens_seen": 34328544,
      "step": 10320
    },
    {
      "epoch": 2.0658779981492135,
      "grad_norm": 0.08332104980945587,
      "learning_rate": 1.1029444808190859e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 34346880,
      "step": 10325
    },
    {
      "epoch": 2.066878423329915,
      "grad_norm": 0.5802815556526184,
      "learning_rate": 1.1007728751636403e-05,
      "loss": 0.0077,
      "num_input_tokens_seen": 34363712,
      "step": 10330
    },
    {
      "epoch": 2.067878848510617,
      "grad_norm": 1.655693769454956,
      "learning_rate": 1.0986028057752662e-05,
      "loss": 0.0099,
      "num_input_tokens_seen": 34379232,
      "step": 10335
    },
    {
      "epoch": 2.068879273691319,
      "grad_norm": 0.04671124368906021,
      "learning_rate": 1.0964342750365667e-05,
      "loss": 0.0071,
      "num_input_tokens_seen": 34395936,
      "step": 10340
    },
    {
      "epoch": 2.0698796988720205,
      "grad_norm": 0.47704026103019714,
      "learning_rate": 1.0942672853284592e-05,
      "loss": 0.0097,
      "num_input_tokens_seen": 34412032,
      "step": 10345
    },
    {
      "epoch": 2.0708801240527226,
      "grad_norm": 0.32408711314201355,
      "learning_rate": 1.0921018390301683e-05,
      "loss": 0.0096,
      "num_input_tokens_seen": 34428352,
      "step": 10350
    },
    {
      "epoch": 2.0718805492334242,
      "grad_norm": 0.0068885162472724915,
      "learning_rate": 1.0899379385192221e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 34445344,
      "step": 10355
    },
    {
      "epoch": 2.072880974414126,
      "grad_norm": 1.3240070343017578,
      "learning_rate": 1.0877755861714539e-05,
      "loss": 0.0122,
      "num_input_tokens_seen": 34463680,
      "step": 10360
    },
    {
      "epoch": 2.073881399594828,
      "grad_norm": 0.024377919733524323,
      "learning_rate": 1.0856147843609968e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 34481088,
      "step": 10365
    },
    {
      "epoch": 2.0748818247755296,
      "grad_norm": 0.23395225405693054,
      "learning_rate": 1.0834555354602796e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 34497440,
      "step": 10370
    },
    {
      "epoch": 2.0758822499562313,
      "grad_norm": 0.13731922209262848,
      "learning_rate": 1.0812978418400282e-05,
      "loss": 0.0113,
      "num_input_tokens_seen": 34514400,
      "step": 10375
    },
    {
      "epoch": 2.0768826751369334,
      "grad_norm": 0.10927022993564606,
      "learning_rate": 1.0791417058692601e-05,
      "loss": 0.0091,
      "num_input_tokens_seen": 34531680,
      "step": 10380
    },
    {
      "epoch": 2.077883100317635,
      "grad_norm": 0.002200738526880741,
      "learning_rate": 1.0769871299152829e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 34549376,
      "step": 10385
    },
    {
      "epoch": 2.0788835254983367,
      "grad_norm": 0.07721970230340958,
      "learning_rate": 1.074834116343692e-05,
      "loss": 0.0146,
      "num_input_tokens_seen": 34565440,
      "step": 10390
    },
    {
      "epoch": 2.0798839506790388,
      "grad_norm": 0.17168541252613068,
      "learning_rate": 1.0726826675183647e-05,
      "loss": 0.0082,
      "num_input_tokens_seen": 34582560,
      "step": 10395
    },
    {
      "epoch": 2.0808843758597404,
      "grad_norm": 0.6241939663887024,
      "learning_rate": 1.0705327858014627e-05,
      "loss": 0.0101,
      "num_input_tokens_seen": 34599456,
      "step": 10400
    },
    {
      "epoch": 2.081884801040442,
      "grad_norm": 0.056994225829839706,
      "learning_rate": 1.0683844735534282e-05,
      "loss": 0.0102,
      "num_input_tokens_seen": 34616160,
      "step": 10405
    },
    {
      "epoch": 2.082885226221144,
      "grad_norm": 0.20421095192432404,
      "learning_rate": 1.0662377331329765e-05,
      "loss": 0.0081,
      "num_input_tokens_seen": 34632352,
      "step": 10410
    },
    {
      "epoch": 2.083885651401846,
      "grad_norm": 0.42918482422828674,
      "learning_rate": 1.0640925668971e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 34648800,
      "step": 10415
    },
    {
      "epoch": 2.0848860765825474,
      "grad_norm": 0.022502677515149117,
      "learning_rate": 1.0619489772010626e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 34664864,
      "step": 10420
    },
    {
      "epoch": 2.0858865017632495,
      "grad_norm": 0.12481766939163208,
      "learning_rate": 1.0598069663983948e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 34680736,
      "step": 10425
    },
    {
      "epoch": 2.086886926943951,
      "grad_norm": 0.04550350829958916,
      "learning_rate": 1.057666536840898e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 34697664,
      "step": 10430
    },
    {
      "epoch": 2.087887352124653,
      "grad_norm": 0.00910136103630066,
      "learning_rate": 1.055527690878633e-05,
      "loss": 0.0013,
      "num_input_tokens_seen": 34713952,
      "step": 10435
    },
    {
      "epoch": 2.088887777305355,
      "grad_norm": 0.4201542139053345,
      "learning_rate": 1.0533904308599241e-05,
      "loss": 0.0072,
      "num_input_tokens_seen": 34730752,
      "step": 10440
    },
    {
      "epoch": 2.0898882024860566,
      "grad_norm": 0.9132028222084045,
      "learning_rate": 1.0512547591313555e-05,
      "loss": 0.0389,
      "num_input_tokens_seen": 34748096,
      "step": 10445
    },
    {
      "epoch": 2.090888627666758,
      "grad_norm": 0.2836069166660309,
      "learning_rate": 1.0491206780377636e-05,
      "loss": 0.0161,
      "num_input_tokens_seen": 34765600,
      "step": 10450
    },
    {
      "epoch": 2.0918890528474603,
      "grad_norm": 0.186019167304039,
      "learning_rate": 1.0469881899222419e-05,
      "loss": 0.0088,
      "num_input_tokens_seen": 34782176,
      "step": 10455
    },
    {
      "epoch": 2.092889478028162,
      "grad_norm": 0.476346492767334,
      "learning_rate": 1.0448572971261353e-05,
      "loss": 0.0057,
      "num_input_tokens_seen": 34798144,
      "step": 10460
    },
    {
      "epoch": 2.0938899032088636,
      "grad_norm": 0.48843252658843994,
      "learning_rate": 1.0427280019890332e-05,
      "loss": 0.0093,
      "num_input_tokens_seen": 34814592,
      "step": 10465
    },
    {
      "epoch": 2.0948903283895657,
      "grad_norm": 0.641743540763855,
      "learning_rate": 1.0406003068487747e-05,
      "loss": 0.0038,
      "num_input_tokens_seen": 34832416,
      "step": 10470
    },
    {
      "epoch": 2.0958907535702673,
      "grad_norm": 0.03875632956624031,
      "learning_rate": 1.0384742140414416e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 34848960,
      "step": 10475
    },
    {
      "epoch": 2.096891178750969,
      "grad_norm": 0.8721215128898621,
      "learning_rate": 1.0363497259013529e-05,
      "loss": 0.0091,
      "num_input_tokens_seen": 34864960,
      "step": 10480
    },
    {
      "epoch": 2.097891603931671,
      "grad_norm": 0.8210321664810181,
      "learning_rate": 1.0342268447610722e-05,
      "loss": 0.0087,
      "num_input_tokens_seen": 34883264,
      "step": 10485
    },
    {
      "epoch": 2.0988920291123727,
      "grad_norm": 0.5005853772163391,
      "learning_rate": 1.0321055729513928e-05,
      "loss": 0.0161,
      "num_input_tokens_seen": 34899488,
      "step": 10490
    },
    {
      "epoch": 2.0998924542930744,
      "grad_norm": 0.21855968236923218,
      "learning_rate": 1.0299859128013437e-05,
      "loss": 0.0051,
      "num_input_tokens_seen": 34917664,
      "step": 10495
    },
    {
      "epoch": 2.1008928794737765,
      "grad_norm": 0.03365173190832138,
      "learning_rate": 1.027867866638185e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 34933024,
      "step": 10500
    },
    {
      "epoch": 2.101893304654478,
      "grad_norm": 0.046729862689971924,
      "learning_rate": 1.0257514367874022e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 34950080,
      "step": 10505
    },
    {
      "epoch": 2.1028937298351797,
      "grad_norm": 0.2730543315410614,
      "learning_rate": 1.0236366255727085e-05,
      "loss": 0.0212,
      "num_input_tokens_seen": 34966208,
      "step": 10510
    },
    {
      "epoch": 2.103894155015882,
      "grad_norm": 0.02403431199491024,
      "learning_rate": 1.0215234353160399e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 34983360,
      "step": 10515
    },
    {
      "epoch": 2.1048945801965835,
      "grad_norm": 0.06615236401557922,
      "learning_rate": 1.0194118683375503e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 35000160,
      "step": 10520
    },
    {
      "epoch": 2.105895005377285,
      "grad_norm": 0.13943585753440857,
      "learning_rate": 1.017301926955614e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 35016288,
      "step": 10525
    },
    {
      "epoch": 2.1068954305579872,
      "grad_norm": 0.6373512148857117,
      "learning_rate": 1.0151936134868198e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 35032544,
      "step": 10530
    },
    {
      "epoch": 2.107895855738689,
      "grad_norm": 0.19955801963806152,
      "learning_rate": 1.0130869302459684e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 35048064,
      "step": 10535
    },
    {
      "epoch": 2.1088962809193905,
      "grad_norm": 0.06729407608509064,
      "learning_rate": 1.0109818795460718e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 35064032,
      "step": 10540
    },
    {
      "epoch": 2.1098967061000926,
      "grad_norm": 0.6858115196228027,
      "learning_rate": 1.0088784636983473e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 35080352,
      "step": 10545
    },
    {
      "epoch": 2.1108971312807943,
      "grad_norm": 0.11107315868139267,
      "learning_rate": 1.0067766850122199e-05,
      "loss": 0.0094,
      "num_input_tokens_seen": 35096480,
      "step": 10550
    },
    {
      "epoch": 2.1118975564614964,
      "grad_norm": 0.18054699897766113,
      "learning_rate": 1.0046765457953163e-05,
      "loss": 0.0058,
      "num_input_tokens_seen": 35113248,
      "step": 10555
    },
    {
      "epoch": 2.112897981642198,
      "grad_norm": 0.3239651918411255,
      "learning_rate": 1.0025780483534616e-05,
      "loss": 0.0112,
      "num_input_tokens_seen": 35129728,
      "step": 10560
    },
    {
      "epoch": 2.1138984068228996,
      "grad_norm": 0.35435789823532104,
      "learning_rate": 1.00048119499068e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 35146464,
      "step": 10565
    },
    {
      "epoch": 2.1148988320036017,
      "grad_norm": 0.1789417862892151,
      "learning_rate": 9.983859880091912e-06,
      "loss": 0.0175,
      "num_input_tokens_seen": 35163872,
      "step": 10570
    },
    {
      "epoch": 2.1158992571843034,
      "grad_norm": 0.28334328532218933,
      "learning_rate": 9.962924297094043e-06,
      "loss": 0.0066,
      "num_input_tokens_seen": 35180512,
      "step": 10575
    },
    {
      "epoch": 2.116899682365005,
      "grad_norm": 0.7784582376480103,
      "learning_rate": 9.942005223899212e-06,
      "loss": 0.0052,
      "num_input_tokens_seen": 35196832,
      "step": 10580
    },
    {
      "epoch": 2.117900107545707,
      "grad_norm": 0.05406097322702408,
      "learning_rate": 9.921102683475303e-06,
      "loss": 0.0034,
      "num_input_tokens_seen": 35213792,
      "step": 10585
    },
    {
      "epoch": 2.1189005327264088,
      "grad_norm": 0.17640522122383118,
      "learning_rate": 9.900216698772042e-06,
      "loss": 0.004,
      "num_input_tokens_seen": 35231008,
      "step": 10590
    },
    {
      "epoch": 2.1199009579071104,
      "grad_norm": 0.11796704679727554,
      "learning_rate": 9.879347292720989e-06,
      "loss": 0.004,
      "num_input_tokens_seen": 35247968,
      "step": 10595
    },
    {
      "epoch": 2.1209013830878125,
      "grad_norm": 0.06427179276943207,
      "learning_rate": 9.858494488235484e-06,
      "loss": 0.0194,
      "num_input_tokens_seen": 35264128,
      "step": 10600
    },
    {
      "epoch": 2.121901808268514,
      "grad_norm": 0.7149617075920105,
      "learning_rate": 9.837658308210652e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 35282304,
      "step": 10605
    },
    {
      "epoch": 2.122902233449216,
      "grad_norm": 0.14961405098438263,
      "learning_rate": 9.816838775523374e-06,
      "loss": 0.0048,
      "num_input_tokens_seen": 35297792,
      "step": 10610
    },
    {
      "epoch": 2.123902658629918,
      "grad_norm": 0.10325565189123154,
      "learning_rate": 9.796035913032226e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 35313792,
      "step": 10615
    },
    {
      "epoch": 2.1249030838106195,
      "grad_norm": 0.04255054146051407,
      "learning_rate": 9.775249743577505e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 35329856,
      "step": 10620
    },
    {
      "epoch": 2.125903508991321,
      "grad_norm": 0.03326898068189621,
      "learning_rate": 9.75448028998118e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 35347520,
      "step": 10625
    },
    {
      "epoch": 2.1269039341720233,
      "grad_norm": 0.05380835011601448,
      "learning_rate": 9.733727575046845e-06,
      "loss": 0.0115,
      "num_input_tokens_seen": 35363648,
      "step": 10630
    },
    {
      "epoch": 2.127904359352725,
      "grad_norm": 0.23670682311058044,
      "learning_rate": 9.712991621559738e-06,
      "loss": 0.0082,
      "num_input_tokens_seen": 35380896,
      "step": 10635
    },
    {
      "epoch": 2.1289047845334266,
      "grad_norm": 0.13752107322216034,
      "learning_rate": 9.692272452286685e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 35397120,
      "step": 10640
    },
    {
      "epoch": 2.1299052097141287,
      "grad_norm": 1.3863645792007446,
      "learning_rate": 9.67157008997609e-06,
      "loss": 0.0201,
      "num_input_tokens_seen": 35414112,
      "step": 10645
    },
    {
      "epoch": 2.1309056348948303,
      "grad_norm": 1.1318001747131348,
      "learning_rate": 9.6508845573579e-06,
      "loss": 0.0149,
      "num_input_tokens_seen": 35430464,
      "step": 10650
    },
    {
      "epoch": 2.131906060075532,
      "grad_norm": 0.30626359581947327,
      "learning_rate": 9.630215877143575e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 35447584,
      "step": 10655
    },
    {
      "epoch": 2.132906485256234,
      "grad_norm": 0.2643603980541229,
      "learning_rate": 9.609564072026083e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 35464064,
      "step": 10660
    },
    {
      "epoch": 2.1339069104369357,
      "grad_norm": 0.03090248629450798,
      "learning_rate": 9.588929164679871e-06,
      "loss": 0.0063,
      "num_input_tokens_seen": 35480704,
      "step": 10665
    },
    {
      "epoch": 2.1349073356176373,
      "grad_norm": 0.8625773787498474,
      "learning_rate": 9.56831117776081e-06,
      "loss": 0.0047,
      "num_input_tokens_seen": 35499136,
      "step": 10670
    },
    {
      "epoch": 2.1359077607983394,
      "grad_norm": 0.42030104994773865,
      "learning_rate": 9.547710133906213e-06,
      "loss": 0.0171,
      "num_input_tokens_seen": 35515616,
      "step": 10675
    },
    {
      "epoch": 2.136908185979041,
      "grad_norm": 0.5008081793785095,
      "learning_rate": 9.527126055734795e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 35534688,
      "step": 10680
    },
    {
      "epoch": 2.1379086111597427,
      "grad_norm": 0.25674349069595337,
      "learning_rate": 9.506558965846604e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 35551904,
      "step": 10685
    },
    {
      "epoch": 2.138909036340445,
      "grad_norm": 0.038256000727415085,
      "learning_rate": 9.486008886823095e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 35568992,
      "step": 10690
    },
    {
      "epoch": 2.1399094615211465,
      "grad_norm": 0.1449941247701645,
      "learning_rate": 9.465475841227017e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 35585440,
      "step": 10695
    },
    {
      "epoch": 2.140909886701848,
      "grad_norm": 0.37136611342430115,
      "learning_rate": 9.444959851602397e-06,
      "loss": 0.0047,
      "num_input_tokens_seen": 35601888,
      "step": 10700
    },
    {
      "epoch": 2.14191031188255,
      "grad_norm": 0.022761758416891098,
      "learning_rate": 9.424460940474573e-06,
      "loss": 0.0232,
      "num_input_tokens_seen": 35618400,
      "step": 10705
    },
    {
      "epoch": 2.142910737063252,
      "grad_norm": 1.0737857818603516,
      "learning_rate": 9.4039791303501e-06,
      "loss": 0.0138,
      "num_input_tokens_seen": 35635872,
      "step": 10710
    },
    {
      "epoch": 2.1439111622439535,
      "grad_norm": 0.10832738131284714,
      "learning_rate": 9.383514443716774e-06,
      "loss": 0.004,
      "num_input_tokens_seen": 35652352,
      "step": 10715
    },
    {
      "epoch": 2.1449115874246556,
      "grad_norm": 0.0014874746557325125,
      "learning_rate": 9.363066903043601e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 35669120,
      "step": 10720
    },
    {
      "epoch": 2.1459120126053572,
      "grad_norm": 0.30137619376182556,
      "learning_rate": 9.342636530780733e-06,
      "loss": 0.0075,
      "num_input_tokens_seen": 35685728,
      "step": 10725
    },
    {
      "epoch": 2.146912437786059,
      "grad_norm": 0.03261270001530647,
      "learning_rate": 9.322223349359491e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 35703392,
      "step": 10730
    },
    {
      "epoch": 2.147912862966761,
      "grad_norm": 0.08350440859794617,
      "learning_rate": 9.301827381192321e-06,
      "loss": 0.004,
      "num_input_tokens_seen": 35720480,
      "step": 10735
    },
    {
      "epoch": 2.1489132881474626,
      "grad_norm": 0.08948685228824615,
      "learning_rate": 9.28144864867277e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 35737728,
      "step": 10740
    },
    {
      "epoch": 2.1499137133281643,
      "grad_norm": 0.060130346566438675,
      "learning_rate": 9.26108717417545e-06,
      "loss": 0.0086,
      "num_input_tokens_seen": 35753664,
      "step": 10745
    },
    {
      "epoch": 2.1509141385088664,
      "grad_norm": 1.0318474769592285,
      "learning_rate": 9.240742980056049e-06,
      "loss": 0.0042,
      "num_input_tokens_seen": 35770688,
      "step": 10750
    },
    {
      "epoch": 2.151914563689568,
      "grad_norm": 0.5060439109802246,
      "learning_rate": 9.220416088651249e-06,
      "loss": 0.0085,
      "num_input_tokens_seen": 35787712,
      "step": 10755
    },
    {
      "epoch": 2.1529149888702697,
      "grad_norm": 0.07923345267772675,
      "learning_rate": 9.200106522278765e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 35804192,
      "step": 10760
    },
    {
      "epoch": 2.1539154140509718,
      "grad_norm": 0.09731190651655197,
      "learning_rate": 9.179814303237263e-06,
      "loss": 0.0073,
      "num_input_tokens_seen": 35820864,
      "step": 10765
    },
    {
      "epoch": 2.1549158392316734,
      "grad_norm": 0.12063892930746078,
      "learning_rate": 9.159539453806385e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 35836896,
      "step": 10770
    },
    {
      "epoch": 2.155916264412375,
      "grad_norm": 0.8915739059448242,
      "learning_rate": 9.139281996246699e-06,
      "loss": 0.007,
      "num_input_tokens_seen": 35853728,
      "step": 10775
    },
    {
      "epoch": 2.156916689593077,
      "grad_norm": 0.9608829021453857,
      "learning_rate": 9.119041952799662e-06,
      "loss": 0.0077,
      "num_input_tokens_seen": 35870848,
      "step": 10780
    },
    {
      "epoch": 2.157917114773779,
      "grad_norm": 0.0003298874362371862,
      "learning_rate": 9.098819345687622e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 35888128,
      "step": 10785
    },
    {
      "epoch": 2.158917539954481,
      "grad_norm": 0.026527972891926765,
      "learning_rate": 9.078614197113794e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 35904992,
      "step": 10790
    },
    {
      "epoch": 2.1599179651351825,
      "grad_norm": 0.1923072785139084,
      "learning_rate": 9.058426529262207e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 35920896,
      "step": 10795
    },
    {
      "epoch": 2.160918390315884,
      "grad_norm": 0.00016449119721073657,
      "learning_rate": 9.0382563642977e-06,
      "loss": 0.0009,
      "num_input_tokens_seen": 35937760,
      "step": 10800
    },
    {
      "epoch": 2.161918815496586,
      "grad_norm": 0.6942543983459473,
      "learning_rate": 9.018103724365918e-06,
      "loss": 0.0097,
      "num_input_tokens_seen": 35954752,
      "step": 10805
    },
    {
      "epoch": 2.162919240677288,
      "grad_norm": 0.0013587274588644505,
      "learning_rate": 8.997968631593217e-06,
      "loss": 0.0069,
      "num_input_tokens_seen": 35970432,
      "step": 10810
    },
    {
      "epoch": 2.1639196658579896,
      "grad_norm": 0.7722641229629517,
      "learning_rate": 8.977851108086735e-06,
      "loss": 0.0105,
      "num_input_tokens_seen": 35986496,
      "step": 10815
    },
    {
      "epoch": 2.1649200910386917,
      "grad_norm": 0.07212720066308975,
      "learning_rate": 8.957751175934302e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 36003072,
      "step": 10820
    },
    {
      "epoch": 2.1659205162193933,
      "grad_norm": 0.03139474615454674,
      "learning_rate": 8.937668857204412e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 36019680,
      "step": 10825
    },
    {
      "epoch": 2.166920941400095,
      "grad_norm": 0.331551730632782,
      "learning_rate": 8.917604173946268e-06,
      "loss": 0.0037,
      "num_input_tokens_seen": 36036832,
      "step": 10830
    },
    {
      "epoch": 2.167921366580797,
      "grad_norm": 0.05217951908707619,
      "learning_rate": 8.89755714818966e-06,
      "loss": 0.0067,
      "num_input_tokens_seen": 36053184,
      "step": 10835
    },
    {
      "epoch": 2.1689217917614987,
      "grad_norm": 0.08116187900304794,
      "learning_rate": 8.877527801945012e-06,
      "loss": 0.0034,
      "num_input_tokens_seen": 36070688,
      "step": 10840
    },
    {
      "epoch": 2.1699222169422003,
      "grad_norm": 0.14644820988178253,
      "learning_rate": 8.857516157203372e-06,
      "loss": 0.0045,
      "num_input_tokens_seen": 36086432,
      "step": 10845
    },
    {
      "epoch": 2.1709226421229024,
      "grad_norm": 1.253024697303772,
      "learning_rate": 8.837522235936287e-06,
      "loss": 0.011,
      "num_input_tokens_seen": 36102976,
      "step": 10850
    },
    {
      "epoch": 2.171923067303604,
      "grad_norm": 0.13459981977939606,
      "learning_rate": 8.81754606009589e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 36119616,
      "step": 10855
    },
    {
      "epoch": 2.1729234924843057,
      "grad_norm": 0.00015339571109507233,
      "learning_rate": 8.79758765161483e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 36136064,
      "step": 10860
    },
    {
      "epoch": 2.173923917665008,
      "grad_norm": 0.02076278254389763,
      "learning_rate": 8.777647032406214e-06,
      "loss": 0.0059,
      "num_input_tokens_seen": 36152576,
      "step": 10865
    },
    {
      "epoch": 2.1749243428457095,
      "grad_norm": 0.4944254159927368,
      "learning_rate": 8.757724224363654e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 36168320,
      "step": 10870
    },
    {
      "epoch": 2.175924768026411,
      "grad_norm": 0.014887170866131783,
      "learning_rate": 8.737819249361195e-06,
      "loss": 0.0141,
      "num_input_tokens_seen": 36185792,
      "step": 10875
    },
    {
      "epoch": 2.176925193207113,
      "grad_norm": 0.007804557681083679,
      "learning_rate": 8.71793212925329e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 36202048,
      "step": 10880
    },
    {
      "epoch": 2.177925618387815,
      "grad_norm": 0.16051238775253296,
      "learning_rate": 8.69806288587481e-06,
      "loss": 0.0034,
      "num_input_tokens_seen": 36219168,
      "step": 10885
    },
    {
      "epoch": 2.1789260435685165,
      "grad_norm": 0.12735103070735931,
      "learning_rate": 8.678211541040968e-06,
      "loss": 0.0035,
      "num_input_tokens_seen": 36236288,
      "step": 10890
    },
    {
      "epoch": 2.1799264687492186,
      "grad_norm": 0.007964778691530228,
      "learning_rate": 8.658378116547361e-06,
      "loss": 0.0003,
      "num_input_tokens_seen": 36252992,
      "step": 10895
    },
    {
      "epoch": 2.1809268939299202,
      "grad_norm": 0.00014435694902203977,
      "learning_rate": 8.638562634169903e-06,
      "loss": 0.0003,
      "num_input_tokens_seen": 36269440,
      "step": 10900
    },
    {
      "epoch": 2.181927319110622,
      "grad_norm": 0.11460757255554199,
      "learning_rate": 8.618765115664775e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 36285568,
      "step": 10905
    },
    {
      "epoch": 2.182927744291324,
      "grad_norm": 0.5251172780990601,
      "learning_rate": 8.598985582768474e-06,
      "loss": 0.0147,
      "num_input_tokens_seen": 36302560,
      "step": 10910
    },
    {
      "epoch": 2.1839281694720256,
      "grad_norm": 0.01360283326357603,
      "learning_rate": 8.579224057197738e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 36319616,
      "step": 10915
    },
    {
      "epoch": 2.1849285946527273,
      "grad_norm": 2.9669125080108643,
      "learning_rate": 8.559480560649519e-06,
      "loss": 0.0098,
      "num_input_tokens_seen": 36335808,
      "step": 10920
    },
    {
      "epoch": 2.1859290198334294,
      "grad_norm": 0.08378507941961288,
      "learning_rate": 8.539755114800995e-06,
      "loss": 0.0035,
      "num_input_tokens_seen": 36351808,
      "step": 10925
    },
    {
      "epoch": 2.186929445014131,
      "grad_norm": 0.003294085618108511,
      "learning_rate": 8.520047741309522e-06,
      "loss": 0.0007,
      "num_input_tokens_seen": 36368192,
      "step": 10930
    },
    {
      "epoch": 2.1879298701948326,
      "grad_norm": 1.2479307651519775,
      "learning_rate": 8.500358461812594e-06,
      "loss": 0.0115,
      "num_input_tokens_seen": 36384096,
      "step": 10935
    },
    {
      "epoch": 2.1889302953755347,
      "grad_norm": 0.0866975262761116,
      "learning_rate": 8.480687297927862e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 36400512,
      "step": 10940
    },
    {
      "epoch": 2.1899307205562364,
      "grad_norm": 0.13979388773441315,
      "learning_rate": 8.461034271253076e-06,
      "loss": 0.0015,
      "num_input_tokens_seen": 36416608,
      "step": 10945
    },
    {
      "epoch": 2.190931145736938,
      "grad_norm": 0.07170166820287704,
      "learning_rate": 8.441399403366079e-06,
      "loss": 0.0032,
      "num_input_tokens_seen": 36433088,
      "step": 10950
    },
    {
      "epoch": 2.19193157091764,
      "grad_norm": 0.08745460212230682,
      "learning_rate": 8.421782715824781e-06,
      "loss": 0.0059,
      "num_input_tokens_seen": 36450304,
      "step": 10955
    },
    {
      "epoch": 2.1929319960983418,
      "grad_norm": 0.1059829369187355,
      "learning_rate": 8.4021842301671e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 36466528,
      "step": 10960
    },
    {
      "epoch": 2.1939324212790434,
      "grad_norm": 1.6656731367111206,
      "learning_rate": 8.38260396791101e-06,
      "loss": 0.0179,
      "num_input_tokens_seen": 36483360,
      "step": 10965
    },
    {
      "epoch": 2.1949328464597455,
      "grad_norm": 0.9303325414657593,
      "learning_rate": 8.363041950554463e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 36500128,
      "step": 10970
    },
    {
      "epoch": 2.195933271640447,
      "grad_norm": 1.052852749824524,
      "learning_rate": 8.343498199575354e-06,
      "loss": 0.0048,
      "num_input_tokens_seen": 36515648,
      "step": 10975
    },
    {
      "epoch": 2.196933696821149,
      "grad_norm": 1.1024091243743896,
      "learning_rate": 8.323972736431556e-06,
      "loss": 0.0189,
      "num_input_tokens_seen": 36532576,
      "step": 10980
    },
    {
      "epoch": 2.197934122001851,
      "grad_norm": 0.06839191168546677,
      "learning_rate": 8.304465582560864e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 36548832,
      "step": 10985
    },
    {
      "epoch": 2.1989345471825525,
      "grad_norm": 0.03600554168224335,
      "learning_rate": 8.284976759380937e-06,
      "loss": 0.0086,
      "num_input_tokens_seen": 36565088,
      "step": 10990
    },
    {
      "epoch": 2.199934972363254,
      "grad_norm": 0.3253752589225769,
      "learning_rate": 8.265506288289333e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 36580448,
      "step": 10995
    },
    {
      "epoch": 2.2009353975439563,
      "grad_norm": 0.13283377885818481,
      "learning_rate": 8.24605419066346e-06,
      "loss": 0.0065,
      "num_input_tokens_seen": 36596704,
      "step": 11000
    },
    {
      "epoch": 2.201935822724658,
      "grad_norm": 1.415050745010376,
      "learning_rate": 8.22662048786055e-06,
      "loss": 0.0076,
      "num_input_tokens_seen": 36613792,
      "step": 11005
    },
    {
      "epoch": 2.2029362479053596,
      "grad_norm": 0.2603066861629486,
      "learning_rate": 8.207205201217644e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 36630240,
      "step": 11010
    },
    {
      "epoch": 2.2039366730860617,
      "grad_norm": 0.22146141529083252,
      "learning_rate": 8.187808352051543e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 36647136,
      "step": 11015
    },
    {
      "epoch": 2.2049370982667633,
      "grad_norm": 0.5399295091629028,
      "learning_rate": 8.168429961658822e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 36662880,
      "step": 11020
    },
    {
      "epoch": 2.205937523447465,
      "grad_norm": 0.7557527422904968,
      "learning_rate": 8.1490700513158e-06,
      "loss": 0.0141,
      "num_input_tokens_seen": 36678784,
      "step": 11025
    },
    {
      "epoch": 2.206937948628167,
      "grad_norm": 0.2671133577823639,
      "learning_rate": 8.12972864227847e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 36695808,
      "step": 11030
    },
    {
      "epoch": 2.2079383738088687,
      "grad_norm": 0.17605052888393402,
      "learning_rate": 8.110405755782543e-06,
      "loss": 0.0055,
      "num_input_tokens_seen": 36710752,
      "step": 11035
    },
    {
      "epoch": 2.2089387989895704,
      "grad_norm": 0.21284525096416473,
      "learning_rate": 8.091101413043395e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 36727232,
      "step": 11040
    },
    {
      "epoch": 2.2099392241702724,
      "grad_norm": 0.05113387480378151,
      "learning_rate": 8.071815635256009e-06,
      "loss": 0.005,
      "num_input_tokens_seen": 36743904,
      "step": 11045
    },
    {
      "epoch": 2.210939649350974,
      "grad_norm": 0.6746780872344971,
      "learning_rate": 8.052548443595031e-06,
      "loss": 0.0064,
      "num_input_tokens_seen": 36762944,
      "step": 11050
    },
    {
      "epoch": 2.211940074531676,
      "grad_norm": 0.5726050734519958,
      "learning_rate": 8.033299859214657e-06,
      "loss": 0.0055,
      "num_input_tokens_seen": 36780224,
      "step": 11055
    },
    {
      "epoch": 2.212940499712378,
      "grad_norm": 0.5448541641235352,
      "learning_rate": 8.014069903248683e-06,
      "loss": 0.0072,
      "num_input_tokens_seen": 36797568,
      "step": 11060
    },
    {
      "epoch": 2.2139409248930795,
      "grad_norm": 0.078267902135849,
      "learning_rate": 7.994858596810448e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 36814432,
      "step": 11065
    },
    {
      "epoch": 2.214941350073781,
      "grad_norm": 0.40484005212783813,
      "learning_rate": 7.97566596099279e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 36832096,
      "step": 11070
    },
    {
      "epoch": 2.215941775254483,
      "grad_norm": 0.2648407220840454,
      "learning_rate": 7.956492016868077e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 36847936,
      "step": 11075
    },
    {
      "epoch": 2.216942200435185,
      "grad_norm": 0.037626806646585464,
      "learning_rate": 7.937336785488156e-06,
      "loss": 0.0004,
      "num_input_tokens_seen": 36863616,
      "step": 11080
    },
    {
      "epoch": 2.217942625615887,
      "grad_norm": 0.0008298809407278895,
      "learning_rate": 7.918200287884294e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 36879648,
      "step": 11085
    },
    {
      "epoch": 2.2189430507965886,
      "grad_norm": 0.01786094717681408,
      "learning_rate": 7.899082545067227e-06,
      "loss": 0.0007,
      "num_input_tokens_seen": 36896896,
      "step": 11090
    },
    {
      "epoch": 2.2199434759772902,
      "grad_norm": 0.058479081839323044,
      "learning_rate": 7.879983578027078e-06,
      "loss": 0.0064,
      "num_input_tokens_seen": 36913760,
      "step": 11095
    },
    {
      "epoch": 2.2209439011579923,
      "grad_norm": 0.10139820724725723,
      "learning_rate": 7.860903407733366e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 36930304,
      "step": 11100
    },
    {
      "epoch": 2.221944326338694,
      "grad_norm": 1.0013636350631714,
      "learning_rate": 7.841842055134977e-06,
      "loss": 0.0047,
      "num_input_tokens_seen": 36947136,
      "step": 11105
    },
    {
      "epoch": 2.2229447515193956,
      "grad_norm": 0.07671966403722763,
      "learning_rate": 7.822799541160109e-06,
      "loss": 0.0079,
      "num_input_tokens_seen": 36962496,
      "step": 11110
    },
    {
      "epoch": 2.2239451767000977,
      "grad_norm": 0.21829085052013397,
      "learning_rate": 7.803775886716303e-06,
      "loss": 0.0044,
      "num_input_tokens_seen": 36979392,
      "step": 11115
    },
    {
      "epoch": 2.2249456018807994,
      "grad_norm": 0.16538646817207336,
      "learning_rate": 7.784771112690393e-06,
      "loss": 0.0073,
      "num_input_tokens_seen": 36996736,
      "step": 11120
    },
    {
      "epoch": 2.225946027061501,
      "grad_norm": 0.448978066444397,
      "learning_rate": 7.765785239948462e-06,
      "loss": 0.0033,
      "num_input_tokens_seen": 37012608,
      "step": 11125
    },
    {
      "epoch": 2.226946452242203,
      "grad_norm": 1.5056158304214478,
      "learning_rate": 7.746818289335861e-06,
      "loss": 0.0115,
      "num_input_tokens_seen": 37029280,
      "step": 11130
    },
    {
      "epoch": 2.2279468774229048,
      "grad_norm": 0.013198950327932835,
      "learning_rate": 7.727870281677168e-06,
      "loss": 0.0038,
      "num_input_tokens_seen": 37045760,
      "step": 11135
    },
    {
      "epoch": 2.2289473026036064,
      "grad_norm": 0.2678176760673523,
      "learning_rate": 7.708941237776138e-06,
      "loss": 0.0054,
      "num_input_tokens_seen": 37063392,
      "step": 11140
    },
    {
      "epoch": 2.2299477277843085,
      "grad_norm": 0.024856559932231903,
      "learning_rate": 7.690031178415724e-06,
      "loss": 0.0174,
      "num_input_tokens_seen": 37079936,
      "step": 11145
    },
    {
      "epoch": 2.23094815296501,
      "grad_norm": 0.018685590475797653,
      "learning_rate": 7.67114012435804e-06,
      "loss": 0.0015,
      "num_input_tokens_seen": 37096224,
      "step": 11150
    },
    {
      "epoch": 2.231948578145712,
      "grad_norm": 0.20162878930568695,
      "learning_rate": 7.652268096344315e-06,
      "loss": 0.0104,
      "num_input_tokens_seen": 37112896,
      "step": 11155
    },
    {
      "epoch": 2.232949003326414,
      "grad_norm": 1.602971076965332,
      "learning_rate": 7.633415115094913e-06,
      "loss": 0.0122,
      "num_input_tokens_seen": 37130464,
      "step": 11160
    },
    {
      "epoch": 2.2339494285071155,
      "grad_norm": 1.0054552555084229,
      "learning_rate": 7.614581201309254e-06,
      "loss": 0.0114,
      "num_input_tokens_seen": 37148096,
      "step": 11165
    },
    {
      "epoch": 2.234949853687817,
      "grad_norm": 0.8621109127998352,
      "learning_rate": 7.59576637566585e-06,
      "loss": 0.0101,
      "num_input_tokens_seen": 37165856,
      "step": 11170
    },
    {
      "epoch": 2.2359502788685193,
      "grad_norm": 0.13474692404270172,
      "learning_rate": 7.576970658822252e-06,
      "loss": 0.0024,
      "num_input_tokens_seen": 37182400,
      "step": 11175
    },
    {
      "epoch": 2.236950704049221,
      "grad_norm": 0.7479958534240723,
      "learning_rate": 7.558194071415009e-06,
      "loss": 0.0078,
      "num_input_tokens_seen": 37198304,
      "step": 11180
    },
    {
      "epoch": 2.2379511292299226,
      "grad_norm": 0.8761096000671387,
      "learning_rate": 7.539436634059688e-06,
      "loss": 0.0035,
      "num_input_tokens_seen": 37215136,
      "step": 11185
    },
    {
      "epoch": 2.2389515544106247,
      "grad_norm": 0.5879620909690857,
      "learning_rate": 7.520698367350837e-06,
      "loss": 0.004,
      "num_input_tokens_seen": 37232416,
      "step": 11190
    },
    {
      "epoch": 2.2399519795913263,
      "grad_norm": 0.04014047607779503,
      "learning_rate": 7.501979291861927e-06,
      "loss": 0.0038,
      "num_input_tokens_seen": 37248768,
      "step": 11195
    },
    {
      "epoch": 2.240952404772028,
      "grad_norm": 0.0336315892636776,
      "learning_rate": 7.483279428145382e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 37265984,
      "step": 11200
    },
    {
      "epoch": 2.24195282995273,
      "grad_norm": 0.35822203755378723,
      "learning_rate": 7.464598796732525e-06,
      "loss": 0.0012,
      "num_input_tokens_seen": 37281632,
      "step": 11205
    },
    {
      "epoch": 2.2429532551334317,
      "grad_norm": 0.014730866998434067,
      "learning_rate": 7.445937418133564e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 37297920,
      "step": 11210
    },
    {
      "epoch": 2.2439536803141333,
      "grad_norm": 1.2691651582717896,
      "learning_rate": 7.427295312837576e-06,
      "loss": 0.0142,
      "num_input_tokens_seen": 37315040,
      "step": 11215
    },
    {
      "epoch": 2.2449541054948354,
      "grad_norm": 0.8418992757797241,
      "learning_rate": 7.408672501312455e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 37332864,
      "step": 11220
    },
    {
      "epoch": 2.245954530675537,
      "grad_norm": 1.1171756982803345,
      "learning_rate": 7.390069004004932e-06,
      "loss": 0.0126,
      "num_input_tokens_seen": 37350112,
      "step": 11225
    },
    {
      "epoch": 2.2469549558562387,
      "grad_norm": 0.37548863887786865,
      "learning_rate": 7.371484841340537e-06,
      "loss": 0.0098,
      "num_input_tokens_seen": 37366752,
      "step": 11230
    },
    {
      "epoch": 2.247955381036941,
      "grad_norm": 0.3306463658809662,
      "learning_rate": 7.35292003372354e-06,
      "loss": 0.0037,
      "num_input_tokens_seen": 37383008,
      "step": 11235
    },
    {
      "epoch": 2.2489558062176425,
      "grad_norm": 0.09577401727437973,
      "learning_rate": 7.334374601536992e-06,
      "loss": 0.0127,
      "num_input_tokens_seen": 37400064,
      "step": 11240
    },
    {
      "epoch": 2.249956231398344,
      "grad_norm": 1.3612958192825317,
      "learning_rate": 7.3158485651426685e-06,
      "loss": 0.0118,
      "num_input_tokens_seen": 37417376,
      "step": 11245
    },
    {
      "epoch": 2.250956656579046,
      "grad_norm": 0.19891683757305145,
      "learning_rate": 7.297341944881017e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 37433696,
      "step": 11250
    },
    {
      "epoch": 2.251957081759748,
      "grad_norm": 0.17722542583942413,
      "learning_rate": 7.2788547610712215e-06,
      "loss": 0.0042,
      "num_input_tokens_seen": 37450752,
      "step": 11255
    },
    {
      "epoch": 2.25295750694045,
      "grad_norm": 0.6168380379676819,
      "learning_rate": 7.2603870340110734e-06,
      "loss": 0.0121,
      "num_input_tokens_seen": 37467648,
      "step": 11260
    },
    {
      "epoch": 2.2539579321211516,
      "grad_norm": 0.46064338088035583,
      "learning_rate": 7.2419387839770345e-06,
      "loss": 0.011,
      "num_input_tokens_seen": 37483808,
      "step": 11265
    },
    {
      "epoch": 2.2549583573018532,
      "grad_norm": 0.0065690092742443085,
      "learning_rate": 7.223510031224173e-06,
      "loss": 0.0033,
      "num_input_tokens_seen": 37500416,
      "step": 11270
    },
    {
      "epoch": 2.255958782482555,
      "grad_norm": 0.28575095534324646,
      "learning_rate": 7.205100795986139e-06,
      "loss": 0.0072,
      "num_input_tokens_seen": 37517344,
      "step": 11275
    },
    {
      "epoch": 2.256959207663257,
      "grad_norm": 0.6938003897666931,
      "learning_rate": 7.1867110984751715e-06,
      "loss": 0.0032,
      "num_input_tokens_seen": 37533824,
      "step": 11280
    },
    {
      "epoch": 2.2579596328439586,
      "grad_norm": 0.454542338848114,
      "learning_rate": 7.168340958882053e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 37549568,
      "step": 11285
    },
    {
      "epoch": 2.2589600580246607,
      "grad_norm": 0.028962474316358566,
      "learning_rate": 7.14999039737608e-06,
      "loss": 0.0128,
      "num_input_tokens_seen": 37566656,
      "step": 11290
    },
    {
      "epoch": 2.2599604832053624,
      "grad_norm": 0.02451271563768387,
      "learning_rate": 7.1316594341050675e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 37583424,
      "step": 11295
    },
    {
      "epoch": 2.260960908386064,
      "grad_norm": 0.48972272872924805,
      "learning_rate": 7.11334808919531e-06,
      "loss": 0.0078,
      "num_input_tokens_seen": 37599360,
      "step": 11300
    },
    {
      "epoch": 2.2619613335667657,
      "grad_norm": 0.10499460250139236,
      "learning_rate": 7.095056382751558e-06,
      "loss": 0.0117,
      "num_input_tokens_seen": 37615552,
      "step": 11305
    },
    {
      "epoch": 2.2629617587474677,
      "grad_norm": 0.054841239005327225,
      "learning_rate": 7.076784334857018e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 37632736,
      "step": 11310
    },
    {
      "epoch": 2.2639621839281694,
      "grad_norm": 0.31568577885627747,
      "learning_rate": 7.058531965573273e-06,
      "loss": 0.0067,
      "num_input_tokens_seen": 37649440,
      "step": 11315
    },
    {
      "epoch": 2.2649626091088715,
      "grad_norm": 0.027038007974624634,
      "learning_rate": 7.040299294940339e-06,
      "loss": 0.003,
      "num_input_tokens_seen": 37665888,
      "step": 11320
    },
    {
      "epoch": 2.265963034289573,
      "grad_norm": 0.35627177357673645,
      "learning_rate": 7.022086342976594e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 37682816,
      "step": 11325
    },
    {
      "epoch": 2.2669634594702748,
      "grad_norm": 0.6563451290130615,
      "learning_rate": 7.003893129678749e-06,
      "loss": 0.0105,
      "num_input_tokens_seen": 37699008,
      "step": 11330
    },
    {
      "epoch": 2.2679638846509764,
      "grad_norm": 0.029778748750686646,
      "learning_rate": 6.98571967502186e-06,
      "loss": 0.0039,
      "num_input_tokens_seen": 37716288,
      "step": 11335
    },
    {
      "epoch": 2.2689643098316785,
      "grad_norm": 0.818820059299469,
      "learning_rate": 6.967565998959297e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 37732704,
      "step": 11340
    },
    {
      "epoch": 2.26996473501238,
      "grad_norm": 2.850776195526123,
      "learning_rate": 6.949432121422683e-06,
      "loss": 0.0131,
      "num_input_tokens_seen": 37749216,
      "step": 11345
    },
    {
      "epoch": 2.2709651601930823,
      "grad_norm": 0.8199514150619507,
      "learning_rate": 6.931318062321934e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 37765184,
      "step": 11350
    },
    {
      "epoch": 2.271965585373784,
      "grad_norm": 1.1873388290405273,
      "learning_rate": 6.913223841545194e-06,
      "loss": 0.0081,
      "num_input_tokens_seen": 37781152,
      "step": 11355
    },
    {
      "epoch": 2.2729660105544855,
      "grad_norm": 1.3290454149246216,
      "learning_rate": 6.8951494789588264e-06,
      "loss": 0.007,
      "num_input_tokens_seen": 37798304,
      "step": 11360
    },
    {
      "epoch": 2.2739664357351876,
      "grad_norm": 0.0264931358397007,
      "learning_rate": 6.877094994407402e-06,
      "loss": 0.0086,
      "num_input_tokens_seen": 37814080,
      "step": 11365
    },
    {
      "epoch": 2.2749668609158893,
      "grad_norm": 0.7083085179328918,
      "learning_rate": 6.859060407713638e-06,
      "loss": 0.0051,
      "num_input_tokens_seen": 37829856,
      "step": 11370
    },
    {
      "epoch": 2.275967286096591,
      "grad_norm": 1.5754274129867554,
      "learning_rate": 6.8410457386784335e-06,
      "loss": 0.0077,
      "num_input_tokens_seen": 37846912,
      "step": 11375
    },
    {
      "epoch": 2.276967711277293,
      "grad_norm": 1.5928505659103394,
      "learning_rate": 6.823051007080822e-06,
      "loss": 0.0235,
      "num_input_tokens_seen": 37864416,
      "step": 11380
    },
    {
      "epoch": 2.2779681364579947,
      "grad_norm": 0.2157992571592331,
      "learning_rate": 6.80507623267791e-06,
      "loss": 0.003,
      "num_input_tokens_seen": 37881184,
      "step": 11385
    },
    {
      "epoch": 2.2789685616386963,
      "grad_norm": 0.4091234803199768,
      "learning_rate": 6.7871214352049326e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 37896480,
      "step": 11390
    },
    {
      "epoch": 2.2799689868193984,
      "grad_norm": 0.056644193828105927,
      "learning_rate": 6.7691866343751745e-06,
      "loss": 0.0081,
      "num_input_tokens_seen": 37912224,
      "step": 11395
    },
    {
      "epoch": 2.2809694120001,
      "grad_norm": 1.0580017566680908,
      "learning_rate": 6.751271849879961e-06,
      "loss": 0.0047,
      "num_input_tokens_seen": 37927744,
      "step": 11400
    },
    {
      "epoch": 2.2819698371808017,
      "grad_norm": 0.44456249475479126,
      "learning_rate": 6.733377101388646e-06,
      "loss": 0.0117,
      "num_input_tokens_seen": 37943712,
      "step": 11405
    },
    {
      "epoch": 2.282970262361504,
      "grad_norm": 0.5159664750099182,
      "learning_rate": 6.715502408548588e-06,
      "loss": 0.0103,
      "num_input_tokens_seen": 37960864,
      "step": 11410
    },
    {
      "epoch": 2.2839706875422054,
      "grad_norm": 0.04254378750920296,
      "learning_rate": 6.697647790985123e-06,
      "loss": 0.0064,
      "num_input_tokens_seen": 37978880,
      "step": 11415
    },
    {
      "epoch": 2.284971112722907,
      "grad_norm": 0.09852634370326996,
      "learning_rate": 6.679813268301552e-06,
      "loss": 0.0075,
      "num_input_tokens_seen": 37994624,
      "step": 11420
    },
    {
      "epoch": 2.285971537903609,
      "grad_norm": 1.8152574300765991,
      "learning_rate": 6.661998860079094e-06,
      "loss": 0.0158,
      "num_input_tokens_seen": 38012288,
      "step": 11425
    },
    {
      "epoch": 2.286971963084311,
      "grad_norm": 0.040850501507520676,
      "learning_rate": 6.644204585876898e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 38029728,
      "step": 11430
    },
    {
      "epoch": 2.2879723882650125,
      "grad_norm": 0.6898013949394226,
      "learning_rate": 6.626430465232014e-06,
      "loss": 0.0054,
      "num_input_tokens_seen": 38046112,
      "step": 11435
    },
    {
      "epoch": 2.2889728134457146,
      "grad_norm": 2.1513373851776123,
      "learning_rate": 6.608676517659346e-06,
      "loss": 0.003,
      "num_input_tokens_seen": 38062368,
      "step": 11440
    },
    {
      "epoch": 2.289973238626416,
      "grad_norm": 0.37828773260116577,
      "learning_rate": 6.590942762651659e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 38079520,
      "step": 11445
    },
    {
      "epoch": 2.290973663807118,
      "grad_norm": 0.4980011284351349,
      "learning_rate": 6.573229219679555e-06,
      "loss": 0.0042,
      "num_input_tokens_seen": 38096320,
      "step": 11450
    },
    {
      "epoch": 2.29197408898782,
      "grad_norm": 0.7982547879219055,
      "learning_rate": 6.5555359081914205e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 38112384,
      "step": 11455
    },
    {
      "epoch": 2.2929745141685216,
      "grad_norm": 0.02630678378045559,
      "learning_rate": 6.537862847613463e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 38130432,
      "step": 11460
    },
    {
      "epoch": 2.2939749393492233,
      "grad_norm": 0.009921961463987827,
      "learning_rate": 6.52021005734964e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 38145344,
      "step": 11465
    },
    {
      "epoch": 2.2949753645299253,
      "grad_norm": 0.8031901121139526,
      "learning_rate": 6.502577556781636e-06,
      "loss": 0.0074,
      "num_input_tokens_seen": 38161376,
      "step": 11470
    },
    {
      "epoch": 2.295975789710627,
      "grad_norm": 0.6120681166648865,
      "learning_rate": 6.484965365268888e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 38178944,
      "step": 11475
    },
    {
      "epoch": 2.2969762148913286,
      "grad_norm": 0.6634009480476379,
      "learning_rate": 6.46737350214851e-06,
      "loss": 0.0091,
      "num_input_tokens_seen": 38195968,
      "step": 11480
    },
    {
      "epoch": 2.2979766400720307,
      "grad_norm": 0.17244035005569458,
      "learning_rate": 6.44980198673531e-06,
      "loss": 0.0075,
      "num_input_tokens_seen": 38213280,
      "step": 11485
    },
    {
      "epoch": 2.2989770652527324,
      "grad_norm": 1.0651710033416748,
      "learning_rate": 6.432250838321762e-06,
      "loss": 0.0051,
      "num_input_tokens_seen": 38229504,
      "step": 11490
    },
    {
      "epoch": 2.299977490433434,
      "grad_norm": 1.1801623106002808,
      "learning_rate": 6.4147200761779575e-06,
      "loss": 0.0239,
      "num_input_tokens_seen": 38246240,
      "step": 11495
    },
    {
      "epoch": 2.300977915614136,
      "grad_norm": 1.4246106147766113,
      "learning_rate": 6.39720971955162e-06,
      "loss": 0.02,
      "num_input_tokens_seen": 38262144,
      "step": 11500
    },
    {
      "epoch": 2.3019783407948378,
      "grad_norm": 0.26787394285202026,
      "learning_rate": 6.379719787668079e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 38279744,
      "step": 11505
    },
    {
      "epoch": 2.3029787659755394,
      "grad_norm": 0.01519018318504095,
      "learning_rate": 6.362250299730196e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 38297184,
      "step": 11510
    },
    {
      "epoch": 2.3039791911562415,
      "grad_norm": 0.12452206760644913,
      "learning_rate": 6.344801274918446e-06,
      "loss": 0.0071,
      "num_input_tokens_seen": 38313280,
      "step": 11515
    },
    {
      "epoch": 2.304979616336943,
      "grad_norm": 1.0121663808822632,
      "learning_rate": 6.327372732390802e-06,
      "loss": 0.0065,
      "num_input_tokens_seen": 38329824,
      "step": 11520
    },
    {
      "epoch": 2.3059800415176452,
      "grad_norm": 0.31872105598449707,
      "learning_rate": 6.309964691282738e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 38348512,
      "step": 11525
    },
    {
      "epoch": 2.306980466698347,
      "grad_norm": 0.16815949976444244,
      "learning_rate": 6.292577170707256e-06,
      "loss": 0.0097,
      "num_input_tokens_seen": 38365664,
      "step": 11530
    },
    {
      "epoch": 2.3079808918790485,
      "grad_norm": 0.09678701311349869,
      "learning_rate": 6.275210189754782e-06,
      "loss": 0.012,
      "num_input_tokens_seen": 38382304,
      "step": 11535
    },
    {
      "epoch": 2.30898131705975,
      "grad_norm": 0.6526739001274109,
      "learning_rate": 6.2578637674932275e-06,
      "loss": 0.0042,
      "num_input_tokens_seen": 38399584,
      "step": 11540
    },
    {
      "epoch": 2.3099817422404523,
      "grad_norm": 0.11651992797851562,
      "learning_rate": 6.2405379229679265e-06,
      "loss": 0.0155,
      "num_input_tokens_seen": 38416064,
      "step": 11545
    },
    {
      "epoch": 2.310982167421154,
      "grad_norm": 0.03568281978368759,
      "learning_rate": 6.2232326752015956e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 38432640,
      "step": 11550
    },
    {
      "epoch": 2.311982592601856,
      "grad_norm": 0.10640271008014679,
      "learning_rate": 6.205948043194359e-06,
      "loss": 0.0038,
      "num_input_tokens_seen": 38449440,
      "step": 11555
    },
    {
      "epoch": 2.3129830177825577,
      "grad_norm": 0.18235915899276733,
      "learning_rate": 6.1886840459237065e-06,
      "loss": 0.0059,
      "num_input_tokens_seen": 38466848,
      "step": 11560
    },
    {
      "epoch": 2.3139834429632593,
      "grad_norm": 0.034905605018138885,
      "learning_rate": 6.171440702344464e-06,
      "loss": 0.0083,
      "num_input_tokens_seen": 38483552,
      "step": 11565
    },
    {
      "epoch": 2.314983868143961,
      "grad_norm": 0.21466754376888275,
      "learning_rate": 6.154218031388784e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 38500832,
      "step": 11570
    },
    {
      "epoch": 2.315984293324663,
      "grad_norm": 0.5182342529296875,
      "learning_rate": 6.137016051966127e-06,
      "loss": 0.0077,
      "num_input_tokens_seen": 38516672,
      "step": 11575
    },
    {
      "epoch": 2.3169847185053647,
      "grad_norm": 0.022903772071003914,
      "learning_rate": 6.119834782963218e-06,
      "loss": 0.0054,
      "num_input_tokens_seen": 38534592,
      "step": 11580
    },
    {
      "epoch": 2.317985143686067,
      "grad_norm": 0.06478303670883179,
      "learning_rate": 6.102674243244064e-06,
      "loss": 0.0044,
      "num_input_tokens_seen": 38551136,
      "step": 11585
    },
    {
      "epoch": 2.3189855688667684,
      "grad_norm": 1.263426661491394,
      "learning_rate": 6.085534451649905e-06,
      "loss": 0.0083,
      "num_input_tokens_seen": 38567744,
      "step": 11590
    },
    {
      "epoch": 2.31998599404747,
      "grad_norm": 0.597636878490448,
      "learning_rate": 6.0684154269991915e-06,
      "loss": 0.0102,
      "num_input_tokens_seen": 38584768,
      "step": 11595
    },
    {
      "epoch": 2.3209864192281717,
      "grad_norm": 1.0190768241882324,
      "learning_rate": 6.051317188087591e-06,
      "loss": 0.0168,
      "num_input_tokens_seen": 38601632,
      "step": 11600
    },
    {
      "epoch": 2.321986844408874,
      "grad_norm": 0.002115818904712796,
      "learning_rate": 6.034239753687923e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 38617664,
      "step": 11605
    },
    {
      "epoch": 2.3229872695895755,
      "grad_norm": 0.2652653753757477,
      "learning_rate": 6.017183142550187e-06,
      "loss": 0.005,
      "num_input_tokens_seen": 38633952,
      "step": 11610
    },
    {
      "epoch": 2.3239876947702776,
      "grad_norm": 0.32525134086608887,
      "learning_rate": 6.000147373401535e-06,
      "loss": 0.0052,
      "num_input_tokens_seen": 38649952,
      "step": 11615
    },
    {
      "epoch": 2.324988119950979,
      "grad_norm": 0.05901453271508217,
      "learning_rate": 5.9831324649461896e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 38665152,
      "step": 11620
    },
    {
      "epoch": 2.325988545131681,
      "grad_norm": 0.978908121585846,
      "learning_rate": 5.966138435865507e-06,
      "loss": 0.0116,
      "num_input_tokens_seen": 38681568,
      "step": 11625
    },
    {
      "epoch": 2.326988970312383,
      "grad_norm": 0.013850186951458454,
      "learning_rate": 5.949165304817908e-06,
      "loss": 0.0082,
      "num_input_tokens_seen": 38697792,
      "step": 11630
    },
    {
      "epoch": 2.3279893954930846,
      "grad_norm": 1.3319257497787476,
      "learning_rate": 5.932213090438859e-06,
      "loss": 0.0048,
      "num_input_tokens_seen": 38714752,
      "step": 11635
    },
    {
      "epoch": 2.3289898206737862,
      "grad_norm": 0.5809262990951538,
      "learning_rate": 5.915281811340878e-06,
      "loss": 0.0045,
      "num_input_tokens_seen": 38731744,
      "step": 11640
    },
    {
      "epoch": 2.3299902458544883,
      "grad_norm": 0.34603744745254517,
      "learning_rate": 5.898371486113491e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 38748224,
      "step": 11645
    },
    {
      "epoch": 2.33099067103519,
      "grad_norm": 0.34748753905296326,
      "learning_rate": 5.881482133323205e-06,
      "loss": 0.0074,
      "num_input_tokens_seen": 38763808,
      "step": 11650
    },
    {
      "epoch": 2.3319910962158916,
      "grad_norm": 0.2618560194969177,
      "learning_rate": 5.864613771513527e-06,
      "loss": 0.0077,
      "num_input_tokens_seen": 38779648,
      "step": 11655
    },
    {
      "epoch": 2.3329915213965937,
      "grad_norm": 0.12622158229351044,
      "learning_rate": 5.847766419204883e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 38795264,
      "step": 11660
    },
    {
      "epoch": 2.3339919465772954,
      "grad_norm": 0.0028299703262746334,
      "learning_rate": 5.830940094894658e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 38812064,
      "step": 11665
    },
    {
      "epoch": 2.334992371757997,
      "grad_norm": 0.08322950452566147,
      "learning_rate": 5.814134817057157e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 38828256,
      "step": 11670
    },
    {
      "epoch": 2.335992796938699,
      "grad_norm": 0.7389622330665588,
      "learning_rate": 5.797350604143542e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 38845024,
      "step": 11675
    },
    {
      "epoch": 2.3369932221194007,
      "grad_norm": 0.0005778248887509108,
      "learning_rate": 5.780587474581878e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 38861280,
      "step": 11680
    },
    {
      "epoch": 2.3379936473001024,
      "grad_norm": 0.1739816516637802,
      "learning_rate": 5.763845446777078e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 38879392,
      "step": 11685
    },
    {
      "epoch": 2.3389940724808045,
      "grad_norm": 0.23818816244602203,
      "learning_rate": 5.7471245391108594e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 38896704,
      "step": 11690
    },
    {
      "epoch": 2.339994497661506,
      "grad_norm": 0.20521743595600128,
      "learning_rate": 5.730424769941786e-06,
      "loss": 0.0062,
      "num_input_tokens_seen": 38913248,
      "step": 11695
    },
    {
      "epoch": 2.340994922842208,
      "grad_norm": 1.6082990169525146,
      "learning_rate": 5.713746157605199e-06,
      "loss": 0.0089,
      "num_input_tokens_seen": 38930240,
      "step": 11700
    },
    {
      "epoch": 2.34199534802291,
      "grad_norm": 0.7052076458930969,
      "learning_rate": 5.697088720413202e-06,
      "loss": 0.007,
      "num_input_tokens_seen": 38946784,
      "step": 11705
    },
    {
      "epoch": 2.3429957732036115,
      "grad_norm": 0.014426125213503838,
      "learning_rate": 5.680452476654657e-06,
      "loss": 0.0034,
      "num_input_tokens_seen": 38964352,
      "step": 11710
    },
    {
      "epoch": 2.343996198384313,
      "grad_norm": 0.027191922068595886,
      "learning_rate": 5.663837444595163e-06,
      "loss": 0.0073,
      "num_input_tokens_seen": 38980288,
      "step": 11715
    },
    {
      "epoch": 2.3449966235650153,
      "grad_norm": 0.4602048099040985,
      "learning_rate": 5.647243642477024e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 38996576,
      "step": 11720
    },
    {
      "epoch": 2.345997048745717,
      "grad_norm": 0.12057361751794815,
      "learning_rate": 5.6306710885192374e-06,
      "loss": 0.0009,
      "num_input_tokens_seen": 39012576,
      "step": 11725
    },
    {
      "epoch": 2.3469974739264186,
      "grad_norm": 0.13868285715579987,
      "learning_rate": 5.614119800917459e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 39028768,
      "step": 11730
    },
    {
      "epoch": 2.3479978991071206,
      "grad_norm": 0.05260397493839264,
      "learning_rate": 5.597589797844008e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 39044640,
      "step": 11735
    },
    {
      "epoch": 2.3489983242878223,
      "grad_norm": 0.05297001078724861,
      "learning_rate": 5.5810810974478434e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 39061184,
      "step": 11740
    },
    {
      "epoch": 2.349998749468524,
      "grad_norm": 0.0046495902352035046,
      "learning_rate": 5.564593717854505e-06,
      "loss": 0.0089,
      "num_input_tokens_seen": 39077312,
      "step": 11745
    },
    {
      "epoch": 2.350999174649226,
      "grad_norm": 2.4701688289642334,
      "learning_rate": 5.548127677166154e-06,
      "loss": 0.0042,
      "num_input_tokens_seen": 39094176,
      "step": 11750
    },
    {
      "epoch": 2.3519995998299277,
      "grad_norm": 0.048507627099752426,
      "learning_rate": 5.531682993461515e-06,
      "loss": 0.0005,
      "num_input_tokens_seen": 39111424,
      "step": 11755
    },
    {
      "epoch": 2.3530000250106293,
      "grad_norm": 1.5936695337295532,
      "learning_rate": 5.515259684795845e-06,
      "loss": 0.0081,
      "num_input_tokens_seen": 39127488,
      "step": 11760
    },
    {
      "epoch": 2.3540004501913314,
      "grad_norm": 0.12213210016489029,
      "learning_rate": 5.498857769200957e-06,
      "loss": 0.0136,
      "num_input_tokens_seen": 39144192,
      "step": 11765
    },
    {
      "epoch": 2.355000875372033,
      "grad_norm": 0.7076942920684814,
      "learning_rate": 5.482477264685165e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 39161920,
      "step": 11770
    },
    {
      "epoch": 2.3560013005527347,
      "grad_norm": 0.5250160098075867,
      "learning_rate": 5.4661181892332744e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 39178784,
      "step": 11775
    },
    {
      "epoch": 2.357001725733437,
      "grad_norm": 0.22318120300769806,
      "learning_rate": 5.449780560806572e-06,
      "loss": 0.0039,
      "num_input_tokens_seen": 39194976,
      "step": 11780
    },
    {
      "epoch": 2.3580021509141385,
      "grad_norm": 0.07826335728168488,
      "learning_rate": 5.433464397342777e-06,
      "loss": 0.0059,
      "num_input_tokens_seen": 39210752,
      "step": 11785
    },
    {
      "epoch": 2.3590025760948405,
      "grad_norm": 0.021109146997332573,
      "learning_rate": 5.41716971675606e-06,
      "loss": 0.0039,
      "num_input_tokens_seen": 39227200,
      "step": 11790
    },
    {
      "epoch": 2.360003001275542,
      "grad_norm": 0.7640042304992676,
      "learning_rate": 5.400896536937003e-06,
      "loss": 0.0033,
      "num_input_tokens_seen": 39243296,
      "step": 11795
    },
    {
      "epoch": 2.361003426456244,
      "grad_norm": 1.1045968532562256,
      "learning_rate": 5.3846448757525645e-06,
      "loss": 0.0067,
      "num_input_tokens_seen": 39260960,
      "step": 11800
    },
    {
      "epoch": 2.3620038516369455,
      "grad_norm": 0.10084889829158783,
      "learning_rate": 5.368414751046097e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 39276416,
      "step": 11805
    },
    {
      "epoch": 2.3630042768176476,
      "grad_norm": 0.0003986780939158052,
      "learning_rate": 5.352206180637306e-06,
      "loss": 0.0104,
      "num_input_tokens_seen": 39292928,
      "step": 11810
    },
    {
      "epoch": 2.3640047019983492,
      "grad_norm": 0.014915438368916512,
      "learning_rate": 5.336019182322205e-06,
      "loss": 0.0065,
      "num_input_tokens_seen": 39309184,
      "step": 11815
    },
    {
      "epoch": 2.3650051271790513,
      "grad_norm": 0.2112547606229782,
      "learning_rate": 5.319853773873168e-06,
      "loss": 0.0105,
      "num_input_tokens_seen": 39326400,
      "step": 11820
    },
    {
      "epoch": 2.366005552359753,
      "grad_norm": 0.9200066328048706,
      "learning_rate": 5.303709973038823e-06,
      "loss": 0.0032,
      "num_input_tokens_seen": 39343552,
      "step": 11825
    },
    {
      "epoch": 2.3670059775404546,
      "grad_norm": 1.281941533088684,
      "learning_rate": 5.287587797544091e-06,
      "loss": 0.0068,
      "num_input_tokens_seen": 39361280,
      "step": 11830
    },
    {
      "epoch": 2.3680064027211563,
      "grad_norm": 0.4280705153942108,
      "learning_rate": 5.271487265090163e-06,
      "loss": 0.0076,
      "num_input_tokens_seen": 39379360,
      "step": 11835
    },
    {
      "epoch": 2.3690068279018583,
      "grad_norm": 0.6366599798202515,
      "learning_rate": 5.255408393354436e-06,
      "loss": 0.0047,
      "num_input_tokens_seen": 39395904,
      "step": 11840
    },
    {
      "epoch": 2.37000725308256,
      "grad_norm": 0.7315851449966431,
      "learning_rate": 5.23935119999055e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 39411552,
      "step": 11845
    },
    {
      "epoch": 2.371007678263262,
      "grad_norm": 0.9730765223503113,
      "learning_rate": 5.22331570262834e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 39427776,
      "step": 11850
    },
    {
      "epoch": 2.3720081034439637,
      "grad_norm": 0.05999741330742836,
      "learning_rate": 5.207301918873806e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 39443200,
      "step": 11855
    },
    {
      "epoch": 2.3730085286246654,
      "grad_norm": 1.359703779220581,
      "learning_rate": 5.191309866309124e-06,
      "loss": 0.0131,
      "num_input_tokens_seen": 39459808,
      "step": 11860
    },
    {
      "epoch": 2.3740089538053675,
      "grad_norm": 0.2668972909450531,
      "learning_rate": 5.175339562492609e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 39477152,
      "step": 11865
    },
    {
      "epoch": 2.375009378986069,
      "grad_norm": 0.11682423949241638,
      "learning_rate": 5.159391024958673e-06,
      "loss": 0.0135,
      "num_input_tokens_seen": 39493312,
      "step": 11870
    },
    {
      "epoch": 2.3760098041667708,
      "grad_norm": 0.002341238083317876,
      "learning_rate": 5.143464271217877e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 39510176,
      "step": 11875
    },
    {
      "epoch": 2.377010229347473,
      "grad_norm": 0.8494821190834045,
      "learning_rate": 5.127559318756814e-06,
      "loss": 0.007,
      "num_input_tokens_seen": 39526848,
      "step": 11880
    },
    {
      "epoch": 2.3780106545281745,
      "grad_norm": 0.15606018900871277,
      "learning_rate": 5.111676185038172e-06,
      "loss": 0.0122,
      "num_input_tokens_seen": 39544288,
      "step": 11885
    },
    {
      "epoch": 2.379011079708876,
      "grad_norm": 0.10261460393667221,
      "learning_rate": 5.095814887500683e-06,
      "loss": 0.0037,
      "num_input_tokens_seen": 39560320,
      "step": 11890
    },
    {
      "epoch": 2.3800115048895782,
      "grad_norm": 0.6049176454544067,
      "learning_rate": 5.0799754435590795e-06,
      "loss": 0.0045,
      "num_input_tokens_seen": 39576544,
      "step": 11895
    },
    {
      "epoch": 2.38101193007028,
      "grad_norm": 0.6998255848884583,
      "learning_rate": 5.064157870604125e-06,
      "loss": 0.0069,
      "num_input_tokens_seen": 39593312,
      "step": 11900
    },
    {
      "epoch": 2.3820123552509815,
      "grad_norm": 1.1377514600753784,
      "learning_rate": 5.048362186002567e-06,
      "loss": 0.0088,
      "num_input_tokens_seen": 39610624,
      "step": 11905
    },
    {
      "epoch": 2.3830127804316836,
      "grad_norm": 0.0668625608086586,
      "learning_rate": 5.0325884070971054e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 39628384,
      "step": 11910
    },
    {
      "epoch": 2.3840132056123853,
      "grad_norm": 0.583050549030304,
      "learning_rate": 5.0168365512064e-06,
      "loss": 0.004,
      "num_input_tokens_seen": 39643936,
      "step": 11915
    },
    {
      "epoch": 2.385013630793087,
      "grad_norm": 0.17105671763420105,
      "learning_rate": 5.001106635625044e-06,
      "loss": 0.0126,
      "num_input_tokens_seen": 39660224,
      "step": 11920
    },
    {
      "epoch": 2.386014055973789,
      "grad_norm": 0.11508852243423462,
      "learning_rate": 4.985398677623535e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 39676800,
      "step": 11925
    },
    {
      "epoch": 2.3870144811544907,
      "grad_norm": 1.0608491897583008,
      "learning_rate": 4.969712694448272e-06,
      "loss": 0.0167,
      "num_input_tokens_seen": 39693856,
      "step": 11930
    },
    {
      "epoch": 2.3880149063351923,
      "grad_norm": 0.2027820646762848,
      "learning_rate": 4.954048703321506e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 39709984,
      "step": 11935
    },
    {
      "epoch": 2.3890153315158944,
      "grad_norm": 0.12901784479618073,
      "learning_rate": 4.93840672144136e-06,
      "loss": 0.01,
      "num_input_tokens_seen": 39727040,
      "step": 11940
    },
    {
      "epoch": 2.390015756696596,
      "grad_norm": 0.0070501226000487804,
      "learning_rate": 4.922786765981796e-06,
      "loss": 0.0064,
      "num_input_tokens_seen": 39742848,
      "step": 11945
    },
    {
      "epoch": 2.3910161818772977,
      "grad_norm": 0.12959033250808716,
      "learning_rate": 4.90718885409257e-06,
      "loss": 0.0048,
      "num_input_tokens_seen": 39759712,
      "step": 11950
    },
    {
      "epoch": 2.392016607058,
      "grad_norm": 0.19388934969902039,
      "learning_rate": 4.891613002899259e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 39775200,
      "step": 11955
    },
    {
      "epoch": 2.3930170322387014,
      "grad_norm": 1.7149473428726196,
      "learning_rate": 4.8760592295032104e-06,
      "loss": 0.0102,
      "num_input_tokens_seen": 39791648,
      "step": 11960
    },
    {
      "epoch": 2.394017457419403,
      "grad_norm": 0.009398255497217178,
      "learning_rate": 4.860527550981519e-06,
      "loss": 0.013,
      "num_input_tokens_seen": 39808192,
      "step": 11965
    },
    {
      "epoch": 2.395017882600105,
      "grad_norm": 0.08744099736213684,
      "learning_rate": 4.84501798438704e-06,
      "loss": 0.0076,
      "num_input_tokens_seen": 39826816,
      "step": 11970
    },
    {
      "epoch": 2.396018307780807,
      "grad_norm": 0.014029473066329956,
      "learning_rate": 4.829530546748342e-06,
      "loss": 0.0095,
      "num_input_tokens_seen": 39843104,
      "step": 11975
    },
    {
      "epoch": 2.3970187329615085,
      "grad_norm": 0.044142477214336395,
      "learning_rate": 4.8140652550697e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 39859584,
      "step": 11980
    },
    {
      "epoch": 2.3980191581422106,
      "grad_norm": 0.3335197865962982,
      "learning_rate": 4.798622126331076e-06,
      "loss": 0.0045,
      "num_input_tokens_seen": 39876160,
      "step": 11985
    },
    {
      "epoch": 2.399019583322912,
      "grad_norm": 0.4007914662361145,
      "learning_rate": 4.783201177488084e-06,
      "loss": 0.0063,
      "num_input_tokens_seen": 39892448,
      "step": 11990
    },
    {
      "epoch": 2.400020008503614,
      "grad_norm": 0.4795380234718323,
      "learning_rate": 4.767802425471998e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 39908576,
      "step": 11995
    },
    {
      "epoch": 2.401020433684316,
      "grad_norm": 0.6965258717536926,
      "learning_rate": 4.7524258871897345e-06,
      "loss": 0.003,
      "num_input_tokens_seen": 39925184,
      "step": 12000
    },
    {
      "epoch": 2.4020208588650176,
      "grad_norm": 0.07378388941287994,
      "learning_rate": 4.7370715795237885e-06,
      "loss": 0.0063,
      "num_input_tokens_seen": 39941664,
      "step": 12005
    },
    {
      "epoch": 2.4030212840457192,
      "grad_norm": 2.771740436553955,
      "learning_rate": 4.7217395193322715e-06,
      "loss": 0.0115,
      "num_input_tokens_seen": 39958560,
      "step": 12010
    },
    {
      "epoch": 2.4040217092264213,
      "grad_norm": 0.05147058516740799,
      "learning_rate": 4.706429723448869e-06,
      "loss": 0.0089,
      "num_input_tokens_seen": 39974272,
      "step": 12015
    },
    {
      "epoch": 2.405022134407123,
      "grad_norm": 0.015699077397584915,
      "learning_rate": 4.691142208682797e-06,
      "loss": 0.0074,
      "num_input_tokens_seen": 39989632,
      "step": 12020
    },
    {
      "epoch": 2.406022559587825,
      "grad_norm": 0.048658337444067,
      "learning_rate": 4.675876991818839e-06,
      "loss": 0.0138,
      "num_input_tokens_seen": 40005696,
      "step": 12025
    },
    {
      "epoch": 2.4070229847685267,
      "grad_norm": 1.0035970211029053,
      "learning_rate": 4.66063408961728e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 40023520,
      "step": 12030
    },
    {
      "epoch": 2.4080234099492284,
      "grad_norm": 0.024283969774842262,
      "learning_rate": 4.645413518813907e-06,
      "loss": 0.0038,
      "num_input_tokens_seen": 40039584,
      "step": 12035
    },
    {
      "epoch": 2.40902383512993,
      "grad_norm": 0.01331272255629301,
      "learning_rate": 4.63021529612e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 40056800,
      "step": 12040
    },
    {
      "epoch": 2.410024260310632,
      "grad_norm": 0.07054154574871063,
      "learning_rate": 4.615039438222271e-06,
      "loss": 0.0121,
      "num_input_tokens_seen": 40073152,
      "step": 12045
    },
    {
      "epoch": 2.4110246854913338,
      "grad_norm": 0.25928425788879395,
      "learning_rate": 4.5998859617829126e-06,
      "loss": 0.0094,
      "num_input_tokens_seen": 40089568,
      "step": 12050
    },
    {
      "epoch": 2.412025110672036,
      "grad_norm": 0.1377958059310913,
      "learning_rate": 4.584754883439532e-06,
      "loss": 0.006,
      "num_input_tokens_seen": 40104736,
      "step": 12055
    },
    {
      "epoch": 2.4130255358527375,
      "grad_norm": 0.2971647083759308,
      "learning_rate": 4.569646219805126e-06,
      "loss": 0.0102,
      "num_input_tokens_seen": 40122496,
      "step": 12060
    },
    {
      "epoch": 2.414025961033439,
      "grad_norm": 0.0399310477077961,
      "learning_rate": 4.554559987468107e-06,
      "loss": 0.0134,
      "num_input_tokens_seen": 40139040,
      "step": 12065
    },
    {
      "epoch": 2.415026386214141,
      "grad_norm": 0.8164964318275452,
      "learning_rate": 4.539496202992252e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 40155392,
      "step": 12070
    },
    {
      "epoch": 2.416026811394843,
      "grad_norm": 2.913259983062744,
      "learning_rate": 4.524454882916676e-06,
      "loss": 0.0114,
      "num_input_tokens_seen": 40171808,
      "step": 12075
    },
    {
      "epoch": 2.4170272365755445,
      "grad_norm": 0.16079522669315338,
      "learning_rate": 4.50943604375586e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 40187200,
      "step": 12080
    },
    {
      "epoch": 2.4180276617562466,
      "grad_norm": 0.0017121377168223262,
      "learning_rate": 4.494439701999567e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 40204992,
      "step": 12085
    },
    {
      "epoch": 2.4190280869369483,
      "grad_norm": 0.07056961953639984,
      "learning_rate": 4.479465874112887e-06,
      "loss": 0.0083,
      "num_input_tokens_seen": 40222432,
      "step": 12090
    },
    {
      "epoch": 2.42002851211765,
      "grad_norm": 1.3982396125793457,
      "learning_rate": 4.464514576536183e-06,
      "loss": 0.0116,
      "num_input_tokens_seen": 40238976,
      "step": 12095
    },
    {
      "epoch": 2.4210289372983516,
      "grad_norm": 0.018442688509821892,
      "learning_rate": 4.449585825685071e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 40255424,
      "step": 12100
    },
    {
      "epoch": 2.4220293624790536,
      "grad_norm": 0.058521077036857605,
      "learning_rate": 4.434679637950417e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 40271456,
      "step": 12105
    },
    {
      "epoch": 2.4230297876597553,
      "grad_norm": 0.002892683958634734,
      "learning_rate": 4.419796029698334e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 40286912,
      "step": 12110
    },
    {
      "epoch": 2.4240302128404574,
      "grad_norm": 0.00801822729408741,
      "learning_rate": 4.404935017270109e-06,
      "loss": 0.0077,
      "num_input_tokens_seen": 40303072,
      "step": 12115
    },
    {
      "epoch": 2.425030638021159,
      "grad_norm": 0.17238368093967438,
      "learning_rate": 4.390096616982242e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 40320352,
      "step": 12120
    },
    {
      "epoch": 2.4260310632018607,
      "grad_norm": 0.005229899659752846,
      "learning_rate": 4.375280845126409e-06,
      "loss": 0.0045,
      "num_input_tokens_seen": 40337792,
      "step": 12125
    },
    {
      "epoch": 2.4270314883825628,
      "grad_norm": 0.5783362984657288,
      "learning_rate": 4.36048771796943e-06,
      "loss": 0.0068,
      "num_input_tokens_seen": 40355264,
      "step": 12130
    },
    {
      "epoch": 2.4280319135632644,
      "grad_norm": 0.23066847026348114,
      "learning_rate": 4.345717251753273e-06,
      "loss": 0.0099,
      "num_input_tokens_seen": 40371936,
      "step": 12135
    },
    {
      "epoch": 2.429032338743966,
      "grad_norm": 1.4478811025619507,
      "learning_rate": 4.330969462695012e-06,
      "loss": 0.0024,
      "num_input_tokens_seen": 40387968,
      "step": 12140
    },
    {
      "epoch": 2.430032763924668,
      "grad_norm": 0.05630865320563316,
      "learning_rate": 4.316244366986833e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 40404224,
      "step": 12145
    },
    {
      "epoch": 2.43103318910537,
      "grad_norm": 1.3600881099700928,
      "learning_rate": 4.301541980796009e-06,
      "loss": 0.0113,
      "num_input_tokens_seen": 40420320,
      "step": 12150
    },
    {
      "epoch": 2.4320336142860715,
      "grad_norm": 2.776556968688965,
      "learning_rate": 4.286862320264865e-06,
      "loss": 0.0255,
      "num_input_tokens_seen": 40439008,
      "step": 12155
    },
    {
      "epoch": 2.4330340394667735,
      "grad_norm": 0.07221867889165878,
      "learning_rate": 4.272205401510787e-06,
      "loss": 0.0082,
      "num_input_tokens_seen": 40455040,
      "step": 12160
    },
    {
      "epoch": 2.434034464647475,
      "grad_norm": 0.12634849548339844,
      "learning_rate": 4.257571240626198e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 40471808,
      "step": 12165
    },
    {
      "epoch": 2.435034889828177,
      "grad_norm": 0.7212656736373901,
      "learning_rate": 4.242959853678513e-06,
      "loss": 0.0198,
      "num_input_tokens_seen": 40489344,
      "step": 12170
    },
    {
      "epoch": 2.436035315008879,
      "grad_norm": 0.3485330641269684,
      "learning_rate": 4.228371256710157e-06,
      "loss": 0.0098,
      "num_input_tokens_seen": 40505600,
      "step": 12175
    },
    {
      "epoch": 2.4370357401895806,
      "grad_norm": 0.12140169739723206,
      "learning_rate": 4.21380546573853e-06,
      "loss": 0.0007,
      "num_input_tokens_seen": 40522080,
      "step": 12180
    },
    {
      "epoch": 2.4380361653702822,
      "grad_norm": 0.07376435399055481,
      "learning_rate": 4.199262496755998e-06,
      "loss": 0.0144,
      "num_input_tokens_seen": 40538368,
      "step": 12185
    },
    {
      "epoch": 2.4390365905509843,
      "grad_norm": 0.10095193237066269,
      "learning_rate": 4.184742365729866e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 40554784,
      "step": 12190
    },
    {
      "epoch": 2.440037015731686,
      "grad_norm": 0.14332710206508636,
      "learning_rate": 4.170245088602351e-06,
      "loss": 0.0007,
      "num_input_tokens_seen": 40570688,
      "step": 12195
    },
    {
      "epoch": 2.4410374409123876,
      "grad_norm": 0.6795973181724548,
      "learning_rate": 4.155770681290599e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 40586272,
      "step": 12200
    },
    {
      "epoch": 2.4420378660930897,
      "grad_norm": 1.7163009643554688,
      "learning_rate": 4.141319159686641e-06,
      "loss": 0.0113,
      "num_input_tokens_seen": 40602400,
      "step": 12205
    },
    {
      "epoch": 2.4430382912737914,
      "grad_norm": 1.4630216360092163,
      "learning_rate": 4.1268905396573636e-06,
      "loss": 0.003,
      "num_input_tokens_seen": 40618336,
      "step": 12210
    },
    {
      "epoch": 2.444038716454493,
      "grad_norm": 0.5265514850616455,
      "learning_rate": 4.112484837044531e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 40635008,
      "step": 12215
    },
    {
      "epoch": 2.445039141635195,
      "grad_norm": 0.6192420125007629,
      "learning_rate": 4.0981020676647385e-06,
      "loss": 0.0037,
      "num_input_tokens_seen": 40651648,
      "step": 12220
    },
    {
      "epoch": 2.4460395668158967,
      "grad_norm": 0.3160333037376404,
      "learning_rate": 4.083742247309394e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 40669184,
      "step": 12225
    },
    {
      "epoch": 2.4470399919965984,
      "grad_norm": 0.0337165929377079,
      "learning_rate": 4.069405391744718e-06,
      "loss": 0.0039,
      "num_input_tokens_seen": 40685568,
      "step": 12230
    },
    {
      "epoch": 2.4480404171773005,
      "grad_norm": 0.027851616963744164,
      "learning_rate": 4.055091516711712e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 40702944,
      "step": 12235
    },
    {
      "epoch": 2.449040842358002,
      "grad_norm": 1.460172176361084,
      "learning_rate": 4.040800637926151e-06,
      "loss": 0.0063,
      "num_input_tokens_seen": 40720576,
      "step": 12240
    },
    {
      "epoch": 2.4500412675387038,
      "grad_norm": 0.32231810688972473,
      "learning_rate": 4.026532771078565e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 40736896,
      "step": 12245
    },
    {
      "epoch": 2.451041692719406,
      "grad_norm": 1.7847610712051392,
      "learning_rate": 4.012287931834196e-06,
      "loss": 0.0052,
      "num_input_tokens_seen": 40752576,
      "step": 12250
    },
    {
      "epoch": 2.4520421179001075,
      "grad_norm": 0.21667858958244324,
      "learning_rate": 3.998066135833031e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 40770432,
      "step": 12255
    },
    {
      "epoch": 2.453042543080809,
      "grad_norm": 1.9653438329696655,
      "learning_rate": 3.983867398689747e-06,
      "loss": 0.0206,
      "num_input_tokens_seen": 40787360,
      "step": 12260
    },
    {
      "epoch": 2.4540429682615112,
      "grad_norm": 0.27307531237602234,
      "learning_rate": 3.969691735993691e-06,
      "loss": 0.0039,
      "num_input_tokens_seen": 40804128,
      "step": 12265
    },
    {
      "epoch": 2.455043393442213,
      "grad_norm": 0.0512768030166626,
      "learning_rate": 3.955539163308894e-06,
      "loss": 0.0089,
      "num_input_tokens_seen": 40820704,
      "step": 12270
    },
    {
      "epoch": 2.4560438186229145,
      "grad_norm": 0.2971963584423065,
      "learning_rate": 3.941409696174034e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 40837344,
      "step": 12275
    },
    {
      "epoch": 2.4570442438036166,
      "grad_norm": 0.05244804173707962,
      "learning_rate": 3.927303350102393e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 40853824,
      "step": 12280
    },
    {
      "epoch": 2.4580446689843183,
      "grad_norm": 0.10713346302509308,
      "learning_rate": 3.913220140581908e-06,
      "loss": 0.0104,
      "num_input_tokens_seen": 40870560,
      "step": 12285
    },
    {
      "epoch": 2.4590450941650204,
      "grad_norm": 1.085828185081482,
      "learning_rate": 3.899160083075095e-06,
      "loss": 0.0175,
      "num_input_tokens_seen": 40886336,
      "step": 12290
    },
    {
      "epoch": 2.460045519345722,
      "grad_norm": 0.4680262804031372,
      "learning_rate": 3.885123193019041e-06,
      "loss": 0.0032,
      "num_input_tokens_seen": 40902912,
      "step": 12295
    },
    {
      "epoch": 2.4610459445264237,
      "grad_norm": 0.21633762121200562,
      "learning_rate": 3.871109485825414e-06,
      "loss": 0.0165,
      "num_input_tokens_seen": 40919968,
      "step": 12300
    },
    {
      "epoch": 2.4620463697071253,
      "grad_norm": 2.2871382236480713,
      "learning_rate": 3.857118976880409e-06,
      "loss": 0.0156,
      "num_input_tokens_seen": 40937184,
      "step": 12305
    },
    {
      "epoch": 2.4630467948878274,
      "grad_norm": 0.4637632966041565,
      "learning_rate": 3.843151681544768e-06,
      "loss": 0.0024,
      "num_input_tokens_seen": 40955872,
      "step": 12310
    },
    {
      "epoch": 2.464047220068529,
      "grad_norm": 0.09708676487207413,
      "learning_rate": 3.829207615153743e-06,
      "loss": 0.0092,
      "num_input_tokens_seen": 40972000,
      "step": 12315
    },
    {
      "epoch": 2.465047645249231,
      "grad_norm": 0.13777290284633636,
      "learning_rate": 3.815286793017072e-06,
      "loss": 0.0049,
      "num_input_tokens_seen": 40989056,
      "step": 12320
    },
    {
      "epoch": 2.466048070429933,
      "grad_norm": 1.4429576396942139,
      "learning_rate": 3.801389230418981e-06,
      "loss": 0.0055,
      "num_input_tokens_seen": 41005984,
      "step": 12325
    },
    {
      "epoch": 2.4670484956106344,
      "grad_norm": 0.30174750089645386,
      "learning_rate": 3.787514942618159e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 41022976,
      "step": 12330
    },
    {
      "epoch": 2.468048920791336,
      "grad_norm": 0.25705936551094055,
      "learning_rate": 3.7736639448477356e-06,
      "loss": 0.006,
      "num_input_tokens_seen": 41040672,
      "step": 12335
    },
    {
      "epoch": 2.469049345972038,
      "grad_norm": 0.9293633103370667,
      "learning_rate": 3.7598362523152714e-06,
      "loss": 0.0112,
      "num_input_tokens_seen": 41056832,
      "step": 12340
    },
    {
      "epoch": 2.47004977115274,
      "grad_norm": 0.7727804183959961,
      "learning_rate": 3.7460318802027465e-06,
      "loss": 0.0065,
      "num_input_tokens_seen": 41073056,
      "step": 12345
    },
    {
      "epoch": 2.471050196333442,
      "grad_norm": 0.9562500715255737,
      "learning_rate": 3.7322508436665183e-06,
      "loss": 0.0081,
      "num_input_tokens_seen": 41088640,
      "step": 12350
    },
    {
      "epoch": 2.4720506215141436,
      "grad_norm": 0.004255808889865875,
      "learning_rate": 3.718493157837344e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 41106144,
      "step": 12355
    },
    {
      "epoch": 2.473051046694845,
      "grad_norm": 0.43987080454826355,
      "learning_rate": 3.704758837820327e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 41122176,
      "step": 12360
    },
    {
      "epoch": 2.474051471875547,
      "grad_norm": 0.2720640301704407,
      "learning_rate": 3.6910478986949243e-06,
      "loss": 0.003,
      "num_input_tokens_seen": 41139072,
      "step": 12365
    },
    {
      "epoch": 2.475051897056249,
      "grad_norm": 0.7602618932723999,
      "learning_rate": 3.677360355514928e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 41155616,
      "step": 12370
    },
    {
      "epoch": 2.4760523222369506,
      "grad_norm": 0.5711652636528015,
      "learning_rate": 3.6636962233084243e-06,
      "loss": 0.0044,
      "num_input_tokens_seen": 41171680,
      "step": 12375
    },
    {
      "epoch": 2.4770527474176527,
      "grad_norm": 0.3574143648147583,
      "learning_rate": 3.6500555170778134e-06,
      "loss": 0.0034,
      "num_input_tokens_seen": 41188608,
      "step": 12380
    },
    {
      "epoch": 2.4780531725983543,
      "grad_norm": 0.02313903719186783,
      "learning_rate": 3.6364382517997676e-06,
      "loss": 0.0057,
      "num_input_tokens_seen": 41205184,
      "step": 12385
    },
    {
      "epoch": 2.479053597779056,
      "grad_norm": 0.022654995322227478,
      "learning_rate": 3.6228444424252244e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 41221888,
      "step": 12390
    },
    {
      "epoch": 2.480054022959758,
      "grad_norm": 0.04455249384045601,
      "learning_rate": 3.6092741038793677e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 41237472,
      "step": 12395
    },
    {
      "epoch": 2.4810544481404597,
      "grad_norm": 0.005392350722104311,
      "learning_rate": 3.595727251061615e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 41255104,
      "step": 12400
    },
    {
      "epoch": 2.4820548733211614,
      "grad_norm": 1.2513427734375,
      "learning_rate": 3.5822038988455876e-06,
      "loss": 0.011,
      "num_input_tokens_seen": 41272032,
      "step": 12405
    },
    {
      "epoch": 2.4830552985018635,
      "grad_norm": 0.09189336001873016,
      "learning_rate": 3.568704062079112e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 41288288,
      "step": 12410
    },
    {
      "epoch": 2.484055723682565,
      "grad_norm": 1.106570839881897,
      "learning_rate": 3.5552277555842036e-06,
      "loss": 0.0099,
      "num_input_tokens_seen": 41303616,
      "step": 12415
    },
    {
      "epoch": 2.4850561488632668,
      "grad_norm": 1.25088369846344,
      "learning_rate": 3.5417749941570236e-06,
      "loss": 0.0034,
      "num_input_tokens_seen": 41320416,
      "step": 12420
    },
    {
      "epoch": 2.486056574043969,
      "grad_norm": 0.39379605650901794,
      "learning_rate": 3.5283457925679054e-06,
      "loss": 0.0174,
      "num_input_tokens_seen": 41337760,
      "step": 12425
    },
    {
      "epoch": 2.4870569992246705,
      "grad_norm": 0.5777842998504639,
      "learning_rate": 3.5149401655612897e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 41353568,
      "step": 12430
    },
    {
      "epoch": 2.488057424405372,
      "grad_norm": 0.09307357668876648,
      "learning_rate": 3.50155812785575e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 41370496,
      "step": 12435
    },
    {
      "epoch": 2.4890578495860742,
      "grad_norm": 0.009802114218473434,
      "learning_rate": 3.4881996941439705e-06,
      "loss": 0.0037,
      "num_input_tokens_seen": 41386432,
      "step": 12440
    },
    {
      "epoch": 2.490058274766776,
      "grad_norm": 0.0019708112813532352,
      "learning_rate": 3.474864879092693e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 41403296,
      "step": 12445
    },
    {
      "epoch": 2.4910586999474775,
      "grad_norm": 0.3863993287086487,
      "learning_rate": 3.461553697342745e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 41420032,
      "step": 12450
    },
    {
      "epoch": 2.4920591251281796,
      "grad_norm": 0.7204772233963013,
      "learning_rate": 3.448266163509009e-06,
      "loss": 0.0074,
      "num_input_tokens_seen": 41437792,
      "step": 12455
    },
    {
      "epoch": 2.4930595503088813,
      "grad_norm": 0.0746840089559555,
      "learning_rate": 3.4350022921803807e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 41453632,
      "step": 12460
    },
    {
      "epoch": 2.494059975489583,
      "grad_norm": 0.3205074965953827,
      "learning_rate": 3.4217620979198063e-06,
      "loss": 0.0038,
      "num_input_tokens_seen": 41470336,
      "step": 12465
    },
    {
      "epoch": 2.495060400670285,
      "grad_norm": 0.013899927958846092,
      "learning_rate": 3.4085455952642175e-06,
      "loss": 0.0042,
      "num_input_tokens_seen": 41486624,
      "step": 12470
    },
    {
      "epoch": 2.4960608258509867,
      "grad_norm": 0.2298094779253006,
      "learning_rate": 3.3953527987245298e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 41502528,
      "step": 12475
    },
    {
      "epoch": 2.4970612510316883,
      "grad_norm": 1.8205831050872803,
      "learning_rate": 3.382183722785645e-06,
      "loss": 0.0048,
      "num_input_tokens_seen": 41519200,
      "step": 12480
    },
    {
      "epoch": 2.4980616762123904,
      "grad_norm": 0.0019053603755310178,
      "learning_rate": 3.3690383819064168e-06,
      "loss": 0.0032,
      "num_input_tokens_seen": 41536224,
      "step": 12485
    },
    {
      "epoch": 2.499062101393092,
      "grad_norm": 0.08693382889032364,
      "learning_rate": 3.3559167905196366e-06,
      "loss": 0.0093,
      "num_input_tokens_seen": 41554048,
      "step": 12490
    },
    {
      "epoch": 2.500062526573794,
      "grad_norm": 0.004477310460060835,
      "learning_rate": 3.342818963032024e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 41570688,
      "step": 12495
    },
    {
      "epoch": 2.5010629517544958,
      "grad_norm": 0.614901065826416,
      "learning_rate": 3.3297449138241925e-06,
      "loss": 0.0141,
      "num_input_tokens_seen": 41586944,
      "step": 12500
    },
    {
      "epoch": 2.5020633769351974,
      "grad_norm": 0.206019788980484,
      "learning_rate": 3.3166946572506713e-06,
      "loss": 0.0084,
      "num_input_tokens_seen": 41603456,
      "step": 12505
    },
    {
      "epoch": 2.503063802115899,
      "grad_norm": 0.05732341855764389,
      "learning_rate": 3.303668207639857e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 41619872,
      "step": 12510
    },
    {
      "epoch": 2.504064227296601,
      "grad_norm": 0.0009395777597092092,
      "learning_rate": 3.2906655792939974e-06,
      "loss": 0.0177,
      "num_input_tokens_seen": 41636832,
      "step": 12515
    },
    {
      "epoch": 2.505064652477303,
      "grad_norm": 0.8623054623603821,
      "learning_rate": 3.2776867864892006e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 41653376,
      "step": 12520
    },
    {
      "epoch": 2.506065077658005,
      "grad_norm": 0.0076326909475028515,
      "learning_rate": 3.264731843475402e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 41669792,
      "step": 12525
    },
    {
      "epoch": 2.5070655028387065,
      "grad_norm": 0.265615850687027,
      "learning_rate": 3.251800764476337e-06,
      "loss": 0.0069,
      "num_input_tokens_seen": 41686880,
      "step": 12530
    },
    {
      "epoch": 2.508065928019408,
      "grad_norm": 1.4407261610031128,
      "learning_rate": 3.2388935636895603e-06,
      "loss": 0.0093,
      "num_input_tokens_seen": 41702752,
      "step": 12535
    },
    {
      "epoch": 2.50906635320011,
      "grad_norm": 2.3761744499206543,
      "learning_rate": 3.226010255286399e-06,
      "loss": 0.008,
      "num_input_tokens_seen": 41720416,
      "step": 12540
    },
    {
      "epoch": 2.510066778380812,
      "grad_norm": 0.1614091694355011,
      "learning_rate": 3.21315085341195e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 41736224,
      "step": 12545
    },
    {
      "epoch": 2.5110672035615136,
      "grad_norm": 1.1754673719406128,
      "learning_rate": 3.2003153721850644e-06,
      "loss": 0.0103,
      "num_input_tokens_seen": 41753248,
      "step": 12550
    },
    {
      "epoch": 2.5120676287422157,
      "grad_norm": 0.12314176559448242,
      "learning_rate": 3.187503825698321e-06,
      "loss": 0.0118,
      "num_input_tokens_seen": 41769696,
      "step": 12555
    },
    {
      "epoch": 2.5130680539229173,
      "grad_norm": 0.06970703601837158,
      "learning_rate": 3.1747162280180264e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 41787328,
      "step": 12560
    },
    {
      "epoch": 2.514068479103619,
      "grad_norm": 0.03441135585308075,
      "learning_rate": 3.1619525931842003e-06,
      "loss": 0.0078,
      "num_input_tokens_seen": 41805376,
      "step": 12565
    },
    {
      "epoch": 2.5150689042843206,
      "grad_norm": 0.3269253671169281,
      "learning_rate": 3.1492129352105354e-06,
      "loss": 0.0135,
      "num_input_tokens_seen": 41821888,
      "step": 12570
    },
    {
      "epoch": 2.5160693294650227,
      "grad_norm": 0.6984061598777771,
      "learning_rate": 3.1364972680844096e-06,
      "loss": 0.011,
      "num_input_tokens_seen": 41838976,
      "step": 12575
    },
    {
      "epoch": 2.5170697546457244,
      "grad_norm": 0.31647729873657227,
      "learning_rate": 3.1238056057668693e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 41855168,
      "step": 12580
    },
    {
      "epoch": 2.5180701798264264,
      "grad_norm": 0.31902697682380676,
      "learning_rate": 3.1111379621925785e-06,
      "loss": 0.0044,
      "num_input_tokens_seen": 41872128,
      "step": 12585
    },
    {
      "epoch": 2.519070605007128,
      "grad_norm": 0.19838306307792664,
      "learning_rate": 3.0984943512698567e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 41887936,
      "step": 12590
    },
    {
      "epoch": 2.5200710301878297,
      "grad_norm": 0.5044189095497131,
      "learning_rate": 3.085874786880627e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 41904480,
      "step": 12595
    },
    {
      "epoch": 2.5210714553685314,
      "grad_norm": 0.015526696108281612,
      "learning_rate": 3.0732792828804093e-06,
      "loss": 0.0003,
      "num_input_tokens_seen": 41920608,
      "step": 12600
    },
    {
      "epoch": 2.5220718805492335,
      "grad_norm": 0.22375400364398956,
      "learning_rate": 3.0607078530983102e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 41936224,
      "step": 12605
    },
    {
      "epoch": 2.523072305729935,
      "grad_norm": 0.10406365990638733,
      "learning_rate": 3.048160511336995e-06,
      "loss": 0.0009,
      "num_input_tokens_seen": 41953280,
      "step": 12610
    },
    {
      "epoch": 2.524072730910637,
      "grad_norm": 0.635587215423584,
      "learning_rate": 3.035637271372696e-06,
      "loss": 0.0139,
      "num_input_tokens_seen": 41971232,
      "step": 12615
    },
    {
      "epoch": 2.525073156091339,
      "grad_norm": 1.0286800861358643,
      "learning_rate": 3.023138146955176e-06,
      "loss": 0.0099,
      "num_input_tokens_seen": 41987616,
      "step": 12620
    },
    {
      "epoch": 2.5260735812720405,
      "grad_norm": 0.3848174512386322,
      "learning_rate": 3.010663151807716e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 42004448,
      "step": 12625
    },
    {
      "epoch": 2.527074006452742,
      "grad_norm": 0.5619518756866455,
      "learning_rate": 2.9982122996271095e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 42021120,
      "step": 12630
    },
    {
      "epoch": 2.5280744316334443,
      "grad_norm": 0.016514690592885017,
      "learning_rate": 2.985785604083649e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 42037376,
      "step": 12635
    },
    {
      "epoch": 2.529074856814146,
      "grad_norm": 2.134133815765381,
      "learning_rate": 2.973383078821082e-06,
      "loss": 0.0113,
      "num_input_tokens_seen": 42054080,
      "step": 12640
    },
    {
      "epoch": 2.530075281994848,
      "grad_norm": 0.11717314273118973,
      "learning_rate": 2.961004737456652e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 42071616,
      "step": 12645
    },
    {
      "epoch": 2.5310757071755496,
      "grad_norm": 1.4976444244384766,
      "learning_rate": 2.9486505935810206e-06,
      "loss": 0.005,
      "num_input_tokens_seen": 42087456,
      "step": 12650
    },
    {
      "epoch": 2.5320761323562513,
      "grad_norm": 0.11834155023097992,
      "learning_rate": 2.9363206607582962e-06,
      "loss": 0.0124,
      "num_input_tokens_seen": 42103648,
      "step": 12655
    },
    {
      "epoch": 2.533076557536953,
      "grad_norm": 0.11021913588047028,
      "learning_rate": 2.9240149525260078e-06,
      "loss": 0.0039,
      "num_input_tokens_seen": 42119424,
      "step": 12660
    },
    {
      "epoch": 2.534076982717655,
      "grad_norm": 0.011786202900111675,
      "learning_rate": 2.911733482395071e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 42135488,
      "step": 12665
    },
    {
      "epoch": 2.5350774078983567,
      "grad_norm": 1.3259836435317993,
      "learning_rate": 2.899476263849807e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 42152896,
      "step": 12670
    },
    {
      "epoch": 2.5360778330790588,
      "grad_norm": 0.031422726809978485,
      "learning_rate": 2.8872433103479034e-06,
      "loss": 0.0271,
      "num_input_tokens_seen": 42168544,
      "step": 12675
    },
    {
      "epoch": 2.5370782582597604,
      "grad_norm": 0.04686697572469711,
      "learning_rate": 2.8750346353204003e-06,
      "loss": 0.0057,
      "num_input_tokens_seen": 42184832,
      "step": 12680
    },
    {
      "epoch": 2.538078683440462,
      "grad_norm": 0.01803763210773468,
      "learning_rate": 2.8628502521716895e-06,
      "loss": 0.0105,
      "num_input_tokens_seen": 42200416,
      "step": 12685
    },
    {
      "epoch": 2.539079108621164,
      "grad_norm": 1.2098323106765747,
      "learning_rate": 2.850690174279494e-06,
      "loss": 0.0062,
      "num_input_tokens_seen": 42215840,
      "step": 12690
    },
    {
      "epoch": 2.540079533801866,
      "grad_norm": 0.1791813224554062,
      "learning_rate": 2.83855441499484e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 42233184,
      "step": 12695
    },
    {
      "epoch": 2.5410799589825674,
      "grad_norm": 1.6025686264038086,
      "learning_rate": 2.8264429876420695e-06,
      "loss": 0.0112,
      "num_input_tokens_seen": 42249472,
      "step": 12700
    },
    {
      "epoch": 2.5420803841632695,
      "grad_norm": 0.3951073884963989,
      "learning_rate": 2.8143559055187896e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 42265024,
      "step": 12705
    },
    {
      "epoch": 2.543080809343971,
      "grad_norm": 0.5251020193099976,
      "learning_rate": 2.802293181895893e-06,
      "loss": 0.0076,
      "num_input_tokens_seen": 42280640,
      "step": 12710
    },
    {
      "epoch": 2.544081234524673,
      "grad_norm": 0.024615280330181122,
      "learning_rate": 2.7902548300175273e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 42296864,
      "step": 12715
    },
    {
      "epoch": 2.545081659705375,
      "grad_norm": 1.228441834449768,
      "learning_rate": 2.778240863101067e-06,
      "loss": 0.0096,
      "num_input_tokens_seen": 42313344,
      "step": 12720
    },
    {
      "epoch": 2.5460820848860766,
      "grad_norm": 0.009008641354739666,
      "learning_rate": 2.76625129433713e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 42329888,
      "step": 12725
    },
    {
      "epoch": 2.547082510066778,
      "grad_norm": 0.1975419521331787,
      "learning_rate": 2.7542861368895444e-06,
      "loss": 0.004,
      "num_input_tokens_seen": 42346592,
      "step": 12730
    },
    {
      "epoch": 2.5480829352474803,
      "grad_norm": 0.7556296586990356,
      "learning_rate": 2.7423454038953245e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 42362496,
      "step": 12735
    },
    {
      "epoch": 2.549083360428182,
      "grad_norm": 0.1864507496356964,
      "learning_rate": 2.73042910846468e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 42380384,
      "step": 12740
    },
    {
      "epoch": 2.5500837856088836,
      "grad_norm": 1.1749720573425293,
      "learning_rate": 2.7185372636809815e-06,
      "loss": 0.0142,
      "num_input_tokens_seen": 42397760,
      "step": 12745
    },
    {
      "epoch": 2.5510842107895857,
      "grad_norm": 0.46301645040512085,
      "learning_rate": 2.7066698826007624e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 42415264,
      "step": 12750
    },
    {
      "epoch": 2.5520846359702873,
      "grad_norm": 0.0805196762084961,
      "learning_rate": 2.694826978253695e-06,
      "loss": 0.0126,
      "num_input_tokens_seen": 42432960,
      "step": 12755
    },
    {
      "epoch": 2.5530850611509894,
      "grad_norm": 0.6513873934745789,
      "learning_rate": 2.683008563642564e-06,
      "loss": 0.0032,
      "num_input_tokens_seen": 42450176,
      "step": 12760
    },
    {
      "epoch": 2.554085486331691,
      "grad_norm": 0.37311941385269165,
      "learning_rate": 2.671214651743284e-06,
      "loss": 0.0044,
      "num_input_tokens_seen": 42467072,
      "step": 12765
    },
    {
      "epoch": 2.5550859115123927,
      "grad_norm": 0.09965891391038895,
      "learning_rate": 2.6594452555048637e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 42484320,
      "step": 12770
    },
    {
      "epoch": 2.5560863366930944,
      "grad_norm": 0.617047905921936,
      "learning_rate": 2.6477003878493804e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 42502848,
      "step": 12775
    },
    {
      "epoch": 2.5570867618737965,
      "grad_norm": 0.0027189829852432013,
      "learning_rate": 2.6359800616719943e-06,
      "loss": 0.0067,
      "num_input_tokens_seen": 42518880,
      "step": 12780
    },
    {
      "epoch": 2.558087187054498,
      "grad_norm": 0.0006735940696671605,
      "learning_rate": 2.6242842898409255e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 42535424,
      "step": 12785
    },
    {
      "epoch": 2.5590876122352,
      "grad_norm": 0.5481417179107666,
      "learning_rate": 2.612613085197413e-06,
      "loss": 0.006,
      "num_input_tokens_seen": 42553088,
      "step": 12790
    },
    {
      "epoch": 2.560088037415902,
      "grad_norm": 0.0007917046896182001,
      "learning_rate": 2.6009664605557456e-06,
      "loss": 0.0054,
      "num_input_tokens_seen": 42570560,
      "step": 12795
    },
    {
      "epoch": 2.5610884625966035,
      "grad_norm": 0.02182622440159321,
      "learning_rate": 2.5893444287032137e-06,
      "loss": 0.0126,
      "num_input_tokens_seen": 42586688,
      "step": 12800
    },
    {
      "epoch": 2.562088887777305,
      "grad_norm": 0.7775011658668518,
      "learning_rate": 2.577747002400108e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 42603104,
      "step": 12805
    },
    {
      "epoch": 2.5630893129580072,
      "grad_norm": 0.17077545821666718,
      "learning_rate": 2.5661741943797057e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 42621600,
      "step": 12810
    },
    {
      "epoch": 2.564089738138709,
      "grad_norm": 0.03099755011498928,
      "learning_rate": 2.5546260173482465e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 42637536,
      "step": 12815
    },
    {
      "epoch": 2.565090163319411,
      "grad_norm": 2.6456117630004883,
      "learning_rate": 2.543102483984938e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 42654976,
      "step": 12820
    },
    {
      "epoch": 2.5660905885001126,
      "grad_norm": 0.040536269545555115,
      "learning_rate": 2.5316036069419292e-06,
      "loss": 0.0003,
      "num_input_tokens_seen": 42671392,
      "step": 12825
    },
    {
      "epoch": 2.5670910136808143,
      "grad_norm": 0.10133875161409378,
      "learning_rate": 2.5201293988442843e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 42687808,
      "step": 12830
    },
    {
      "epoch": 2.568091438861516,
      "grad_norm": 0.0018810753244906664,
      "learning_rate": 2.5086798722899958e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 42704064,
      "step": 12835
    },
    {
      "epoch": 2.569091864042218,
      "grad_norm": 0.1539742797613144,
      "learning_rate": 2.4972550398499606e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 42720576,
      "step": 12840
    },
    {
      "epoch": 2.5700922892229197,
      "grad_norm": 0.031564582139253616,
      "learning_rate": 2.485854914067942e-06,
      "loss": 0.0004,
      "num_input_tokens_seen": 42737056,
      "step": 12845
    },
    {
      "epoch": 2.5710927144036217,
      "grad_norm": 1.6116400957107544,
      "learning_rate": 2.4744795074606115e-06,
      "loss": 0.0086,
      "num_input_tokens_seen": 42753056,
      "step": 12850
    },
    {
      "epoch": 2.5720931395843234,
      "grad_norm": 0.17667633295059204,
      "learning_rate": 2.463128832517461e-06,
      "loss": 0.0091,
      "num_input_tokens_seen": 42770048,
      "step": 12855
    },
    {
      "epoch": 2.573093564765025,
      "grad_norm": 0.14781901240348816,
      "learning_rate": 2.4518029017008586e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 42786112,
      "step": 12860
    },
    {
      "epoch": 2.5740939899457267,
      "grad_norm": 8.325371163664386e-05,
      "learning_rate": 2.440501727445993e-06,
      "loss": 0.0035,
      "num_input_tokens_seen": 42802208,
      "step": 12865
    },
    {
      "epoch": 2.575094415126429,
      "grad_norm": 0.9688087105751038,
      "learning_rate": 2.429225322160869e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 42819104,
      "step": 12870
    },
    {
      "epoch": 2.5760948403071304,
      "grad_norm": 0.05281443893909454,
      "learning_rate": 2.417973698226297e-06,
      "loss": 0.0009,
      "num_input_tokens_seen": 42835552,
      "step": 12875
    },
    {
      "epoch": 2.5770952654878325,
      "grad_norm": 0.11945304274559021,
      "learning_rate": 2.406746867995896e-06,
      "loss": 0.0106,
      "num_input_tokens_seen": 42852096,
      "step": 12880
    },
    {
      "epoch": 2.578095690668534,
      "grad_norm": 0.06983213126659393,
      "learning_rate": 2.3955448437960326e-06,
      "loss": 0.0129,
      "num_input_tokens_seen": 42870016,
      "step": 12885
    },
    {
      "epoch": 2.579096115849236,
      "grad_norm": 0.5124595761299133,
      "learning_rate": 2.3843676379258636e-06,
      "loss": 0.0078,
      "num_input_tokens_seen": 42887264,
      "step": 12890
    },
    {
      "epoch": 2.5800965410299375,
      "grad_norm": 0.016952697187662125,
      "learning_rate": 2.373215262657286e-06,
      "loss": 0.0103,
      "num_input_tokens_seen": 42903776,
      "step": 12895
    },
    {
      "epoch": 2.5810969662106396,
      "grad_norm": 1.0047266483306885,
      "learning_rate": 2.3620877302349296e-06,
      "loss": 0.0068,
      "num_input_tokens_seen": 42919936,
      "step": 12900
    },
    {
      "epoch": 2.582097391391341,
      "grad_norm": 0.0015937060816213489,
      "learning_rate": 2.3509850528761675e-06,
      "loss": 0.0071,
      "num_input_tokens_seen": 42936864,
      "step": 12905
    },
    {
      "epoch": 2.5830978165720433,
      "grad_norm": 1.1491022109985352,
      "learning_rate": 2.339907242771061e-06,
      "loss": 0.0114,
      "num_input_tokens_seen": 42954976,
      "step": 12910
    },
    {
      "epoch": 2.584098241752745,
      "grad_norm": 0.05709850415587425,
      "learning_rate": 2.328854312082376e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 42971680,
      "step": 12915
    },
    {
      "epoch": 2.5850986669334466,
      "grad_norm": 1.429154872894287,
      "learning_rate": 2.317826272945578e-06,
      "loss": 0.0059,
      "num_input_tokens_seen": 42987712,
      "step": 12920
    },
    {
      "epoch": 2.5860990921141487,
      "grad_norm": 0.738528847694397,
      "learning_rate": 2.3068231374687722e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 43003904,
      "step": 12925
    },
    {
      "epoch": 2.5870995172948503,
      "grad_norm": 0.006645476911216974,
      "learning_rate": 2.2958449177327478e-06,
      "loss": 0.008,
      "num_input_tokens_seen": 43020160,
      "step": 12930
    },
    {
      "epoch": 2.588099942475552,
      "grad_norm": 1.5658634901046753,
      "learning_rate": 2.2848916257909305e-06,
      "loss": 0.0061,
      "num_input_tokens_seen": 43037472,
      "step": 12935
    },
    {
      "epoch": 2.589100367656254,
      "grad_norm": 0.013548973947763443,
      "learning_rate": 2.2739632736693663e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 43053920,
      "step": 12940
    },
    {
      "epoch": 2.5901007928369557,
      "grad_norm": 0.029908064752817154,
      "learning_rate": 2.263059873366735e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 43071552,
      "step": 12945
    },
    {
      "epoch": 2.5911012180176574,
      "grad_norm": 0.03885706141591072,
      "learning_rate": 2.2521814368543088e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 43088096,
      "step": 12950
    },
    {
      "epoch": 2.5921016431983595,
      "grad_norm": 0.28232115507125854,
      "learning_rate": 2.2413279760759576e-06,
      "loss": 0.0052,
      "num_input_tokens_seen": 43105568,
      "step": 12955
    },
    {
      "epoch": 2.593102068379061,
      "grad_norm": 0.15103229880332947,
      "learning_rate": 2.2304995029481303e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 43121792,
      "step": 12960
    },
    {
      "epoch": 2.5941024935597627,
      "grad_norm": 0.05918802320957184,
      "learning_rate": 2.21969602935983e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 43139040,
      "step": 12965
    },
    {
      "epoch": 2.595102918740465,
      "grad_norm": 0.05563802644610405,
      "learning_rate": 2.208917567172625e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 43155264,
      "step": 12970
    },
    {
      "epoch": 2.5961033439211665,
      "grad_norm": 0.25392958521842957,
      "learning_rate": 2.19816412822062e-06,
      "loss": 0.0077,
      "num_input_tokens_seen": 43171200,
      "step": 12975
    },
    {
      "epoch": 2.597103769101868,
      "grad_norm": 0.006377012934535742,
      "learning_rate": 2.1874357243104355e-06,
      "loss": 0.0104,
      "num_input_tokens_seen": 43188704,
      "step": 12980
    },
    {
      "epoch": 2.5981041942825702,
      "grad_norm": 0.21373534202575684,
      "learning_rate": 2.176732367221218e-06,
      "loss": 0.0067,
      "num_input_tokens_seen": 43205248,
      "step": 12985
    },
    {
      "epoch": 2.599104619463272,
      "grad_norm": 0.003723123110830784,
      "learning_rate": 2.166054068704609e-06,
      "loss": 0.0005,
      "num_input_tokens_seen": 43221152,
      "step": 12990
    },
    {
      "epoch": 2.600105044643974,
      "grad_norm": 0.2379627823829651,
      "learning_rate": 2.1554008404847335e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 43237824,
      "step": 12995
    },
    {
      "epoch": 2.6011054698246756,
      "grad_norm": 0.0022142406087368727,
      "learning_rate": 2.144772694258193e-06,
      "loss": 0.0067,
      "num_input_tokens_seen": 43254592,
      "step": 13000
    },
    {
      "epoch": 2.6021058950053773,
      "grad_norm": 1.8083733320236206,
      "learning_rate": 2.134169641694056e-06,
      "loss": 0.0087,
      "num_input_tokens_seen": 43270432,
      "step": 13005
    },
    {
      "epoch": 2.603106320186079,
      "grad_norm": 0.0026089141611009836,
      "learning_rate": 2.1235916944338325e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 43288096,
      "step": 13010
    },
    {
      "epoch": 2.604106745366781,
      "grad_norm": 0.4001295864582062,
      "learning_rate": 2.1130388640914794e-06,
      "loss": 0.0065,
      "num_input_tokens_seen": 43303392,
      "step": 13015
    },
    {
      "epoch": 2.6051071705474826,
      "grad_norm": 0.2899264991283417,
      "learning_rate": 2.1025111622533537e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 43321248,
      "step": 13020
    },
    {
      "epoch": 2.6061075957281847,
      "grad_norm": 1.9828625917434692,
      "learning_rate": 2.0920086004782487e-06,
      "loss": 0.0067,
      "num_input_tokens_seen": 43337472,
      "step": 13025
    },
    {
      "epoch": 2.6071080209088864,
      "grad_norm": 0.3573082983493805,
      "learning_rate": 2.081531190297342e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 43353120,
      "step": 13030
    },
    {
      "epoch": 2.608108446089588,
      "grad_norm": 0.07330165803432465,
      "learning_rate": 2.0710789432141968e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 43368832,
      "step": 13035
    },
    {
      "epoch": 2.6091088712702897,
      "grad_norm": 0.6771386861801147,
      "learning_rate": 2.0606518707047493e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 43385856,
      "step": 13040
    },
    {
      "epoch": 2.6101092964509918,
      "grad_norm": 0.08871681988239288,
      "learning_rate": 2.0502499842173033e-06,
      "loss": 0.0059,
      "num_input_tokens_seen": 43401888,
      "step": 13045
    },
    {
      "epoch": 2.6111097216316934,
      "grad_norm": 0.0013250740012153983,
      "learning_rate": 2.0398732951724952e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 43417664,
      "step": 13050
    },
    {
      "epoch": 2.6121101468123955,
      "grad_norm": 0.0141685726121068,
      "learning_rate": 2.0295218149633034e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 43434624,
      "step": 13055
    },
    {
      "epoch": 2.613110571993097,
      "grad_norm": 0.004939412698149681,
      "learning_rate": 2.0191955549550447e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 43450432,
      "step": 13060
    },
    {
      "epoch": 2.614110997173799,
      "grad_norm": 0.20622195303440094,
      "learning_rate": 2.0088945264853137e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 43467456,
      "step": 13065
    },
    {
      "epoch": 2.6151114223545004,
      "grad_norm": 0.7119448184967041,
      "learning_rate": 1.9986187408640304e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 43484384,
      "step": 13070
    },
    {
      "epoch": 2.6161118475352025,
      "grad_norm": 0.006884606555104256,
      "learning_rate": 1.9883682093733786e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 43500160,
      "step": 13075
    },
    {
      "epoch": 2.617112272715904,
      "grad_norm": 0.4917401373386383,
      "learning_rate": 1.9781429432678274e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 43517472,
      "step": 13080
    },
    {
      "epoch": 2.6181126978966063,
      "grad_norm": 0.22076082229614258,
      "learning_rate": 1.9679429537741103e-06,
      "loss": 0.0136,
      "num_input_tokens_seen": 43533408,
      "step": 13085
    },
    {
      "epoch": 2.619113123077308,
      "grad_norm": 1.2674239873886108,
      "learning_rate": 1.957768252091191e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 43550752,
      "step": 13090
    },
    {
      "epoch": 2.6201135482580096,
      "grad_norm": 0.01584111526608467,
      "learning_rate": 1.9476188493902813e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 43567296,
      "step": 13095
    },
    {
      "epoch": 2.621113973438711,
      "grad_norm": 0.2965255081653595,
      "learning_rate": 1.93749475681482e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 43583232,
      "step": 13100
    },
    {
      "epoch": 2.6221143986194133,
      "grad_norm": 0.2845730483531952,
      "learning_rate": 1.9273959854804347e-06,
      "loss": 0.0045,
      "num_input_tokens_seen": 43599840,
      "step": 13105
    },
    {
      "epoch": 2.623114823800115,
      "grad_norm": 0.8512167930603027,
      "learning_rate": 1.9173225464749865e-06,
      "loss": 0.0076,
      "num_input_tokens_seen": 43616352,
      "step": 13110
    },
    {
      "epoch": 2.624115248980817,
      "grad_norm": 1.9393882751464844,
      "learning_rate": 1.9072744508585011e-06,
      "loss": 0.0153,
      "num_input_tokens_seen": 43632480,
      "step": 13115
    },
    {
      "epoch": 2.6251156741615187,
      "grad_norm": 0.04237949475646019,
      "learning_rate": 1.8972517096631725e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 43649088,
      "step": 13120
    },
    {
      "epoch": 2.6261160993422203,
      "grad_norm": 0.04926257207989693,
      "learning_rate": 1.8872543338933762e-06,
      "loss": 0.0015,
      "num_input_tokens_seen": 43666944,
      "step": 13125
    },
    {
      "epoch": 2.627116524522922,
      "grad_norm": 0.2511679232120514,
      "learning_rate": 1.8772823345256207e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 43683136,
      "step": 13130
    },
    {
      "epoch": 2.628116949703624,
      "grad_norm": 1.242906093597412,
      "learning_rate": 1.8673357225085646e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 43701184,
      "step": 13135
    },
    {
      "epoch": 2.6291173748843257,
      "grad_norm": 0.09425300359725952,
      "learning_rate": 1.857414508762992e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 43717408,
      "step": 13140
    },
    {
      "epoch": 2.630117800065028,
      "grad_norm": 0.008755037561058998,
      "learning_rate": 1.8475187041817887e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 43733312,
      "step": 13145
    },
    {
      "epoch": 2.6311182252457295,
      "grad_norm": 2.027099847793579,
      "learning_rate": 1.837648319629956e-06,
      "loss": 0.005,
      "num_input_tokens_seen": 43750496,
      "step": 13150
    },
    {
      "epoch": 2.632118650426431,
      "grad_norm": 1.6400530338287354,
      "learning_rate": 1.8278033659445831e-06,
      "loss": 0.0052,
      "num_input_tokens_seen": 43767776,
      "step": 13155
    },
    {
      "epoch": 2.6331190756071328,
      "grad_norm": 0.012904511764645576,
      "learning_rate": 1.8179838539348376e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 43785440,
      "step": 13160
    },
    {
      "epoch": 2.634119500787835,
      "grad_norm": 0.36115771532058716,
      "learning_rate": 1.8081897943819453e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 43801856,
      "step": 13165
    },
    {
      "epoch": 2.6351199259685365,
      "grad_norm": 0.31264355778694153,
      "learning_rate": 1.7984211980392045e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 43817696,
      "step": 13170
    },
    {
      "epoch": 2.6361203511492386,
      "grad_norm": 0.008913216181099415,
      "learning_rate": 1.7886780756319333e-06,
      "loss": 0.0005,
      "num_input_tokens_seen": 43835584,
      "step": 13175
    },
    {
      "epoch": 2.6371207763299402,
      "grad_norm": 0.05168372765183449,
      "learning_rate": 1.7789604378574965e-06,
      "loss": 0.0038,
      "num_input_tokens_seen": 43852000,
      "step": 13180
    },
    {
      "epoch": 2.638121201510642,
      "grad_norm": 0.008873915299773216,
      "learning_rate": 1.7692682953852819e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 43868608,
      "step": 13185
    },
    {
      "epoch": 2.639121626691344,
      "grad_norm": 1.931552767753601,
      "learning_rate": 1.759601658856669e-06,
      "loss": 0.0046,
      "num_input_tokens_seen": 43885120,
      "step": 13190
    },
    {
      "epoch": 2.6401220518720456,
      "grad_norm": 0.883804440498352,
      "learning_rate": 1.7499605388850482e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 43901280,
      "step": 13195
    },
    {
      "epoch": 2.6411224770527473,
      "grad_norm": 0.09361104667186737,
      "learning_rate": 1.740344946055783e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 43916736,
      "step": 13200
    },
    {
      "epoch": 2.6421229022334494,
      "grad_norm": 0.7549806237220764,
      "learning_rate": 1.7307548909262117e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 43934464,
      "step": 13205
    },
    {
      "epoch": 2.643123327414151,
      "grad_norm": 0.11326675862073898,
      "learning_rate": 1.7211903840256532e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 43950752,
      "step": 13210
    },
    {
      "epoch": 2.6441237525948527,
      "grad_norm": 0.11757057160139084,
      "learning_rate": 1.7116514358553438e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 43967776,
      "step": 13215
    },
    {
      "epoch": 2.6451241777755548,
      "grad_norm": 0.6157832741737366,
      "learning_rate": 1.7021380568884803e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 43985184,
      "step": 13220
    },
    {
      "epoch": 2.6461246029562564,
      "grad_norm": 0.09997295588254929,
      "learning_rate": 1.692650257570183e-06,
      "loss": 0.0073,
      "num_input_tokens_seen": 44002176,
      "step": 13225
    },
    {
      "epoch": 2.647125028136958,
      "grad_norm": 0.20389370620250702,
      "learning_rate": 1.6831880483174771e-06,
      "loss": 0.0005,
      "num_input_tokens_seen": 44019776,
      "step": 13230
    },
    {
      "epoch": 2.64812545331766,
      "grad_norm": 0.6792962551116943,
      "learning_rate": 1.6737514395192972e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 44036736,
      "step": 13235
    },
    {
      "epoch": 2.649125878498362,
      "grad_norm": 0.35949239134788513,
      "learning_rate": 1.66434044153648e-06,
      "loss": 0.0109,
      "num_input_tokens_seen": 44053472,
      "step": 13240
    },
    {
      "epoch": 2.6501263036790634,
      "grad_norm": 0.00013263363507576287,
      "learning_rate": 1.654955064701727e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 44070016,
      "step": 13245
    },
    {
      "epoch": 2.6511267288597655,
      "grad_norm": 0.24277490377426147,
      "learning_rate": 1.6455953193196205e-06,
      "loss": 0.0049,
      "num_input_tokens_seen": 44087328,
      "step": 13250
    },
    {
      "epoch": 2.652127154040467,
      "grad_norm": 0.12013000249862671,
      "learning_rate": 1.6362612156665903e-06,
      "loss": 0.0093,
      "num_input_tokens_seen": 44104768,
      "step": 13255
    },
    {
      "epoch": 2.6531275792211693,
      "grad_norm": 0.14752374589443207,
      "learning_rate": 1.6269527639909215e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 44120608,
      "step": 13260
    },
    {
      "epoch": 2.654128004401871,
      "grad_norm": 0.9413996934890747,
      "learning_rate": 1.6176699745127422e-06,
      "loss": 0.0076,
      "num_input_tokens_seen": 44136992,
      "step": 13265
    },
    {
      "epoch": 2.6551284295825726,
      "grad_norm": 0.035456474870443344,
      "learning_rate": 1.6084128574239881e-06,
      "loss": 0.0051,
      "num_input_tokens_seen": 44153280,
      "step": 13270
    },
    {
      "epoch": 2.656128854763274,
      "grad_norm": 0.007137295324355364,
      "learning_rate": 1.5991814228884156e-06,
      "loss": 0.0002,
      "num_input_tokens_seen": 44169376,
      "step": 13275
    },
    {
      "epoch": 2.6571292799439763,
      "grad_norm": 0.014517224393785,
      "learning_rate": 1.589975681041589e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 44185472,
      "step": 13280
    },
    {
      "epoch": 2.658129705124678,
      "grad_norm": 1.2557194232940674,
      "learning_rate": 1.580795641990851e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 44202816,
      "step": 13285
    },
    {
      "epoch": 2.65913013030538,
      "grad_norm": 0.022609582170844078,
      "learning_rate": 1.5716413158153338e-06,
      "loss": 0.0071,
      "num_input_tokens_seen": 44219648,
      "step": 13290
    },
    {
      "epoch": 2.6601305554860817,
      "grad_norm": 0.11861256510019302,
      "learning_rate": 1.5625127125659445e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 44235232,
      "step": 13295
    },
    {
      "epoch": 2.6611309806667833,
      "grad_norm": 0.17032299935817719,
      "learning_rate": 1.5534098422653243e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 44251552,
      "step": 13300
    },
    {
      "epoch": 2.662131405847485,
      "grad_norm": 0.22129566967487335,
      "learning_rate": 1.544332714907884e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 44268704,
      "step": 13305
    },
    {
      "epoch": 2.663131831028187,
      "grad_norm": 0.0006550539401359856,
      "learning_rate": 1.5352813404597655e-06,
      "loss": 0.0054,
      "num_input_tokens_seen": 44284768,
      "step": 13310
    },
    {
      "epoch": 2.6641322562088887,
      "grad_norm": 0.47652092576026917,
      "learning_rate": 1.5262557288588275e-06,
      "loss": 0.0075,
      "num_input_tokens_seen": 44301184,
      "step": 13315
    },
    {
      "epoch": 2.665132681389591,
      "grad_norm": 0.6773017048835754,
      "learning_rate": 1.5172558900146517e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 44318208,
      "step": 13320
    },
    {
      "epoch": 2.6661331065702925,
      "grad_norm": 0.3130875527858734,
      "learning_rate": 1.5082818338085142e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 44335872,
      "step": 13325
    },
    {
      "epoch": 2.667133531750994,
      "grad_norm": 1.1312685012817383,
      "learning_rate": 1.499333570093392e-06,
      "loss": 0.0026,
      "num_input_tokens_seen": 44353184,
      "step": 13330
    },
    {
      "epoch": 2.6681339569316957,
      "grad_norm": 0.06901410967111588,
      "learning_rate": 1.4904111086939398e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 44370592,
      "step": 13335
    },
    {
      "epoch": 2.669134382112398,
      "grad_norm": 0.028959838673472404,
      "learning_rate": 1.4815144594064773e-06,
      "loss": 0.0052,
      "num_input_tokens_seen": 44387680,
      "step": 13340
    },
    {
      "epoch": 2.6701348072930995,
      "grad_norm": 0.4723975956439972,
      "learning_rate": 1.4726436319989933e-06,
      "loss": 0.0035,
      "num_input_tokens_seen": 44403680,
      "step": 13345
    },
    {
      "epoch": 2.6711352324738016,
      "grad_norm": 0.8988831043243408,
      "learning_rate": 1.463798636211125e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 44419552,
      "step": 13350
    },
    {
      "epoch": 2.6721356576545032,
      "grad_norm": 0.9595151543617249,
      "learning_rate": 1.4549794817541374e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 44435808,
      "step": 13355
    },
    {
      "epoch": 2.673136082835205,
      "grad_norm": 0.26894503831863403,
      "learning_rate": 1.446186178310932e-06,
      "loss": 0.0011,
      "num_input_tokens_seen": 44452864,
      "step": 13360
    },
    {
      "epoch": 2.6741365080159065,
      "grad_norm": 0.024321796372532845,
      "learning_rate": 1.4374187355360307e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 44469280,
      "step": 13365
    },
    {
      "epoch": 2.6751369331966086,
      "grad_norm": 0.059120357036590576,
      "learning_rate": 1.4286771630555557e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 44486816,
      "step": 13370
    },
    {
      "epoch": 2.6761373583773103,
      "grad_norm": 0.024076877161860466,
      "learning_rate": 1.4199614704672293e-06,
      "loss": 0.0004,
      "num_input_tokens_seen": 44503808,
      "step": 13375
    },
    {
      "epoch": 2.6771377835580124,
      "grad_norm": 0.00032420014031231403,
      "learning_rate": 1.4112716673403497e-06,
      "loss": 0.0042,
      "num_input_tokens_seen": 44520416,
      "step": 13380
    },
    {
      "epoch": 2.678138208738714,
      "grad_norm": 0.47371798753738403,
      "learning_rate": 1.402607763215802e-06,
      "loss": 0.0047,
      "num_input_tokens_seen": 44537248,
      "step": 13385
    },
    {
      "epoch": 2.6791386339194156,
      "grad_norm": 0.649607241153717,
      "learning_rate": 1.393969767606032e-06,
      "loss": 0.0081,
      "num_input_tokens_seen": 44553600,
      "step": 13390
    },
    {
      "epoch": 2.6801390591001173,
      "grad_norm": 0.12219960242509842,
      "learning_rate": 1.3853576899950344e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 44571520,
      "step": 13395
    },
    {
      "epoch": 2.6811394842808194,
      "grad_norm": 0.9294015169143677,
      "learning_rate": 1.3767715398383507e-06,
      "loss": 0.0097,
      "num_input_tokens_seen": 44588000,
      "step": 13400
    },
    {
      "epoch": 2.682139909461521,
      "grad_norm": 0.6352314352989197,
      "learning_rate": 1.3682113265630652e-06,
      "loss": 0.0091,
      "num_input_tokens_seen": 44604960,
      "step": 13405
    },
    {
      "epoch": 2.683140334642223,
      "grad_norm": 0.12153366208076477,
      "learning_rate": 1.3596770595677655e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 44620448,
      "step": 13410
    },
    {
      "epoch": 2.6841407598229248,
      "grad_norm": 0.23107106983661652,
      "learning_rate": 1.35116874822257e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 44638272,
      "step": 13415
    },
    {
      "epoch": 2.6851411850036264,
      "grad_norm": 0.07143131643533707,
      "learning_rate": 1.3426864018690865e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 44655136,
      "step": 13420
    },
    {
      "epoch": 2.686141610184328,
      "grad_norm": 1.5513947010040283,
      "learning_rate": 1.334230029820424e-06,
      "loss": 0.0045,
      "num_input_tokens_seen": 44671808,
      "step": 13425
    },
    {
      "epoch": 2.68714203536503,
      "grad_norm": 0.32228389382362366,
      "learning_rate": 1.3257996413611695e-06,
      "loss": 0.0101,
      "num_input_tokens_seen": 44688448,
      "step": 13430
    },
    {
      "epoch": 2.688142460545732,
      "grad_norm": 0.09518205374479294,
      "learning_rate": 1.3173952457473798e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 44704736,
      "step": 13435
    },
    {
      "epoch": 2.689142885726434,
      "grad_norm": 0.04356449842453003,
      "learning_rate": 1.3090168522065738e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 44722400,
      "step": 13440
    },
    {
      "epoch": 2.6901433109071355,
      "grad_norm": 0.25289589166641235,
      "learning_rate": 1.300664469937729e-06,
      "loss": 0.0033,
      "num_input_tokens_seen": 44740000,
      "step": 13445
    },
    {
      "epoch": 2.691143736087837,
      "grad_norm": 1.0440763235092163,
      "learning_rate": 1.292338108111249e-06,
      "loss": 0.0083,
      "num_input_tokens_seen": 44757248,
      "step": 13450
    },
    {
      "epoch": 2.6921441612685393,
      "grad_norm": 3.475522041320801,
      "learning_rate": 1.284037775868982e-06,
      "loss": 0.0082,
      "num_input_tokens_seen": 44774400,
      "step": 13455
    },
    {
      "epoch": 2.693144586449241,
      "grad_norm": 0.00039189582457765937,
      "learning_rate": 1.2757634823241938e-06,
      "loss": 0.0015,
      "num_input_tokens_seen": 44790560,
      "step": 13460
    },
    {
      "epoch": 2.6941450116299426,
      "grad_norm": 0.11311087757349014,
      "learning_rate": 1.2675152365615533e-06,
      "loss": 0.0007,
      "num_input_tokens_seen": 44808032,
      "step": 13465
    },
    {
      "epoch": 2.6951454368106447,
      "grad_norm": 1.334816813468933,
      "learning_rate": 1.25929304763715e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 44824448,
      "step": 13470
    },
    {
      "epoch": 2.6961458619913463,
      "grad_norm": 0.06325584650039673,
      "learning_rate": 1.2510969245784376e-06,
      "loss": 0.0071,
      "num_input_tokens_seen": 44841440,
      "step": 13475
    },
    {
      "epoch": 2.697146287172048,
      "grad_norm": 0.4612632691860199,
      "learning_rate": 1.242926876384276e-06,
      "loss": 0.0054,
      "num_input_tokens_seen": 44857696,
      "step": 13480
    },
    {
      "epoch": 2.69814671235275,
      "grad_norm": 0.0031939514447003603,
      "learning_rate": 1.2347829120248844e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 44874656,
      "step": 13485
    },
    {
      "epoch": 2.6991471375334517,
      "grad_norm": 0.07405953854322433,
      "learning_rate": 1.2266650404418379e-06,
      "loss": 0.0053,
      "num_input_tokens_seen": 44891328,
      "step": 13490
    },
    {
      "epoch": 2.700147562714154,
      "grad_norm": 0.0022581659723073244,
      "learning_rate": 1.2185732705480762e-06,
      "loss": 0.0041,
      "num_input_tokens_seen": 44908160,
      "step": 13495
    },
    {
      "epoch": 2.7011479878948554,
      "grad_norm": 0.7377486824989319,
      "learning_rate": 1.2105076112278763e-06,
      "loss": 0.0101,
      "num_input_tokens_seen": 44924672,
      "step": 13500
    },
    {
      "epoch": 2.702148413075557,
      "grad_norm": 0.6490694284439087,
      "learning_rate": 1.202468071336843e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 44940608,
      "step": 13505
    },
    {
      "epoch": 2.7031488382562587,
      "grad_norm": 0.9131351709365845,
      "learning_rate": 1.1944546597019046e-06,
      "loss": 0.0032,
      "num_input_tokens_seen": 44956704,
      "step": 13510
    },
    {
      "epoch": 2.704149263436961,
      "grad_norm": 1.0523442029953003,
      "learning_rate": 1.1864673851213065e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 44973280,
      "step": 13515
    },
    {
      "epoch": 2.7051496886176625,
      "grad_norm": 0.2591552734375,
      "learning_rate": 1.1785062563645976e-06,
      "loss": 0.0047,
      "num_input_tokens_seen": 44989984,
      "step": 13520
    },
    {
      "epoch": 2.7061501137983646,
      "grad_norm": 1.9421402215957642,
      "learning_rate": 1.1705712821726195e-06,
      "loss": 0.0084,
      "num_input_tokens_seen": 45005920,
      "step": 13525
    },
    {
      "epoch": 2.707150538979066,
      "grad_norm": 0.03383360430598259,
      "learning_rate": 1.1626624712574862e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 45022080,
      "step": 13530
    },
    {
      "epoch": 2.708150964159768,
      "grad_norm": 0.04191024228930473,
      "learning_rate": 1.1547798323025994e-06,
      "loss": 0.0063,
      "num_input_tokens_seen": 45038624,
      "step": 13535
    },
    {
      "epoch": 2.7091513893404695,
      "grad_norm": 0.12199781090021133,
      "learning_rate": 1.14692337396263e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 45055072,
      "step": 13540
    },
    {
      "epoch": 2.7101518145211716,
      "grad_norm": 0.04170893877744675,
      "learning_rate": 1.139093104863484e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 45072096,
      "step": 13545
    },
    {
      "epoch": 2.7111522397018732,
      "grad_norm": 0.03867083787918091,
      "learning_rate": 1.1312890336023285e-06,
      "loss": 0.0047,
      "num_input_tokens_seen": 45087264,
      "step": 13550
    },
    {
      "epoch": 2.7121526648825753,
      "grad_norm": 0.011973058804869652,
      "learning_rate": 1.1235111687475707e-06,
      "loss": 0.009,
      "num_input_tokens_seen": 45103136,
      "step": 13555
    },
    {
      "epoch": 2.713153090063277,
      "grad_norm": 0.05495055764913559,
      "learning_rate": 1.115759518838827e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 45119872,
      "step": 13560
    },
    {
      "epoch": 2.7141535152439786,
      "grad_norm": 0.5237078666687012,
      "learning_rate": 1.1080340923869475e-06,
      "loss": 0.0054,
      "num_input_tokens_seen": 45136736,
      "step": 13565
    },
    {
      "epoch": 2.7151539404246803,
      "grad_norm": 1.1536104679107666,
      "learning_rate": 1.1003348978739836e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 45153312,
      "step": 13570
    },
    {
      "epoch": 2.7161543656053824,
      "grad_norm": 1.8903820514678955,
      "learning_rate": 1.0926619437531898e-06,
      "loss": 0.0069,
      "num_input_tokens_seen": 45169632,
      "step": 13575
    },
    {
      "epoch": 2.717154790786084,
      "grad_norm": 0.9461204409599304,
      "learning_rate": 1.0850152384490081e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 45186368,
      "step": 13580
    },
    {
      "epoch": 2.718155215966786,
      "grad_norm": 0.07230707257986069,
      "learning_rate": 1.077394790357053e-06,
      "loss": 0.0072,
      "num_input_tokens_seen": 45202592,
      "step": 13585
    },
    {
      "epoch": 2.7191556411474878,
      "grad_norm": 0.8314633965492249,
      "learning_rate": 1.069800607844121e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 45220000,
      "step": 13590
    },
    {
      "epoch": 2.7201560663281894,
      "grad_norm": 0.5452732443809509,
      "learning_rate": 1.0622326992481695e-06,
      "loss": 0.0059,
      "num_input_tokens_seen": 45236960,
      "step": 13595
    },
    {
      "epoch": 2.721156491508891,
      "grad_norm": 0.2344525009393692,
      "learning_rate": 1.0546910728783027e-06,
      "loss": 0.0015,
      "num_input_tokens_seen": 45255232,
      "step": 13600
    },
    {
      "epoch": 2.722156916689593,
      "grad_norm": 1.2953059673309326,
      "learning_rate": 1.0471757370147745e-06,
      "loss": 0.0193,
      "num_input_tokens_seen": 45272576,
      "step": 13605
    },
    {
      "epoch": 2.723157341870295,
      "grad_norm": 0.07832599431276321,
      "learning_rate": 1.0396866999089704e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 45288352,
      "step": 13610
    },
    {
      "epoch": 2.724157767050997,
      "grad_norm": 0.4092762768268585,
      "learning_rate": 1.0322239697833997e-06,
      "loss": 0.0058,
      "num_input_tokens_seen": 45305280,
      "step": 13615
    },
    {
      "epoch": 2.7251581922316985,
      "grad_norm": 0.7260483503341675,
      "learning_rate": 1.024787554831691e-06,
      "loss": 0.0096,
      "num_input_tokens_seen": 45320320,
      "step": 13620
    },
    {
      "epoch": 2.7261586174124,
      "grad_norm": 0.014826206490397453,
      "learning_rate": 1.0173774632185801e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 45337760,
      "step": 13625
    },
    {
      "epoch": 2.727159042593102,
      "grad_norm": 0.0008843060932122171,
      "learning_rate": 1.0099937030799034e-06,
      "loss": 0.004,
      "num_input_tokens_seen": 45355744,
      "step": 13630
    },
    {
      "epoch": 2.728159467773804,
      "grad_norm": 0.1502399891614914,
      "learning_rate": 1.0026362825225844e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 45371488,
      "step": 13635
    },
    {
      "epoch": 2.7291598929545056,
      "grad_norm": 0.02906009927392006,
      "learning_rate": 9.953052096246224e-07,
      "loss": 0.0056,
      "num_input_tokens_seen": 45387584,
      "step": 13640
    },
    {
      "epoch": 2.7301603181352077,
      "grad_norm": 0.042441483587026596,
      "learning_rate": 9.880004924350984e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 45403968,
      "step": 13645
    },
    {
      "epoch": 2.7311607433159093,
      "grad_norm": 0.10809578746557236,
      "learning_rate": 9.80722138974155e-07,
      "loss": 0.0058,
      "num_input_tokens_seen": 45420320,
      "step": 13650
    },
    {
      "epoch": 2.732161168496611,
      "grad_norm": 0.007705936208367348,
      "learning_rate": 9.734701572329753e-07,
      "loss": 0.0015,
      "num_input_tokens_seen": 45436672,
      "step": 13655
    },
    {
      "epoch": 2.7331615936773126,
      "grad_norm": 0.012891904450953007,
      "learning_rate": 9.662445551738064e-07,
      "loss": 0.0065,
      "num_input_tokens_seen": 45453440,
      "step": 13660
    },
    {
      "epoch": 2.7341620188580147,
      "grad_norm": 0.10762783885002136,
      "learning_rate": 9.59045340729925e-07,
      "loss": 0.0055,
      "num_input_tokens_seen": 45469312,
      "step": 13665
    },
    {
      "epoch": 2.7351624440387163,
      "grad_norm": 1.0634167194366455,
      "learning_rate": 9.518725218056246e-07,
      "loss": 0.0036,
      "num_input_tokens_seen": 45484416,
      "step": 13670
    },
    {
      "epoch": 2.7361628692194184,
      "grad_norm": 1.4050567150115967,
      "learning_rate": 9.447261062762419e-07,
      "loss": 0.0052,
      "num_input_tokens_seen": 45500544,
      "step": 13675
    },
    {
      "epoch": 2.73716329440012,
      "grad_norm": 1.091619849205017,
      "learning_rate": 9.376061019881005e-07,
      "loss": 0.0081,
      "num_input_tokens_seen": 45517664,
      "step": 13680
    },
    {
      "epoch": 2.7381637195808217,
      "grad_norm": 0.01584814302623272,
      "learning_rate": 9.305125167585388e-07,
      "loss": 0.0036,
      "num_input_tokens_seen": 45535296,
      "step": 13685
    },
    {
      "epoch": 2.739164144761524,
      "grad_norm": 0.1886490136384964,
      "learning_rate": 9.234453583758906e-07,
      "loss": 0.001,
      "num_input_tokens_seen": 45552128,
      "step": 13690
    },
    {
      "epoch": 2.7401645699422255,
      "grad_norm": 0.009118461050093174,
      "learning_rate": 9.164046345994604e-07,
      "loss": 0.0085,
      "num_input_tokens_seen": 45568352,
      "step": 13695
    },
    {
      "epoch": 2.741164995122927,
      "grad_norm": 0.0091153709217906,
      "learning_rate": 9.093903531595476e-07,
      "loss": 0.0023,
      "num_input_tokens_seen": 45584512,
      "step": 13700
    },
    {
      "epoch": 2.742165420303629,
      "grad_norm": 0.3516223132610321,
      "learning_rate": 9.024025217574089e-07,
      "loss": 0.0065,
      "num_input_tokens_seen": 45600352,
      "step": 13705
    },
    {
      "epoch": 2.743165845484331,
      "grad_norm": 1.48143470287323,
      "learning_rate": 8.954411480652569e-07,
      "loss": 0.0201,
      "num_input_tokens_seen": 45616160,
      "step": 13710
    },
    {
      "epoch": 2.7441662706650325,
      "grad_norm": 0.9060990810394287,
      "learning_rate": 8.885062397262639e-07,
      "loss": 0.0026,
      "num_input_tokens_seen": 45632736,
      "step": 13715
    },
    {
      "epoch": 2.7451666958457346,
      "grad_norm": 0.10566138476133347,
      "learning_rate": 8.815978043545392e-07,
      "loss": 0.0035,
      "num_input_tokens_seen": 45649696,
      "step": 13720
    },
    {
      "epoch": 2.7461671210264362,
      "grad_norm": 0.3342597186565399,
      "learning_rate": 8.747158495351348e-07,
      "loss": 0.0007,
      "num_input_tokens_seen": 45665856,
      "step": 13725
    },
    {
      "epoch": 2.747167546207138,
      "grad_norm": 0.643731415271759,
      "learning_rate": 8.678603828240178e-07,
      "loss": 0.0034,
      "num_input_tokens_seen": 45682464,
      "step": 13730
    },
    {
      "epoch": 2.74816797138784,
      "grad_norm": 1.8434544801712036,
      "learning_rate": 8.610314117480783e-07,
      "loss": 0.0098,
      "num_input_tokens_seen": 45699456,
      "step": 13735
    },
    {
      "epoch": 2.7491683965685416,
      "grad_norm": 3.394533634185791,
      "learning_rate": 8.542289438051132e-07,
      "loss": 0.0203,
      "num_input_tokens_seen": 45716512,
      "step": 13740
    },
    {
      "epoch": 2.7501688217492433,
      "grad_norm": 0.34342724084854126,
      "learning_rate": 8.474529864638259e-07,
      "loss": 0.0071,
      "num_input_tokens_seen": 45733312,
      "step": 13745
    },
    {
      "epoch": 2.7511692469299454,
      "grad_norm": 0.0001394088176311925,
      "learning_rate": 8.40703547163807e-07,
      "loss": 0.0039,
      "num_input_tokens_seen": 45749280,
      "step": 13750
    },
    {
      "epoch": 2.752169672110647,
      "grad_norm": 0.177780881524086,
      "learning_rate": 8.339806333155342e-07,
      "loss": 0.0046,
      "num_input_tokens_seen": 45766208,
      "step": 13755
    },
    {
      "epoch": 2.753170097291349,
      "grad_norm": 0.6227426528930664,
      "learning_rate": 8.272842523003643e-07,
      "loss": 0.0019,
      "num_input_tokens_seen": 45781952,
      "step": 13760
    },
    {
      "epoch": 2.7541705224720507,
      "grad_norm": 0.005509722046554089,
      "learning_rate": 8.206144114705134e-07,
      "loss": 0.0038,
      "num_input_tokens_seen": 45798560,
      "step": 13765
    },
    {
      "epoch": 2.7551709476527524,
      "grad_norm": 0.0030598777811974287,
      "learning_rate": 8.139711181490706e-07,
      "loss": 0.0003,
      "num_input_tokens_seen": 45814656,
      "step": 13770
    },
    {
      "epoch": 2.756171372833454,
      "grad_norm": 1.6217525005340576,
      "learning_rate": 8.07354379629971e-07,
      "loss": 0.0151,
      "num_input_tokens_seen": 45832032,
      "step": 13775
    },
    {
      "epoch": 2.757171798014156,
      "grad_norm": 0.009558570571243763,
      "learning_rate": 8.00764203177995e-07,
      "loss": 0.0071,
      "num_input_tokens_seen": 45849120,
      "step": 13780
    },
    {
      "epoch": 2.7581722231948578,
      "grad_norm": 0.3835294246673584,
      "learning_rate": 7.942005960287602e-07,
      "loss": 0.0015,
      "num_input_tokens_seen": 45865856,
      "step": 13785
    },
    {
      "epoch": 2.75917264837556,
      "grad_norm": 0.00481353048235178,
      "learning_rate": 7.876635653887076e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 45882784,
      "step": 13790
    },
    {
      "epoch": 2.7601730735562615,
      "grad_norm": 0.06251227110624313,
      "learning_rate": 7.811531184351073e-07,
      "loss": 0.0002,
      "num_input_tokens_seen": 45899136,
      "step": 13795
    },
    {
      "epoch": 2.761173498736963,
      "grad_norm": 0.011212627403438091,
      "learning_rate": 7.746692623160357e-07,
      "loss": 0.0062,
      "num_input_tokens_seen": 45916512,
      "step": 13800
    },
    {
      "epoch": 2.762173923917665,
      "grad_norm": 0.6980820894241333,
      "learning_rate": 7.68212004150376e-07,
      "loss": 0.0042,
      "num_input_tokens_seen": 45933120,
      "step": 13805
    },
    {
      "epoch": 2.763174349098367,
      "grad_norm": 0.00017238684813492,
      "learning_rate": 7.617813510278071e-07,
      "loss": 0.0039,
      "num_input_tokens_seen": 45950368,
      "step": 13810
    },
    {
      "epoch": 2.7641747742790685,
      "grad_norm": 0.021036557853221893,
      "learning_rate": 7.553773100088035e-07,
      "loss": 0.0014,
      "num_input_tokens_seen": 45965664,
      "step": 13815
    },
    {
      "epoch": 2.7651751994597706,
      "grad_norm": 0.16205120086669922,
      "learning_rate": 7.489998881246102e-07,
      "loss": 0.0071,
      "num_input_tokens_seen": 45981760,
      "step": 13820
    },
    {
      "epoch": 2.7661756246404723,
      "grad_norm": 1.0853997468948364,
      "learning_rate": 7.426490923772567e-07,
      "loss": 0.0086,
      "num_input_tokens_seen": 45997216,
      "step": 13825
    },
    {
      "epoch": 2.767176049821174,
      "grad_norm": 0.02892439253628254,
      "learning_rate": 7.363249297395319e-07,
      "loss": 0.0038,
      "num_input_tokens_seen": 46015552,
      "step": 13830
    },
    {
      "epoch": 2.7681764750018756,
      "grad_norm": 0.5620378255844116,
      "learning_rate": 7.300274071549873e-07,
      "loss": 0.0077,
      "num_input_tokens_seen": 46032256,
      "step": 13835
    },
    {
      "epoch": 2.7691769001825777,
      "grad_norm": 2.367673397064209,
      "learning_rate": 7.237565315379225e-07,
      "loss": 0.0063,
      "num_input_tokens_seen": 46048928,
      "step": 13840
    },
    {
      "epoch": 2.7701773253632793,
      "grad_norm": 0.2110394984483719,
      "learning_rate": 7.175123097733772e-07,
      "loss": 0.0006,
      "num_input_tokens_seen": 46065184,
      "step": 13845
    },
    {
      "epoch": 2.7711777505439814,
      "grad_norm": 0.0275700893253088,
      "learning_rate": 7.112947487171367e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 46081472,
      "step": 13850
    },
    {
      "epoch": 2.772178175724683,
      "grad_norm": 0.027688276022672653,
      "learning_rate": 7.051038551957073e-07,
      "loss": 0.0007,
      "num_input_tokens_seen": 46098368,
      "step": 13855
    },
    {
      "epoch": 2.7731786009053847,
      "grad_norm": 0.29458293318748474,
      "learning_rate": 6.9893963600631e-07,
      "loss": 0.0146,
      "num_input_tokens_seen": 46115424,
      "step": 13860
    },
    {
      "epoch": 2.7741790260860864,
      "grad_norm": 0.04649446904659271,
      "learning_rate": 6.928020979168948e-07,
      "loss": 0.0007,
      "num_input_tokens_seen": 46132352,
      "step": 13865
    },
    {
      "epoch": 2.7751794512667884,
      "grad_norm": 0.3606918454170227,
      "learning_rate": 6.866912476661075e-07,
      "loss": 0.0023,
      "num_input_tokens_seen": 46149792,
      "step": 13870
    },
    {
      "epoch": 2.77617987644749,
      "grad_norm": 0.058070261031389236,
      "learning_rate": 6.80607091963284e-07,
      "loss": 0.0031,
      "num_input_tokens_seen": 46166400,
      "step": 13875
    },
    {
      "epoch": 2.777180301628192,
      "grad_norm": 0.01889689825475216,
      "learning_rate": 6.745496374884725e-07,
      "loss": 0.0053,
      "num_input_tokens_seen": 46183104,
      "step": 13880
    },
    {
      "epoch": 2.778180726808894,
      "grad_norm": 0.8481465578079224,
      "learning_rate": 6.685188908923917e-07,
      "loss": 0.0125,
      "num_input_tokens_seen": 46200160,
      "step": 13885
    },
    {
      "epoch": 2.7791811519895955,
      "grad_norm": 1.760446310043335,
      "learning_rate": 6.625148587964286e-07,
      "loss": 0.0069,
      "num_input_tokens_seen": 46217792,
      "step": 13890
    },
    {
      "epoch": 2.780181577170297,
      "grad_norm": 0.020744048058986664,
      "learning_rate": 6.5653754779266e-07,
      "loss": 0.004,
      "num_input_tokens_seen": 46232832,
      "step": 13895
    },
    {
      "epoch": 2.781182002350999,
      "grad_norm": 0.008959859609603882,
      "learning_rate": 6.505869644438057e-07,
      "loss": 0.0004,
      "num_input_tokens_seen": 46250304,
      "step": 13900
    },
    {
      "epoch": 2.782182427531701,
      "grad_norm": 0.021957986056804657,
      "learning_rate": 6.446631152832478e-07,
      "loss": 0.0038,
      "num_input_tokens_seen": 46266752,
      "step": 13905
    },
    {
      "epoch": 2.783182852712403,
      "grad_norm": 0.4723031520843506,
      "learning_rate": 6.387660068150225e-07,
      "loss": 0.0049,
      "num_input_tokens_seen": 46283552,
      "step": 13910
    },
    {
      "epoch": 2.7841832778931046,
      "grad_norm": 0.7204026579856873,
      "learning_rate": 6.328956455137924e-07,
      "loss": 0.002,
      "num_input_tokens_seen": 46300768,
      "step": 13915
    },
    {
      "epoch": 2.7851837030738062,
      "grad_norm": 0.003451480297371745,
      "learning_rate": 6.270520378248601e-07,
      "loss": 0.0132,
      "num_input_tokens_seen": 46317984,
      "step": 13920
    },
    {
      "epoch": 2.786184128254508,
      "grad_norm": 0.012446947395801544,
      "learning_rate": 6.212351901641628e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 46333856,
      "step": 13925
    },
    {
      "epoch": 2.78718455343521,
      "grad_norm": 0.005464858375489712,
      "learning_rate": 6.154451089182389e-07,
      "loss": 0.001,
      "num_input_tokens_seen": 46350912,
      "step": 13930
    },
    {
      "epoch": 2.7881849786159116,
      "grad_norm": 0.0008582032169215381,
      "learning_rate": 6.096818004442533e-07,
      "loss": 0.0002,
      "num_input_tokens_seen": 46367840,
      "step": 13935
    },
    {
      "epoch": 2.7891854037966137,
      "grad_norm": 0.08223076909780502,
      "learning_rate": 6.03945271069975e-07,
      "loss": 0.0097,
      "num_input_tokens_seen": 46383968,
      "step": 13940
    },
    {
      "epoch": 2.7901858289773154,
      "grad_norm": 0.7391047477722168,
      "learning_rate": 5.982355270937628e-07,
      "loss": 0.0039,
      "num_input_tokens_seen": 46399744,
      "step": 13945
    },
    {
      "epoch": 2.791186254158017,
      "grad_norm": 1.6347475051879883,
      "learning_rate": 5.925525747845717e-07,
      "loss": 0.0047,
      "num_input_tokens_seen": 46415968,
      "step": 13950
    },
    {
      "epoch": 2.792186679338719,
      "grad_norm": 1.4310767650604248,
      "learning_rate": 5.868964203819439e-07,
      "loss": 0.0064,
      "num_input_tokens_seen": 46433280,
      "step": 13955
    },
    {
      "epoch": 2.7931871045194208,
      "grad_norm": 0.6000454425811768,
      "learning_rate": 5.812670700959922e-07,
      "loss": 0.0093,
      "num_input_tokens_seen": 46451040,
      "step": 13960
    },
    {
      "epoch": 2.7941875297001224,
      "grad_norm": 0.0004939927603118122,
      "learning_rate": 5.756645301074087e-07,
      "loss": 0.0038,
      "num_input_tokens_seen": 46467648,
      "step": 13965
    },
    {
      "epoch": 2.7951879548808245,
      "grad_norm": 0.04235506057739258,
      "learning_rate": 5.700888065674398e-07,
      "loss": 0.0082,
      "num_input_tokens_seen": 46482656,
      "step": 13970
    },
    {
      "epoch": 2.796188380061526,
      "grad_norm": 0.432803213596344,
      "learning_rate": 5.645399055978967e-07,
      "loss": 0.0033,
      "num_input_tokens_seen": 46498784,
      "step": 13975
    },
    {
      "epoch": 2.797188805242228,
      "grad_norm": 0.3252566158771515,
      "learning_rate": 5.590178332911366e-07,
      "loss": 0.0009,
      "num_input_tokens_seen": 46516064,
      "step": 13980
    },
    {
      "epoch": 2.79818923042293,
      "grad_norm": 0.01758207008242607,
      "learning_rate": 5.535225957100654e-07,
      "loss": 0.0053,
      "num_input_tokens_seen": 46532704,
      "step": 13985
    },
    {
      "epoch": 2.7991896556036315,
      "grad_norm": 0.01629800535738468,
      "learning_rate": 5.480541988881177e-07,
      "loss": 0.0044,
      "num_input_tokens_seen": 46549184,
      "step": 13990
    },
    {
      "epoch": 2.800190080784333,
      "grad_norm": 0.0037658638320863247,
      "learning_rate": 5.426126488292715e-07,
      "loss": 0.0082,
      "num_input_tokens_seen": 46564864,
      "step": 13995
    },
    {
      "epoch": 2.8011905059650353,
      "grad_norm": 1.712538242340088,
      "learning_rate": 5.371979515080145e-07,
      "loss": 0.0059,
      "num_input_tokens_seen": 46581440,
      "step": 14000
    },
    {
      "epoch": 2.802190931145737,
      "grad_norm": 0.7744560837745667,
      "learning_rate": 5.318101128693575e-07,
      "loss": 0.0085,
      "num_input_tokens_seen": 46598592,
      "step": 14005
    },
    {
      "epoch": 2.8031913563264386,
      "grad_norm": 0.03285866603255272,
      "learning_rate": 5.264491388288272e-07,
      "loss": 0.0019,
      "num_input_tokens_seen": 46614752,
      "step": 14010
    },
    {
      "epoch": 2.8041917815071407,
      "grad_norm": 0.07157700508832932,
      "learning_rate": 5.211150352724459e-07,
      "loss": 0.0014,
      "num_input_tokens_seen": 46631232,
      "step": 14015
    },
    {
      "epoch": 2.8051922066878423,
      "grad_norm": 0.025913609191775322,
      "learning_rate": 5.158078080567425e-07,
      "loss": 0.0082,
      "num_input_tokens_seen": 46647328,
      "step": 14020
    },
    {
      "epoch": 2.8061926318685444,
      "grad_norm": 0.022490929812192917,
      "learning_rate": 5.105274630087282e-07,
      "loss": 0.0153,
      "num_input_tokens_seen": 46663840,
      "step": 14025
    },
    {
      "epoch": 2.807193057049246,
      "grad_norm": 0.15464244782924652,
      "learning_rate": 5.052740059259042e-07,
      "loss": 0.0031,
      "num_input_tokens_seen": 46679904,
      "step": 14030
    },
    {
      "epoch": 2.8081934822299477,
      "grad_norm": 0.16805796325206757,
      "learning_rate": 5.000474425762541e-07,
      "loss": 0.0016,
      "num_input_tokens_seen": 46697792,
      "step": 14035
    },
    {
      "epoch": 2.8091939074106493,
      "grad_norm": 0.7833674550056458,
      "learning_rate": 4.948477786982237e-07,
      "loss": 0.0095,
      "num_input_tokens_seen": 46714784,
      "step": 14040
    },
    {
      "epoch": 2.8101943325913514,
      "grad_norm": 1.0255088806152344,
      "learning_rate": 4.89675020000735e-07,
      "loss": 0.0044,
      "num_input_tokens_seen": 46732320,
      "step": 14045
    },
    {
      "epoch": 2.811194757772053,
      "grad_norm": 0.20480923354625702,
      "learning_rate": 4.845291721631645e-07,
      "loss": 0.0039,
      "num_input_tokens_seen": 46747648,
      "step": 14050
    },
    {
      "epoch": 2.812195182952755,
      "grad_norm": 0.0014816083712503314,
      "learning_rate": 4.794102408353402e-07,
      "loss": 0.0071,
      "num_input_tokens_seen": 46764928,
      "step": 14055
    },
    {
      "epoch": 2.813195608133457,
      "grad_norm": 0.789983332157135,
      "learning_rate": 4.743182316375439e-07,
      "loss": 0.0094,
      "num_input_tokens_seen": 46781472,
      "step": 14060
    },
    {
      "epoch": 2.8141960333141585,
      "grad_norm": 1.3696625232696533,
      "learning_rate": 4.692531501604924e-07,
      "loss": 0.0135,
      "num_input_tokens_seen": 46798656,
      "step": 14065
    },
    {
      "epoch": 2.81519645849486,
      "grad_norm": 1.2937527894973755,
      "learning_rate": 4.6421500196534294e-07,
      "loss": 0.0062,
      "num_input_tokens_seen": 46815232,
      "step": 14070
    },
    {
      "epoch": 2.816196883675562,
      "grad_norm": 0.0012796290684491396,
      "learning_rate": 4.592037925836734e-07,
      "loss": 0.0039,
      "num_input_tokens_seen": 46831712,
      "step": 14075
    },
    {
      "epoch": 2.817197308856264,
      "grad_norm": 0.41710588335990906,
      "learning_rate": 4.5421952751749375e-07,
      "loss": 0.0024,
      "num_input_tokens_seen": 46848288,
      "step": 14080
    },
    {
      "epoch": 2.818197734036966,
      "grad_norm": 0.05578610301017761,
      "learning_rate": 4.492622122392265e-07,
      "loss": 0.0036,
      "num_input_tokens_seen": 46864864,
      "step": 14085
    },
    {
      "epoch": 2.8191981592176676,
      "grad_norm": 0.005146725103259087,
      "learning_rate": 4.443318521917067e-07,
      "loss": 0.0042,
      "num_input_tokens_seen": 46881536,
      "step": 14090
    },
    {
      "epoch": 2.8201985843983692,
      "grad_norm": 0.015356880612671375,
      "learning_rate": 4.3942845278816814e-07,
      "loss": 0.0023,
      "num_input_tokens_seen": 46898048,
      "step": 14095
    },
    {
      "epoch": 2.821199009579071,
      "grad_norm": 0.0015450617065653205,
      "learning_rate": 4.3455201941224877e-07,
      "loss": 0.0029,
      "num_input_tokens_seen": 46914624,
      "step": 14100
    },
    {
      "epoch": 2.822199434759773,
      "grad_norm": 0.04543983191251755,
      "learning_rate": 4.2970255741798524e-07,
      "loss": 0.0025,
      "num_input_tokens_seen": 46930624,
      "step": 14105
    },
    {
      "epoch": 2.8231998599404746,
      "grad_norm": 0.2659851014614105,
      "learning_rate": 4.248800721297852e-07,
      "loss": 0.0021,
      "num_input_tokens_seen": 46947296,
      "step": 14110
    },
    {
      "epoch": 2.8242002851211767,
      "grad_norm": 0.2597131133079529,
      "learning_rate": 4.2008456884244927e-07,
      "loss": 0.0211,
      "num_input_tokens_seen": 46963232,
      "step": 14115
    },
    {
      "epoch": 2.8252007103018784,
      "grad_norm": 1.0440688133239746,
      "learning_rate": 4.153160528211575e-07,
      "loss": 0.0019,
      "num_input_tokens_seen": 46978688,
      "step": 14120
    },
    {
      "epoch": 2.82620113548258,
      "grad_norm": 0.08481547981500626,
      "learning_rate": 4.105745293014468e-07,
      "loss": 0.0031,
      "num_input_tokens_seen": 46995744,
      "step": 14125
    },
    {
      "epoch": 2.8272015606632817,
      "grad_norm": 0.5087813138961792,
      "learning_rate": 4.0586000348922513e-07,
      "loss": 0.0032,
      "num_input_tokens_seen": 47012128,
      "step": 14130
    },
    {
      "epoch": 2.8282019858439837,
      "grad_norm": 1.1093705892562866,
      "learning_rate": 4.011724805607575e-07,
      "loss": 0.0023,
      "num_input_tokens_seen": 47029408,
      "step": 14135
    },
    {
      "epoch": 2.8292024110246854,
      "grad_norm": 0.6753928065299988,
      "learning_rate": 3.9651196566266593e-07,
      "loss": 0.0064,
      "num_input_tokens_seen": 47046112,
      "step": 14140
    },
    {
      "epoch": 2.8302028362053875,
      "grad_norm": 0.0030104496981948614,
      "learning_rate": 3.9187846391191273e-07,
      "loss": 0.0024,
      "num_input_tokens_seen": 47063072,
      "step": 14145
    },
    {
      "epoch": 2.831203261386089,
      "grad_norm": 0.06650929152965546,
      "learning_rate": 3.8727198039579804e-07,
      "loss": 0.0093,
      "num_input_tokens_seen": 47080160,
      "step": 14150
    },
    {
      "epoch": 2.832203686566791,
      "grad_norm": 0.28551894426345825,
      "learning_rate": 3.8269252017197054e-07,
      "loss": 0.0043,
      "num_input_tokens_seen": 47096448,
      "step": 14155
    },
    {
      "epoch": 2.8332041117474924,
      "grad_norm": 0.018239686265587807,
      "learning_rate": 3.781400882684e-07,
      "loss": 0.0007,
      "num_input_tokens_seen": 47113664,
      "step": 14160
    },
    {
      "epoch": 2.8342045369281945,
      "grad_norm": 0.34065768122673035,
      "learning_rate": 3.7361468968337976e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 47129984,
      "step": 14165
    },
    {
      "epoch": 2.835204962108896,
      "grad_norm": 0.03855415806174278,
      "learning_rate": 3.6911632938552433e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 47145920,
      "step": 14170
    },
    {
      "epoch": 2.8362053872895983,
      "grad_norm": 1.1146639585494995,
      "learning_rate": 3.64645012313769e-07,
      "loss": 0.003,
      "num_input_tokens_seen": 47162752,
      "step": 14175
    },
    {
      "epoch": 2.8372058124703,
      "grad_norm": 0.008468233048915863,
      "learning_rate": 3.6020074337734244e-07,
      "loss": 0.0077,
      "num_input_tokens_seen": 47178656,
      "step": 14180
    },
    {
      "epoch": 2.8382062376510016,
      "grad_norm": 0.07861342281103134,
      "learning_rate": 3.557835274557886e-07,
      "loss": 0.0011,
      "num_input_tokens_seen": 47195104,
      "step": 14185
    },
    {
      "epoch": 2.839206662831703,
      "grad_norm": 0.5510034561157227,
      "learning_rate": 3.5139336939894184e-07,
      "loss": 0.0048,
      "num_input_tokens_seen": 47212704,
      "step": 14190
    },
    {
      "epoch": 2.8402070880124053,
      "grad_norm": 0.02640078216791153,
      "learning_rate": 3.470302740269327e-07,
      "loss": 0.0061,
      "num_input_tokens_seen": 47229824,
      "step": 14195
    },
    {
      "epoch": 2.841207513193107,
      "grad_norm": 0.05584491044282913,
      "learning_rate": 3.4269424613017907e-07,
      "loss": 0.0061,
      "num_input_tokens_seen": 47247072,
      "step": 14200
    },
    {
      "epoch": 2.842207938373809,
      "grad_norm": 0.2541450560092926,
      "learning_rate": 3.3838529046937295e-07,
      "loss": 0.0026,
      "num_input_tokens_seen": 47263648,
      "step": 14205
    },
    {
      "epoch": 2.8432083635545107,
      "grad_norm": 2.0473499298095703,
      "learning_rate": 3.34103411775491e-07,
      "loss": 0.0107,
      "num_input_tokens_seen": 47281056,
      "step": 14210
    },
    {
      "epoch": 2.8442087887352123,
      "grad_norm": 1.1724183559417725,
      "learning_rate": 3.298486147497809e-07,
      "loss": 0.0028,
      "num_input_tokens_seen": 47297024,
      "step": 14215
    },
    {
      "epoch": 2.8452092139159144,
      "grad_norm": 0.006476056762039661,
      "learning_rate": 3.256209040637503e-07,
      "loss": 0.0037,
      "num_input_tokens_seen": 47313440,
      "step": 14220
    },
    {
      "epoch": 2.846209639096616,
      "grad_norm": 0.0024868613108992577,
      "learning_rate": 3.214202843591696e-07,
      "loss": 0.0004,
      "num_input_tokens_seen": 47330656,
      "step": 14225
    },
    {
      "epoch": 2.8472100642773177,
      "grad_norm": 0.1444324404001236,
      "learning_rate": 3.1724676024806886e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 47346880,
      "step": 14230
    },
    {
      "epoch": 2.84821048945802,
      "grad_norm": 1.7168079614639282,
      "learning_rate": 3.131003363127216e-07,
      "loss": 0.0051,
      "num_input_tokens_seen": 47363296,
      "step": 14235
    },
    {
      "epoch": 2.8492109146387214,
      "grad_norm": 0.03592948615550995,
      "learning_rate": 3.089810171056584e-07,
      "loss": 0.0025,
      "num_input_tokens_seen": 47379744,
      "step": 14240
    },
    {
      "epoch": 2.850211339819423,
      "grad_norm": 0.31552281975746155,
      "learning_rate": 3.048888071496392e-07,
      "loss": 0.0055,
      "num_input_tokens_seen": 47395712,
      "step": 14245
    },
    {
      "epoch": 2.851211765000125,
      "grad_norm": 0.0047025843523442745,
      "learning_rate": 3.008237109376644e-07,
      "loss": 0.0003,
      "num_input_tokens_seen": 47412352,
      "step": 14250
    },
    {
      "epoch": 2.852212190180827,
      "grad_norm": 0.34631943702697754,
      "learning_rate": 2.9678573293296653e-07,
      "loss": 0.0012,
      "num_input_tokens_seen": 47429120,
      "step": 14255
    },
    {
      "epoch": 2.853212615361529,
      "grad_norm": 1.796602487564087,
      "learning_rate": 2.9277487756899645e-07,
      "loss": 0.009,
      "num_input_tokens_seen": 47445856,
      "step": 14260
    },
    {
      "epoch": 2.8542130405422306,
      "grad_norm": 0.028868287801742554,
      "learning_rate": 2.8879114924943995e-07,
      "loss": 0.0026,
      "num_input_tokens_seen": 47463264,
      "step": 14265
    },
    {
      "epoch": 2.855213465722932,
      "grad_norm": 0.18896105885505676,
      "learning_rate": 2.8483455234818437e-07,
      "loss": 0.0091,
      "num_input_tokens_seen": 47480672,
      "step": 14270
    },
    {
      "epoch": 2.856213890903634,
      "grad_norm": 0.0005116119282320142,
      "learning_rate": 2.8090509120933527e-07,
      "loss": 0.0039,
      "num_input_tokens_seen": 47496800,
      "step": 14275
    },
    {
      "epoch": 2.857214316084336,
      "grad_norm": 0.37647685408592224,
      "learning_rate": 2.7700277014720834e-07,
      "loss": 0.0026,
      "num_input_tokens_seen": 47512192,
      "step": 14280
    },
    {
      "epoch": 2.8582147412650376,
      "grad_norm": 0.03440157696604729,
      "learning_rate": 2.731275934463151e-07,
      "loss": 0.0034,
      "num_input_tokens_seen": 47529152,
      "step": 14285
    },
    {
      "epoch": 2.8592151664457397,
      "grad_norm": 0.0038820921909064054,
      "learning_rate": 2.6927956536136056e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 47544928,
      "step": 14290
    },
    {
      "epoch": 2.8602155916264413,
      "grad_norm": 1.7431918382644653,
      "learning_rate": 2.654586901172568e-07,
      "loss": 0.0097,
      "num_input_tokens_seen": 47561984,
      "step": 14295
    },
    {
      "epoch": 2.861216016807143,
      "grad_norm": 0.2928279936313629,
      "learning_rate": 2.616649719090869e-07,
      "loss": 0.0102,
      "num_input_tokens_seen": 47577984,
      "step": 14300
    },
    {
      "epoch": 2.8622164419878446,
      "grad_norm": 0.6232368350028992,
      "learning_rate": 2.578984149021302e-07,
      "loss": 0.0145,
      "num_input_tokens_seen": 47594112,
      "step": 14305
    },
    {
      "epoch": 2.8632168671685467,
      "grad_norm": 0.014803355559706688,
      "learning_rate": 2.5415902323183694e-07,
      "loss": 0.0007,
      "num_input_tokens_seen": 47610368,
      "step": 14310
    },
    {
      "epoch": 2.8642172923492484,
      "grad_norm": 2.0789053440093994,
      "learning_rate": 2.504468010038341e-07,
      "loss": 0.0059,
      "num_input_tokens_seen": 47627520,
      "step": 14315
    },
    {
      "epoch": 2.8652177175299505,
      "grad_norm": 1.0814423561096191,
      "learning_rate": 2.467617522939197e-07,
      "loss": 0.0078,
      "num_input_tokens_seen": 47642656,
      "step": 14320
    },
    {
      "epoch": 2.866218142710652,
      "grad_norm": 0.5673795342445374,
      "learning_rate": 2.431038811480518e-07,
      "loss": 0.0035,
      "num_input_tokens_seen": 47659552,
      "step": 14325
    },
    {
      "epoch": 2.8672185678913538,
      "grad_norm": 0.2489253282546997,
      "learning_rate": 2.3947319158235394e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 47676672,
      "step": 14330
    },
    {
      "epoch": 2.8682189930720554,
      "grad_norm": 0.14608818292617798,
      "learning_rate": 2.3586968758310678e-07,
      "loss": 0.0072,
      "num_input_tokens_seen": 47693568,
      "step": 14335
    },
    {
      "epoch": 2.8692194182527575,
      "grad_norm": 0.09965641051530838,
      "learning_rate": 2.3229337310673727e-07,
      "loss": 0.0032,
      "num_input_tokens_seen": 47710048,
      "step": 14340
    },
    {
      "epoch": 2.870219843433459,
      "grad_norm": 0.02677687630057335,
      "learning_rate": 2.2874425207982387e-07,
      "loss": 0.0013,
      "num_input_tokens_seen": 47727936,
      "step": 14345
    },
    {
      "epoch": 2.8712202686141612,
      "grad_norm": 0.01108902134001255,
      "learning_rate": 2.2522232839908842e-07,
      "loss": 0.0034,
      "num_input_tokens_seen": 47745056,
      "step": 14350
    },
    {
      "epoch": 2.872220693794863,
      "grad_norm": 0.08057229220867157,
      "learning_rate": 2.217276059313933e-07,
      "loss": 0.0184,
      "num_input_tokens_seen": 47761536,
      "step": 14355
    },
    {
      "epoch": 2.8732211189755645,
      "grad_norm": 0.33636265993118286,
      "learning_rate": 2.182600885137276e-07,
      "loss": 0.0078,
      "num_input_tokens_seen": 47778304,
      "step": 14360
    },
    {
      "epoch": 2.874221544156266,
      "grad_norm": 0.23361791670322418,
      "learning_rate": 2.148197799532209e-07,
      "loss": 0.0023,
      "num_input_tokens_seen": 47794656,
      "step": 14365
    },
    {
      "epoch": 2.8752219693369683,
      "grad_norm": 3.437770128250122,
      "learning_rate": 2.1140668402712393e-07,
      "loss": 0.0197,
      "num_input_tokens_seen": 47811200,
      "step": 14370
    },
    {
      "epoch": 2.87622239451767,
      "grad_norm": 0.43796396255493164,
      "learning_rate": 2.0802080448280858e-07,
      "loss": 0.0034,
      "num_input_tokens_seen": 47828096,
      "step": 14375
    },
    {
      "epoch": 2.877222819698372,
      "grad_norm": 0.09619075804948807,
      "learning_rate": 2.0466214503777058e-07,
      "loss": 0.0064,
      "num_input_tokens_seen": 47844672,
      "step": 14380
    },
    {
      "epoch": 2.8782232448790737,
      "grad_norm": 0.2009647935628891,
      "learning_rate": 2.0133070937961017e-07,
      "loss": 0.0012,
      "num_input_tokens_seen": 47859968,
      "step": 14385
    },
    {
      "epoch": 2.8792236700597753,
      "grad_norm": 0.015301394276320934,
      "learning_rate": 1.9802650116604593e-07,
      "loss": 0.0033,
      "num_input_tokens_seen": 47876608,
      "step": 14390
    },
    {
      "epoch": 2.880224095240477,
      "grad_norm": 1.40193772315979,
      "learning_rate": 1.947495240248981e-07,
      "loss": 0.0102,
      "num_input_tokens_seen": 47893024,
      "step": 14395
    },
    {
      "epoch": 2.881224520421179,
      "grad_norm": 0.0199880488216877,
      "learning_rate": 1.9149978155409143e-07,
      "loss": 0.0068,
      "num_input_tokens_seen": 47909984,
      "step": 14400
    },
    {
      "epoch": 2.8822249456018807,
      "grad_norm": 0.359917551279068,
      "learning_rate": 1.8827727732164958e-07,
      "loss": 0.0015,
      "num_input_tokens_seen": 47926784,
      "step": 14405
    },
    {
      "epoch": 2.883225370782583,
      "grad_norm": 0.023170724511146545,
      "learning_rate": 1.85082014865684e-07,
      "loss": 0.0012,
      "num_input_tokens_seen": 47943360,
      "step": 14410
    },
    {
      "epoch": 2.8842257959632844,
      "grad_norm": 0.05834521725773811,
      "learning_rate": 1.819139976944023e-07,
      "loss": 0.0015,
      "num_input_tokens_seen": 47959584,
      "step": 14415
    },
    {
      "epoch": 2.885226221143986,
      "grad_norm": 0.3212447166442871,
      "learning_rate": 1.7877322928609442e-07,
      "loss": 0.002,
      "num_input_tokens_seen": 47977120,
      "step": 14420
    },
    {
      "epoch": 2.8862266463246877,
      "grad_norm": 0.005522616673260927,
      "learning_rate": 1.7565971308913798e-07,
      "loss": 0.0027,
      "num_input_tokens_seen": 47995136,
      "step": 14425
    },
    {
      "epoch": 2.88722707150539,
      "grad_norm": 0.008083946071565151,
      "learning_rate": 1.7257345252198187e-07,
      "loss": 0.0169,
      "num_input_tokens_seen": 48011968,
      "step": 14430
    },
    {
      "epoch": 2.8882274966860915,
      "grad_norm": 0.00019500641792546958,
      "learning_rate": 1.6951445097316e-07,
      "loss": 0.0026,
      "num_input_tokens_seen": 48029408,
      "step": 14435
    },
    {
      "epoch": 2.8892279218667936,
      "grad_norm": 0.06523102521896362,
      "learning_rate": 1.6648271180126628e-07,
      "loss": 0.0192,
      "num_input_tokens_seen": 48046432,
      "step": 14440
    },
    {
      "epoch": 2.890228347047495,
      "grad_norm": 0.27350881695747375,
      "learning_rate": 1.6347823833497145e-07,
      "loss": 0.0293,
      "num_input_tokens_seen": 48063168,
      "step": 14445
    },
    {
      "epoch": 2.891228772228197,
      "grad_norm": 0.044810544699430466,
      "learning_rate": 1.605010338730034e-07,
      "loss": 0.0028,
      "num_input_tokens_seen": 48080128,
      "step": 14450
    },
    {
      "epoch": 2.892229197408899,
      "grad_norm": 0.0014945862349122763,
      "learning_rate": 1.5755110168415577e-07,
      "loss": 0.0012,
      "num_input_tokens_seen": 48096608,
      "step": 14455
    },
    {
      "epoch": 2.8932296225896006,
      "grad_norm": 0.08826776593923569,
      "learning_rate": 1.5462844500727657e-07,
      "loss": 0.0093,
      "num_input_tokens_seen": 48113920,
      "step": 14460
    },
    {
      "epoch": 2.8942300477703022,
      "grad_norm": 0.10818234086036682,
      "learning_rate": 1.517330670512629e-07,
      "loss": 0.002,
      "num_input_tokens_seen": 48131360,
      "step": 14465
    },
    {
      "epoch": 2.8952304729510043,
      "grad_norm": 0.09250593930482864,
      "learning_rate": 1.4886497099506903e-07,
      "loss": 0.0086,
      "num_input_tokens_seen": 48147104,
      "step": 14470
    },
    {
      "epoch": 2.896230898131706,
      "grad_norm": 0.7708470821380615,
      "learning_rate": 1.4602415998768992e-07,
      "loss": 0.0231,
      "num_input_tokens_seen": 48163520,
      "step": 14475
    },
    {
      "epoch": 2.8972313233124076,
      "grad_norm": 0.01487722061574459,
      "learning_rate": 1.4321063714816674e-07,
      "loss": 0.0146,
      "num_input_tokens_seen": 48180192,
      "step": 14480
    },
    {
      "epoch": 2.8982317484931097,
      "grad_norm": 1.3531252145767212,
      "learning_rate": 1.4042440556557568e-07,
      "loss": 0.0094,
      "num_input_tokens_seen": 48196992,
      "step": 14485
    },
    {
      "epoch": 2.8992321736738114,
      "grad_norm": 0.3768194615840912,
      "learning_rate": 1.376654682990308e-07,
      "loss": 0.0014,
      "num_input_tokens_seen": 48213792,
      "step": 14490
    },
    {
      "epoch": 2.900232598854513,
      "grad_norm": 0.49699288606643677,
      "learning_rate": 1.3493382837768132e-07,
      "loss": 0.0012,
      "num_input_tokens_seen": 48229376,
      "step": 14495
    },
    {
      "epoch": 2.901233024035215,
      "grad_norm": 0.07719047367572784,
      "learning_rate": 1.3222948880070306e-07,
      "loss": 0.0097,
      "num_input_tokens_seen": 48245280,
      "step": 14500
    },
    {
      "epoch": 2.9022334492159167,
      "grad_norm": 0.12778985500335693,
      "learning_rate": 1.2955245253729597e-07,
      "loss": 0.0008,
      "num_input_tokens_seen": 48262432,
      "step": 14505
    },
    {
      "epoch": 2.9032338743966184,
      "grad_norm": 0.7701250910758972,
      "learning_rate": 1.269027225266839e-07,
      "loss": 0.0031,
      "num_input_tokens_seen": 48279584,
      "step": 14510
    },
    {
      "epoch": 2.9042342995773205,
      "grad_norm": 0.27826884388923645,
      "learning_rate": 1.2428030167811743e-07,
      "loss": 0.0044,
      "num_input_tokens_seen": 48296000,
      "step": 14515
    },
    {
      "epoch": 2.905234724758022,
      "grad_norm": 0.00854239147156477,
      "learning_rate": 1.2168519287084622e-07,
      "loss": 0.0028,
      "num_input_tokens_seen": 48313920,
      "step": 14520
    },
    {
      "epoch": 2.9062351499387242,
      "grad_norm": 0.8932532072067261,
      "learning_rate": 1.191173989541522e-07,
      "loss": 0.0067,
      "num_input_tokens_seen": 48330176,
      "step": 14525
    },
    {
      "epoch": 2.907235575119426,
      "grad_norm": 0.014440479688346386,
      "learning_rate": 1.1657692274731624e-07,
      "loss": 0.0062,
      "num_input_tokens_seen": 48346752,
      "step": 14530
    },
    {
      "epoch": 2.9082360003001275,
      "grad_norm": 1.0627107620239258,
      "learning_rate": 1.1406376703962385e-07,
      "loss": 0.0144,
      "num_input_tokens_seen": 48364384,
      "step": 14535
    },
    {
      "epoch": 2.909236425480829,
      "grad_norm": 0.957610011100769,
      "learning_rate": 1.1157793459037058e-07,
      "loss": 0.0041,
      "num_input_tokens_seen": 48380800,
      "step": 14540
    },
    {
      "epoch": 2.9102368506615313,
      "grad_norm": 0.008517167530953884,
      "learning_rate": 1.0911942812885378e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 48397312,
      "step": 14545
    },
    {
      "epoch": 2.911237275842233,
      "grad_norm": 0.00145952426828444,
      "learning_rate": 1.0668825035436426e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 48414592,
      "step": 14550
    },
    {
      "epoch": 2.912237701022935,
      "grad_norm": 0.00036152126267552376,
      "learning_rate": 1.0428440393618621e-07,
      "loss": 0.0009,
      "num_input_tokens_seen": 48431040,
      "step": 14555
    },
    {
      "epoch": 2.9132381262036366,
      "grad_norm": 0.01873910427093506,
      "learning_rate": 1.0190789151360014e-07,
      "loss": 0.0042,
      "num_input_tokens_seen": 48447424,
      "step": 14560
    },
    {
      "epoch": 2.9142385513843383,
      "grad_norm": 0.04497993737459183,
      "learning_rate": 9.955871569587438e-08,
      "loss": 0.0067,
      "num_input_tokens_seen": 48465856,
      "step": 14565
    },
    {
      "epoch": 2.91523897656504,
      "grad_norm": 0.6971708536148071,
      "learning_rate": 9.723687906225964e-08,
      "loss": 0.0102,
      "num_input_tokens_seen": 48482176,
      "step": 14570
    },
    {
      "epoch": 2.916239401745742,
      "grad_norm": 0.5259365439414978,
      "learning_rate": 9.494238416199453e-08,
      "loss": 0.001,
      "num_input_tokens_seen": 48498016,
      "step": 14575
    },
    {
      "epoch": 2.9172398269264437,
      "grad_norm": 0.30338263511657715,
      "learning_rate": 9.267523351429441e-08,
      "loss": 0.0019,
      "num_input_tokens_seen": 48514464,
      "step": 14580
    },
    {
      "epoch": 2.9182402521071458,
      "grad_norm": 1.076026201248169,
      "learning_rate": 9.043542960835705e-08,
      "loss": 0.008,
      "num_input_tokens_seen": 48530240,
      "step": 14585
    },
    {
      "epoch": 2.9192406772878474,
      "grad_norm": 0.5502411723136902,
      "learning_rate": 8.822297490335141e-08,
      "loss": 0.0047,
      "num_input_tokens_seen": 48546624,
      "step": 14590
    },
    {
      "epoch": 2.920241102468549,
      "grad_norm": 1.4354311227798462,
      "learning_rate": 8.603787182841772e-08,
      "loss": 0.013,
      "num_input_tokens_seen": 48563776,
      "step": 14595
    },
    {
      "epoch": 2.9212415276492507,
      "grad_norm": 0.009782413020730019,
      "learning_rate": 8.388012278266744e-08,
      "loss": 0.0054,
      "num_input_tokens_seen": 48580032,
      "step": 14600
    },
    {
      "epoch": 2.922241952829953,
      "grad_norm": 0.03704375773668289,
      "learning_rate": 8.17497301351805e-08,
      "loss": 0.0056,
      "num_input_tokens_seen": 48597760,
      "step": 14605
    },
    {
      "epoch": 2.9232423780106545,
      "grad_norm": 1.8117647171020508,
      "learning_rate": 7.964669622499976e-08,
      "loss": 0.0092,
      "num_input_tokens_seen": 48615104,
      "step": 14610
    },
    {
      "epoch": 2.9242428031913565,
      "grad_norm": 0.5814581513404846,
      "learning_rate": 7.75710233611282e-08,
      "loss": 0.0024,
      "num_input_tokens_seen": 48631648,
      "step": 14615
    },
    {
      "epoch": 2.925243228372058,
      "grad_norm": 0.46254706382751465,
      "learning_rate": 7.552271382253174e-08,
      "loss": 0.0095,
      "num_input_tokens_seen": 48647584,
      "step": 14620
    },
    {
      "epoch": 2.92624365355276,
      "grad_norm": 0.010827763006091118,
      "learning_rate": 7.350176985812806e-08,
      "loss": 0.0034,
      "num_input_tokens_seen": 48662944,
      "step": 14625
    },
    {
      "epoch": 2.9272440787334615,
      "grad_norm": 0.15264484286308289,
      "learning_rate": 7.150819368679229e-08,
      "loss": 0.0042,
      "num_input_tokens_seen": 48680096,
      "step": 14630
    },
    {
      "epoch": 2.9282445039141636,
      "grad_norm": 0.5206581950187683,
      "learning_rate": 6.954198749735408e-08,
      "loss": 0.0056,
      "num_input_tokens_seen": 48697888,
      "step": 14635
    },
    {
      "epoch": 2.9292449290948652,
      "grad_norm": 0.6589008569717407,
      "learning_rate": 6.760315344858658e-08,
      "loss": 0.0047,
      "num_input_tokens_seen": 48715360,
      "step": 14640
    },
    {
      "epoch": 2.9302453542755673,
      "grad_norm": 1.2055493593215942,
      "learning_rate": 6.569169366921479e-08,
      "loss": 0.0018,
      "num_input_tokens_seen": 48731808,
      "step": 14645
    },
    {
      "epoch": 2.931245779456269,
      "grad_norm": 0.0015922492602840066,
      "learning_rate": 6.38076102579016e-08,
      "loss": 0.0017,
      "num_input_tokens_seen": 48748000,
      "step": 14650
    },
    {
      "epoch": 2.9322462046369706,
      "grad_norm": 1.6169053316116333,
      "learning_rate": 6.195090528326453e-08,
      "loss": 0.0194,
      "num_input_tokens_seen": 48763872,
      "step": 14655
    },
    {
      "epoch": 2.9332466298176723,
      "grad_norm": 0.033600740134716034,
      "learning_rate": 6.012158078384511e-08,
      "loss": 0.0092,
      "num_input_tokens_seen": 48780320,
      "step": 14660
    },
    {
      "epoch": 2.9342470549983743,
      "grad_norm": 0.010385665111243725,
      "learning_rate": 5.8319638768139504e-08,
      "loss": 0.0034,
      "num_input_tokens_seen": 48797952,
      "step": 14665
    },
    {
      "epoch": 2.935247480179076,
      "grad_norm": 1.4164212942123413,
      "learning_rate": 5.654508121456792e-08,
      "loss": 0.019,
      "num_input_tokens_seen": 48815008,
      "step": 14670
    },
    {
      "epoch": 2.936247905359778,
      "grad_norm": 0.10302630066871643,
      "learning_rate": 5.479791007148571e-08,
      "loss": 0.0066,
      "num_input_tokens_seen": 48831936,
      "step": 14675
    },
    {
      "epoch": 2.9372483305404797,
      "grad_norm": 0.0018415686208754778,
      "learning_rate": 5.307812725718342e-08,
      "loss": 0.0108,
      "num_input_tokens_seen": 48848544,
      "step": 14680
    },
    {
      "epoch": 2.9382487557211814,
      "grad_norm": 0.749054491519928,
      "learning_rate": 5.1385734659881165e-08,
      "loss": 0.0065,
      "num_input_tokens_seen": 48865344,
      "step": 14685
    },
    {
      "epoch": 2.939249180901883,
      "grad_norm": 0.0004770423984155059,
      "learning_rate": 4.9720734137720384e-08,
      "loss": 0.0002,
      "num_input_tokens_seen": 48882656,
      "step": 14690
    },
    {
      "epoch": 2.940249606082585,
      "grad_norm": 0.12848667800426483,
      "learning_rate": 4.8083127518772084e-08,
      "loss": 0.0013,
      "num_input_tokens_seen": 48899168,
      "step": 14695
    },
    {
      "epoch": 2.9412500312632868,
      "grad_norm": 0.00011170543439220637,
      "learning_rate": 4.647291660103137e-08,
      "loss": 0.0046,
      "num_input_tokens_seen": 48914624,
      "step": 14700
    },
    {
      "epoch": 2.942250456443989,
      "grad_norm": 0.10075289011001587,
      "learning_rate": 4.489010315241182e-08,
      "loss": 0.0015,
      "num_input_tokens_seen": 48931040,
      "step": 14705
    },
    {
      "epoch": 2.9432508816246905,
      "grad_norm": 0.8776507377624512,
      "learning_rate": 4.3334688910748324e-08,
      "loss": 0.0066,
      "num_input_tokens_seen": 48946592,
      "step": 14710
    },
    {
      "epoch": 2.944251306805392,
      "grad_norm": 0.0035674963146448135,
      "learning_rate": 4.1806675583788703e-08,
      "loss": 0.0051,
      "num_input_tokens_seen": 48962688,
      "step": 14715
    },
    {
      "epoch": 2.9452517319860942,
      "grad_norm": 0.020074935629963875,
      "learning_rate": 4.030606484920208e-08,
      "loss": 0.0009,
      "num_input_tokens_seen": 48979616,
      "step": 14720
    },
    {
      "epoch": 2.946252157166796,
      "grad_norm": 0.017800666391849518,
      "learning_rate": 3.8832858354567736e-08,
      "loss": 0.0011,
      "num_input_tokens_seen": 48996000,
      "step": 14725
    },
    {
      "epoch": 2.9472525823474975,
      "grad_norm": 0.3195970952510834,
      "learning_rate": 3.7387057717380715e-08,
      "loss": 0.0012,
      "num_input_tokens_seen": 49011776,
      "step": 14730
    },
    {
      "epoch": 2.9482530075281996,
      "grad_norm": 0.0029207849875092506,
      "learning_rate": 3.596866452503789e-08,
      "loss": 0.0008,
      "num_input_tokens_seen": 49028512,
      "step": 14735
    },
    {
      "epoch": 2.9492534327089013,
      "grad_norm": 0.014407744631171227,
      "learning_rate": 3.457768033485465e-08,
      "loss": 0.0144,
      "num_input_tokens_seen": 49045824,
      "step": 14740
    },
    {
      "epoch": 2.950253857889603,
      "grad_norm": 0.3978143632411957,
      "learning_rate": 3.321410667404268e-08,
      "loss": 0.0129,
      "num_input_tokens_seen": 49062368,
      "step": 14745
    },
    {
      "epoch": 2.951254283070305,
      "grad_norm": 1.726647973060608,
      "learning_rate": 3.1877945039726634e-08,
      "loss": 0.0086,
      "num_input_tokens_seen": 49079520,
      "step": 14750
    },
    {
      "epoch": 2.9522547082510067,
      "grad_norm": 0.22471491992473602,
      "learning_rate": 3.056919689893023e-08,
      "loss": 0.0042,
      "num_input_tokens_seen": 49096800,
      "step": 14755
    },
    {
      "epoch": 2.9532551334317083,
      "grad_norm": 0.6711318492889404,
      "learning_rate": 2.928786368858183e-08,
      "loss": 0.0052,
      "num_input_tokens_seen": 49113792,
      "step": 14760
    },
    {
      "epoch": 2.9542555586124104,
      "grad_norm": 0.6739404201507568,
      "learning_rate": 2.803394681550886e-08,
      "loss": 0.0009,
      "num_input_tokens_seen": 49131040,
      "step": 14765
    },
    {
      "epoch": 2.955255983793112,
      "grad_norm": 0.3748348653316498,
      "learning_rate": 2.6807447656432283e-08,
      "loss": 0.0021,
      "num_input_tokens_seen": 49148160,
      "step": 14770
    },
    {
      "epoch": 2.9562564089738137,
      "grad_norm": 0.09970913082361221,
      "learning_rate": 2.5608367557977684e-08,
      "loss": 0.0069,
      "num_input_tokens_seen": 49164512,
      "step": 14775
    },
    {
      "epoch": 2.957256834154516,
      "grad_norm": 1.040163278579712,
      "learning_rate": 2.443670783666141e-08,
      "loss": 0.0061,
      "num_input_tokens_seen": 49181024,
      "step": 14780
    },
    {
      "epoch": 2.9582572593352174,
      "grad_norm": 0.02159752883017063,
      "learning_rate": 2.3292469778896097e-08,
      "loss": 0.0003,
      "num_input_tokens_seen": 49196992,
      "step": 14785
    },
    {
      "epoch": 2.9592576845159195,
      "grad_norm": 1.130287528038025,
      "learning_rate": 2.217565464098792e-08,
      "loss": 0.0073,
      "num_input_tokens_seen": 49213440,
      "step": 14790
    },
    {
      "epoch": 2.960258109696621,
      "grad_norm": 0.032430630177259445,
      "learning_rate": 2.1086263649128247e-08,
      "loss": 0.0013,
      "num_input_tokens_seen": 49230528,
      "step": 14795
    },
    {
      "epoch": 2.961258534877323,
      "grad_norm": 0.04397590830922127,
      "learning_rate": 2.002429799940475e-08,
      "loss": 0.0042,
      "num_input_tokens_seen": 49247840,
      "step": 14800
    },
    {
      "epoch": 2.9622589600580245,
      "grad_norm": 0.9333756566047668,
      "learning_rate": 1.8989758857790308e-08,
      "loss": 0.0113,
      "num_input_tokens_seen": 49264896,
      "step": 14805
    },
    {
      "epoch": 2.9632593852387266,
      "grad_norm": 1.2465392351150513,
      "learning_rate": 1.7982647360151318e-08,
      "loss": 0.007,
      "num_input_tokens_seen": 49282016,
      "step": 14810
    },
    {
      "epoch": 2.964259810419428,
      "grad_norm": 0.4767018258571625,
      "learning_rate": 1.7002964612231055e-08,
      "loss": 0.0031,
      "num_input_tokens_seen": 49298624,
      "step": 14815
    },
    {
      "epoch": 2.9652602356001303,
      "grad_norm": 2.005401134490967,
      "learning_rate": 1.6050711689663546e-08,
      "loss": 0.0087,
      "num_input_tokens_seen": 49316320,
      "step": 14820
    },
    {
      "epoch": 2.966260660780832,
      "grad_norm": 1.832701563835144,
      "learning_rate": 1.512588963796524e-08,
      "loss": 0.0092,
      "num_input_tokens_seen": 49332448,
      "step": 14825
    },
    {
      "epoch": 2.9672610859615336,
      "grad_norm": 1.6767756938934326,
      "learning_rate": 1.4228499472535017e-08,
      "loss": 0.0068,
      "num_input_tokens_seen": 49350496,
      "step": 14830
    },
    {
      "epoch": 2.9682615111422352,
      "grad_norm": 0.027200087904930115,
      "learning_rate": 1.33585421786514e-08,
      "loss": 0.0007,
      "num_input_tokens_seen": 49366592,
      "step": 14835
    },
    {
      "epoch": 2.9692619363229373,
      "grad_norm": 0.2780955135822296,
      "learning_rate": 1.2516018711478117e-08,
      "loss": 0.0034,
      "num_input_tokens_seen": 49383872,
      "step": 14840
    },
    {
      "epoch": 2.970262361503639,
      "grad_norm": 0.06683382391929626,
      "learning_rate": 1.170092999605299e-08,
      "loss": 0.0163,
      "num_input_tokens_seen": 49400288,
      "step": 14845
    },
    {
      "epoch": 2.971262786684341,
      "grad_norm": 0.013099157251417637,
      "learning_rate": 1.0913276927296267e-08,
      "loss": 0.0015,
      "num_input_tokens_seen": 49416800,
      "step": 14850
    },
    {
      "epoch": 2.9722632118650427,
      "grad_norm": 0.014247686602175236,
      "learning_rate": 1.0153060369999523e-08,
      "loss": 0.0027,
      "num_input_tokens_seen": 49432800,
      "step": 14855
    },
    {
      "epoch": 2.9732636370457444,
      "grad_norm": 0.0005714791477657855,
      "learning_rate": 9.420281158839527e-09,
      "loss": 0.0019,
      "num_input_tokens_seen": 49449440,
      "step": 14860
    },
    {
      "epoch": 2.974264062226446,
      "grad_norm": 2.1527199745178223,
      "learning_rate": 8.714940098361601e-09,
      "loss": 0.0105,
      "num_input_tokens_seen": 49467072,
      "step": 14865
    },
    {
      "epoch": 2.975264487407148,
      "grad_norm": 2.3097634315490723,
      "learning_rate": 8.037037962987937e-09,
      "loss": 0.0097,
      "num_input_tokens_seen": 49482976,
      "step": 14870
    },
    {
      "epoch": 2.9762649125878498,
      "grad_norm": 0.021265488117933273,
      "learning_rate": 7.386575497014825e-09,
      "loss": 0.0065,
      "num_input_tokens_seen": 49498400,
      "step": 14875
    },
    {
      "epoch": 2.977265337768552,
      "grad_norm": 2.3299357891082764,
      "learning_rate": 6.76355341460988e-09,
      "loss": 0.0093,
      "num_input_tokens_seen": 49514880,
      "step": 14880
    },
    {
      "epoch": 2.9782657629492535,
      "grad_norm": 1.0414243936538696,
      "learning_rate": 6.167972399814815e-09,
      "loss": 0.0043,
      "num_input_tokens_seen": 49531072,
      "step": 14885
    },
    {
      "epoch": 2.979266188129955,
      "grad_norm": 0.4314385652542114,
      "learning_rate": 5.599833106539887e-09,
      "loss": 0.001,
      "num_input_tokens_seen": 49547904,
      "step": 14890
    },
    {
      "epoch": 2.980266613310657,
      "grad_norm": 0.07597378641366959,
      "learning_rate": 5.05913615856668e-09,
      "loss": 0.0085,
      "num_input_tokens_seen": 49565024,
      "step": 14895
    },
    {
      "epoch": 2.981267038491359,
      "grad_norm": 0.001904193079099059,
      "learning_rate": 4.545882149553649e-09,
      "loss": 0.0008,
      "num_input_tokens_seen": 49582720,
      "step": 14900
    },
    {
      "epoch": 2.9822674636720605,
      "grad_norm": 0.3590216040611267,
      "learning_rate": 4.060071643016694e-09,
      "loss": 0.0024,
      "num_input_tokens_seen": 49599680,
      "step": 14905
    },
    {
      "epoch": 2.9832678888527626,
      "grad_norm": 0.30538246035575867,
      "learning_rate": 3.601705172351366e-09,
      "loss": 0.0088,
      "num_input_tokens_seen": 49616352,
      "step": 14910
    },
    {
      "epoch": 2.9842683140334643,
      "grad_norm": 0.7103545069694519,
      "learning_rate": 3.1707832408134354e-09,
      "loss": 0.0015,
      "num_input_tokens_seen": 49632160,
      "step": 14915
    },
    {
      "epoch": 2.985268739214166,
      "grad_norm": 1.2359659671783447,
      "learning_rate": 2.7673063215272188e-09,
      "loss": 0.0045,
      "num_input_tokens_seen": 49649120,
      "step": 14920
    },
    {
      "epoch": 2.9862691643948676,
      "grad_norm": 0.8294517397880554,
      "learning_rate": 2.3912748574911326e-09,
      "loss": 0.0026,
      "num_input_tokens_seen": 49665216,
      "step": 14925
    },
    {
      "epoch": 2.9872695895755697,
      "grad_norm": 0.1594506800174713,
      "learning_rate": 2.042689261561037e-09,
      "loss": 0.0023,
      "num_input_tokens_seen": 49681824,
      "step": 14930
    },
    {
      "epoch": 2.9882700147562713,
      "grad_norm": 0.11420899629592896,
      "learning_rate": 1.7215499164668913e-09,
      "loss": 0.0028,
      "num_input_tokens_seen": 49698464,
      "step": 14935
    },
    {
      "epoch": 2.9892704399369734,
      "grad_norm": 0.029725858941674232,
      "learning_rate": 1.427857174796099e-09,
      "loss": 0.0049,
      "num_input_tokens_seen": 49714944,
      "step": 14940
    },
    {
      "epoch": 2.990270865117675,
      "grad_norm": 0.0007896269671618938,
      "learning_rate": 1.1616113590073863e-09,
      "loss": 0.0065,
      "num_input_tokens_seen": 49730816,
      "step": 14945
    },
    {
      "epoch": 2.9912712902983767,
      "grad_norm": 1.536004662513733,
      "learning_rate": 9.228127614252513e-10,
      "loss": 0.0089,
      "num_input_tokens_seen": 49745664,
      "step": 14950
    },
    {
      "epoch": 2.9922717154790783,
      "grad_norm": 1.1751419305801392,
      "learning_rate": 7.114616442288613e-10,
      "loss": 0.013,
      "num_input_tokens_seen": 49761792,
      "step": 14955
    },
    {
      "epoch": 2.9932721406597804,
      "grad_norm": 0.3211577832698822,
      "learning_rate": 5.275582394770328e-10,
      "loss": 0.0072,
      "num_input_tokens_seen": 49778240,
      "step": 14960
    },
    {
      "epoch": 2.994272565840482,
      "grad_norm": 0.012990407645702362,
      "learning_rate": 3.711027490804764e-10,
      "loss": 0.0008,
      "num_input_tokens_seen": 49795104,
      "step": 14965
    },
    {
      "epoch": 2.995272991021184,
      "grad_norm": 0.07811839133501053,
      "learning_rate": 2.4209534481844933e-10,
      "loss": 0.0095,
      "num_input_tokens_seen": 49812672,
      "step": 14970
    },
    {
      "epoch": 2.996273416201886,
      "grad_norm": 0.0054620616137981415,
      "learning_rate": 1.405361683359807e-10,
      "loss": 0.0044,
      "num_input_tokens_seen": 49828416,
      "step": 14975
    },
    {
      "epoch": 2.9972738413825875,
      "grad_norm": 0.0696403980255127,
      "learning_rate": 6.642533113276895e-11,
      "loss": 0.0021,
      "num_input_tokens_seen": 49844512,
      "step": 14980
    },
    {
      "epoch": 2.9982742665632895,
      "grad_norm": 0.018436448648571968,
      "learning_rate": 1.976291458538615e-11,
      "loss": 0.0004,
      "num_input_tokens_seen": 49861664,
      "step": 14985
    },
    {
      "epoch": 2.999274691743991,
      "grad_norm": 0.10342025756835938,
      "learning_rate": 5.489699195226905e-13,
      "loss": 0.0008,
      "num_input_tokens_seen": 49878144,
      "step": 14990
    },
    {
      "epoch": 2.9994747767801315,
      "num_input_tokens_seen": 49881152,
      "step": 14991,
      "total_flos": 2.4994081540447273e+18,
      "train_loss": 0.027549093207103522,
      "train_runtime": 856295.3813,
      "train_samples_per_second": 0.56,
      "train_steps_per_second": 0.018
    }
  ],
  "logging_steps": 5,
  "max_steps": 14991,
  "num_input_tokens_seen": 49881152,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.4994081540447273e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}