{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6666666666666666,
  "eval_steps": 500,
  "global_step": 2744,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012147716229348883,
      "grad_norm": 22.0,
      "learning_rate": 3.380914285714285e-05,
      "loss": 2.1764,
      "step": 5
    },
    {
      "epoch": 0.0024295432458697765,
      "grad_norm": 13.1875,
      "learning_rate": 7.607057142857141e-05,
      "loss": 1.8601,
      "step": 10
    },
    {
      "epoch": 0.0036443148688046646,
      "grad_norm": 6.6875,
      "learning_rate": 0.00011833199999999999,
      "loss": 1.1947,
      "step": 15
    },
    {
      "epoch": 0.004859086491739553,
      "grad_norm": 2.65625,
      "learning_rate": 0.00016059342857142853,
      "loss": 0.7752,
      "step": 20
    },
    {
      "epoch": 0.0060738581146744415,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00020285485714285713,
      "loss": 0.546,
      "step": 25
    },
    {
      "epoch": 0.007288629737609329,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002451162857142857,
      "loss": 0.4864,
      "step": 30
    },
    {
      "epoch": 0.008503401360544218,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00028737771428571425,
      "loss": 0.4397,
      "step": 35
    },
    {
      "epoch": 0.009718172983479106,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029582994222560353,
      "loss": 0.466,
      "step": 40
    },
    {
      "epoch": 0.010932944606413994,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029582970751722125,
      "loss": 0.4189,
      "step": 45
    },
    {
      "epoch": 0.012147716229348883,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029582929226433497,
      "loss": 0.4224,
      "step": 50
    },
    {
      "epoch": 0.01336248785228377,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0002958286964676205,
      "loss": 0.4094,
      "step": 55
    },
    {
      "epoch": 0.014577259475218658,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002958279201280475,
      "loss": 0.3838,
      "step": 60
    },
    {
      "epoch": 0.015792031098153546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029582696324687946,
      "loss": 0.389,
      "step": 65
    },
    {
      "epoch": 0.017006802721088437,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002958258258256736,
      "loss": 0.3934,
      "step": 70
    },
    {
      "epoch": 0.017006802721088437,
      "eval_loss": 0.363860160112381,
      "eval_runtime": 2.4278,
      "eval_samples_per_second": 82.378,
      "eval_steps_per_second": 82.378,
      "step": 70
    },
    {
      "epoch": 0.018221574344023325,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002958245078662811,
      "loss": 0.3635,
      "step": 75
    },
    {
      "epoch": 0.019436345966958212,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029582300937084694,
      "loss": 0.354,
      "step": 80
    },
    {
      "epoch": 0.0206511175898931,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00029582133034180983,
      "loss": 0.3154,
      "step": 85
    },
    {
      "epoch": 0.021865889212827987,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002958194707819022,
      "loss": 0.3332,
      "step": 90
    },
    {
      "epoch": 0.023080660835762875,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00029581743069415057,
      "loss": 0.3235,
      "step": 95
    },
    {
      "epoch": 0.024295432458697766,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029581521008187507,
      "loss": 0.3195,
      "step": 100
    },
    {
      "epoch": 0.025510204081632654,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00029581280894868967,
      "loss": 0.3159,
      "step": 105
    },
    {
      "epoch": 0.02672497570456754,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029581022729850215,
      "loss": 0.3072,
      "step": 110
    },
    {
      "epoch": 0.02793974732750243,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000295807465135514,
      "loss": 0.3268,
      "step": 115
    },
    {
      "epoch": 0.029154518950437316,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00029580452246422064,
      "loss": 0.2857,
      "step": 120
    },
    {
      "epoch": 0.030369290573372208,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002958013992894111,
      "loss": 0.2897,
      "step": 125
    },
    {
      "epoch": 0.03158406219630709,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0002957980956161683,
      "loss": 0.285,
      "step": 130
    },
    {
      "epoch": 0.03279883381924198,
      "grad_norm": 0.76953125,
      "learning_rate": 0.00029579461144986876,
      "loss": 0.2759,
      "step": 135
    },
    {
      "epoch": 0.034013605442176874,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00029579094679618294,
      "loss": 0.2859,
      "step": 140
    },
    {
      "epoch": 0.03522837706511176,
      "grad_norm": 0.78125,
      "learning_rate": 0.00029578710166107484,
      "loss": 0.2717,
      "step": 145
    },
    {
      "epoch": 0.03644314868804665,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002957830760508024,
      "loss": 0.2828,
      "step": 150
    },
    {
      "epoch": 0.03765792031098154,
      "grad_norm": 0.875,
      "learning_rate": 0.0002957788699719171,
      "loss": 0.2809,
      "step": 155
    },
    {
      "epoch": 0.038872691933916424,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029577448343126415,
      "loss": 0.2889,
      "step": 160
    },
    {
      "epoch": 0.04008746355685131,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00029576991643598264,
      "loss": 0.2642,
      "step": 165
    },
    {
      "epoch": 0.0413022351797862,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029576516899350503,
      "loss": 0.2701,
      "step": 170
    },
    {
      "epoch": 0.04251700680272109,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002957602411115577,
      "loss": 0.2729,
      "step": 175
    },
    {
      "epoch": 0.043731778425655975,
      "grad_norm": 1.5,
      "learning_rate": 0.00029575513279816063,
      "loss": 0.2628,
      "step": 180
    },
    {
      "epoch": 0.04494655004859086,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0002957498440616274,
      "loss": 0.2432,
      "step": 185
    },
    {
      "epoch": 0.04616132167152575,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00029574437491056523,
      "loss": 0.2525,
      "step": 190
    },
    {
      "epoch": 0.047376093294460644,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002957387253538751,
      "loss": 0.265,
      "step": 195
    },
    {
      "epoch": 0.04859086491739553,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00029573289540075126,
      "loss": 0.2574,
      "step": 200
    },
    {
      "epoch": 0.04980563654033042,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002957268850606819,
      "loss": 0.2546,
      "step": 205
    },
    {
      "epoch": 0.05102040816326531,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00029572069434344864,
      "loss": 0.2483,
      "step": 210
    },
    {
      "epoch": 0.052235179786200195,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00029571432325912664,
      "loss": 0.2547,
      "step": 215
    },
    {
      "epoch": 0.05344995140913508,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0002957077718180845,
      "loss": 0.2413,
      "step": 220
    },
    {
      "epoch": 0.05466472303206997,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029570104003098476,
      "loss": 0.2385,
      "step": 225
    },
    {
      "epoch": 0.05587949465500486,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029569412790878295,
      "loss": 0.2319,
      "step": 230
    },
    {
      "epoch": 0.057094266277939745,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0002956870354627284,
      "loss": 0.2327,
      "step": 235
    },
    {
      "epoch": 0.05830903790087463,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029567976270436367,
      "loss": 0.2172,
      "step": 240
    },
    {
      "epoch": 0.05952380952380952,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002956723096455252,
      "loss": 0.2508,
      "step": 245
    },
    {
      "epoch": 0.060738581146744415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029566467629834237,
      "loss": 0.2238,
      "step": 250
    },
    {
      "epoch": 0.0619533527696793,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002956568626752382,
      "loss": 0.2169,
      "step": 255
    },
    {
      "epoch": 0.06316812439261418,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002956488687889292,
      "loss": 0.2381,
      "step": 260
    },
    {
      "epoch": 0.06438289601554907,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0002956406946524252,
      "loss": 0.2084,
      "step": 265
    },
    {
      "epoch": 0.06559766763848396,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002956323402790292,
      "loss": 0.2272,
      "step": 270
    },
    {
      "epoch": 0.06681243926141886,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002956238056823377,
      "loss": 0.2233,
      "step": 275
    },
    {
      "epoch": 0.06802721088435375,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029561509087624045,
      "loss": 0.2112,
      "step": 280
    },
    {
      "epoch": 0.06924198250728864,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029560619587492047,
      "loss": 0.2077,
      "step": 285
    },
    {
      "epoch": 0.07045675413022352,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002955971206928542,
      "loss": 0.2171,
      "step": 290
    },
    {
      "epoch": 0.07167152575315841,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029558786534481104,
      "loss": 0.2065,
      "step": 295
    },
    {
      "epoch": 0.0728862973760933,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029557842984585385,
      "loss": 0.2142,
      "step": 300
    },
    {
      "epoch": 0.07410106899902819,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029556881421133856,
      "loss": 0.2114,
      "step": 305
    },
    {
      "epoch": 0.07531584062196307,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002955590184569143,
      "loss": 0.2145,
      "step": 310
    },
    {
      "epoch": 0.07653061224489796,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002955490425985233,
      "loss": 0.1923,
      "step": 315
    },
    {
      "epoch": 0.07774538386783285,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00029553888665240094,
      "loss": 0.189,
      "step": 320
    },
    {
      "epoch": 0.07896015549076774,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002955285506350757,
      "loss": 0.1808,
      "step": 325
    },
    {
      "epoch": 0.08017492711370262,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002955180345633691,
      "loss": 0.1807,
      "step": 330
    },
    {
      "epoch": 0.08138969873663751,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002955073384543956,
      "loss": 0.1841,
      "step": 335
    },
    {
      "epoch": 0.0826044703595724,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002954964623255629,
      "loss": 0.1794,
      "step": 340
    },
    {
      "epoch": 0.08381924198250729,
      "grad_norm": 1.0,
      "learning_rate": 0.0002954854061945714,
      "loss": 0.1816,
      "step": 345
    },
    {
      "epoch": 0.08503401360544217,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002954741700794147,
      "loss": 0.1929,
      "step": 350
    },
    {
      "epoch": 0.08624878522837706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029546275399837906,
      "loss": 0.1828,
      "step": 355
    },
    {
      "epoch": 0.08746355685131195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029545115797004385,
      "loss": 0.18,
      "step": 360
    },
    {
      "epoch": 0.08867832847424684,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002954393820132812,
      "loss": 0.1868,
      "step": 365
    },
    {
      "epoch": 0.08989310009718172,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00029542742614725606,
      "loss": 0.1722,
      "step": 370
    },
    {
      "epoch": 0.09110787172011661,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002954152903914262,
      "loss": 0.1766,
      "step": 375
    },
    {
      "epoch": 0.0923226433430515,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002954029747655422,
      "loss": 0.1594,
      "step": 380
    },
    {
      "epoch": 0.0935374149659864,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002953904792896473,
      "loss": 0.188,
      "step": 385
    },
    {
      "epoch": 0.09475218658892129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002953778039840774,
      "loss": 0.1933,
      "step": 390
    },
    {
      "epoch": 0.09596695821185618,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002953649488694612,
      "loss": 0.1601,
      "step": 395
    },
    {
      "epoch": 0.09718172983479106,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002953519139667199,
      "loss": 0.1704,
      "step": 400
    },
    {
      "epoch": 0.09839650145772595,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00029533869929706743,
      "loss": 0.1622,
      "step": 405
    },
    {
      "epoch": 0.09961127308066084,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002953253048820102,
      "loss": 0.1602,
      "step": 410
    },
    {
      "epoch": 0.10082604470359573,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029531173074334716,
      "loss": 0.1599,
      "step": 415
    },
    {
      "epoch": 0.10204081632653061,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002952979769031697,
      "loss": 0.1603,
      "step": 420
    },
    {
      "epoch": 0.1032555879494655,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002952840433838618,
      "loss": 0.1642,
      "step": 425
    },
    {
      "epoch": 0.10447035957240039,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029526993020809975,
      "loss": 0.1506,
      "step": 430
    },
    {
      "epoch": 0.10568513119533528,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029525563739885236,
      "loss": 0.1698,
      "step": 435
    },
    {
      "epoch": 0.10689990281827016,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029524116497938054,
      "loss": 0.1702,
      "step": 440
    },
    {
      "epoch": 0.10811467444120505,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0002952265129732378,
      "loss": 0.1371,
      "step": 445
    },
    {
      "epoch": 0.10932944606413994,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00029521168140426963,
      "loss": 0.1422,
      "step": 450
    },
    {
      "epoch": 0.11054421768707483,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00029519667029661407,
      "loss": 0.1522,
      "step": 455
    },
    {
      "epoch": 0.11175898931000972,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002951814796747011,
      "loss": 0.1556,
      "step": 460
    },
    {
      "epoch": 0.1129737609329446,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002951661095632529,
      "loss": 0.1396,
      "step": 465
    },
    {
      "epoch": 0.11418853255587949,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002951505599872839,
      "loss": 0.1425,
      "step": 470
    },
    {
      "epoch": 0.11540330417881438,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002951348309721005,
      "loss": 0.1539,
      "step": 475
    },
    {
      "epoch": 0.11661807580174927,
      "grad_norm": 1.09375,
      "learning_rate": 0.000295118922543301,
      "loss": 0.1441,
      "step": 480
    },
    {
      "epoch": 0.11783284742468415,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029510283472677603,
      "loss": 0.1353,
      "step": 485
    },
    {
      "epoch": 0.11904761904761904,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00029508656754870786,
      "loss": 0.1262,
      "step": 490
    },
    {
      "epoch": 0.12026239067055394,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029507012103557076,
      "loss": 0.1378,
      "step": 495
    },
    {
      "epoch": 0.12147716229348883,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029505349521413086,
      "loss": 0.1396,
      "step": 500
    },
    {
      "epoch": 0.12147716229348883,
      "eval_loss": 0.1296369582414627,
      "eval_runtime": 2.4106,
      "eval_samples_per_second": 82.967,
      "eval_steps_per_second": 82.967,
      "step": 500
    },
    {
      "epoch": 0.12269193391642372,
      "grad_norm": 0.9375,
      "learning_rate": 0.00029503669011144616,
      "loss": 0.1333,
      "step": 505
    },
    {
      "epoch": 0.1239067055393586,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002950197057548664,
      "loss": 0.1352,
      "step": 510
    },
    {
      "epoch": 0.1251214771622935,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029500254217203296,
      "loss": 0.1329,
      "step": 515
    },
    {
      "epoch": 0.12633624878522837,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002949851993908791,
      "loss": 0.1254,
      "step": 520
    },
    {
      "epoch": 0.12755102040816327,
      "grad_norm": 1.0,
      "learning_rate": 0.00029496767743962955,
      "loss": 0.1302,
      "step": 525
    },
    {
      "epoch": 0.12876579203109814,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00029494997634680074,
      "loss": 0.1328,
      "step": 530
    },
    {
      "epoch": 0.12998056365403304,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002949320961412005,
      "loss": 0.1416,
      "step": 535
    },
    {
      "epoch": 0.13119533527696792,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029491403685192837,
      "loss": 0.1254,
      "step": 540
    },
    {
      "epoch": 0.13241010689990282,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002948957985083752,
      "loss": 0.1286,
      "step": 545
    },
    {
      "epoch": 0.13362487852283772,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029487738114022314,
      "loss": 0.1353,
      "step": 550
    },
    {
      "epoch": 0.1348396501457726,
      "grad_norm": 0.953125,
      "learning_rate": 0.000294858784777446,
      "loss": 0.1263,
      "step": 555
    },
    {
      "epoch": 0.1360544217687075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002948400094503087,
      "loss": 0.121,
      "step": 560
    },
    {
      "epoch": 0.13726919339164237,
      "grad_norm": 0.9375,
      "learning_rate": 0.00029482105518936736,
      "loss": 0.112,
      "step": 565
    },
    {
      "epoch": 0.13848396501457727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002948019220254695,
      "loss": 0.1174,
      "step": 570
    },
    {
      "epoch": 0.13969873663751214,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029478260998975364,
      "loss": 0.1225,
      "step": 575
    },
    {
      "epoch": 0.14091350826044705,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00029476311911364947,
      "loss": 0.1152,
      "step": 580
    },
    {
      "epoch": 0.14212827988338192,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002947434494288777,
      "loss": 0.125,
      "step": 585
    },
    {
      "epoch": 0.14334305150631682,
      "grad_norm": 1.0,
      "learning_rate": 0.00029472360096745016,
      "loss": 0.1102,
      "step": 590
    },
    {
      "epoch": 0.1445578231292517,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00029470357376166947,
      "loss": 0.1159,
      "step": 595
    },
    {
      "epoch": 0.1457725947521866,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002946833678441292,
      "loss": 0.1125,
      "step": 600
    },
    {
      "epoch": 0.14698736637512147,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00029466298324771394,
      "loss": 0.0989,
      "step": 605
    },
    {
      "epoch": 0.14820213799805637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002946424200055988,
      "loss": 0.1196,
      "step": 610
    },
    {
      "epoch": 0.14941690962099125,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029462167815124974,
      "loss": 0.1095,
      "step": 615
    },
    {
      "epoch": 0.15063168124392615,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0002946007577184235,
      "loss": 0.1134,
      "step": 620
    },
    {
      "epoch": 0.15184645286686102,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029457965874116735,
      "loss": 0.121,
      "step": 625
    },
    {
      "epoch": 0.15306122448979592,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002945583812538192,
      "loss": 0.1169,
      "step": 630
    },
    {
      "epoch": 0.1542759961127308,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002945369252910073,
      "loss": 0.1033,
      "step": 635
    },
    {
      "epoch": 0.1554907677356657,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002945152908876507,
      "loss": 0.1065,
      "step": 640
    },
    {
      "epoch": 0.15670553935860057,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0002944934780789585,
      "loss": 0.0958,
      "step": 645
    },
    {
      "epoch": 0.15792031098153547,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029447148690043033,
      "loss": 0.1126,
      "step": 650
    },
    {
      "epoch": 0.15913508260447035,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002944493173878562,
      "loss": 0.1003,
      "step": 655
    },
    {
      "epoch": 0.16034985422740525,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002944269695773161,
      "loss": 0.1002,
      "step": 660
    },
    {
      "epoch": 0.16156462585034015,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002944044435051804,
      "loss": 0.1039,
      "step": 665
    },
    {
      "epoch": 0.16277939747327502,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029438173920810946,
      "loss": 0.1068,
      "step": 670
    },
    {
      "epoch": 0.16399416909620992,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00029435885672305387,
      "loss": 0.1064,
      "step": 675
    },
    {
      "epoch": 0.1652089407191448,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00029433579608725396,
      "loss": 0.0894,
      "step": 680
    },
    {
      "epoch": 0.1664237123420797,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029431255733824025,
      "loss": 0.0958,
      "step": 685
    },
    {
      "epoch": 0.16763848396501457,
      "grad_norm": 0.84375,
      "learning_rate": 0.0002942891405138329,
      "loss": 0.0839,
      "step": 690
    },
    {
      "epoch": 0.16885325558794947,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002942655456521421,
      "loss": 0.0967,
      "step": 695
    },
    {
      "epoch": 0.17006802721088435,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029424177279156765,
      "loss": 0.0935,
      "step": 700
    },
    {
      "epoch": 0.17128279883381925,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00029421782197079903,
      "loss": 0.0914,
      "step": 705
    },
    {
      "epoch": 0.17249757045675412,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002941936932288155,
      "loss": 0.099,
      "step": 710
    },
    {
      "epoch": 0.17371234207968902,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00029416938660488563,
      "loss": 0.0886,
      "step": 715
    },
    {
      "epoch": 0.1749271137026239,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002941449021385677,
      "loss": 0.0918,
      "step": 720
    },
    {
      "epoch": 0.1761418853255588,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029412023986970936,
      "loss": 0.0925,
      "step": 725
    },
    {
      "epoch": 0.17735665694849367,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00029409539983844753,
      "loss": 0.0897,
      "step": 730
    },
    {
      "epoch": 0.17857142857142858,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00029407038208520864,
      "loss": 0.0943,
      "step": 735
    },
    {
      "epoch": 0.17978620019436345,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002940451866507082,
      "loss": 0.0898,
      "step": 740
    },
    {
      "epoch": 0.18100097181729835,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029401981357595077,
      "loss": 0.0875,
      "step": 745
    },
    {
      "epoch": 0.18221574344023322,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029399426290223037,
      "loss": 0.0888,
      "step": 750
    },
    {
      "epoch": 0.18343051506316813,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002939685346711298,
      "loss": 0.086,
      "step": 755
    },
    {
      "epoch": 0.184645286686103,
      "grad_norm": 1.125,
      "learning_rate": 0.0002939426289245208,
      "loss": 0.0829,
      "step": 760
    },
    {
      "epoch": 0.1858600583090379,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029391654570456424,
      "loss": 0.0945,
      "step": 765
    },
    {
      "epoch": 0.1870748299319728,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002938902850537095,
      "loss": 0.0976,
      "step": 770
    },
    {
      "epoch": 0.18828960155490768,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00029386384701469504,
      "loss": 0.0869,
      "step": 775
    },
    {
      "epoch": 0.18950437317784258,
      "grad_norm": 0.890625,
      "learning_rate": 0.0002938372316305478,
      "loss": 0.0758,
      "step": 780
    },
    {
      "epoch": 0.19071914480077745,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002938104389445834,
      "loss": 0.0853,
      "step": 785
    },
    {
      "epoch": 0.19193391642371235,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002937834690004061,
      "loss": 0.0896,
      "step": 790
    },
    {
      "epoch": 0.19314868804664723,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002937563218419085,
      "loss": 0.0852,
      "step": 795
    },
    {
      "epoch": 0.19436345966958213,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029372899751327173,
      "loss": 0.0794,
      "step": 800
    },
    {
      "epoch": 0.195578231292517,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00029370149605896517,
      "loss": 0.0718,
      "step": 805
    },
    {
      "epoch": 0.1967930029154519,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029367381752374655,
      "loss": 0.083,
      "step": 810
    },
    {
      "epoch": 0.19800777453838678,
      "grad_norm": 0.9375,
      "learning_rate": 0.00029364596195266166,
      "loss": 0.079,
      "step": 815
    },
    {
      "epoch": 0.19922254616132168,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002936179293910445,
      "loss": 0.0723,
      "step": 820
    },
    {
      "epoch": 0.20043731778425655,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00029358971988451726,
      "loss": 0.0724,
      "step": 825
    },
    {
      "epoch": 0.20165208940719145,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029356133347898983,
      "loss": 0.079,
      "step": 830
    },
    {
      "epoch": 0.20286686103012633,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029353277022066014,
      "loss": 0.0748,
      "step": 835
    },
    {
      "epoch": 0.20408163265306123,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002935040301560139,
      "loss": 0.0798,
      "step": 840
    },
    {
      "epoch": 0.2052964042759961,
      "grad_norm": 0.984375,
      "learning_rate": 0.00029347511333182463,
      "loss": 0.0737,
      "step": 845
    },
    {
      "epoch": 0.206511175898931,
      "grad_norm": 0.953125,
      "learning_rate": 0.00029344601979515347,
      "loss": 0.0766,
      "step": 850
    },
    {
      "epoch": 0.20772594752186588,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00029341674959334915,
      "loss": 0.0697,
      "step": 855
    },
    {
      "epoch": 0.20894071914480078,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0002933873027740479,
      "loss": 0.073,
      "step": 860
    },
    {
      "epoch": 0.21015549076773565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029335767938517356,
      "loss": 0.0694,
      "step": 865
    },
    {
      "epoch": 0.21137026239067055,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000293327879474937,
      "loss": 0.0736,
      "step": 870
    },
    {
      "epoch": 0.21258503401360543,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002932979030918367,
      "loss": 0.0774,
      "step": 875
    },
    {
      "epoch": 0.21379980563654033,
      "grad_norm": 0.921875,
      "learning_rate": 0.0002932677502846581,
      "loss": 0.0586,
      "step": 880
    },
    {
      "epoch": 0.21501457725947523,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000293237421102474,
      "loss": 0.0706,
      "step": 885
    },
    {
      "epoch": 0.2162293488824101,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029320691559464395,
      "loss": 0.0729,
      "step": 890
    },
    {
      "epoch": 0.217444120505345,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00029317623381081476,
      "loss": 0.0626,
      "step": 895
    },
    {
      "epoch": 0.21865889212827988,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0002931453758009199,
      "loss": 0.0691,
      "step": 900
    },
    {
      "epoch": 0.21987366375121478,
      "grad_norm": 0.953125,
      "learning_rate": 0.00029311434161517986,
      "loss": 0.0737,
      "step": 905
    },
    {
      "epoch": 0.22108843537414966,
      "grad_norm": 0.875,
      "learning_rate": 0.0002930831313041016,
      "loss": 0.0594,
      "step": 910
    },
    {
      "epoch": 0.22230320699708456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002930517449184789,
      "loss": 0.0633,
      "step": 915
    },
    {
      "epoch": 0.22351797862001943,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00029302018250939205,
      "loss": 0.0693,
      "step": 920
    },
    {
      "epoch": 0.22473275024295433,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0002929884441282077,
      "loss": 0.0637,
      "step": 925
    },
    {
      "epoch": 0.2259475218658892,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002929565298265791,
      "loss": 0.062,
      "step": 930
    },
    {
      "epoch": 0.2271622934888241,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00029292443965644565,
      "loss": 0.0686,
      "step": 935
    },
    {
      "epoch": 0.22837706511175898,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000292892173670033,
      "loss": 0.064,
      "step": 940
    },
    {
      "epoch": 0.22959183673469388,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000292859731919853,
      "loss": 0.0624,
      "step": 945
    },
    {
      "epoch": 0.23080660835762876,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00029282711445870343,
      "loss": 0.0718,
      "step": 950
    },
    {
      "epoch": 0.23202137998056366,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002927943213396681,
      "loss": 0.0692,
      "step": 955
    },
    {
      "epoch": 0.23323615160349853,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0002927613526161168,
      "loss": 0.0661,
      "step": 960
    },
    {
      "epoch": 0.23445092322643343,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0002927282083417049,
      "loss": 0.0504,
      "step": 965
    },
    {
      "epoch": 0.2356656948493683,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0002926948885703736,
      "loss": 0.0645,
      "step": 970
    },
    {
      "epoch": 0.2368804664723032,
      "grad_norm": 0.921875,
      "learning_rate": 0.00029266139335634977,
      "loss": 0.0632,
      "step": 975
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002926277227541456,
      "loss": 0.0665,
      "step": 980
    },
    {
      "epoch": 0.23931000971817298,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00029259387681855896,
      "loss": 0.0593,
      "step": 985
    },
    {
      "epoch": 0.24052478134110788,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002925598556046729,
      "loss": 0.0673,
      "step": 990
    },
    {
      "epoch": 0.24173955296404276,
      "grad_norm": 0.9375,
      "learning_rate": 0.00029252565916785585,
      "loss": 0.062,
      "step": 995
    },
    {
      "epoch": 0.24295432458697766,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00029249128756376135,
      "loss": 0.0641,
      "step": 1000
    },
    {
      "epoch": 0.24295432458697766,
      "eval_loss": 0.05327101796865463,
      "eval_runtime": 2.4201,
      "eval_samples_per_second": 82.641,
      "eval_steps_per_second": 82.641,
      "step": 1000
    },
    {
      "epoch": 0.24416909620991253,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002924567408483279,
      "loss": 0.0633,
      "step": 1005
    },
    {
      "epoch": 0.24538386783284744,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0002924220190777792,
      "loss": 0.0604,
      "step": 1010
    },
    {
      "epoch": 0.2465986394557823,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029238712230862386,
      "loss": 0.0571,
      "step": 1015
    },
    {
      "epoch": 0.2478134110787172,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000292352050597655,
      "loss": 0.0551,
      "step": 1020
    },
    {
      "epoch": 0.24902818270165208,
      "grad_norm": 0.9375,
      "learning_rate": 0.00029231680400195067,
      "loss": 0.0587,
      "step": 1025
    },
    {
      "epoch": 0.250242954324587,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0002922813825788736,
      "loss": 0.0534,
      "step": 1030
    },
    {
      "epoch": 0.25145772594752186,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002922457863860709,
      "loss": 0.0528,
      "step": 1035
    },
    {
      "epoch": 0.25267249757045673,
      "grad_norm": 0.71875,
      "learning_rate": 0.00029221001548147417,
      "loss": 0.0476,
      "step": 1040
    },
    {
      "epoch": 0.25388726919339166,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0002921740699232993,
      "loss": 0.0463,
      "step": 1045
    },
    {
      "epoch": 0.25510204081632654,
      "grad_norm": 0.796875,
      "learning_rate": 0.0002921379497700466,
      "loss": 0.0499,
      "step": 1050
    },
    {
      "epoch": 0.2563168124392614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002921016550805002,
      "loss": 0.0552,
      "step": 1055
    },
    {
      "epoch": 0.2575315840621963,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00029206518591372863,
      "loss": 0.0452,
      "step": 1060
    },
    {
      "epoch": 0.2587463556851312,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0002920285423290841,
      "loss": 0.0529,
      "step": 1065
    },
    {
      "epoch": 0.2599611273080661,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00029199172438620295,
      "loss": 0.0509,
      "step": 1070
    },
    {
      "epoch": 0.26117589893100096,
      "grad_norm": 0.859375,
      "learning_rate": 0.000291954732145005,
      "loss": 0.0534,
      "step": 1075
    },
    {
      "epoch": 0.26239067055393583,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002919175656656939,
      "loss": 0.045,
      "step": 1080
    },
    {
      "epoch": 0.26360544217687076,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00029188022500875686,
      "loss": 0.0567,
      "step": 1085
    },
    {
      "epoch": 0.26482021379980564,
      "grad_norm": 0.796875,
      "learning_rate": 0.00029184271023496454,
      "loss": 0.053,
      "step": 1090
    },
    {
      "epoch": 0.2660349854227405,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029180502140537087,
      "loss": 0.0564,
      "step": 1095
    },
    {
      "epoch": 0.26724975704567544,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00029176715858131324,
      "loss": 0.0545,
      "step": 1100
    },
    {
      "epoch": 0.2684645286686103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002917291218244121,
      "loss": 0.0427,
      "step": 1105
    },
    {
      "epoch": 0.2696793002915452,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000291690911196571,
      "loss": 0.0471,
      "step": 1110
    },
    {
      "epoch": 0.27089407191448006,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00029165252675997634,
      "loss": 0.0505,
      "step": 1115
    },
    {
      "epoch": 0.272108843537415,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0002916139685770975,
      "loss": 0.0503,
      "step": 1120
    },
    {
      "epoch": 0.27332361516034986,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0002915752367106867,
      "loss": 0.0404,
      "step": 1125
    },
    {
      "epoch": 0.27453838678328474,
      "grad_norm": 0.9375,
      "learning_rate": 0.0002915363312237787,
      "loss": 0.047,
      "step": 1130
    },
    {
      "epoch": 0.2757531584062196,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00029149725217969086,
      "loss": 0.0436,
      "step": 1135
    },
    {
      "epoch": 0.27696793002915454,
      "grad_norm": 0.796875,
      "learning_rate": 0.00029145799964202303,
      "loss": 0.0486,
      "step": 1140
    },
    {
      "epoch": 0.2781827016520894,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0002914185736746573,
      "loss": 0.0471,
      "step": 1145
    },
    {
      "epoch": 0.2793974732750243,
      "grad_norm": 0.71875,
      "learning_rate": 0.0002913789743417581,
      "loss": 0.0449,
      "step": 1150
    },
    {
      "epoch": 0.28061224489795916,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0002913392017077721,
      "loss": 0.0425,
      "step": 1155
    },
    {
      "epoch": 0.2818270165208941,
      "grad_norm": 0.7109375,
      "learning_rate": 0.00029129925583742787,
      "loss": 0.045,
      "step": 1160
    },
    {
      "epoch": 0.28304178814382897,
      "grad_norm": 0.78125,
      "learning_rate": 0.00029125913679573586,
      "loss": 0.0427,
      "step": 1165
    },
    {
      "epoch": 0.28425655976676384,
      "grad_norm": 0.75,
      "learning_rate": 0.0002912188446479886,
      "loss": 0.0459,
      "step": 1170
    },
    {
      "epoch": 0.2854713313896987,
      "grad_norm": 0.62890625,
      "learning_rate": 0.00029117837945976,
      "loss": 0.0408,
      "step": 1175
    },
    {
      "epoch": 0.28668610301263364,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000291137741296906,
      "loss": 0.0501,
      "step": 1180
    },
    {
      "epoch": 0.2879008746355685,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0002910969302255637,
      "loss": 0.0502,
      "step": 1185
    },
    {
      "epoch": 0.2891156462585034,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0002910559463121517,
      "loss": 0.0431,
      "step": 1190
    },
    {
      "epoch": 0.29033041788143826,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0002910147896233701,
      "loss": 0.0465,
      "step": 1195
    },
    {
      "epoch": 0.2915451895043732,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0002909734602261998,
      "loss": 0.0445,
      "step": 1200
    },
    {
      "epoch": 0.29275996112730807,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002909319581879032,
      "loss": 0.0385,
      "step": 1205
    },
    {
      "epoch": 0.29397473275024294,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0002908902835760232,
      "loss": 0.0299,
      "step": 1210
    },
    {
      "epoch": 0.29518950437317787,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00029084843645838405,
      "loss": 0.0442,
      "step": 1215
    },
    {
      "epoch": 0.29640427599611274,
      "grad_norm": 0.8203125,
      "learning_rate": 0.00029080641690309045,
      "loss": 0.0457,
      "step": 1220
    },
    {
      "epoch": 0.2976190476190476,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0002907642249785277,
      "loss": 0.0496,
      "step": 1225
    },
    {
      "epoch": 0.2988338192419825,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0002907218607533619,
      "loss": 0.0401,
      "step": 1230
    },
    {
      "epoch": 0.3000485908649174,
      "grad_norm": 0.66015625,
      "learning_rate": 0.00029067932429653923,
      "loss": 0.0423,
      "step": 1235
    },
    {
      "epoch": 0.3012633624878523,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0002906366156772864,
      "loss": 0.0423,
      "step": 1240
    },
    {
      "epoch": 0.30247813411078717,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0002905937349651102,
      "loss": 0.0416,
      "step": 1245
    },
    {
      "epoch": 0.30369290573372204,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00029055068222979755,
      "loss": 0.0393,
      "step": 1250
    },
    {
      "epoch": 0.30490767735665697,
      "grad_norm": 0.90625,
      "learning_rate": 0.00029050745754141536,
      "loss": 0.0428,
      "step": 1255
    },
    {
      "epoch": 0.30612244897959184,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0002904640609703103,
      "loss": 0.0445,
      "step": 1260
    },
    {
      "epoch": 0.3073372206025267,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002904204925871088,
      "loss": 0.0456,
      "step": 1265
    },
    {
      "epoch": 0.3085519922254616,
      "grad_norm": 0.77734375,
      "learning_rate": 0.000290376752462717,
      "loss": 0.0419,
      "step": 1270
    },
    {
      "epoch": 0.3097667638483965,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0002903328406683204,
      "loss": 0.0449,
      "step": 1275
    },
    {
      "epoch": 0.3109815354713314,
      "grad_norm": 0.71875,
      "learning_rate": 0.00029028875727538393,
      "loss": 0.0408,
      "step": 1280
    },
    {
      "epoch": 0.31219630709426627,
      "grad_norm": 0.6015625,
      "learning_rate": 0.00029024450235565186,
      "loss": 0.0386,
      "step": 1285
    },
    {
      "epoch": 0.31341107871720114,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0002902000759811476,
      "loss": 0.0335,
      "step": 1290
    },
    {
      "epoch": 0.31462585034013607,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00029015547822417353,
      "loss": 0.0414,
      "step": 1295
    },
    {
      "epoch": 0.31584062196307094,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00029011070915731094,
      "loss": 0.0399,
      "step": 1300
    },
    {
      "epoch": 0.3170553935860058,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00029006576885342005,
      "loss": 0.0373,
      "step": 1305
    },
    {
      "epoch": 0.3182701652089407,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00029002065738563957,
      "loss": 0.0357,
      "step": 1310
    },
    {
      "epoch": 0.3194849368318756,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000289975374827387,
      "loss": 0.037,
      "step": 1315
    },
    {
      "epoch": 0.3206997084548105,
      "grad_norm": 0.80078125,
      "learning_rate": 0.00028992992125235817,
      "loss": 0.0347,
      "step": 1320
    },
    {
      "epoch": 0.32191448007774537,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0002898842967345271,
      "loss": 0.037,
      "step": 1325
    },
    {
      "epoch": 0.3231292517006803,
      "grad_norm": 0.65625,
      "learning_rate": 0.0002898385013481462,
      "loss": 0.0344,
      "step": 1330
    },
    {
      "epoch": 0.32434402332361517,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00028979253516774604,
      "loss": 0.0356,
      "step": 1335
    },
    {
      "epoch": 0.32555879494655005,
      "grad_norm": 0.75,
      "learning_rate": 0.0002897463982681348,
      "loss": 0.0288,
      "step": 1340
    },
    {
      "epoch": 0.3267735665694849,
      "grad_norm": 0.6796875,
      "learning_rate": 0.00028970009072439885,
      "loss": 0.0359,
      "step": 1345
    },
    {
      "epoch": 0.32798833819241985,
      "grad_norm": 0.828125,
      "learning_rate": 0.00028965361261190214,
      "loss": 0.0403,
      "step": 1350
    },
    {
      "epoch": 0.3292031098153547,
      "grad_norm": 0.7265625,
      "learning_rate": 0.0002896069640062862,
      "loss": 0.03,
      "step": 1355
    },
    {
      "epoch": 0.3304178814382896,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00028956014498346996,
      "loss": 0.0383,
      "step": 1360
    },
    {
      "epoch": 0.33163265306122447,
      "grad_norm": 0.703125,
      "learning_rate": 0.00028951315561964996,
      "loss": 0.0373,
      "step": 1365
    },
    {
      "epoch": 0.3328474246841594,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0002894659959912997,
      "loss": 0.0419,
      "step": 1370
    },
    {
      "epoch": 0.3333333333333333,
      "eval_loss": 0.03247929736971855,
      "eval_runtime": 2.4187,
      "eval_samples_per_second": 82.689,
      "eval_steps_per_second": 82.689,
      "step": 1372
    },
    {
      "epoch": 0.3340621963070943,
      "grad_norm": 0.7421875,
      "learning_rate": 0.00028941866617516993,
      "loss": 0.0322,
      "step": 1375
    },
    {
      "epoch": 0.33527696793002915,
      "grad_norm": 0.765625,
      "learning_rate": 0.0002893711662482882,
      "loss": 0.0366,
      "step": 1380
    },
    {
      "epoch": 0.336491739552964,
      "grad_norm": 0.6875,
      "learning_rate": 0.0002893234962879591,
      "loss": 0.0274,
      "step": 1385
    },
    {
      "epoch": 0.33770651117589895,
      "grad_norm": 0.63671875,
      "learning_rate": 0.00028927565637176393,
      "loss": 0.0335,
      "step": 1390
    },
    {
      "epoch": 0.3389212827988338,
      "grad_norm": 0.859375,
      "learning_rate": 0.00028922764657756045,
      "loss": 0.0383,
      "step": 1395
    },
    {
      "epoch": 0.3401360544217687,
      "grad_norm": 0.71875,
      "learning_rate": 0.00028917946698348295,
      "loss": 0.0365,
      "step": 1400
    },
    {
      "epoch": 0.34135082604470357,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0002891311176679422,
      "loss": 0.0331,
      "step": 1405
    },
    {
      "epoch": 0.3425655976676385,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000289082598709625,
      "loss": 0.0403,
      "step": 1410
    },
    {
      "epoch": 0.3437803692905734,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0002890339101874943,
      "loss": 0.0318,
      "step": 1415
    },
    {
      "epoch": 0.34499514091350825,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00028898505218078894,
      "loss": 0.0335,
      "step": 1420
    },
    {
      "epoch": 0.3462099125364432,
      "grad_norm": 0.734375,
      "learning_rate": 0.0002889360247690238,
      "loss": 0.0329,
      "step": 1425
    },
    {
      "epoch": 0.34742468415937805,
      "grad_norm": 0.7421875,
      "learning_rate": 0.00028888682803198925,
      "loss": 0.033,
      "step": 1430
    },
    {
      "epoch": 0.3486394557823129,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0002888374620497514,
      "loss": 0.0299,
      "step": 1435
    },
    {
      "epoch": 0.3498542274052478,
      "grad_norm": 0.5546875,
      "learning_rate": 0.00028878792690265157,
      "loss": 0.0296,
      "step": 1440
    },
    {
      "epoch": 0.3510689990281827,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00028873822267130666,
      "loss": 0.035,
      "step": 1445
    },
    {
      "epoch": 0.3522837706511176,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00028868834943660855,
      "loss": 0.0295,
      "step": 1450
    },
    {
      "epoch": 0.3534985422740525,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002886383072797242,
      "loss": 0.0346,
      "step": 1455
    },
    {
      "epoch": 0.35471331389698735,
      "grad_norm": 0.69921875,
      "learning_rate": 0.00028858809628209563,
      "loss": 0.0364,
      "step": 1460
    },
    {
      "epoch": 0.3559280855199223,
      "grad_norm": 0.71484375,
      "learning_rate": 0.00028853771652543946,
      "loss": 0.0333,
      "step": 1465
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 0.671875,
      "learning_rate": 0.00028848716809174704,
      "loss": 0.0299,
      "step": 1470
    },
    {
      "epoch": 0.358357628765792,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0002884364510632842,
      "loss": 0.025,
      "step": 1475
    },
    {
      "epoch": 0.3595724003887269,
      "grad_norm": 0.75390625,
      "learning_rate": 0.00028838556552259126,
      "loss": 0.0328,
      "step": 1480
    },
    {
      "epoch": 0.3607871720116618,
      "grad_norm": 0.65625,
      "learning_rate": 0.0002883345115524826,
      "loss": 0.0354,
      "step": 1485
    },
    {
      "epoch": 0.3620019436345967,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000288283289236047,
      "loss": 0.0309,
      "step": 1490
    },
    {
      "epoch": 0.3632167152575316,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0002882318986566469,
      "loss": 0.0319,
      "step": 1495
    },
    {
      "epoch": 0.36443148688046645,
      "grad_norm": 0.6953125,
      "learning_rate": 0.00028818033989791875,
      "loss": 0.0291,
      "step": 1500
    },
    {
      "epoch": 0.36443148688046645,
      "eval_loss": 0.0316455252468586,
      "eval_runtime": 2.4199,
      "eval_samples_per_second": 82.648,
      "eval_steps_per_second": 82.648,
      "step": 1500
    },
    {
      "epoch": 0.3656462585034014,
      "grad_norm": 0.59375,
      "learning_rate": 0.0002881286130437726,
      "loss": 0.021,
      "step": 1505
    },
    {
      "epoch": 0.36686103012633625,
      "grad_norm": 0.69921875,
      "learning_rate": 0.00028807671817839236,
      "loss": 0.0333,
      "step": 1510
    },
    {
      "epoch": 0.3680758017492711,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002880246553862349,
      "loss": 0.0282,
      "step": 1515
    },
    {
      "epoch": 0.369290573372206,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0002879724247520309,
      "loss": 0.0339,
      "step": 1520
    },
    {
      "epoch": 0.37050534499514093,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0002879200263607837,
      "loss": 0.0292,
      "step": 1525
    },
    {
      "epoch": 0.3717201166180758,
      "grad_norm": 0.5,
      "learning_rate": 0.00028786746029777014,
      "loss": 0.0277,
      "step": 1530
    },
    {
      "epoch": 0.3729348882410107,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0002878147266485396,
      "loss": 0.0303,
      "step": 1535
    },
    {
      "epoch": 0.3741496598639456,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00028776182549891416,
      "loss": 0.0294,
      "step": 1540
    },
    {
      "epoch": 0.3753644314868805,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0002877087569349889,
      "loss": 0.0309,
      "step": 1545
    },
    {
      "epoch": 0.37657920310981535,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00028765552104313095,
      "loss": 0.0336,
      "step": 1550
    },
    {
      "epoch": 0.3777939747327502,
      "grad_norm": 0.625,
      "learning_rate": 0.00028760211790997993,
      "loss": 0.029,
      "step": 1555
    },
    {
      "epoch": 0.37900874635568516,
      "grad_norm": 0.921875,
      "learning_rate": 0.00028754854762244765,
      "loss": 0.0274,
      "step": 1560
    },
    {
      "epoch": 0.38022351797862003,
      "grad_norm": 0.71484375,
      "learning_rate": 0.0002874948102677179,
      "loss": 0.0264,
      "step": 1565
    },
    {
      "epoch": 0.3814382896015549,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00028744090593324647,
      "loss": 0.0225,
      "step": 1570
    },
    {
      "epoch": 0.3826530612244898,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0002873868347067608,
      "loss": 0.0271,
      "step": 1575
    },
    {
      "epoch": 0.3838678328474247,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00028733259667625995,
      "loss": 0.0314,
      "step": 1580
    },
    {
      "epoch": 0.3850826044703596,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0002872781919300145,
      "loss": 0.0316,
      "step": 1585
    },
    {
      "epoch": 0.38629737609329445,
      "grad_norm": 0.578125,
      "learning_rate": 0.0002872236205565663,
      "loss": 0.0212,
      "step": 1590
    },
    {
      "epoch": 0.3875121477162293,
      "grad_norm": 0.68359375,
      "learning_rate": 0.00028716888264472844,
      "loss": 0.03,
      "step": 1595
    },
    {
      "epoch": 0.38872691933916426,
      "grad_norm": 0.9375,
      "learning_rate": 0.000287113978283585,
      "loss": 0.0252,
      "step": 1600
    },
    {
      "epoch": 0.38994169096209913,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0002870589075624909,
      "loss": 0.0251,
      "step": 1605
    },
    {
      "epoch": 0.391156462585034,
      "grad_norm": 0.7421875,
      "learning_rate": 0.000287003670571072,
      "loss": 0.0313,
      "step": 1610
    },
    {
      "epoch": 0.3923712342079689,
      "grad_norm": 0.65625,
      "learning_rate": 0.00028694826739922454,
      "loss": 0.0243,
      "step": 1615
    },
    {
      "epoch": 0.3935860058309038,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00028689269813711536,
      "loss": 0.0312,
      "step": 1620
    },
    {
      "epoch": 0.3948007774538387,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00028683696287518145,
      "loss": 0.0295,
      "step": 1625
    },
    {
      "epoch": 0.39601554907677355,
      "grad_norm": 0.921875,
      "learning_rate": 0.00028678106170413015,
      "loss": 0.0269,
      "step": 1630
    },
    {
      "epoch": 0.39723032069970843,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00028672499471493865,
      "loss": 0.0226,
      "step": 1635
    },
    {
      "epoch": 0.39844509232264336,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002866687619988541,
      "loss": 0.025,
      "step": 1640
    },
    {
      "epoch": 0.39965986394557823,
      "grad_norm": 0.7109375,
      "learning_rate": 0.00028661236364739334,
      "loss": 0.0307,
      "step": 1645
    },
    {
      "epoch": 0.4008746355685131,
      "grad_norm": 0.65625,
      "learning_rate": 0.00028655579975234277,
      "loss": 0.0252,
      "step": 1650
    },
    {
      "epoch": 0.40208940719144803,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0002864990704057582,
      "loss": 0.03,
      "step": 1655
    },
    {
      "epoch": 0.4033041788143829,
      "grad_norm": 0.466796875,
      "learning_rate": 0.00028644217569996476,
      "loss": 0.0269,
      "step": 1660
    },
    {
      "epoch": 0.4045189504373178,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0002863851157275566,
      "loss": 0.0242,
      "step": 1665
    },
    {
      "epoch": 0.40573372206025266,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0002863278905813969,
      "loss": 0.0259,
      "step": 1670
    },
    {
      "epoch": 0.4069484936831876,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0002862705003546177,
      "loss": 0.0299,
      "step": 1675
    },
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 0.64453125,
      "learning_rate": 0.00028621294514061946,
      "loss": 0.0244,
      "step": 1680
    },
    {
      "epoch": 0.40937803692905733,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00028615522503307156,
      "loss": 0.0307,
      "step": 1685
    },
    {
      "epoch": 0.4105928085519922,
      "grad_norm": 0.640625,
      "learning_rate": 0.0002860973401259114,
      "loss": 0.0276,
      "step": 1690
    },
    {
      "epoch": 0.41180758017492713,
      "grad_norm": 0.7109375,
      "learning_rate": 0.0002860392905133447,
      "loss": 0.0225,
      "step": 1695
    },
    {
      "epoch": 0.413022351797862,
      "grad_norm": 0.72265625,
      "learning_rate": 0.00028598107628984536,
      "loss": 0.0299,
      "step": 1700
    },
    {
      "epoch": 0.4142371234207969,
      "grad_norm": 0.66796875,
      "learning_rate": 0.00028592269755015477,
      "loss": 0.0252,
      "step": 1705
    },
    {
      "epoch": 0.41545189504373176,
      "grad_norm": 0.73046875,
      "learning_rate": 0.0002858641543892827,
      "loss": 0.0294,
      "step": 1710
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 0.474609375,
      "learning_rate": 0.00028580544690250587,
      "loss": 0.0236,
      "step": 1715
    },
    {
      "epoch": 0.41788143828960156,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00028574657518536887,
      "loss": 0.0313,
      "step": 1720
    },
    {
      "epoch": 0.41909620991253643,
      "grad_norm": 0.80078125,
      "learning_rate": 0.00028568753933368336,
      "loss": 0.0307,
      "step": 1725
    },
    {
      "epoch": 0.4203109815354713,
      "grad_norm": 0.5859375,
      "learning_rate": 0.00028562833944352825,
      "loss": 0.0189,
      "step": 1730
    },
    {
      "epoch": 0.42152575315840624,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00028556897561124927,
      "loss": 0.0266,
      "step": 1735
    },
    {
      "epoch": 0.4227405247813411,
      "grad_norm": 0.58203125,
      "learning_rate": 0.00028550944793345913,
      "loss": 0.0252,
      "step": 1740
    },
    {
      "epoch": 0.423955296404276,
      "grad_norm": 0.5625,
      "learning_rate": 0.000285449756507037,
      "loss": 0.0242,
      "step": 1745
    },
    {
      "epoch": 0.42517006802721086,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0002853899014291287,
      "loss": 0.0195,
      "step": 1750
    },
    {
      "epoch": 0.4263848396501458,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0002853298827971464,
      "loss": 0.0226,
      "step": 1755
    },
    {
      "epoch": 0.42759961127308066,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0002852697007087683,
      "loss": 0.0215,
      "step": 1760
    },
    {
      "epoch": 0.42881438289601553,
      "grad_norm": 0.4140625,
      "learning_rate": 0.00028520935526193865,
      "loss": 0.0201,
      "step": 1765
    },
    {
      "epoch": 0.43002915451895046,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0002851488465548678,
      "loss": 0.0234,
      "step": 1770
    },
    {
      "epoch": 0.43124392614188534,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0002850881746860315,
      "loss": 0.0176,
      "step": 1775
    },
    {
      "epoch": 0.4324586977648202,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0002850273397541711,
      "loss": 0.025,
      "step": 1780
    },
    {
      "epoch": 0.4336734693877551,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00028496634185829357,
      "loss": 0.0306,
      "step": 1785
    },
    {
      "epoch": 0.43488824101069,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0002849051810976707,
      "loss": 0.0239,
      "step": 1790
    },
    {
      "epoch": 0.4361030126336249,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0002848438575718397,
      "loss": 0.0209,
      "step": 1795
    },
    {
      "epoch": 0.43731778425655976,
      "grad_norm": 0.6015625,
      "learning_rate": 0.00028478237138060246,
      "loss": 0.0218,
      "step": 1800
    },
    {
      "epoch": 0.43853255587949463,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0002847207226240257,
      "loss": 0.0267,
      "step": 1805
    },
    {
      "epoch": 0.43974732750242956,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0002846589114024407,
      "loss": 0.0206,
      "step": 1810
    },
    {
      "epoch": 0.44096209912536444,
      "grad_norm": 0.671875,
      "learning_rate": 0.000284596937816443,
      "loss": 0.0239,
      "step": 1815
    },
    {
      "epoch": 0.4421768707482993,
      "grad_norm": 0.68359375,
      "learning_rate": 0.00028453480196689266,
      "loss": 0.0287,
      "step": 1820
    },
    {
      "epoch": 0.4433916423712342,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0002844725039549135,
      "loss": 0.0277,
      "step": 1825
    },
    {
      "epoch": 0.4446064139941691,
      "grad_norm": 0.84375,
      "learning_rate": 0.00028441004388189364,
      "loss": 0.0246,
      "step": 1830
    },
    {
      "epoch": 0.445821185617104,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00028434742184948455,
      "loss": 0.0227,
      "step": 1835
    },
    {
      "epoch": 0.44703595724003886,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00028428463795960153,
      "loss": 0.0263,
      "step": 1840
    },
    {
      "epoch": 0.44825072886297374,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00028422169231442327,
      "loss": 0.0279,
      "step": 1845
    },
    {
      "epoch": 0.44946550048590866,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0002841585850163915,
      "loss": 0.0265,
      "step": 1850
    },
    {
      "epoch": 0.45068027210884354,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0002840953161682114,
      "loss": 0.0249,
      "step": 1855
    },
    {
      "epoch": 0.4518950437317784,
      "grad_norm": 0.75390625,
      "learning_rate": 0.00028403188587285084,
      "loss": 0.0273,
      "step": 1860
    },
    {
      "epoch": 0.45310981535471334,
      "grad_norm": 0.55078125,
      "learning_rate": 0.00028396829423354037,
      "loss": 0.0193,
      "step": 1865
    },
    {
      "epoch": 0.4543245869776482,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0002839045413537732,
      "loss": 0.0254,
      "step": 1870
    },
    {
      "epoch": 0.4555393586005831,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00028384062733730514,
      "loss": 0.0209,
      "step": 1875
    },
    {
      "epoch": 0.45675413022351796,
      "grad_norm": 0.55859375,
      "learning_rate": 0.00028377655228815403,
      "loss": 0.0226,
      "step": 1880
    },
    {
      "epoch": 0.4579689018464529,
      "grad_norm": 0.55859375,
      "learning_rate": 0.00028371231631059977,
      "loss": 0.0244,
      "step": 1885
    },
    {
      "epoch": 0.45918367346938777,
      "grad_norm": 0.578125,
      "learning_rate": 0.0002836479195091843,
      "loss": 0.0235,
      "step": 1890
    },
    {
      "epoch": 0.46039844509232264,
      "grad_norm": 0.52734375,
      "learning_rate": 0.00028358336198871115,
      "loss": 0.0209,
      "step": 1895
    },
    {
      "epoch": 0.4616132167152575,
      "grad_norm": 0.6171875,
      "learning_rate": 0.00028351864385424555,
      "loss": 0.025,
      "step": 1900
    },
    {
      "epoch": 0.46282798833819244,
      "grad_norm": 0.953125,
      "learning_rate": 0.000283453765211114,
      "loss": 0.024,
      "step": 1905
    },
    {
      "epoch": 0.4640427599611273,
      "grad_norm": 0.859375,
      "learning_rate": 0.00028338872616490444,
      "loss": 0.0234,
      "step": 1910
    },
    {
      "epoch": 0.4652575315840622,
      "grad_norm": 0.73046875,
      "learning_rate": 0.00028332352682146556,
      "loss": 0.0248,
      "step": 1915
    },
    {
      "epoch": 0.46647230320699706,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00028325816728690714,
      "loss": 0.0265,
      "step": 1920
    },
    {
      "epoch": 0.467687074829932,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0002831926476675996,
      "loss": 0.0222,
      "step": 1925
    },
    {
      "epoch": 0.46890184645286687,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00028312696807017385,
      "loss": 0.0257,
      "step": 1930
    },
    {
      "epoch": 0.47011661807580174,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0002830611286015213,
      "loss": 0.0257,
      "step": 1935
    },
    {
      "epoch": 0.4713313896987366,
      "grad_norm": 0.5,
      "learning_rate": 0.00028299512936879337,
      "loss": 0.0208,
      "step": 1940
    },
    {
      "epoch": 0.47254616132167154,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0002829289704794016,
      "loss": 0.0234,
      "step": 1945
    },
    {
      "epoch": 0.4737609329446064,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0002828626520410173,
      "loss": 0.0221,
      "step": 1950
    },
    {
      "epoch": 0.4749757045675413,
      "grad_norm": 0.71484375,
      "learning_rate": 0.0002827961741615716,
      "loss": 0.0278,
      "step": 1955
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 0.79296875,
      "learning_rate": 0.00028272953694925476,
      "loss": 0.0216,
      "step": 1960
    },
    {
      "epoch": 0.4774052478134111,
      "grad_norm": 0.765625,
      "learning_rate": 0.0002826627405125168,
      "loss": 0.0264,
      "step": 1965
    },
    {
      "epoch": 0.47862001943634597,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0002825957849600665,
      "loss": 0.02,
      "step": 1970
    },
    {
      "epoch": 0.47983479105928084,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0002825286704008719,
      "loss": 0.0234,
      "step": 1975
    },
    {
      "epoch": 0.48104956268221577,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0002824613969441596,
      "loss": 0.0221,
      "step": 1980
    },
    {
      "epoch": 0.48226433430515064,
      "grad_norm": 0.75,
      "learning_rate": 0.0002823939646994148,
      "loss": 0.0195,
      "step": 1985
    },
    {
      "epoch": 0.4834791059280855,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00028232637377638116,
      "loss": 0.0209,
      "step": 1990
    },
    {
      "epoch": 0.4846938775510204,
      "grad_norm": 0.671875,
      "learning_rate": 0.0002822586242850608,
      "loss": 0.0238,
      "step": 1995
    },
    {
      "epoch": 0.4859086491739553,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0002821907163357136,
      "loss": 0.0269,
      "step": 2000
    },
    {
      "epoch": 0.4859086491739553,
      "eval_loss": 0.022741403430700302,
      "eval_runtime": 2.4208,
      "eval_samples_per_second": 82.618,
      "eval_steps_per_second": 82.618,
      "step": 2000
    },
    {
      "epoch": 0.4871234207968902,
      "grad_norm": 0.671875,
      "learning_rate": 0.0002821226500388574,
      "loss": 0.0271,
      "step": 2005
    },
    {
      "epoch": 0.48833819241982507,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0002820544255052679,
      "loss": 0.0212,
      "step": 2010
    },
    {
      "epoch": 0.48955296404275994,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00028198604284597814,
      "loss": 0.0236,
      "step": 2015
    },
    {
      "epoch": 0.49076773566569487,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00028191750217227855,
      "loss": 0.0208,
      "step": 2020
    },
    {
      "epoch": 0.49198250728862974,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00028184880359571684,
      "loss": 0.0298,
      "step": 2025
    },
    {
      "epoch": 0.4931972789115646,
      "grad_norm": 0.71875,
      "learning_rate": 0.00028177994722809746,
      "loss": 0.0217,
      "step": 2030
    },
    {
      "epoch": 0.4944120505344995,
      "grad_norm": 0.640625,
      "learning_rate": 0.00028171093318148197,
      "loss": 0.0219,
      "step": 2035
    },
    {
      "epoch": 0.4956268221574344,
      "grad_norm": 0.734375,
      "learning_rate": 0.00028164176156818835,
      "loss": 0.0247,
      "step": 2040
    },
    {
      "epoch": 0.4968415937803693,
      "grad_norm": 0.79296875,
      "learning_rate": 0.00028157243250079094,
      "loss": 0.0282,
      "step": 2045
    },
    {
      "epoch": 0.49805636540330417,
      "grad_norm": 0.640625,
      "learning_rate": 0.0002815029460921206,
      "loss": 0.0199,
      "step": 2050
    },
    {
      "epoch": 0.49927113702623904,
      "grad_norm": 0.55859375,
      "learning_rate": 0.000281433302455264,
      "loss": 0.0215,
      "step": 2055
    },
    {
      "epoch": 0.500485908649174,
      "grad_norm": 0.6875,
      "learning_rate": 0.0002813635017035638,
      "loss": 0.0192,
      "step": 2060
    },
    {
      "epoch": 0.5017006802721088,
      "grad_norm": 0.578125,
      "learning_rate": 0.00028129354395061844,
      "loss": 0.0251,
      "step": 2065
    },
    {
      "epoch": 0.5029154518950437,
      "grad_norm": 0.453125,
      "learning_rate": 0.0002812234293102817,
      "loss": 0.0207,
      "step": 2070
    },
    {
      "epoch": 0.5041302235179786,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0002811531578966628,
      "loss": 0.0212,
      "step": 2075
    },
    {
      "epoch": 0.5053449951409135,
      "grad_norm": 0.70703125,
      "learning_rate": 0.00028108272982412604,
      "loss": 0.0252,
      "step": 2080
    },
    {
      "epoch": 0.5065597667638484,
      "grad_norm": 0.423828125,
      "learning_rate": 0.00028101214520729077,
      "loss": 0.0211,
      "step": 2085
    },
    {
      "epoch": 0.5077745383867833,
      "grad_norm": 0.46875,
      "learning_rate": 0.000280941404161031,
      "loss": 0.0222,
      "step": 2090
    },
    {
      "epoch": 0.5089893100097181,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0002808705068004755,
      "loss": 0.0236,
      "step": 2095
    },
    {
      "epoch": 0.5102040816326531,
      "grad_norm": 0.427734375,
      "learning_rate": 0.00028079945324100726,
      "loss": 0.0218,
      "step": 2100
    },
    {
      "epoch": 0.511418853255588,
      "grad_norm": 0.59375,
      "learning_rate": 0.00028072824359826347,
      "loss": 0.0191,
      "step": 2105
    },
    {
      "epoch": 0.5126336248785228,
      "grad_norm": 0.69921875,
      "learning_rate": 0.00028065687798813543,
      "loss": 0.027,
      "step": 2110
    },
    {
      "epoch": 0.5138483965014577,
      "grad_norm": 0.671875,
      "learning_rate": 0.00028058535652676826,
      "loss": 0.0151,
      "step": 2115
    },
    {
      "epoch": 0.5150631681243926,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00028051367933056075,
      "loss": 0.0233,
      "step": 2120
    },
    {
      "epoch": 0.5162779397473275,
      "grad_norm": 0.5,
      "learning_rate": 0.0002804418465161651,
      "loss": 0.0148,
      "step": 2125
    },
    {
      "epoch": 0.5174927113702624,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00028036985820048673,
      "loss": 0.0183,
      "step": 2130
    },
    {
      "epoch": 0.5187074829931972,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00028029771450068414,
      "loss": 0.0242,
      "step": 2135
    },
    {
      "epoch": 0.5199222546161322,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0002802254155341689,
      "loss": 0.0198,
      "step": 2140
    },
    {
      "epoch": 0.5211370262390671,
      "grad_norm": 0.53125,
      "learning_rate": 0.0002801529614186049,
      "loss": 0.0173,
      "step": 2145
    },
    {
      "epoch": 0.5223517978620019,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00028008035227190906,
      "loss": 0.0207,
      "step": 2150
    },
    {
      "epoch": 0.5235665694849369,
      "grad_norm": 0.4921875,
      "learning_rate": 0.00028000758821225,
      "loss": 0.0188,
      "step": 2155
    },
    {
      "epoch": 0.5247813411078717,
      "grad_norm": 0.81640625,
      "learning_rate": 0.00027993466935804893,
      "loss": 0.025,
      "step": 2160
    },
    {
      "epoch": 0.5259961127308066,
      "grad_norm": 0.58984375,
      "learning_rate": 0.00027986159582797876,
      "loss": 0.0199,
      "step": 2165
    },
    {
      "epoch": 0.5272108843537415,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0002797883677409642,
      "loss": 0.0216,
      "step": 2170
    },
    {
      "epoch": 0.5284256559766763,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0002797149852161814,
      "loss": 0.0209,
      "step": 2175
    },
    {
      "epoch": 0.5296404275996113,
      "grad_norm": 0.77734375,
      "learning_rate": 0.00027964144837305804,
      "loss": 0.0255,
      "step": 2180
    },
    {
      "epoch": 0.5308551992225462,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0002795677573312727,
      "loss": 0.017,
      "step": 2185
    },
    {
      "epoch": 0.532069970845481,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0002794939122107552,
      "loss": 0.0221,
      "step": 2190
    },
    {
      "epoch": 0.533284742468416,
      "grad_norm": 0.64453125,
      "learning_rate": 0.00027941991313168587,
      "loss": 0.0174,
      "step": 2195
    },
    {
      "epoch": 0.5344995140913509,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0002793457602144957,
      "loss": 0.0163,
      "step": 2200
    },
    {
      "epoch": 0.5357142857142857,
      "grad_norm": 0.46875,
      "learning_rate": 0.0002792714535798661,
      "loss": 0.0187,
      "step": 2205
    },
    {
      "epoch": 0.5369290573372206,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0002791969933487285,
      "loss": 0.0236,
      "step": 2210
    },
    {
      "epoch": 0.5381438289601554,
      "grad_norm": 0.828125,
      "learning_rate": 0.00027912237964226454,
      "loss": 0.0178,
      "step": 2215
    },
    {
      "epoch": 0.5393586005830904,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0002790476125819054,
      "loss": 0.0222,
      "step": 2220
    },
    {
      "epoch": 0.5405733722060253,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0002789726922893319,
      "loss": 0.0184,
      "step": 2225
    },
    {
      "epoch": 0.5417881438289601,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00027889761888647436,
      "loss": 0.0211,
      "step": 2230
    },
    {
      "epoch": 0.543002915451895,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002788223924955121,
      "loss": 0.0206,
      "step": 2235
    },
    {
      "epoch": 0.54421768707483,
      "grad_norm": 0.7421875,
      "learning_rate": 0.00027874701323887357,
      "loss": 0.0176,
      "step": 2240
    },
    {
      "epoch": 0.5454324586977648,
      "grad_norm": 0.703125,
      "learning_rate": 0.000278671481239236,
      "loss": 0.0176,
      "step": 2245
    },
    {
      "epoch": 0.5466472303206997,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00027859579661952504,
      "loss": 0.018,
      "step": 2250
    },
    {
      "epoch": 0.5478620019436345,
      "grad_norm": 0.6796875,
      "learning_rate": 0.00027851995950291497,
      "loss": 0.0188,
      "step": 2255
    },
    {
      "epoch": 0.5490767735665695,
      "grad_norm": 0.5546875,
      "learning_rate": 0.000278443970012828,
      "loss": 0.0177,
      "step": 2260
    },
    {
      "epoch": 0.5502915451895044,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0002783678282729345,
      "loss": 0.0192,
      "step": 2265
    },
    {
      "epoch": 0.5515063168124392,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0002782915344071526,
      "loss": 0.0239,
      "step": 2270
    },
    {
      "epoch": 0.5527210884353742,
      "grad_norm": 0.375,
      "learning_rate": 0.0002782150885396479,
      "loss": 0.0213,
      "step": 2275
    },
    {
      "epoch": 0.5539358600583091,
      "grad_norm": 0.515625,
      "learning_rate": 0.0002781384907948335,
      "loss": 0.0182,
      "step": 2280
    },
    {
      "epoch": 0.5551506316812439,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0002780617412973697,
      "loss": 0.0173,
      "step": 2285
    },
    {
      "epoch": 0.5563654033041788,
      "grad_norm": 0.59375,
      "learning_rate": 0.0002779848401721635,
      "loss": 0.0196,
      "step": 2290
    },
    {
      "epoch": 0.5575801749271136,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0002779077875443691,
      "loss": 0.0175,
      "step": 2295
    },
    {
      "epoch": 0.5587949465500486,
      "grad_norm": 0.6796875,
      "learning_rate": 0.00027783058353938686,
      "loss": 0.0234,
      "step": 2300
    },
    {
      "epoch": 0.5600097181729835,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0002777532282828638,
      "loss": 0.0151,
      "step": 2305
    },
    {
      "epoch": 0.5612244897959183,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0002776757219006929,
      "loss": 0.0173,
      "step": 2310
    },
    {
      "epoch": 0.5624392614188533,
      "grad_norm": 0.375,
      "learning_rate": 0.0002775980645190131,
      "loss": 0.0159,
      "step": 2315
    },
    {
      "epoch": 0.5636540330417882,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0002775202562642094,
      "loss": 0.0169,
      "step": 2320
    },
    {
      "epoch": 0.564868804664723,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0002774422972629118,
      "loss": 0.0171,
      "step": 2325
    },
    {
      "epoch": 0.5660835762876579,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0002773641876419961,
      "loss": 0.0123,
      "step": 2330
    },
    {
      "epoch": 0.5672983479105929,
      "grad_norm": 0.490234375,
      "learning_rate": 0.000277285927528583,
      "loss": 0.0117,
      "step": 2335
    },
    {
      "epoch": 0.5685131195335277,
      "grad_norm": 0.703125,
      "learning_rate": 0.0002772075170500382,
      "loss": 0.018,
      "step": 2340
    },
    {
      "epoch": 0.5697278911564626,
      "grad_norm": 0.359375,
      "learning_rate": 0.0002771289563339721,
      "loss": 0.0144,
      "step": 2345
    },
    {
      "epoch": 0.5709426627793974,
      "grad_norm": 0.48828125,
      "learning_rate": 0.00027705024550823954,
      "loss": 0.0156,
      "step": 2350
    },
    {
      "epoch": 0.5721574344023324,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00027697138470093977,
      "loss": 0.02,
      "step": 2355
    },
    {
      "epoch": 0.5733722060252673,
      "grad_norm": 0.4609375,
      "learning_rate": 0.0002768923740404161,
      "loss": 0.0165,
      "step": 2360
    },
    {
      "epoch": 0.5745869776482021,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0002768132136552556,
      "loss": 0.0261,
      "step": 2365
    },
    {
      "epoch": 0.575801749271137,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002767339036742892,
      "loss": 0.0174,
      "step": 2370
    },
    {
      "epoch": 0.577016520894072,
      "grad_norm": 0.609375,
      "learning_rate": 0.00027665444422659115,
      "loss": 0.0154,
      "step": 2375
    },
    {
      "epoch": 0.5782312925170068,
      "grad_norm": 0.70703125,
      "learning_rate": 0.00027657483544147904,
      "loss": 0.021,
      "step": 2380
    },
    {
      "epoch": 0.5794460641399417,
      "grad_norm": 0.63671875,
      "learning_rate": 0.00027649507744851345,
      "loss": 0.0182,
      "step": 2385
    },
    {
      "epoch": 0.5806608357628765,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0002764151703774977,
      "loss": 0.0146,
      "step": 2390
    },
    {
      "epoch": 0.5818756073858115,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0002763351143584781,
      "loss": 0.0179,
      "step": 2395
    },
    {
      "epoch": 0.5830903790087464,
      "grad_norm": 0.53125,
      "learning_rate": 0.00027625490952174283,
      "loss": 0.0193,
      "step": 2400
    },
    {
      "epoch": 0.5843051506316812,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0002761745559978227,
      "loss": 0.0162,
      "step": 2405
    },
    {
      "epoch": 0.5855199222546161,
      "grad_norm": 0.384765625,
      "learning_rate": 0.00027609405391749027,
      "loss": 0.0199,
      "step": 2410
    },
    {
      "epoch": 0.5867346938775511,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0002760134034117599,
      "loss": 0.0157,
      "step": 2415
    },
    {
      "epoch": 0.5879494655004859,
      "grad_norm": 0.515625,
      "learning_rate": 0.00027593260461188774,
      "loss": 0.0159,
      "step": 2420
    },
    {
      "epoch": 0.5891642371234208,
      "grad_norm": 0.828125,
      "learning_rate": 0.00027585165764937086,
      "loss": 0.0211,
      "step": 2425
    },
    {
      "epoch": 0.5903790087463557,
      "grad_norm": 0.50390625,
      "learning_rate": 0.00027577056265594787,
      "loss": 0.0147,
      "step": 2430
    },
    {
      "epoch": 0.5915937803692906,
      "grad_norm": 0.578125,
      "learning_rate": 0.000275689319763598,
      "loss": 0.0162,
      "step": 2435
    },
    {
      "epoch": 0.5928085519922255,
      "grad_norm": 0.59765625,
      "learning_rate": 0.00027560792910454133,
      "loss": 0.0216,
      "step": 2440
    },
    {
      "epoch": 0.5940233236151603,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0002755263908112384,
      "loss": 0.0137,
      "step": 2445
    },
    {
      "epoch": 0.5952380952380952,
      "grad_norm": 0.49609375,
      "learning_rate": 0.00027544470501639006,
      "loss": 0.0208,
      "step": 2450
    },
    {
      "epoch": 0.5964528668610302,
      "grad_norm": 0.490234375,
      "learning_rate": 0.00027536287185293715,
      "loss": 0.0194,
      "step": 2455
    },
    {
      "epoch": 0.597667638483965,
      "grad_norm": 0.33984375,
      "learning_rate": 0.0002752808914540603,
      "loss": 0.0174,
      "step": 2460
    },
    {
      "epoch": 0.5988824101068999,
      "grad_norm": 0.474609375,
      "learning_rate": 0.00027519876395317984,
      "loss": 0.023,
      "step": 2465
    },
    {
      "epoch": 0.6000971817298348,
      "grad_norm": 0.6171875,
      "learning_rate": 0.00027511648948395554,
      "loss": 0.0184,
      "step": 2470
    },
    {
      "epoch": 0.6013119533527697,
      "grad_norm": 0.451171875,
      "learning_rate": 0.00027503406818028635,
      "loss": 0.0166,
      "step": 2475
    },
    {
      "epoch": 0.6025267249757046,
      "grad_norm": 0.47265625,
      "learning_rate": 0.00027495150017631004,
      "loss": 0.0173,
      "step": 2480
    },
    {
      "epoch": 0.6037414965986394,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0002748687856064032,
      "loss": 0.0177,
      "step": 2485
    },
    {
      "epoch": 0.6049562682215743,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0002747859246051811,
      "loss": 0.0188,
      "step": 2490
    },
    {
      "epoch": 0.6061710398445093,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0002747029173074972,
      "loss": 0.015,
      "step": 2495
    },
    {
      "epoch": 0.6073858114674441,
      "grad_norm": 0.482421875,
      "learning_rate": 0.000274619763848443,
      "loss": 0.0182,
      "step": 2500
    },
    {
      "epoch": 0.6073858114674441,
      "eval_loss": 0.01776907965540886,
      "eval_runtime": 2.4156,
      "eval_samples_per_second": 82.796,
      "eval_steps_per_second": 82.796,
      "step": 2500
    },
    {
      "epoch": 0.608600583090379,
      "grad_norm": 0.546875,
      "learning_rate": 0.0002745364643633479,
      "loss": 0.0186,
      "step": 2505
    },
    {
      "epoch": 0.6098153547133139,
      "grad_norm": 0.5,
      "learning_rate": 0.0002744530189877791,
      "loss": 0.018,
      "step": 2510
    },
    {
      "epoch": 0.6110301263362488,
      "grad_norm": 0.4453125,
      "learning_rate": 0.00027436942785754096,
      "loss": 0.0178,
      "step": 2515
    },
    {
      "epoch": 0.6122448979591837,
      "grad_norm": 0.390625,
      "learning_rate": 0.0002742856911086753,
      "loss": 0.0201,
      "step": 2520
    },
    {
      "epoch": 0.6134596695821186,
      "grad_norm": 0.6796875,
      "learning_rate": 0.00027420180887746096,
      "loss": 0.0158,
      "step": 2525
    },
    {
      "epoch": 0.6146744412050534,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002741177813004132,
      "loss": 0.0191,
      "step": 2530
    },
    {
      "epoch": 0.6158892128279884,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0002740336085142841,
      "loss": 0.0199,
      "step": 2535
    },
    {
      "epoch": 0.6171039844509232,
      "grad_norm": 0.515625,
      "learning_rate": 0.0002739492906560621,
      "loss": 0.0152,
      "step": 2540
    },
    {
      "epoch": 0.6183187560738581,
      "grad_norm": 0.50390625,
      "learning_rate": 0.00027386482786297166,
      "loss": 0.0198,
      "step": 2545
    },
    {
      "epoch": 0.619533527696793,
      "grad_norm": 0.51953125,
      "learning_rate": 0.00027378022027247295,
      "loss": 0.0191,
      "step": 2550
    },
    {
      "epoch": 0.6207482993197279,
      "grad_norm": 0.419921875,
      "learning_rate": 0.0002736954680222621,
      "loss": 0.0132,
      "step": 2555
    },
    {
      "epoch": 0.6219630709426628,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002736105712502704,
      "loss": 0.017,
      "step": 2560
    },
    {
      "epoch": 0.6231778425655977,
      "grad_norm": 0.515625,
      "learning_rate": 0.0002735255300946646,
      "loss": 0.016,
      "step": 2565
    },
    {
      "epoch": 0.6243926141885325,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0002734403446938463,
      "loss": 0.0181,
      "step": 2570
    },
    {
      "epoch": 0.6256073858114675,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0002733550151864516,
      "loss": 0.0137,
      "step": 2575
    },
    {
      "epoch": 0.6268221574344023,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0002732695417113517,
      "loss": 0.0148,
      "step": 2580
    },
    {
      "epoch": 0.6280369290573372,
      "grad_norm": 0.72265625,
      "learning_rate": 0.0002731839244076516,
      "loss": 0.0156,
      "step": 2585
    },
    {
      "epoch": 0.6292517006802721,
      "grad_norm": 0.404296875,
      "learning_rate": 0.0002730981634146905,
      "loss": 0.0174,
      "step": 2590
    },
    {
      "epoch": 0.630466472303207,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0002730122588720417,
      "loss": 0.0199,
      "step": 2595
    },
    {
      "epoch": 0.6316812439261419,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0002729262109195118,
      "loss": 0.0153,
      "step": 2600
    },
    {
      "epoch": 0.6328960155490768,
      "grad_norm": 0.486328125,
      "learning_rate": 0.00027284001969714083,
      "loss": 0.0196,
      "step": 2605
    },
    {
      "epoch": 0.6341107871720116,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0002727536853452023,
      "loss": 0.0158,
      "step": 2610
    },
    {
      "epoch": 0.6353255587949466,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0002726672080042022,
      "loss": 0.018,
      "step": 2615
    },
    {
      "epoch": 0.6365403304178814,
      "grad_norm": 0.4296875,
      "learning_rate": 0.0002725805878148797,
      "loss": 0.0156,
      "step": 2620
    },
    {
      "epoch": 0.6377551020408163,
      "grad_norm": 0.52734375,
      "learning_rate": 0.00027249382491820607,
      "loss": 0.0175,
      "step": 2625
    },
    {
      "epoch": 0.6389698736637512,
      "grad_norm": 0.57421875,
      "learning_rate": 0.000272406919455385,
      "loss": 0.0157,
      "step": 2630
    },
    {
      "epoch": 0.6401846452866861,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0002723198715678521,
      "loss": 0.0139,
      "step": 2635
    },
    {
      "epoch": 0.641399416909621,
      "grad_norm": 0.3828125,
      "learning_rate": 0.00027223268139727505,
      "loss": 0.0131,
      "step": 2640
    },
    {
      "epoch": 0.6426141885325559,
      "grad_norm": 0.380859375,
      "learning_rate": 0.00027214534908555273,
      "loss": 0.0126,
      "step": 2645
    },
    {
      "epoch": 0.6438289601554907,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002720578747748156,
      "loss": 0.0154,
      "step": 2650
    },
    {
      "epoch": 0.6450437317784257,
      "grad_norm": 0.369140625,
      "learning_rate": 0.0002719702586074251,
      "loss": 0.0125,
      "step": 2655
    },
    {
      "epoch": 0.6462585034013606,
      "grad_norm": 0.546875,
      "learning_rate": 0.0002718825007259736,
      "loss": 0.0148,
      "step": 2660
    },
    {
      "epoch": 0.6474732750242954,
      "grad_norm": 0.470703125,
      "learning_rate": 0.00027179460127328394,
      "loss": 0.016,
      "step": 2665
    },
    {
      "epoch": 0.6486880466472303,
      "grad_norm": 0.490234375,
      "learning_rate": 0.00027170656039240964,
      "loss": 0.0166,
      "step": 2670
    },
    {
      "epoch": 0.6499028182701652,
      "grad_norm": 0.59375,
      "learning_rate": 0.00027161837822663415,
      "loss": 0.0193,
      "step": 2675
    },
    {
      "epoch": 0.6511175898931001,
      "grad_norm": 0.4453125,
      "learning_rate": 0.00027153005491947107,
      "loss": 0.0194,
      "step": 2680
    },
    {
      "epoch": 0.652332361516035,
      "grad_norm": 0.37890625,
      "learning_rate": 0.0002714415906146635,
      "loss": 0.0181,
      "step": 2685
    },
    {
      "epoch": 0.652332361516035,
      "eval_loss": 0.017361324280500412,
      "eval_runtime": 2.3804,
      "eval_samples_per_second": 84.021,
      "eval_steps_per_second": 84.021,
      "step": 2685
    },
    {
      "epoch": 0.6535471331389698,
      "grad_norm": 0.5390625,
      "learning_rate": 0.00027135298545618404,
      "loss": 0.0164,
      "step": 2690
    },
    {
      "epoch": 0.6547619047619048,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0002712642395882346,
      "loss": 0.0186,
      "step": 2695
    },
    {
      "epoch": 0.6559766763848397,
      "grad_norm": 0.5625,
      "learning_rate": 0.0002711753531552462,
      "loss": 0.0165,
      "step": 2700
    },
    {
      "epoch": 0.6571914480077745,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0002710863263018784,
      "loss": 0.0184,
      "step": 2705
    },
    {
      "epoch": 0.6584062196307094,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0002709971591730193,
      "loss": 0.0171,
      "step": 2710
    },
    {
      "epoch": 0.6596209912536443,
      "grad_norm": 0.62109375,
      "learning_rate": 0.00027090785191378553,
      "loss": 0.0125,
      "step": 2715
    },
    {
      "epoch": 0.6608357628765792,
      "grad_norm": 0.40625,
      "learning_rate": 0.0002708184046695215,
      "loss": 0.0118,
      "step": 2720
    },
    {
      "epoch": 0.6620505344995141,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0002707288175857996,
      "loss": 0.0209,
      "step": 2725
    },
    {
      "epoch": 0.6632653061224489,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0002706390908084198,
      "loss": 0.0161,
      "step": 2730
    },
    {
      "epoch": 0.6644800777453839,
      "grad_norm": 0.453125,
      "learning_rate": 0.0002705492244834093,
      "loss": 0.0141,
      "step": 2735
    },
    {
      "epoch": 0.6656948493683188,
      "grad_norm": 0.6875,
      "learning_rate": 0.0002704592187570226,
      "loss": 0.0207,
      "step": 2740
    },
    {
      "epoch": 0.6666666666666666,
      "eval_loss": 0.015805033966898918,
      "eval_runtime": 2.4212,
      "eval_samples_per_second": 82.603,
      "eval_steps_per_second": 82.603,
      "step": 2744
    }
  ],
  "logging_steps": 5,
  "max_steps": 12348,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3188241862059622e+17,
  "train_batch_size": 46,
  "trial_name": null,
  "trial_params": null
}