{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.979591836734694,
  "eval_steps": 500,
  "global_step": 610,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 312.0,
      "learning_rate": 3.278688524590164e-06,
      "loss": 33.9196,
      "step": 1
    },
    {
      "epoch": 0.04,
      "grad_norm": 354.0,
      "learning_rate": 1.6393442622950818e-05,
      "loss": 40.6749,
      "step": 5
    },
    {
      "epoch": 0.08,
      "grad_norm": 125.0,
      "learning_rate": 3.2786885245901635e-05,
      "loss": 27.6429,
      "step": 10
    },
    {
      "epoch": 0.12,
      "grad_norm": 29.875,
      "learning_rate": 4.918032786885246e-05,
      "loss": 19.7282,
      "step": 15
    },
    {
      "epoch": 0.16,
      "grad_norm": 10.1875,
      "learning_rate": 6.557377049180327e-05,
      "loss": 16.6712,
      "step": 20
    },
    {
      "epoch": 0.2,
      "grad_norm": 5.53125,
      "learning_rate": 8.19672131147541e-05,
      "loss": 14.7944,
      "step": 25
    },
    {
      "epoch": 0.24,
      "grad_norm": 7.09375,
      "learning_rate": 9.836065573770493e-05,
      "loss": 14.0597,
      "step": 30
    },
    {
      "epoch": 0.29,
      "grad_norm": 11.875,
      "learning_rate": 0.00011475409836065574,
      "loss": 12.5767,
      "step": 35
    },
    {
      "epoch": 0.33,
      "grad_norm": 55.75,
      "learning_rate": 0.00013114754098360654,
      "loss": 9.9687,
      "step": 40
    },
    {
      "epoch": 0.37,
      "grad_norm": 9.5,
      "learning_rate": 0.00014754098360655738,
      "loss": 2.8926,
      "step": 45
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.78125,
      "learning_rate": 0.0001639344262295082,
      "loss": 1.5807,
      "step": 50
    },
    {
      "epoch": 0.45,
      "grad_norm": 2.609375,
      "learning_rate": 0.00018032786885245904,
      "loss": 1.3009,
      "step": 55
    },
    {
      "epoch": 0.49,
      "grad_norm": 2.109375,
      "learning_rate": 0.00019672131147540985,
      "loss": 1.1905,
      "step": 60
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019997380453683512,
      "loss": 1.1472,
      "step": 65
    },
    {
      "epoch": 0.57,
      "grad_norm": 3.5,
      "learning_rate": 0.00019986740898848306,
      "loss": 1.0454,
      "step": 70
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.84375,
      "learning_rate": 0.00019967926316561134,
      "loss": 0.9724,
      "step": 75
    },
    {
      "epoch": 0.65,
      "grad_norm": 2.421875,
      "learning_rate": 0.00019940952108215526,
      "loss": 0.9521,
      "step": 80
    },
    {
      "epoch": 0.69,
      "grad_norm": 6.9375,
      "learning_rate": 0.00019905840354577972,
      "loss": 0.9019,
      "step": 85
    },
    {
      "epoch": 0.73,
      "grad_norm": 3.78125,
      "learning_rate": 0.00019862619797712846,
      "loss": 0.8665,
      "step": 90
    },
    {
      "epoch": 0.78,
      "grad_norm": 2.34375,
      "learning_rate": 0.00019811325817454517,
      "loss": 0.8629,
      "step": 95
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019752000402445825,
      "loss": 0.7888,
      "step": 100
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001968469211576668,
      "loss": 0.8141,
      "step": 105
    },
    {
      "epoch": 0.9,
      "grad_norm": 3.65625,
      "learning_rate": 0.00019609456055180897,
      "loss": 0.8212,
      "step": 110
    },
    {
      "epoch": 0.94,
      "grad_norm": 6.96875,
      "learning_rate": 0.00019526353808033825,
      "loss": 0.8073,
      "step": 115
    },
    {
      "epoch": 0.98,
      "grad_norm": 2.171875,
      "learning_rate": 0.00019435453400837684,
      "loss": 0.8192,
      "step": 120
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.2006253004074097,
      "eval_runtime": 1.1043,
      "eval_samples_per_second": 1.811,
      "eval_steps_per_second": 0.906,
      "step": 122
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.62890625,
      "learning_rate": 0.00019336829243585856,
      "loss": 0.7633,
      "step": 125
    },
    {
      "epoch": 1.06,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001923056206884176,
      "loss": 0.7091,
      "step": 130
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.578125,
      "learning_rate": 0.00019116738865652134,
      "loss": 0.6773,
      "step": 135
    },
    {
      "epoch": 1.14,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018995452808338823,
      "loss": 0.7058,
      "step": 140
    },
    {
      "epoch": 1.18,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018866803180227402,
      "loss": 0.7189,
      "step": 145
    },
    {
      "epoch": 1.22,
      "grad_norm": 1.71875,
      "learning_rate": 0.00018730895292375018,
      "loss": 0.6698,
      "step": 150
    },
    {
      "epoch": 1.27,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018587840397364007,
      "loss": 0.6941,
      "step": 155
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.482421875,
      "learning_rate": 0.00018437755598231856,
      "loss": 0.6511,
      "step": 160
    },
    {
      "epoch": 1.35,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018280763752612052,
      "loss": 0.6672,
      "step": 165
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.72265625,
      "learning_rate": 0.00018116993372164263,
      "loss": 0.698,
      "step": 170
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0001794657851737625,
      "loss": 0.6736,
      "step": 175
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.5234375,
      "learning_rate": 0.00017769658687823525,
      "loss": 0.651,
      "step": 180
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001758637870797665,
      "loss": 0.6809,
      "step": 185
    },
    {
      "epoch": 1.55,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001739688860864967,
      "loss": 0.6331,
      "step": 190
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00017201343504186643,
      "loss": 0.6562,
      "step": 195
    },
    {
      "epoch": 1.63,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016999903465486914,
      "loss": 0.6416,
      "step": 200
    },
    {
      "epoch": 1.67,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016792733388972932,
      "loss": 0.6957,
      "step": 205
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0001658000286160805,
      "loss": 0.6857,
      "step": 210
    },
    {
      "epoch": 1.76,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016361886022074612,
      "loss": 0.6589,
      "step": 215
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0001613856141822612,
      "loss": 0.6635,
      "step": 220
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.38671875,
      "learning_rate": 0.00015910211860930065,
      "loss": 0.6252,
      "step": 225
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0001567702427442113,
      "loss": 0.6187,
      "step": 230
    },
    {
      "epoch": 1.92,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00015439189543287247,
      "loss": 0.6194,
      "step": 235
    },
    {
      "epoch": 1.96,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015196902356213714,
      "loss": 0.6396,
      "step": 240
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0001495036104661337,
      "loss": 0.6377,
      "step": 245
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.1303967237472534,
      "eval_runtime": 1.0903,
      "eval_samples_per_second": 1.834,
      "eval_steps_per_second": 0.917,
      "step": 245
    },
    {
      "epoch": 2.04,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000146997674302732,
      "loss": 0.5446,
      "step": 250
    },
    {
      "epoch": 2.08,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0001444532664015037,
      "loss": 0.5915,
      "step": 255
    },
    {
      "epoch": 2.12,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001418724695845277,
      "loss": 0.5521,
      "step": 260
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001392573964614172,
      "loss": 0.5279,
      "step": 265
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.470703125,
      "learning_rate": 0.00013661018769996227,
      "loss": 0.558,
      "step": 270
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00013393301027380476,
      "loss": 0.5422,
      "step": 275
    },
    {
      "epoch": 2.29,
      "grad_norm": 0.53125,
      "learning_rate": 0.00013122805568857948,
      "loss": 0.5401,
      "step": 280
    },
    {
      "epoch": 2.33,
      "grad_norm": 0.6171875,
      "learning_rate": 0.00012849753818797353,
      "loss": 0.5458,
      "step": 285
    },
    {
      "epoch": 2.37,
      "grad_norm": 0.47265625,
      "learning_rate": 0.00012574369294117297,
      "loss": 0.5309,
      "step": 290
    },
    {
      "epoch": 2.41,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001229687742131796,
      "loss": 0.5276,
      "step": 295
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.6796875,
      "learning_rate": 0.00012017505351949661,
      "loss": 0.5322,
      "step": 300
    },
    {
      "epoch": 2.49,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.5242,
      "step": 305
    },
    {
      "epoch": 2.53,
      "grad_norm": 0.55859375,
      "learning_rate": 0.00011454036738036899,
      "loss": 0.5104,
      "step": 310
    },
    {
      "epoch": 2.57,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00011170401442205474,
      "loss": 0.6306,
      "step": 315
    },
    {
      "epoch": 2.61,
      "grad_norm": 0.5078125,
      "learning_rate": 0.00010885808069658451,
      "loss": 0.5246,
      "step": 320
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00010600489585149484,
      "loss": 0.5307,
      "step": 325
    },
    {
      "epoch": 2.69,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001031467954700025,
      "loss": 0.523,
      "step": 330
    },
    {
      "epoch": 2.73,
      "grad_norm": 0.44921875,
      "learning_rate": 0.00010028611915912405,
      "loss": 0.5449,
      "step": 335
    },
    {
      "epoch": 2.78,
      "grad_norm": 0.443359375,
      "learning_rate": 9.742520863450115e-05,
      "loss": 0.5226,
      "step": 340
    },
    {
      "epoch": 2.82,
      "grad_norm": 0.466796875,
      "learning_rate": 9.456640580350018e-05,
      "loss": 0.5173,
      "step": 345
    },
    {
      "epoch": 2.86,
      "grad_norm": 0.76953125,
      "learning_rate": 9.171205084815444e-05,
      "loss": 0.5281,
      "step": 350
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.6796875,
      "learning_rate": 8.886448030951912e-05,
      "loss": 0.543,
      "step": 355
    },
    {
      "epoch": 2.94,
      "grad_norm": 0.427734375,
      "learning_rate": 8.602602517500651e-05,
      "loss": 0.5459,
      "step": 360
    },
    {
      "epoch": 2.98,
      "grad_norm": 0.5,
      "learning_rate": 8.319900897026733e-05,
      "loss": 0.5334,
      "step": 365
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.1456480026245117,
      "eval_runtime": 1.0913,
      "eval_samples_per_second": 1.833,
      "eval_steps_per_second": 0.916,
      "step": 367
    },
    {
      "epoch": 3.02,
      "grad_norm": 0.494140625,
      "learning_rate": 8.038574585718032e-05,
      "loss": 0.5025,
      "step": 370
    },
    {
      "epoch": 3.06,
      "grad_norm": 0.69921875,
      "learning_rate": 7.758853873950676e-05,
      "loss": 0.4247,
      "step": 375
    },
    {
      "epoch": 3.1,
      "grad_norm": 0.48828125,
      "learning_rate": 7.480967737776089e-05,
      "loss": 0.4551,
      "step": 380
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.58984375,
      "learning_rate": 7.205143651483906e-05,
      "loss": 0.4667,
      "step": 385
    },
    {
      "epoch": 3.18,
      "grad_norm": 0.59375,
      "learning_rate": 6.931607401394229e-05,
      "loss": 0.4581,
      "step": 390
    },
    {
      "epoch": 3.22,
      "grad_norm": 0.7421875,
      "learning_rate": 6.660582901031621e-05,
      "loss": 0.4441,
      "step": 395
    },
    {
      "epoch": 3.27,
      "grad_norm": 0.546875,
      "learning_rate": 6.392292007832168e-05,
      "loss": 0.4247,
      "step": 400
    },
    {
      "epoch": 3.31,
      "grad_norm": 0.66796875,
      "learning_rate": 6.126954341533599e-05,
      "loss": 0.4214,
      "step": 405
    },
    {
      "epoch": 3.35,
      "grad_norm": 0.48046875,
      "learning_rate": 5.864787104397194e-05,
      "loss": 0.4413,
      "step": 410
    },
    {
      "epoch": 3.39,
      "grad_norm": 0.65234375,
      "learning_rate": 5.6060049034085815e-05,
      "loss": 0.4648,
      "step": 415
    },
    {
      "epoch": 3.43,
      "grad_norm": 0.53125,
      "learning_rate": 5.3508195746030296e-05,
      "loss": 0.5039,
      "step": 420
    },
    {
      "epoch": 3.47,
      "grad_norm": 0.498046875,
      "learning_rate": 5.099440009658957e-05,
      "loss": 0.4493,
      "step": 425
    },
    {
      "epoch": 3.51,
      "grad_norm": 0.43359375,
      "learning_rate": 4.852071984901696e-05,
      "loss": 0.4343,
      "step": 430
    },
    {
      "epoch": 3.55,
      "grad_norm": 0.52734375,
      "learning_rate": 4.6089179928574487e-05,
      "loss": 0.4354,
      "step": 435
    },
    {
      "epoch": 3.59,
      "grad_norm": 0.48046875,
      "learning_rate": 4.3701770764953124e-05,
      "loss": 0.4694,
      "step": 440
    },
    {
      "epoch": 3.63,
      "grad_norm": 0.431640625,
      "learning_rate": 4.136044666293044e-05,
      "loss": 0.4446,
      "step": 445
    },
    {
      "epoch": 3.67,
      "grad_norm": 0.447265625,
      "learning_rate": 3.9067124202600194e-05,
      "loss": 0.4398,
      "step": 450
    },
    {
      "epoch": 3.71,
      "grad_norm": 0.4609375,
      "learning_rate": 3.6823680670482485e-05,
      "loss": 0.4449,
      "step": 455
    },
    {
      "epoch": 3.76,
      "grad_norm": 0.4765625,
      "learning_rate": 3.463195252279939e-05,
      "loss": 0.4443,
      "step": 460
    },
    {
      "epoch": 3.8,
      "grad_norm": 0.423828125,
      "learning_rate": 3.24937338821738e-05,
      "loss": 0.4346,
      "step": 465
    },
    {
      "epoch": 3.84,
      "grad_norm": 0.4453125,
      "learning_rate": 3.0410775068981613e-05,
      "loss": 0.4423,
      "step": 470
    },
    {
      "epoch": 3.88,
      "grad_norm": 0.828125,
      "learning_rate": 2.8384781168560693e-05,
      "loss": 0.4304,
      "step": 475
    },
    {
      "epoch": 3.92,
      "grad_norm": 0.490234375,
      "learning_rate": 2.6417410635448013e-05,
      "loss": 0.443,
      "step": 480
    },
    {
      "epoch": 3.96,
      "grad_norm": 0.46875,
      "learning_rate": 2.451027393578863e-05,
      "loss": 0.4392,
      "step": 485
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5,
      "learning_rate": 2.2664932229027024e-05,
      "loss": 0.4454,
      "step": 490
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.1934518814086914,
      "eval_runtime": 1.0901,
      "eval_samples_per_second": 1.835,
      "eval_steps_per_second": 0.917,
      "step": 490
    },
    {
      "epoch": 4.04,
      "grad_norm": 0.45703125,
      "learning_rate": 2.088289608996071e-05,
      "loss": 0.3925,
      "step": 495
    },
    {
      "epoch": 4.08,
      "grad_norm": 0.4921875,
      "learning_rate": 1.9165624272201567e-05,
      "loss": 0.403,
      "step": 500
    },
    {
      "epoch": 4.12,
      "grad_norm": 0.5,
      "learning_rate": 1.7514522514057553e-05,
      "loss": 0.3824,
      "step": 505
    },
    {
      "epoch": 4.16,
      "grad_norm": 0.462890625,
      "learning_rate": 1.593094238781191e-05,
      "loss": 0.3818,
      "step": 510
    },
    {
      "epoch": 4.2,
      "grad_norm": 0.47265625,
      "learning_rate": 1.4416180193342144e-05,
      "loss": 0.4017,
      "step": 515
    },
    {
      "epoch": 4.24,
      "grad_norm": 0.4609375,
      "learning_rate": 1.2971475896984475e-05,
      "loss": 0.4047,
      "step": 520
    },
    {
      "epoch": 4.29,
      "grad_norm": 0.466796875,
      "learning_rate": 1.1598012116512092e-05,
      "loss": 0.3735,
      "step": 525
    },
    {
      "epoch": 4.33,
      "grad_norm": 0.47265625,
      "learning_rate": 1.0296913153058163e-05,
      "loss": 0.3954,
      "step": 530
    },
    {
      "epoch": 4.37,
      "grad_norm": 0.49609375,
      "learning_rate": 9.069244070776428e-06,
      "loss": 0.3955,
      "step": 535
    },
    {
      "epoch": 4.41,
      "grad_norm": 0.5,
      "learning_rate": 7.916009824992254e-06,
      "loss": 0.3824,
      "step": 540
    },
    {
      "epoch": 4.45,
      "grad_norm": 0.53125,
      "learning_rate": 6.838154439558253e-06,
      "loss": 0.3846,
      "step": 545
    },
    {
      "epoch": 4.49,
      "grad_norm": 0.455078125,
      "learning_rate": 5.836560234087418e-06,
      "loss": 0.4087,
      "step": 550
    },
    {
      "epoch": 4.53,
      "grad_norm": 0.455078125,
      "learning_rate": 4.912047101696848e-06,
      "loss": 0.3785,
      "step": 555
    },
    {
      "epoch": 4.57,
      "grad_norm": 0.48828125,
      "learning_rate": 4.065371837853016e-06,
      "loss": 0.4013,
      "step": 560
    },
    {
      "epoch": 4.61,
      "grad_norm": 0.55859375,
      "learning_rate": 3.2972275208679625e-06,
      "loss": 0.4005,
      "step": 565
    },
    {
      "epoch": 4.65,
      "grad_norm": 0.498046875,
      "learning_rate": 2.608242944553607e-06,
      "loss": 0.3842,
      "step": 570
    },
    {
      "epoch": 4.69,
      "grad_norm": 0.470703125,
      "learning_rate": 1.9989821034986032e-06,
      "loss": 0.3791,
      "step": 575
    },
    {
      "epoch": 4.73,
      "grad_norm": 0.494140625,
      "learning_rate": 1.4699437313891007e-06,
      "loss": 0.398,
      "step": 580
    },
    {
      "epoch": 4.78,
      "grad_norm": 0.482421875,
      "learning_rate": 1.02156089275125e-06,
      "loss": 0.3836,
      "step": 585
    },
    {
      "epoch": 4.82,
      "grad_norm": 0.55078125,
      "learning_rate": 6.542006284497304e-07,
      "loss": 0.3827,
      "step": 590
    },
    {
      "epoch": 4.86,
      "grad_norm": 0.4296875,
      "learning_rate": 3.681636552324452e-07,
      "loss": 0.3846,
      "step": 595
    },
    {
      "epoch": 4.9,
      "grad_norm": 0.4609375,
      "learning_rate": 1.6368411956741991e-07,
      "loss": 0.4347,
      "step": 600
    },
    {
      "epoch": 4.94,
      "grad_norm": 0.47265625,
      "learning_rate": 4.092940597322237e-08,
      "loss": 0.3899,
      "step": 605
    },
    {
      "epoch": 4.98,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0,
      "loss": 0.408,
      "step": 610
    },
    {
      "epoch": 4.98,
      "eval_loss": 1.2246639728546143,
      "eval_runtime": 1.1185,
      "eval_samples_per_second": 1.788,
      "eval_steps_per_second": 0.894,
      "step": 610
    },
    {
      "epoch": 4.98,
      "step": 610,
      "total_flos": 4.678055725535068e+17,
      "train_loss": 1.826815680011374,
      "train_runtime": 4235.156,
      "train_samples_per_second": 1.156,
      "train_steps_per_second": 0.144
    }
  ],
  "logging_steps": 5,
  "max_steps": 610,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "total_flos": 4.678055725535068e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}