68d9c03b17
Model: fpadovani/nld-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
19244 lines
526 KiB
JSON
19244 lines
526 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 8.628805088596092,
|
|
"eval_steps": 500,
|
|
"global_step": 9500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 6.896420526504516,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.6746,
|
|
"mean_token_accuracy": 0.10462436228990554,
|
|
"num_tokens": 10156.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 6.9291552066802975,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.6298,
|
|
"mean_token_accuracy": 0.10596030503511429,
|
|
"num_tokens": 20669.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 6.921386194229126,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.7926,
|
|
"mean_token_accuracy": 0.1035456344485283,
|
|
"num_tokens": 30952.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 6.880351543426514,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.5598,
|
|
"mean_token_accuracy": 0.11079400256276131,
|
|
"num_tokens": 41429.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 6.825059843063355,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.6021,
|
|
"mean_token_accuracy": 0.11143513917922973,
|
|
"num_tokens": 52166.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.847425651550293,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.6887,
|
|
"mean_token_accuracy": 0.10952995195984841,
|
|
"num_tokens": 61670.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 6.82064414024353,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.6304,
|
|
"mean_token_accuracy": 0.10759195610880852,
|
|
"num_tokens": 71702.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 6.8523636817932125,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.5578,
|
|
"mean_token_accuracy": 0.1165112502872944,
|
|
"num_tokens": 82840.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 6.877813386917114,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.6065,
|
|
"mean_token_accuracy": 0.11428144350647926,
|
|
"num_tokens": 92290.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 6.919889783859253,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.5831,
|
|
"mean_token_accuracy": 0.11021584272384644,
|
|
"num_tokens": 101842.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 6.893999290466309,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.7614,
|
|
"mean_token_accuracy": 0.1027369037270546,
|
|
"num_tokens": 114030.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 6.846065711975098,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.6197,
|
|
"mean_token_accuracy": 0.11213333159685135,
|
|
"num_tokens": 124349.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 6.895073652267456,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.5608,
|
|
"mean_token_accuracy": 0.1175434947013855,
|
|
"num_tokens": 133534.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 6.788026571273804,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.5385,
|
|
"mean_token_accuracy": 0.11116221472620964,
|
|
"num_tokens": 144383.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 6.856563472747803,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.63,
|
|
"mean_token_accuracy": 0.1141400121152401,
|
|
"num_tokens": 155467.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 6.913156938552857,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.6797,
|
|
"mean_token_accuracy": 0.11325036436319351,
|
|
"num_tokens": 165903.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 6.983823299407959,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.6351,
|
|
"mean_token_accuracy": 0.11170147731900215,
|
|
"num_tokens": 175812.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 6.8215875148773195,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.522,
|
|
"mean_token_accuracy": 0.11236928179860114,
|
|
"num_tokens": 186357.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 6.770737648010254,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.5487,
|
|
"mean_token_accuracy": 0.12053044140338898,
|
|
"num_tokens": 196124.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 6.868503904342651,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.5161,
|
|
"mean_token_accuracy": 0.11221285238862037,
|
|
"num_tokens": 205655.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 6.833589649200439,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.6283,
|
|
"mean_token_accuracy": 0.11240326315164566,
|
|
"num_tokens": 215870.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 6.85828800201416,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.598,
|
|
"mean_token_accuracy": 0.11383136883378028,
|
|
"num_tokens": 225027.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 6.8051470756530765,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.6308,
|
|
"mean_token_accuracy": 0.10869170278310776,
|
|
"num_tokens": 236352.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 6.937057971954346,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.5858,
|
|
"mean_token_accuracy": 0.1101492315530777,
|
|
"num_tokens": 246713.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 6.795955419540405,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.4996,
|
|
"mean_token_accuracy": 0.11599990427494049,
|
|
"num_tokens": 256950.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 6.817977476119995,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.6398,
|
|
"mean_token_accuracy": 0.10774285346269608,
|
|
"num_tokens": 267345.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 6.953092241287232,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.4949,
|
|
"mean_token_accuracy": 0.11916223764419556,
|
|
"num_tokens": 276782.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 6.737736177444458,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.4408,
|
|
"mean_token_accuracy": 0.11658801287412643,
|
|
"num_tokens": 287024.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 6.772894716262817,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.6343,
|
|
"mean_token_accuracy": 0.11369729340076447,
|
|
"num_tokens": 296597.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 6.844267416000366,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.5191,
|
|
"mean_token_accuracy": 0.10904297158122063,
|
|
"num_tokens": 307709.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 6.828700256347656,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.5807,
|
|
"mean_token_accuracy": 0.11724420189857483,
|
|
"num_tokens": 317976.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 6.920355224609375,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.4874,
|
|
"mean_token_accuracy": 0.1191753476858139,
|
|
"num_tokens": 327319.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 6.78939299583435,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.597,
|
|
"mean_token_accuracy": 0.11329774484038353,
|
|
"num_tokens": 337337.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 6.905331754684449,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.6082,
|
|
"mean_token_accuracy": 0.10870072841644288,
|
|
"num_tokens": 347779.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.9889932632446286,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.6465,
|
|
"mean_token_accuracy": 0.1119416669011116,
|
|
"num_tokens": 358088.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 6.825058841705323,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.5695,
|
|
"mean_token_accuracy": 0.11772328317165374,
|
|
"num_tokens": 368314.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.870346403121948,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.5375,
|
|
"mean_token_accuracy": 0.12153311371803284,
|
|
"num_tokens": 378406.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 6.821107530593872,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.5483,
|
|
"mean_token_accuracy": 0.11495172157883644,
|
|
"num_tokens": 388912.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 6.928589963912964,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.5553,
|
|
"mean_token_accuracy": 0.11644353196024895,
|
|
"num_tokens": 399290.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 6.713993787765503,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.5336,
|
|
"mean_token_accuracy": 0.116058848798275,
|
|
"num_tokens": 409757.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 6.918181371688843,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.6683,
|
|
"mean_token_accuracy": 0.11516659632325173,
|
|
"num_tokens": 420182.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 6.916316556930542,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.5965,
|
|
"mean_token_accuracy": 0.11291420236229896,
|
|
"num_tokens": 429897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 6.8051811218261715,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.505,
|
|
"mean_token_accuracy": 0.11681886836886406,
|
|
"num_tokens": 441131.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 6.783562803268433,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.479,
|
|
"mean_token_accuracy": 0.11763835921883584,
|
|
"num_tokens": 451894.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.783149242401123,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.4492,
|
|
"mean_token_accuracy": 0.11824885159730911,
|
|
"num_tokens": 462613.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.753207111358643,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.5137,
|
|
"mean_token_accuracy": 0.11897409781813621,
|
|
"num_tokens": 471606.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 6.76159701347351,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.4775,
|
|
"mean_token_accuracy": 0.11979492008686066,
|
|
"num_tokens": 481559.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 6.782271385192871,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.528,
|
|
"mean_token_accuracy": 0.11235171258449554,
|
|
"num_tokens": 492766.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 6.8794763565063475,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.5819,
|
|
"mean_token_accuracy": 0.11366603001952172,
|
|
"num_tokens": 503357.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.7299378395080565,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.3954,
|
|
"mean_token_accuracy": 0.1229715809226036,
|
|
"num_tokens": 513098.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.724137163162231,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.4894,
|
|
"mean_token_accuracy": 0.11080436035990715,
|
|
"num_tokens": 522807.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 6.897167539596557,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.4995,
|
|
"mean_token_accuracy": 0.11499308794736862,
|
|
"num_tokens": 532670.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.709174537658692,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.473,
|
|
"mean_token_accuracy": 0.11593437120318413,
|
|
"num_tokens": 543647.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 6.884047985076904,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.5383,
|
|
"mean_token_accuracy": 0.11711002215743065,
|
|
"num_tokens": 553047.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 6.862821865081787,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.6081,
|
|
"mean_token_accuracy": 0.11437835842370987,
|
|
"num_tokens": 563413.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 6.815651369094849,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.5606,
|
|
"mean_token_accuracy": 0.11053355634212494,
|
|
"num_tokens": 573790.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 6.889989757537842,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.4411,
|
|
"mean_token_accuracy": 0.12015985175967217,
|
|
"num_tokens": 582958.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.679728221893311,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.4211,
|
|
"mean_token_accuracy": 0.12544431537389755,
|
|
"num_tokens": 592886.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.767288637161255,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.474,
|
|
"mean_token_accuracy": 0.11538089066743851,
|
|
"num_tokens": 603692.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 6.835441160202026,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.5155,
|
|
"mean_token_accuracy": 0.11608976647257804,
|
|
"num_tokens": 613534.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 6.816699981689453,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.4827,
|
|
"mean_token_accuracy": 0.11503889933228492,
|
|
"num_tokens": 623862.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.761893367767334,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.5609,
|
|
"mean_token_accuracy": 0.11557264029979705,
|
|
"num_tokens": 633759.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 6.797633409500122,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.4841,
|
|
"mean_token_accuracy": 0.11050511375069619,
|
|
"num_tokens": 643553.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.790513753890991,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.3953,
|
|
"mean_token_accuracy": 0.12165892273187637,
|
|
"num_tokens": 653714.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.650777912139892,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.4126,
|
|
"mean_token_accuracy": 0.11792858242988587,
|
|
"num_tokens": 662817.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.787194871902466,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.4678,
|
|
"mean_token_accuracy": 0.11634065806865693,
|
|
"num_tokens": 672381.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 6.614792156219482,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.3267,
|
|
"mean_token_accuracy": 0.12460481449961662,
|
|
"num_tokens": 682525.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 6.796973419189453,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.4709,
|
|
"mean_token_accuracy": 0.1173401191830635,
|
|
"num_tokens": 691826.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.684580755233765,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.4277,
|
|
"mean_token_accuracy": 0.12432598993182183,
|
|
"num_tokens": 701319.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.745461702346802,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.5259,
|
|
"mean_token_accuracy": 0.11115463376045227,
|
|
"num_tokens": 710888.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.793838691711426,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.4902,
|
|
"mean_token_accuracy": 0.11816843003034591,
|
|
"num_tokens": 722516.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.679027318954468,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.4294,
|
|
"mean_token_accuracy": 0.11942167058587075,
|
|
"num_tokens": 733058.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.8348832607269285,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.5692,
|
|
"mean_token_accuracy": 0.11810065880417824,
|
|
"num_tokens": 744832.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.7176000595092775,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.4604,
|
|
"mean_token_accuracy": 0.11912996396422386,
|
|
"num_tokens": 755387.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.684933090209961,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.4127,
|
|
"mean_token_accuracy": 0.12224102169275283,
|
|
"num_tokens": 764455.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.729821681976318,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.45,
|
|
"mean_token_accuracy": 0.11670342311263085,
|
|
"num_tokens": 774513.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.8765486717224125,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.5615,
|
|
"mean_token_accuracy": 0.11504393219947814,
|
|
"num_tokens": 785540.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.644732666015625,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.11783865094184875,
|
|
"num_tokens": 795233.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.689159107208252,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.473,
|
|
"mean_token_accuracy": 0.11936740279197693,
|
|
"num_tokens": 804855.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.81409478187561,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.435,
|
|
"mean_token_accuracy": 0.12733146026730538,
|
|
"num_tokens": 814818.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.715639543533325,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.4878,
|
|
"mean_token_accuracy": 0.12573158666491507,
|
|
"num_tokens": 824622.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.676964807510376,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.3521,
|
|
"mean_token_accuracy": 0.13041723296046256,
|
|
"num_tokens": 834813.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.649370241165161,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.4981,
|
|
"mean_token_accuracy": 0.11529626548290253,
|
|
"num_tokens": 844887.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.780358791351318,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.4845,
|
|
"mean_token_accuracy": 0.11879636645317078,
|
|
"num_tokens": 855430.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.5404973983764645,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.338,
|
|
"mean_token_accuracy": 0.11892746910452842,
|
|
"num_tokens": 866052.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.659523391723633,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.5016,
|
|
"mean_token_accuracy": 0.11411220878362656,
|
|
"num_tokens": 877961.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.719778823852539,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.4434,
|
|
"mean_token_accuracy": 0.1259594090282917,
|
|
"num_tokens": 888386.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.7781747341156,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.4981,
|
|
"mean_token_accuracy": 0.1193716824054718,
|
|
"num_tokens": 898272.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.670702028274536,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.4401,
|
|
"mean_token_accuracy": 0.1257551923394203,
|
|
"num_tokens": 908317.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.680757284164429,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.4318,
|
|
"mean_token_accuracy": 0.12077017426490784,
|
|
"num_tokens": 917970.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.71429853439331,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.4752,
|
|
"mean_token_accuracy": 0.11400743648409843,
|
|
"num_tokens": 928132.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.641982316970825,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.406,
|
|
"mean_token_accuracy": 0.12408015206456184,
|
|
"num_tokens": 938092.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.781756639480591,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.527,
|
|
"mean_token_accuracy": 0.1168364092707634,
|
|
"num_tokens": 948650.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.634301090240479,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.4467,
|
|
"mean_token_accuracy": 0.12366055399179458,
|
|
"num_tokens": 960195.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.751830911636352,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.4679,
|
|
"mean_token_accuracy": 0.12158388197422028,
|
|
"num_tokens": 970608.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.693758106231689,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4638,
|
|
"mean_token_accuracy": 0.11778140738606453,
|
|
"num_tokens": 981053.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.553822231292725,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.3881,
|
|
"mean_token_accuracy": 0.12884526401758195,
|
|
"num_tokens": 991560.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.685806798934936,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.417,
|
|
"mean_token_accuracy": 0.11735521331429481,
|
|
"num_tokens": 1002008.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.666706895828247,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.4369,
|
|
"mean_token_accuracy": 0.1235019363462925,
|
|
"num_tokens": 1012349.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.629448175430298,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.42,
|
|
"mean_token_accuracy": 0.12076696678996086,
|
|
"num_tokens": 1022886.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 6.595471782462542,
|
|
"eval_loss": 6.509462356567383,
|
|
"eval_mean_token_accuracy": 0.12101959670179112,
|
|
"eval_num_tokens": 1022886.0,
|
|
"eval_runtime": 2.6073,
|
|
"eval_samples_per_second": 1317.432,
|
|
"eval_steps_per_second": 164.919,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.708282375335694,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.4548,
|
|
"mean_token_accuracy": 0.11976599842309951,
|
|
"num_tokens": 1033129.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.655236673355103,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.3747,
|
|
"mean_token_accuracy": 0.1225571446120739,
|
|
"num_tokens": 1043069.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.641728639602661,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4328,
|
|
"mean_token_accuracy": 0.11797027960419655,
|
|
"num_tokens": 1054049.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.556688928604126,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.4019,
|
|
"mean_token_accuracy": 0.11712242439389228,
|
|
"num_tokens": 1065155.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.558606910705566,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.39,
|
|
"mean_token_accuracy": 0.12213995903730393,
|
|
"num_tokens": 1075510.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.728071165084839,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4222,
|
|
"mean_token_accuracy": 0.12322503104805946,
|
|
"num_tokens": 1086494.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.623588752746582,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.4519,
|
|
"mean_token_accuracy": 0.123404211550951,
|
|
"num_tokens": 1097311.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.666962194442749,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.4078,
|
|
"mean_token_accuracy": 0.12043015584349633,
|
|
"num_tokens": 1107757.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.484703302383423,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3036,
|
|
"mean_token_accuracy": 0.1263484887778759,
|
|
"num_tokens": 1118889.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.772952461242676,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5188,
|
|
"mean_token_accuracy": 0.11266010776162147,
|
|
"num_tokens": 1128951.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.567839097976685,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.3179,
|
|
"mean_token_accuracy": 0.12523618042469026,
|
|
"num_tokens": 1138583.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.6205119609832765,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.3908,
|
|
"mean_token_accuracy": 0.12211493328213692,
|
|
"num_tokens": 1149572.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.573495483398437,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.3645,
|
|
"mean_token_accuracy": 0.12388822212815284,
|
|
"num_tokens": 1160532.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.596926069259643,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.3614,
|
|
"mean_token_accuracy": 0.11972147524356842,
|
|
"num_tokens": 1170915.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.570085716247559,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.3211,
|
|
"mean_token_accuracy": 0.13134037628769873,
|
|
"num_tokens": 1180822.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.566324710845947,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.3758,
|
|
"mean_token_accuracy": 0.12053995877504349,
|
|
"num_tokens": 1190447.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.567368268966675,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.12156423777341843,
|
|
"num_tokens": 1200951.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.617525672912597,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.3104,
|
|
"mean_token_accuracy": 0.12355375960469246,
|
|
"num_tokens": 1211846.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.399952840805054,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.12395373061299324,
|
|
"num_tokens": 1222074.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.456777429580688,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.361,
|
|
"mean_token_accuracy": 0.12207802906632423,
|
|
"num_tokens": 1231741.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.6119006156921385,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.2814,
|
|
"mean_token_accuracy": 0.12338746637105942,
|
|
"num_tokens": 1241078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.580363655090332,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4734,
|
|
"mean_token_accuracy": 0.11461173743009567,
|
|
"num_tokens": 1251685.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.506072759628296,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.3256,
|
|
"mean_token_accuracy": 0.12901117354631425,
|
|
"num_tokens": 1262044.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.541091108322144,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3435,
|
|
"mean_token_accuracy": 0.122703967243433,
|
|
"num_tokens": 1273647.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.595802736282349,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.3912,
|
|
"mean_token_accuracy": 0.11969173923134804,
|
|
"num_tokens": 1284396.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.554928493499756,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3926,
|
|
"mean_token_accuracy": 0.1207507885992527,
|
|
"num_tokens": 1294339.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.501711988449097,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.2918,
|
|
"mean_token_accuracy": 0.12425325810909271,
|
|
"num_tokens": 1304572.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.545367765426636,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3387,
|
|
"mean_token_accuracy": 0.12730946987867356,
|
|
"num_tokens": 1314780.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.468374395370484,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.3613,
|
|
"mean_token_accuracy": 0.12463314607739448,
|
|
"num_tokens": 1325215.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.464850759506225,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.2897,
|
|
"mean_token_accuracy": 0.12672503292560577,
|
|
"num_tokens": 1334745.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.6450659275054935,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4059,
|
|
"mean_token_accuracy": 0.12144473567605019,
|
|
"num_tokens": 1345150.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.531317329406738,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3413,
|
|
"mean_token_accuracy": 0.12459630817174912,
|
|
"num_tokens": 1355619.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.399773836135864,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.2798,
|
|
"mean_token_accuracy": 0.1268183633685112,
|
|
"num_tokens": 1365933.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.45189356803894,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.2897,
|
|
"mean_token_accuracy": 0.12644198834896087,
|
|
"num_tokens": 1376867.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.676425504684448,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4651,
|
|
"mean_token_accuracy": 0.12012395113706589,
|
|
"num_tokens": 1386739.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.461312770843506,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3074,
|
|
"mean_token_accuracy": 0.1274910882115364,
|
|
"num_tokens": 1397138.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.583694648742676,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4056,
|
|
"mean_token_accuracy": 0.12489013597369195,
|
|
"num_tokens": 1408171.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.459269571304321,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3432,
|
|
"mean_token_accuracy": 0.11991192623972893,
|
|
"num_tokens": 1418259.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.495850515365601,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3515,
|
|
"mean_token_accuracy": 0.1289968691766262,
|
|
"num_tokens": 1428080.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.587929105758667,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3882,
|
|
"mean_token_accuracy": 0.1224957212805748,
|
|
"num_tokens": 1438502.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.369075345993042,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3218,
|
|
"mean_token_accuracy": 0.11988230571150779,
|
|
"num_tokens": 1448396.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.576228523254395,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3946,
|
|
"mean_token_accuracy": 0.11962528601288795,
|
|
"num_tokens": 1459728.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.55154824256897,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.2787,
|
|
"mean_token_accuracy": 0.1307799093425274,
|
|
"num_tokens": 1469733.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.394459199905396,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.2679,
|
|
"mean_token_accuracy": 0.12678055763244628,
|
|
"num_tokens": 1479004.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.404668664932251,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.2501,
|
|
"mean_token_accuracy": 0.128167537599802,
|
|
"num_tokens": 1489848.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.381428194046021,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.1237479291856289,
|
|
"num_tokens": 1500877.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.574023818969726,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3895,
|
|
"mean_token_accuracy": 0.12464114427566528,
|
|
"num_tokens": 1511840.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.43049988746643,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.2855,
|
|
"mean_token_accuracy": 0.12366045117378235,
|
|
"num_tokens": 1521839.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.393753480911255,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.1946,
|
|
"mean_token_accuracy": 0.13061774522066116,
|
|
"num_tokens": 1531285.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.2839765548706055,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2588,
|
|
"mean_token_accuracy": 0.12291661202907563,
|
|
"num_tokens": 1542027.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.484817457199097,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4047,
|
|
"mean_token_accuracy": 0.12562887594103814,
|
|
"num_tokens": 1552527.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.602127647399902,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4098,
|
|
"mean_token_accuracy": 0.12528286278247833,
|
|
"num_tokens": 1562605.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.374436855316162,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3009,
|
|
"mean_token_accuracy": 0.12750372365117074,
|
|
"num_tokens": 1572128.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.405194664001465,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.2391,
|
|
"mean_token_accuracy": 0.12423495128750801,
|
|
"num_tokens": 1582352.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.473455476760864,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3481,
|
|
"mean_token_accuracy": 0.12409614026546478,
|
|
"num_tokens": 1592887.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.391372776031494,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.2809,
|
|
"mean_token_accuracy": 0.12576968520879744,
|
|
"num_tokens": 1603564.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.292193841934204,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.1248,
|
|
"mean_token_accuracy": 0.1360088810324669,
|
|
"num_tokens": 1613270.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.291995048522949,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.1902,
|
|
"mean_token_accuracy": 0.13638920113444328,
|
|
"num_tokens": 1623009.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.352292394638061,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2393,
|
|
"mean_token_accuracy": 0.12952997758984566,
|
|
"num_tokens": 1633209.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.448199939727783,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2638,
|
|
"mean_token_accuracy": 0.13108772560954093,
|
|
"num_tokens": 1643708.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.398557519912719,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3491,
|
|
"mean_token_accuracy": 0.12394944950938225,
|
|
"num_tokens": 1653973.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.427014875411987,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3201,
|
|
"mean_token_accuracy": 0.1278303563594818,
|
|
"num_tokens": 1664856.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.294073295593262,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.244,
|
|
"mean_token_accuracy": 0.1253332458436489,
|
|
"num_tokens": 1675935.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.440651893615723,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2759,
|
|
"mean_token_accuracy": 0.13156883418560028,
|
|
"num_tokens": 1686597.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.291696548461914,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.229,
|
|
"mean_token_accuracy": 0.13001254424452782,
|
|
"num_tokens": 1697385.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.508813810348511,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2836,
|
|
"mean_token_accuracy": 0.12464521527290344,
|
|
"num_tokens": 1706680.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.340917015075684,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.2856,
|
|
"mean_token_accuracy": 0.1344773754477501,
|
|
"num_tokens": 1716130.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.336133861541748,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2062,
|
|
"mean_token_accuracy": 0.12281531766057015,
|
|
"num_tokens": 1726835.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.3471999168396,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2473,
|
|
"mean_token_accuracy": 0.13400087505578995,
|
|
"num_tokens": 1737708.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.379322052001953,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.2514,
|
|
"mean_token_accuracy": 0.1327501006424427,
|
|
"num_tokens": 1747649.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.334333848953247,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2703,
|
|
"mean_token_accuracy": 0.12708405926823615,
|
|
"num_tokens": 1758327.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.355171585083008,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2188,
|
|
"mean_token_accuracy": 0.12282907068729401,
|
|
"num_tokens": 1769767.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.366466236114502,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3194,
|
|
"mean_token_accuracy": 0.12329577431082725,
|
|
"num_tokens": 1780139.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.390305662155152,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.196,
|
|
"mean_token_accuracy": 0.13310526832938194,
|
|
"num_tokens": 1790783.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.256012010574341,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2141,
|
|
"mean_token_accuracy": 0.1262803591787815,
|
|
"num_tokens": 1801349.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.263790082931519,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1997,
|
|
"mean_token_accuracy": 0.1285887822508812,
|
|
"num_tokens": 1811990.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.283868408203125,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2189,
|
|
"mean_token_accuracy": 0.13228811025619508,
|
|
"num_tokens": 1821725.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.3054393291473385,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.1758,
|
|
"mean_token_accuracy": 0.12552776783704758,
|
|
"num_tokens": 1831794.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.354279804229736,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2583,
|
|
"mean_token_accuracy": 0.1279245376586914,
|
|
"num_tokens": 1842002.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.322077322006225,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2406,
|
|
"mean_token_accuracy": 0.13111493960022927,
|
|
"num_tokens": 1852282.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.277355575561524,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2576,
|
|
"mean_token_accuracy": 0.11843425631523133,
|
|
"num_tokens": 1862830.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.309708213806152,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2305,
|
|
"mean_token_accuracy": 0.12218644246459007,
|
|
"num_tokens": 1872746.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.325345468521118,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2331,
|
|
"mean_token_accuracy": 0.12431942000985145,
|
|
"num_tokens": 1884353.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.335691499710083,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2807,
|
|
"mean_token_accuracy": 0.12483451291918754,
|
|
"num_tokens": 1895719.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.287385368347168,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2187,
|
|
"mean_token_accuracy": 0.13662559390068055,
|
|
"num_tokens": 1905905.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.395937919616699,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3178,
|
|
"mean_token_accuracy": 0.12746959999203683,
|
|
"num_tokens": 1916010.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.280560445785523,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2101,
|
|
"mean_token_accuracy": 0.13113323375582694,
|
|
"num_tokens": 1926429.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.175940752029419,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1531,
|
|
"mean_token_accuracy": 0.12813871949911118,
|
|
"num_tokens": 1936322.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.38573260307312,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.32,
|
|
"mean_token_accuracy": 0.12148248627781869,
|
|
"num_tokens": 1947885.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.343698263168335,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2749,
|
|
"mean_token_accuracy": 0.12465898916125298,
|
|
"num_tokens": 1957876.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.21341495513916,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.2965,
|
|
"mean_token_accuracy": 0.12245200350880622,
|
|
"num_tokens": 1969531.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.302726888656617,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1338,
|
|
"mean_token_accuracy": 0.13125799745321273,
|
|
"num_tokens": 1980787.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.2150201320648195,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2051,
|
|
"mean_token_accuracy": 0.12881923243403434,
|
|
"num_tokens": 1991645.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.302574110031128,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2776,
|
|
"mean_token_accuracy": 0.1285564586520195,
|
|
"num_tokens": 2002544.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.373931741714477,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3267,
|
|
"mean_token_accuracy": 0.12761179283261298,
|
|
"num_tokens": 2012813.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.294265222549439,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.1811,
|
|
"mean_token_accuracy": 0.13078830316662787,
|
|
"num_tokens": 2023756.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.2665367603302,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2755,
|
|
"mean_token_accuracy": 0.1283035881817341,
|
|
"num_tokens": 2034245.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.37894115447998,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3172,
|
|
"mean_token_accuracy": 0.1272338256239891,
|
|
"num_tokens": 2045635.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.2262472152709964,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2816,
|
|
"mean_token_accuracy": 0.12699347138404846,
|
|
"num_tokens": 2056501.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.301205778121949,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2814,
|
|
"mean_token_accuracy": 0.12710349559783934,
|
|
"num_tokens": 2066540.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.264865876353064,
|
|
"eval_loss": 6.313098907470703,
|
|
"eval_mean_token_accuracy": 0.12870088995889176,
|
|
"eval_num_tokens": 2066540.0,
|
|
"eval_runtime": 2.568,
|
|
"eval_samples_per_second": 1337.594,
|
|
"eval_steps_per_second": 167.443,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.239779758453369,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.1535,
|
|
"mean_token_accuracy": 0.13159483671188354,
|
|
"num_tokens": 2077008.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.217110586166382,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.0884,
|
|
"mean_token_accuracy": 0.13319630175828934,
|
|
"num_tokens": 2087314.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.217319345474243,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.2643,
|
|
"mean_token_accuracy": 0.12830559089779853,
|
|
"num_tokens": 2097928.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.274431657791138,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.1881,
|
|
"mean_token_accuracy": 0.12477216497063637,
|
|
"num_tokens": 2108662.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.187909269332886,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.223,
|
|
"mean_token_accuracy": 0.12627384215593337,
|
|
"num_tokens": 2118546.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.349261856079101,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.2414,
|
|
"mean_token_accuracy": 0.12983163222670555,
|
|
"num_tokens": 2128012.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.177357196807861,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.073,
|
|
"mean_token_accuracy": 0.13193537145853043,
|
|
"num_tokens": 2138241.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.3127295017242435,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.2552,
|
|
"mean_token_accuracy": 0.12837711572647095,
|
|
"num_tokens": 2150780.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.215236186981201,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.1408,
|
|
"mean_token_accuracy": 0.13067102953791618,
|
|
"num_tokens": 2161918.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.179879856109619,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.1727,
|
|
"mean_token_accuracy": 0.12578780353069305,
|
|
"num_tokens": 2172742.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.097270870208741,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.13572826236486435,
|
|
"num_tokens": 2182122.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.303631353378296,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.209,
|
|
"mean_token_accuracy": 0.1262480393052101,
|
|
"num_tokens": 2193703.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.157730007171631,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.1135,
|
|
"mean_token_accuracy": 0.12884216010570526,
|
|
"num_tokens": 2204041.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.187658834457397,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.1674,
|
|
"mean_token_accuracy": 0.13455067873001098,
|
|
"num_tokens": 2214276.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.153891324996948,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.1353,
|
|
"mean_token_accuracy": 0.13399909287691117,
|
|
"num_tokens": 2225390.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.200978183746338,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.0922,
|
|
"mean_token_accuracy": 0.13723313212394714,
|
|
"num_tokens": 2235708.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.213730144500732,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.121,
|
|
"mean_token_accuracy": 0.13452123403549193,
|
|
"num_tokens": 2246284.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.166877222061157,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.1369,
|
|
"mean_token_accuracy": 0.130904608219862,
|
|
"num_tokens": 2257219.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.093742799758911,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.1404,
|
|
"mean_token_accuracy": 0.13008194640278817,
|
|
"num_tokens": 2268019.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.205635499954224,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.1544,
|
|
"mean_token_accuracy": 0.13232401981949807,
|
|
"num_tokens": 2278382.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.179733435312907,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 5.9808,
|
|
"mean_token_accuracy": 0.13488776650693682,
|
|
"num_tokens": 2288515.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.1191057682037355,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 5.8044,
|
|
"mean_token_accuracy": 0.142243093252182,
|
|
"num_tokens": 2298396.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.097149085998535,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 5.92,
|
|
"mean_token_accuracy": 0.13672418817877768,
|
|
"num_tokens": 2308707.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.116047430038452,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 5.8593,
|
|
"mean_token_accuracy": 0.14258627519011496,
|
|
"num_tokens": 2318981.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.078009796142578,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 5.9585,
|
|
"mean_token_accuracy": 0.1414831206202507,
|
|
"num_tokens": 2329647.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.079913902282715,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 5.8868,
|
|
"mean_token_accuracy": 0.14288607388734817,
|
|
"num_tokens": 2339449.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.111679124832153,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 5.9847,
|
|
"mean_token_accuracy": 0.1418033130466938,
|
|
"num_tokens": 2349904.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.121826887130737,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 5.907,
|
|
"mean_token_accuracy": 0.1483687862753868,
|
|
"num_tokens": 2359833.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.131973028182983,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 5.922,
|
|
"mean_token_accuracy": 0.14147619977593423,
|
|
"num_tokens": 2369857.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.112235689163208,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.1445896491408348,
|
|
"num_tokens": 2380372.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.056684350967407,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 5.9382,
|
|
"mean_token_accuracy": 0.13677495047450067,
|
|
"num_tokens": 2391239.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.01713809967041,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 5.9239,
|
|
"mean_token_accuracy": 0.14606923013925552,
|
|
"num_tokens": 2401178.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.183258724212647,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 5.9935,
|
|
"mean_token_accuracy": 0.1424830012023449,
|
|
"num_tokens": 2412496.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.0600934505462645,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 5.9925,
|
|
"mean_token_accuracy": 0.14098257049918175,
|
|
"num_tokens": 2423281.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.028135776519775,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 5.917,
|
|
"mean_token_accuracy": 0.1496851161122322,
|
|
"num_tokens": 2434078.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.0243528366088865,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 5.8565,
|
|
"mean_token_accuracy": 0.1433352991938591,
|
|
"num_tokens": 2443808.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.003511810302735,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 5.8932,
|
|
"mean_token_accuracy": 0.14691477864980698,
|
|
"num_tokens": 2453397.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.111886692047119,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 5.9914,
|
|
"mean_token_accuracy": 0.14086213409900666,
|
|
"num_tokens": 2464295.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.094159698486328,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 6.0385,
|
|
"mean_token_accuracy": 0.1400605171918869,
|
|
"num_tokens": 2475311.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.088391304016113,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 5.9265,
|
|
"mean_token_accuracy": 0.14126485511660575,
|
|
"num_tokens": 2485968.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.012838125228882,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 5.9739,
|
|
"mean_token_accuracy": 0.14347559139132499,
|
|
"num_tokens": 2496183.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.082389783859253,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 5.9222,
|
|
"mean_token_accuracy": 0.1374544233083725,
|
|
"num_tokens": 2506242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.026495170593262,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 5.9265,
|
|
"mean_token_accuracy": 0.14492063149809836,
|
|
"num_tokens": 2516443.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.0153117179870605,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 5.8924,
|
|
"mean_token_accuracy": 0.14509287774562835,
|
|
"num_tokens": 2525795.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.053783798217774,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 5.9273,
|
|
"mean_token_accuracy": 0.1454616516828537,
|
|
"num_tokens": 2535046.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.035402059555054,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 5.9314,
|
|
"mean_token_accuracy": 0.1393577352166176,
|
|
"num_tokens": 2546489.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.883663558959961,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 5.8799,
|
|
"mean_token_accuracy": 0.14748345613479613,
|
|
"num_tokens": 2556335.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.1058430671691895,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 5.8974,
|
|
"mean_token_accuracy": 0.13527284041047097,
|
|
"num_tokens": 2567647.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.1065260887146,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 6.0927,
|
|
"mean_token_accuracy": 0.13363656476140023,
|
|
"num_tokens": 2578618.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.936534261703491,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 5.9121,
|
|
"mean_token_accuracy": 0.148319573700428,
|
|
"num_tokens": 2587797.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.004483413696289,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.13952423110604287,
|
|
"num_tokens": 2598403.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.078390455245971,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 5.9584,
|
|
"mean_token_accuracy": 0.13786801546812058,
|
|
"num_tokens": 2608296.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.071820068359375,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 5.9772,
|
|
"mean_token_accuracy": 0.14245357289910315,
|
|
"num_tokens": 2617906.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.060214328765869,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 5.8932,
|
|
"mean_token_accuracy": 0.14361216574907304,
|
|
"num_tokens": 2627699.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.006322240829467,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 5.888,
|
|
"mean_token_accuracy": 0.145387901365757,
|
|
"num_tokens": 2638384.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.947664785385132,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 5.824,
|
|
"mean_token_accuracy": 0.14749597311019896,
|
|
"num_tokens": 2648279.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.99368987083435,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 5.9026,
|
|
"mean_token_accuracy": 0.13739741817116738,
|
|
"num_tokens": 2657000.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 5.978987026214599,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 5.9289,
|
|
"mean_token_accuracy": 0.1422249659895897,
|
|
"num_tokens": 2667760.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.980521869659424,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.14500522017478942,
|
|
"num_tokens": 2678450.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.9780802726745605,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 5.9705,
|
|
"mean_token_accuracy": 0.13771126195788383,
|
|
"num_tokens": 2689695.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.118952322006225,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.1434788480401039,
|
|
"num_tokens": 2700414.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.9196710109710695,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 5.807,
|
|
"mean_token_accuracy": 0.15709998458623886,
|
|
"num_tokens": 2709985.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.966207981109619,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.14382998421788215,
|
|
"num_tokens": 2720990.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.968968629837036,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 5.8729,
|
|
"mean_token_accuracy": 0.1478932410478592,
|
|
"num_tokens": 2731285.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.925943326950073,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 5.9479,
|
|
"mean_token_accuracy": 0.13792588859796523,
|
|
"num_tokens": 2741024.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.055606651306152,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.14819782376289367,
|
|
"num_tokens": 2751447.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.936514568328858,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 5.8866,
|
|
"mean_token_accuracy": 0.14688774794340134,
|
|
"num_tokens": 2761283.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.855565118789673,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.8139,
|
|
"mean_token_accuracy": 0.15138995051383972,
|
|
"num_tokens": 2771354.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.11658091545105,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 6.0104,
|
|
"mean_token_accuracy": 0.14197376891970634,
|
|
"num_tokens": 2781063.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.987172937393188,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 5.9178,
|
|
"mean_token_accuracy": 0.14141748249530792,
|
|
"num_tokens": 2791103.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.037368106842041,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 6.0165,
|
|
"mean_token_accuracy": 0.13971293494105338,
|
|
"num_tokens": 2802382.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.9335362911224365,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 5.887,
|
|
"mean_token_accuracy": 0.1408921793103218,
|
|
"num_tokens": 2813740.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.016415691375732,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 5.9058,
|
|
"mean_token_accuracy": 0.14143060073256492,
|
|
"num_tokens": 2824121.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.976479005813599,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 5.9392,
|
|
"mean_token_accuracy": 0.1393844984471798,
|
|
"num_tokens": 2833356.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.986210346221924,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 5.8845,
|
|
"mean_token_accuracy": 0.14194694459438323,
|
|
"num_tokens": 2843839.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.947541618347168,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 5.8626,
|
|
"mean_token_accuracy": 0.14251384139060974,
|
|
"num_tokens": 2853523.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.972327136993409,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 5.8994,
|
|
"mean_token_accuracy": 0.14213169515132903,
|
|
"num_tokens": 2863226.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.0358611106872555,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.14667337387800217,
|
|
"num_tokens": 2873822.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.929260873794556,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 5.8269,
|
|
"mean_token_accuracy": 0.14827583581209183,
|
|
"num_tokens": 2883645.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.964941835403442,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 5.9531,
|
|
"mean_token_accuracy": 0.13821501806378364,
|
|
"num_tokens": 2894397.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.050010871887207,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 5.9172,
|
|
"mean_token_accuracy": 0.14499335139989852,
|
|
"num_tokens": 2904947.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.8528584957122805,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.14976181611418724,
|
|
"num_tokens": 2916183.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.945002794265747,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.772,
|
|
"mean_token_accuracy": 0.1439489722251892,
|
|
"num_tokens": 2925429.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.932587099075318,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 5.8522,
|
|
"mean_token_accuracy": 0.14796290099620818,
|
|
"num_tokens": 2935142.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.958586883544922,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 5.9441,
|
|
"mean_token_accuracy": 0.14057652130723,
|
|
"num_tokens": 2945151.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.061767864227295,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 6.0356,
|
|
"mean_token_accuracy": 0.138636327534914,
|
|
"num_tokens": 2956015.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.978745079040527,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 5.8497,
|
|
"mean_token_accuracy": 0.14527930766344072,
|
|
"num_tokens": 2965879.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.985322904586792,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 5.9475,
|
|
"mean_token_accuracy": 0.14127622097730635,
|
|
"num_tokens": 2977021.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.905470991134644,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.8336,
|
|
"mean_token_accuracy": 0.14668259024620056,
|
|
"num_tokens": 2987169.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.028813171386719,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.1460015244781971,
|
|
"num_tokens": 2997666.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.955622577667237,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.7945,
|
|
"mean_token_accuracy": 0.14732399433851243,
|
|
"num_tokens": 3007021.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.8474937915802006,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 5.8811,
|
|
"mean_token_accuracy": 0.14818593561649324,
|
|
"num_tokens": 3016557.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.935906982421875,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 5.828,
|
|
"mean_token_accuracy": 0.14482102990150453,
|
|
"num_tokens": 3026146.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.0163074970245365,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 5.905,
|
|
"mean_token_accuracy": 0.14760341346263886,
|
|
"num_tokens": 3037234.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.8928052425384525,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.8464,
|
|
"mean_token_accuracy": 0.14804077446460723,
|
|
"num_tokens": 3047282.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.879059600830078,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.1442468747496605,
|
|
"num_tokens": 3058648.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.058223009109497,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 6.0189,
|
|
"mean_token_accuracy": 0.14330590963363649,
|
|
"num_tokens": 3068996.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.915167236328125,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.8572,
|
|
"mean_token_accuracy": 0.1507183864712715,
|
|
"num_tokens": 3078991.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.908344554901123,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.8149,
|
|
"mean_token_accuracy": 0.14535962045192719,
|
|
"num_tokens": 3090200.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.909794521331787,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 5.8749,
|
|
"mean_token_accuracy": 0.14464468657970428,
|
|
"num_tokens": 3101971.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3625624716038165,
|
|
"eval_entropy": 5.803071750596512,
|
|
"eval_loss": 6.138317584991455,
|
|
"eval_mean_token_accuracy": 0.14087435618736024,
|
|
"eval_num_tokens": 3101971.0,
|
|
"eval_runtime": 2.5872,
|
|
"eval_samples_per_second": 1327.665,
|
|
"eval_steps_per_second": 166.2,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.939899635314942,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 5.9748,
|
|
"mean_token_accuracy": 0.14151915162801743,
|
|
"num_tokens": 3113381.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.992388725280762,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 5.9265,
|
|
"mean_token_accuracy": 0.14152635335922242,
|
|
"num_tokens": 3125099.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.864381885528564,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 5.818,
|
|
"mean_token_accuracy": 0.14353349879384042,
|
|
"num_tokens": 3135007.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.950526857376099,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.8664,
|
|
"mean_token_accuracy": 0.14639711380004883,
|
|
"num_tokens": 3145031.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.970791959762574,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 5.9203,
|
|
"mean_token_accuracy": 0.14393452554941177,
|
|
"num_tokens": 3156044.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.8396875858306885,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.8029,
|
|
"mean_token_accuracy": 0.14731024354696273,
|
|
"num_tokens": 3165124.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.829409646987915,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.7517,
|
|
"mean_token_accuracy": 0.15294980108737946,
|
|
"num_tokens": 3175246.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.9492274761199955,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 5.829,
|
|
"mean_token_accuracy": 0.15482705384492873,
|
|
"num_tokens": 3184817.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.890252304077149,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 5.88,
|
|
"mean_token_accuracy": 0.13766854852437974,
|
|
"num_tokens": 3195142.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.883646726608276,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.8277,
|
|
"mean_token_accuracy": 0.14727459996938705,
|
|
"num_tokens": 3205724.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.947081089019775,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 5.967,
|
|
"mean_token_accuracy": 0.13908318430185318,
|
|
"num_tokens": 3216919.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.022718143463135,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 5.9826,
|
|
"mean_token_accuracy": 0.14413236752152442,
|
|
"num_tokens": 3228929.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.889075803756714,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 5.9151,
|
|
"mean_token_accuracy": 0.1442297466099262,
|
|
"num_tokens": 3239541.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.947975778579712,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.808,
|
|
"mean_token_accuracy": 0.15035119354724885,
|
|
"num_tokens": 3249527.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.80174708366394,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.7229,
|
|
"mean_token_accuracy": 0.14686224609613419,
|
|
"num_tokens": 3260497.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.01086459159851,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 5.9059,
|
|
"mean_token_accuracy": 0.14451203346252442,
|
|
"num_tokens": 3270865.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.827912139892578,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.8388,
|
|
"mean_token_accuracy": 0.14479616284370422,
|
|
"num_tokens": 3281531.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.886781787872314,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 5.8212,
|
|
"mean_token_accuracy": 0.14333095103502275,
|
|
"num_tokens": 3291970.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.896246290206909,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.8317,
|
|
"mean_token_accuracy": 0.15132568031549454,
|
|
"num_tokens": 3301638.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.009463214874268,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 5.9398,
|
|
"mean_token_accuracy": 0.14903505146503448,
|
|
"num_tokens": 3311780.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.8871430397033695,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.7905,
|
|
"mean_token_accuracy": 0.14206099957227708,
|
|
"num_tokens": 3322096.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.984230804443359,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 5.9603,
|
|
"mean_token_accuracy": 0.14079316705465317,
|
|
"num_tokens": 3332324.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.82469220161438,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 5.7421,
|
|
"mean_token_accuracy": 0.15400818437337876,
|
|
"num_tokens": 3342037.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.978690481185913,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.8475,
|
|
"mean_token_accuracy": 0.15072827488183976,
|
|
"num_tokens": 3353047.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.8175068378448485,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.752,
|
|
"mean_token_accuracy": 0.15160765796899794,
|
|
"num_tokens": 3363841.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.861622142791748,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.8531,
|
|
"mean_token_accuracy": 0.15095436573028564,
|
|
"num_tokens": 3374086.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.854400539398194,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.842,
|
|
"mean_token_accuracy": 0.14369781017303468,
|
|
"num_tokens": 3384303.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.885209083557129,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.1515107646584511,
|
|
"num_tokens": 3394858.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.861180448532105,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.8509,
|
|
"mean_token_accuracy": 0.14710102379322051,
|
|
"num_tokens": 3405252.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.002606916427612,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.8905,
|
|
"mean_token_accuracy": 0.1449957713484764,
|
|
"num_tokens": 3416007.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.965933513641358,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 5.9095,
|
|
"mean_token_accuracy": 0.1449468031525612,
|
|
"num_tokens": 3426932.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.788904619216919,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 5.7864,
|
|
"mean_token_accuracy": 0.150228051841259,
|
|
"num_tokens": 3437262.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.870692205429077,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.15172960758209228,
|
|
"num_tokens": 3447525.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.903681898117066,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 5.8706,
|
|
"mean_token_accuracy": 0.15269027948379515,
|
|
"num_tokens": 3458198.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.857991027832031,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.7585,
|
|
"mean_token_accuracy": 0.15112820416688919,
|
|
"num_tokens": 3468149.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.851669645309448,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.7426,
|
|
"mean_token_accuracy": 0.14366238862276076,
|
|
"num_tokens": 3478355.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.863175249099731,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.78,
|
|
"mean_token_accuracy": 0.14647621810436248,
|
|
"num_tokens": 3488726.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.775790691375732,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.1521575778722763,
|
|
"num_tokens": 3498346.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.779529762268067,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.8083,
|
|
"mean_token_accuracy": 0.15214328318834305,
|
|
"num_tokens": 3508213.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.973640441894531,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 5.9498,
|
|
"mean_token_accuracy": 0.15100772976875304,
|
|
"num_tokens": 3518978.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.889350128173828,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.8652,
|
|
"mean_token_accuracy": 0.15074713602662088,
|
|
"num_tokens": 3529328.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.835753965377807,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.8161,
|
|
"mean_token_accuracy": 0.1438861608505249,
|
|
"num_tokens": 3539347.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.890240907669067,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.826,
|
|
"mean_token_accuracy": 0.14054906442761422,
|
|
"num_tokens": 3549798.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.859528207778931,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 5.9107,
|
|
"mean_token_accuracy": 0.13808602094650269,
|
|
"num_tokens": 3560363.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.90363507270813,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.8496,
|
|
"mean_token_accuracy": 0.1475616753101349,
|
|
"num_tokens": 3570080.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.8384379863739015,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.8471,
|
|
"mean_token_accuracy": 0.1519201047718525,
|
|
"num_tokens": 3580053.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.935881853103638,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 5.9349,
|
|
"mean_token_accuracy": 0.1434665158390999,
|
|
"num_tokens": 3590119.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.908384084701538,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.14098186045885086,
|
|
"num_tokens": 3601809.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.996772575378418,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.7734,
|
|
"mean_token_accuracy": 0.15318574160337448,
|
|
"num_tokens": 3611752.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.8177117824554445,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.8364,
|
|
"mean_token_accuracy": 0.14665204137563706,
|
|
"num_tokens": 3621137.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.893672800064087,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.8036,
|
|
"mean_token_accuracy": 0.15030065551400185,
|
|
"num_tokens": 3631567.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.823561716079712,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.7718,
|
|
"mean_token_accuracy": 0.14383847191929816,
|
|
"num_tokens": 3641145.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.878145360946656,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.14076429829001427,
|
|
"num_tokens": 3651716.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.849016904830933,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 5.7568,
|
|
"mean_token_accuracy": 0.1524158664047718,
|
|
"num_tokens": 3662577.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.809049606323242,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.7578,
|
|
"mean_token_accuracy": 0.15062245875597,
|
|
"num_tokens": 3672984.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.860438108444214,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.14729242622852326,
|
|
"num_tokens": 3682634.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.846762800216675,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 5.8157,
|
|
"mean_token_accuracy": 0.146414215862751,
|
|
"num_tokens": 3693022.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.780075168609619,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.6891,
|
|
"mean_token_accuracy": 0.15560250282287597,
|
|
"num_tokens": 3702234.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.848706197738648,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.8747,
|
|
"mean_token_accuracy": 0.1431535691022873,
|
|
"num_tokens": 3712235.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.967378902435303,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.7987,
|
|
"mean_token_accuracy": 0.15522836297750472,
|
|
"num_tokens": 3722745.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.747704410552979,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.8619,
|
|
"mean_token_accuracy": 0.15082454830408096,
|
|
"num_tokens": 3732942.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.92849817276001,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.8359,
|
|
"mean_token_accuracy": 0.1492227330803871,
|
|
"num_tokens": 3743806.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.817096996307373,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.8451,
|
|
"mean_token_accuracy": 0.15074975788593292,
|
|
"num_tokens": 3755115.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.86489577293396,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.7996,
|
|
"mean_token_accuracy": 0.15833696573972703,
|
|
"num_tokens": 3765867.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.829450464248657,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.7553,
|
|
"mean_token_accuracy": 0.15331559181213378,
|
|
"num_tokens": 3777071.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.832288360595703,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.7655,
|
|
"mean_token_accuracy": 0.15164128616452216,
|
|
"num_tokens": 3786554.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.846448755264282,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.8053,
|
|
"mean_token_accuracy": 0.15107975900173187,
|
|
"num_tokens": 3797627.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.79697470664978,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.8023,
|
|
"mean_token_accuracy": 0.14799778312444686,
|
|
"num_tokens": 3808841.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.86922812461853,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.8318,
|
|
"mean_token_accuracy": 0.14217061325907707,
|
|
"num_tokens": 3818995.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.878432178497315,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.8943,
|
|
"mean_token_accuracy": 0.14105214104056357,
|
|
"num_tokens": 3828762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.864979553222656,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.7222,
|
|
"mean_token_accuracy": 0.14912379533052444,
|
|
"num_tokens": 3838752.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.801807594299317,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.7759,
|
|
"mean_token_accuracy": 0.15704613775014878,
|
|
"num_tokens": 3849664.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.7729755401611325,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.7587,
|
|
"mean_token_accuracy": 0.15540159195661546,
|
|
"num_tokens": 3859053.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.846253728866577,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.8154,
|
|
"mean_token_accuracy": 0.14808257222175597,
|
|
"num_tokens": 3869654.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.880489921569824,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.807,
|
|
"mean_token_accuracy": 0.14715632200241088,
|
|
"num_tokens": 3879678.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.855801057815552,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.7184,
|
|
"mean_token_accuracy": 0.15137905180454253,
|
|
"num_tokens": 3889950.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.780152130126953,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.7952,
|
|
"mean_token_accuracy": 0.15263395756483078,
|
|
"num_tokens": 3900095.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.913795232772827,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.8476,
|
|
"mean_token_accuracy": 0.14310912042856216,
|
|
"num_tokens": 3911034.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.836691522598267,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 5.8351,
|
|
"mean_token_accuracy": 0.14468196928501129,
|
|
"num_tokens": 3922219.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.8131739616394045,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.8028,
|
|
"mean_token_accuracy": 0.15152838975191116,
|
|
"num_tokens": 3932235.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.823177719116211,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.7199,
|
|
"mean_token_accuracy": 0.15392016172409057,
|
|
"num_tokens": 3941845.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.868143844604492,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.8383,
|
|
"mean_token_accuracy": 0.14277116879820823,
|
|
"num_tokens": 3953392.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.778489685058593,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.6782,
|
|
"mean_token_accuracy": 0.15229883342981337,
|
|
"num_tokens": 3963294.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.7286993026733395,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.745,
|
|
"mean_token_accuracy": 0.14428074955940245,
|
|
"num_tokens": 3973660.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.847760391235352,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.747,
|
|
"mean_token_accuracy": 0.15036094337701797,
|
|
"num_tokens": 3983587.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.828336906433106,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.14645186960697174,
|
|
"num_tokens": 3994346.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.811273908615112,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.7783,
|
|
"mean_token_accuracy": 0.1508607640862465,
|
|
"num_tokens": 4005234.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.762925434112549,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.7253,
|
|
"mean_token_accuracy": 0.15352076441049575,
|
|
"num_tokens": 4015542.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.760870552062988,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.7122,
|
|
"mean_token_accuracy": 0.15534436702728271,
|
|
"num_tokens": 4024874.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.849366092681885,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.8755,
|
|
"mean_token_accuracy": 0.14911848157644272,
|
|
"num_tokens": 4035880.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.881513500213623,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.759,
|
|
"mean_token_accuracy": 0.15174078643321992,
|
|
"num_tokens": 4046484.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.825903701782226,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.14973154589533805,
|
|
"num_tokens": 4057037.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.858614110946656,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.8966,
|
|
"mean_token_accuracy": 0.145718052983284,
|
|
"num_tokens": 4066169.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.840204524993896,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.14749296009540558,
|
|
"num_tokens": 4076697.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.819388723373413,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.6843,
|
|
"mean_token_accuracy": 0.1580486848950386,
|
|
"num_tokens": 4087372.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.809361839294434,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.7997,
|
|
"mean_token_accuracy": 0.1456289991736412,
|
|
"num_tokens": 4097653.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.86043119430542,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.7225,
|
|
"mean_token_accuracy": 0.15086327865719795,
|
|
"num_tokens": 4107381.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.815886402130127,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.8475,
|
|
"mean_token_accuracy": 0.14718833714723586,
|
|
"num_tokens": 4118762.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.831641626358032,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.7223,
|
|
"mean_token_accuracy": 0.15157042741775512,
|
|
"num_tokens": 4129011.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.853576707839966,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.802,
|
|
"mean_token_accuracy": 0.14652616456151007,
|
|
"num_tokens": 4138975.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.622762672291246,
|
|
"eval_loss": 5.993390083312988,
|
|
"eval_mean_token_accuracy": 0.14741607572623464,
|
|
"eval_num_tokens": 4138975.0,
|
|
"eval_runtime": 2.5897,
|
|
"eval_samples_per_second": 1326.412,
|
|
"eval_steps_per_second": 166.043,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.856018829345703,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.8489,
|
|
"mean_token_accuracy": 0.14588721394538878,
|
|
"num_tokens": 4150164.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.791663122177124,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.7281,
|
|
"mean_token_accuracy": 0.15424934923648834,
|
|
"num_tokens": 4159872.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.8209357261657715,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.8013,
|
|
"mean_token_accuracy": 0.14625373631715774,
|
|
"num_tokens": 4170665.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.854351425170899,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.6779,
|
|
"mean_token_accuracy": 0.1630919858813286,
|
|
"num_tokens": 4180476.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.827850008010865,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.8306,
|
|
"mean_token_accuracy": 0.14828771427273751,
|
|
"num_tokens": 4190740.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.7555688381195065,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.7195,
|
|
"mean_token_accuracy": 0.15400384366512299,
|
|
"num_tokens": 4201504.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.702020502090454,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.6556,
|
|
"mean_token_accuracy": 0.15983152240514756,
|
|
"num_tokens": 4211614.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.818504285812378,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.7759,
|
|
"mean_token_accuracy": 0.14869814217090607,
|
|
"num_tokens": 4221141.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.771631574630737,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.8272,
|
|
"mean_token_accuracy": 0.15053175687789916,
|
|
"num_tokens": 4231877.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.761961126327515,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.6402,
|
|
"mean_token_accuracy": 0.1547040030360222,
|
|
"num_tokens": 4242331.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.747348070144653,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.67,
|
|
"mean_token_accuracy": 0.15063972622156144,
|
|
"num_tokens": 4252406.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.7493750095367435,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.728,
|
|
"mean_token_accuracy": 0.1525947332382202,
|
|
"num_tokens": 4263036.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.7897923469543455,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.7157,
|
|
"mean_token_accuracy": 0.15506067872047424,
|
|
"num_tokens": 4273032.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.7889642238616945,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.784,
|
|
"mean_token_accuracy": 0.15208337604999542,
|
|
"num_tokens": 4283360.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.803781032562256,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.15857744663953782,
|
|
"num_tokens": 4292623.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.749684429168701,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.7489,
|
|
"mean_token_accuracy": 0.15692918449640275,
|
|
"num_tokens": 4302817.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.791387701034546,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.7548,
|
|
"mean_token_accuracy": 0.15455486327409745,
|
|
"num_tokens": 4313049.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.790107250213623,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.6392,
|
|
"mean_token_accuracy": 0.16129859089851378,
|
|
"num_tokens": 4322873.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.705437517166137,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.7853,
|
|
"mean_token_accuracy": 0.14610286056995392,
|
|
"num_tokens": 4333339.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.850047540664673,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.7579,
|
|
"mean_token_accuracy": 0.15404897481203078,
|
|
"num_tokens": 4344917.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.683778953552246,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.607,
|
|
"mean_token_accuracy": 0.15856548696756362,
|
|
"num_tokens": 4354140.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.776776075363159,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.8509,
|
|
"mean_token_accuracy": 0.14904744699597358,
|
|
"num_tokens": 4364976.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.805686569213867,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.6251,
|
|
"mean_token_accuracy": 0.15593284219503403,
|
|
"num_tokens": 4376423.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.781400728225708,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.7879,
|
|
"mean_token_accuracy": 0.1525404378771782,
|
|
"num_tokens": 4387865.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.727698278427124,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.7629,
|
|
"mean_token_accuracy": 0.15718339532613754,
|
|
"num_tokens": 4398275.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.813430452346802,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.676,
|
|
"mean_token_accuracy": 0.15229944735765458,
|
|
"num_tokens": 4407396.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.919112348556519,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 5.9389,
|
|
"mean_token_accuracy": 0.14159240424633027,
|
|
"num_tokens": 4419682.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.802810907363892,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.8104,
|
|
"mean_token_accuracy": 0.1542133241891861,
|
|
"num_tokens": 4430590.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.841368341445923,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.7139,
|
|
"mean_token_accuracy": 0.15902879238128662,
|
|
"num_tokens": 4441979.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.734964179992676,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.7371,
|
|
"mean_token_accuracy": 0.1503818467259407,
|
|
"num_tokens": 4451975.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.70875997543335,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.6886,
|
|
"mean_token_accuracy": 0.15106418430805207,
|
|
"num_tokens": 4462778.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.811322546005249,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.7218,
|
|
"mean_token_accuracy": 0.16250898092985153,
|
|
"num_tokens": 4472461.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.740966129302978,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.6869,
|
|
"mean_token_accuracy": 0.1604561150074005,
|
|
"num_tokens": 4482507.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.7683569431304935,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.6942,
|
|
"mean_token_accuracy": 0.1548524633049965,
|
|
"num_tokens": 4492601.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.791854763031006,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.8206,
|
|
"mean_token_accuracy": 0.15138453990221024,
|
|
"num_tokens": 4503696.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.799863243103028,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.6999,
|
|
"mean_token_accuracy": 0.15267234295606613,
|
|
"num_tokens": 4513649.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.754292249679565,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.7185,
|
|
"mean_token_accuracy": 0.15536670610308648,
|
|
"num_tokens": 4524054.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.798763227462769,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.6983,
|
|
"mean_token_accuracy": 0.15112931579351424,
|
|
"num_tokens": 4535549.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.798023080825805,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.7442,
|
|
"mean_token_accuracy": 0.148808054625988,
|
|
"num_tokens": 4545654.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.770284223556518,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.7483,
|
|
"mean_token_accuracy": 0.15513118505477905,
|
|
"num_tokens": 4556230.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.740370697445339,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.493,
|
|
"mean_token_accuracy": 0.1594390802913242,
|
|
"num_tokens": 4564486.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.732261323928833,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.3839,
|
|
"mean_token_accuracy": 0.1617157354950905,
|
|
"num_tokens": 4575026.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.679157733917236,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.4105,
|
|
"mean_token_accuracy": 0.16303803771734238,
|
|
"num_tokens": 4585902.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.831645584106445,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.5159,
|
|
"mean_token_accuracy": 0.16119294613599777,
|
|
"num_tokens": 4597609.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.643975067138672,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.3878,
|
|
"mean_token_accuracy": 0.16487512588500977,
|
|
"num_tokens": 4607531.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.620351362228393,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.16214648336172105,
|
|
"num_tokens": 4618826.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.80631537437439,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.4185,
|
|
"mean_token_accuracy": 0.16344223469495772,
|
|
"num_tokens": 4629526.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.7110284805297855,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.4574,
|
|
"mean_token_accuracy": 0.15788894891738892,
|
|
"num_tokens": 4639930.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.710696649551392,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.3528,
|
|
"mean_token_accuracy": 0.1725384697318077,
|
|
"num_tokens": 4649661.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.591615819931031,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.3556,
|
|
"mean_token_accuracy": 0.16901922821998597,
|
|
"num_tokens": 4660178.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.752491807937622,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.4072,
|
|
"mean_token_accuracy": 0.16357505470514297,
|
|
"num_tokens": 4670417.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.630772399902344,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.3424,
|
|
"mean_token_accuracy": 0.17232108116149902,
|
|
"num_tokens": 4680221.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.6423460960388185,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.4025,
|
|
"mean_token_accuracy": 0.15915464460849763,
|
|
"num_tokens": 4691136.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.5559093952178955,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.2628,
|
|
"mean_token_accuracy": 0.17410909086465837,
|
|
"num_tokens": 4700879.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.655717039108277,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.4576,
|
|
"mean_token_accuracy": 0.15726039111614226,
|
|
"num_tokens": 4711477.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.700805521011352,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.4118,
|
|
"mean_token_accuracy": 0.16517761647701262,
|
|
"num_tokens": 4721973.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.545358180999756,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.4052,
|
|
"mean_token_accuracy": 0.16079191267490386,
|
|
"num_tokens": 4732425.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.692819213867187,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.3694,
|
|
"mean_token_accuracy": 0.1703648626804352,
|
|
"num_tokens": 4743377.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.660792446136474,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.441,
|
|
"mean_token_accuracy": 0.16637289226055146,
|
|
"num_tokens": 4753534.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.651226997375488,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.3941,
|
|
"mean_token_accuracy": 0.16543074548244477,
|
|
"num_tokens": 4763159.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.612343263626099,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.3612,
|
|
"mean_token_accuracy": 0.17187314927577974,
|
|
"num_tokens": 4773497.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.593225431442261,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.4698,
|
|
"mean_token_accuracy": 0.1683978781104088,
|
|
"num_tokens": 4784291.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.6598340511322025,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.4339,
|
|
"mean_token_accuracy": 0.1617972657084465,
|
|
"num_tokens": 4794762.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.76288595199585,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.5201,
|
|
"mean_token_accuracy": 0.16361771821975707,
|
|
"num_tokens": 4804449.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.570909309387207,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.3906,
|
|
"mean_token_accuracy": 0.16723965257406234,
|
|
"num_tokens": 4814172.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.701880884170532,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.4753,
|
|
"mean_token_accuracy": 0.1626623421907425,
|
|
"num_tokens": 4824909.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.700212097167968,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.5176,
|
|
"mean_token_accuracy": 0.1671539530158043,
|
|
"num_tokens": 4834546.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.6141955852508545,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.4281,
|
|
"mean_token_accuracy": 0.16331773102283478,
|
|
"num_tokens": 4844809.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.623186111450195,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.4252,
|
|
"mean_token_accuracy": 0.16779287457466124,
|
|
"num_tokens": 4855883.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.654774856567383,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.1657600909471512,
|
|
"num_tokens": 4865936.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.6073150634765625,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.3783,
|
|
"mean_token_accuracy": 0.1651308611035347,
|
|
"num_tokens": 4876621.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.534100818634033,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.3585,
|
|
"mean_token_accuracy": 0.15960354954004288,
|
|
"num_tokens": 4886315.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.656412553787232,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.4194,
|
|
"mean_token_accuracy": 0.1665195882320404,
|
|
"num_tokens": 4897214.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.653901052474976,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4112,
|
|
"mean_token_accuracy": 0.16093333512544633,
|
|
"num_tokens": 4907360.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.606322145462036,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.3804,
|
|
"mean_token_accuracy": 0.16915421783924103,
|
|
"num_tokens": 4917591.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.620642137527466,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.487,
|
|
"mean_token_accuracy": 0.15261010229587554,
|
|
"num_tokens": 4928075.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.664631366729736,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.3644,
|
|
"mean_token_accuracy": 0.16735952645540236,
|
|
"num_tokens": 4937898.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.647318983078003,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.1659243404865265,
|
|
"num_tokens": 4948152.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.666592788696289,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.16841500252485275,
|
|
"num_tokens": 4957973.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.553496265411377,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.3786,
|
|
"mean_token_accuracy": 0.16857295632362365,
|
|
"num_tokens": 4967375.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.599955654144287,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.3674,
|
|
"mean_token_accuracy": 0.1698802411556244,
|
|
"num_tokens": 4977286.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.557249259948731,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.4527,
|
|
"mean_token_accuracy": 0.16588815152645112,
|
|
"num_tokens": 4987930.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.679167652130127,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.4992,
|
|
"mean_token_accuracy": 0.1566596269607544,
|
|
"num_tokens": 4998707.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.561664009094239,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.3526,
|
|
"mean_token_accuracy": 0.18218102008104325,
|
|
"num_tokens": 5008540.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.518757581710815,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.4099,
|
|
"mean_token_accuracy": 0.17148143649101258,
|
|
"num_tokens": 5018199.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.724684429168701,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.4428,
|
|
"mean_token_accuracy": 0.1620268665254116,
|
|
"num_tokens": 5028251.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.574983930587768,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.4383,
|
|
"mean_token_accuracy": 0.15690853893756868,
|
|
"num_tokens": 5038290.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.590346145629883,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.3856,
|
|
"mean_token_accuracy": 0.16743428707122804,
|
|
"num_tokens": 5048337.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.65562105178833,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.4225,
|
|
"mean_token_accuracy": 0.1642186239361763,
|
|
"num_tokens": 5058533.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.621915626525879,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.444,
|
|
"mean_token_accuracy": 0.16793086230754853,
|
|
"num_tokens": 5068556.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.591219758987426,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.5367,
|
|
"mean_token_accuracy": 0.1617761433124542,
|
|
"num_tokens": 5079317.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.730090093612671,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.16330357789993286,
|
|
"num_tokens": 5089544.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.598151540756225,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.4701,
|
|
"mean_token_accuracy": 0.1600750893354416,
|
|
"num_tokens": 5099491.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.69893159866333,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.479,
|
|
"mean_token_accuracy": 0.16477042436599731,
|
|
"num_tokens": 5109523.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.626549148559571,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.4391,
|
|
"mean_token_accuracy": 0.16813508719205855,
|
|
"num_tokens": 5119940.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.6414477825164795,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.3984,
|
|
"mean_token_accuracy": 0.16937321275472642,
|
|
"num_tokens": 5129789.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.600001049041748,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.412,
|
|
"mean_token_accuracy": 0.16630619764328003,
|
|
"num_tokens": 5139570.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.57002968788147,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.16205061227083206,
|
|
"num_tokens": 5150526.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.611339998245239,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.3439,
|
|
"mean_token_accuracy": 0.16476985663175583,
|
|
"num_tokens": 5160895.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.564104986190796,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.3411,
|
|
"mean_token_accuracy": 0.16363473534584044,
|
|
"num_tokens": 5171054.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.270786006360745,
|
|
"eval_entropy": 5.547056958841723,
|
|
"eval_loss": 5.934781074523926,
|
|
"eval_mean_token_accuracy": 0.15122652161259983,
|
|
"eval_num_tokens": 5171054.0,
|
|
"eval_runtime": 2.5706,
|
|
"eval_samples_per_second": 1336.263,
|
|
"eval_steps_per_second": 167.276,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.674898481369018,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.4971,
|
|
"mean_token_accuracy": 0.15226430892944337,
|
|
"num_tokens": 5180870.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.635344076156616,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.1611126720905304,
|
|
"num_tokens": 5190388.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.631716966629028,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.4312,
|
|
"mean_token_accuracy": 0.15773276388645172,
|
|
"num_tokens": 5200582.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.615285587310791,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.3961,
|
|
"mean_token_accuracy": 0.16504428684711456,
|
|
"num_tokens": 5211079.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.618725776672363,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.5285,
|
|
"mean_token_accuracy": 0.15380205437541009,
|
|
"num_tokens": 5221481.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.667839670181275,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.4046,
|
|
"mean_token_accuracy": 0.16740840673446655,
|
|
"num_tokens": 5231387.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.51996865272522,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.3627,
|
|
"mean_token_accuracy": 0.1713024988770485,
|
|
"num_tokens": 5240758.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.588564825057984,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.3792,
|
|
"mean_token_accuracy": 0.1652996152639389,
|
|
"num_tokens": 5250839.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.667622900009155,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.16262664049863815,
|
|
"num_tokens": 5261554.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.578097534179688,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.4548,
|
|
"mean_token_accuracy": 0.1629500225186348,
|
|
"num_tokens": 5272422.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.606480073928833,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.4047,
|
|
"mean_token_accuracy": 0.16124797239899635,
|
|
"num_tokens": 5282522.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.642434692382812,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.4315,
|
|
"mean_token_accuracy": 0.1681920886039734,
|
|
"num_tokens": 5293100.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.666210889816284,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.16152834743261338,
|
|
"num_tokens": 5303103.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.656523323059082,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.4604,
|
|
"mean_token_accuracy": 0.16738127470016478,
|
|
"num_tokens": 5313765.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.665504360198975,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.5252,
|
|
"mean_token_accuracy": 0.15994666367769242,
|
|
"num_tokens": 5324816.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.609907913208008,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.4095,
|
|
"mean_token_accuracy": 0.16535257995128633,
|
|
"num_tokens": 5334882.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.5859911918640135,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.4081,
|
|
"mean_token_accuracy": 0.16611809879541398,
|
|
"num_tokens": 5344623.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.594038772583008,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.1647620052099228,
|
|
"num_tokens": 5355786.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.54853401184082,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.4239,
|
|
"mean_token_accuracy": 0.16065974533557892,
|
|
"num_tokens": 5366139.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.7517313957214355,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.5384,
|
|
"mean_token_accuracy": 0.15450456589460373,
|
|
"num_tokens": 5376890.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.629910087585449,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.4217,
|
|
"mean_token_accuracy": 0.17052333652973176,
|
|
"num_tokens": 5386602.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.626346254348755,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.5085,
|
|
"mean_token_accuracy": 0.16329919546842575,
|
|
"num_tokens": 5397749.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.620373153686524,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.4618,
|
|
"mean_token_accuracy": 0.1557668000459671,
|
|
"num_tokens": 5407666.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.622222089767456,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.5403,
|
|
"mean_token_accuracy": 0.15818636864423752,
|
|
"num_tokens": 5418668.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.62026686668396,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.4126,
|
|
"mean_token_accuracy": 0.1663537710905075,
|
|
"num_tokens": 5429098.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.556104850769043,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.3686,
|
|
"mean_token_accuracy": 0.17242256104946135,
|
|
"num_tokens": 5438997.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.593198776245117,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.4362,
|
|
"mean_token_accuracy": 0.16179078668355942,
|
|
"num_tokens": 5449315.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.5610145092010494,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.16201046258211135,
|
|
"num_tokens": 5460523.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.677450180053711,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.4802,
|
|
"mean_token_accuracy": 0.1657104641199112,
|
|
"num_tokens": 5470012.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.68997163772583,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.4371,
|
|
"mean_token_accuracy": 0.16607872843742372,
|
|
"num_tokens": 5480364.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.6389130592346195,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.5685,
|
|
"mean_token_accuracy": 0.1556097000837326,
|
|
"num_tokens": 5492138.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.615856266021728,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.3352,
|
|
"mean_token_accuracy": 0.1789136692881584,
|
|
"num_tokens": 5501935.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.5911217212677,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.15772087574005128,
|
|
"num_tokens": 5513993.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.637281036376953,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.16663948744535445,
|
|
"num_tokens": 5524822.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.629761791229248,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.4412,
|
|
"mean_token_accuracy": 0.16348044127225875,
|
|
"num_tokens": 5535583.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.557689619064331,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.4244,
|
|
"mean_token_accuracy": 0.1707684352993965,
|
|
"num_tokens": 5546199.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.608806896209717,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.3838,
|
|
"mean_token_accuracy": 0.1719855099916458,
|
|
"num_tokens": 5556409.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.594619941711426,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.4463,
|
|
"mean_token_accuracy": 0.16815454065799712,
|
|
"num_tokens": 5565854.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.654301166534424,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.15685327649116515,
|
|
"num_tokens": 5576194.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.6021034717559814,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.3879,
|
|
"mean_token_accuracy": 0.17086174637079238,
|
|
"num_tokens": 5586679.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.59050407409668,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.4596,
|
|
"mean_token_accuracy": 0.1551541805267334,
|
|
"num_tokens": 5597268.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.635110235214233,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.4207,
|
|
"mean_token_accuracy": 0.15990530997514724,
|
|
"num_tokens": 5608046.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.584029912948608,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.3352,
|
|
"mean_token_accuracy": 0.1730578988790512,
|
|
"num_tokens": 5617541.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.574986505508423,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.3879,
|
|
"mean_token_accuracy": 0.16527536809444426,
|
|
"num_tokens": 5628428.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.647528553009034,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.5731,
|
|
"mean_token_accuracy": 0.15523725599050522,
|
|
"num_tokens": 5639557.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.732480430603028,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.5425,
|
|
"mean_token_accuracy": 0.16144262850284577,
|
|
"num_tokens": 5650586.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.622272396087647,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.4677,
|
|
"mean_token_accuracy": 0.16425771415233612,
|
|
"num_tokens": 5661525.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.597211313247681,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.4821,
|
|
"mean_token_accuracy": 0.16563609689474107,
|
|
"num_tokens": 5671272.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.640516662597657,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.4039,
|
|
"mean_token_accuracy": 0.16468022316694259,
|
|
"num_tokens": 5680384.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.594927740097046,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.4274,
|
|
"mean_token_accuracy": 0.16956948786973952,
|
|
"num_tokens": 5690983.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.598019456863403,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.4588,
|
|
"mean_token_accuracy": 0.15710995495319366,
|
|
"num_tokens": 5702144.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.633581495285034,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.5034,
|
|
"mean_token_accuracy": 0.1576123595237732,
|
|
"num_tokens": 5713283.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.562904357910156,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.4781,
|
|
"mean_token_accuracy": 0.1587149292230606,
|
|
"num_tokens": 5724533.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.587670230865479,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.4169,
|
|
"mean_token_accuracy": 0.1602309003472328,
|
|
"num_tokens": 5735186.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.6287859916687015,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.4901,
|
|
"mean_token_accuracy": 0.15671877190470695,
|
|
"num_tokens": 5746537.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.648980951309204,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.4553,
|
|
"mean_token_accuracy": 0.163504695892334,
|
|
"num_tokens": 5756666.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.649062299728394,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.4733,
|
|
"mean_token_accuracy": 0.16278468668460847,
|
|
"num_tokens": 5766870.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.580277347564698,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.4677,
|
|
"mean_token_accuracy": 0.16587430834770203,
|
|
"num_tokens": 5776775.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.691947507858276,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.4753,
|
|
"mean_token_accuracy": 0.16362739652395247,
|
|
"num_tokens": 5786993.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.614350414276123,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.4543,
|
|
"mean_token_accuracy": 0.16665585339069366,
|
|
"num_tokens": 5796230.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.6437500476837155,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.4383,
|
|
"mean_token_accuracy": 0.16608017310500145,
|
|
"num_tokens": 5807086.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.627534961700439,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.4745,
|
|
"mean_token_accuracy": 0.16178110539913176,
|
|
"num_tokens": 5817897.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.639148235321045,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.3828,
|
|
"mean_token_accuracy": 0.17533257454633713,
|
|
"num_tokens": 5827794.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.519614839553833,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.4549,
|
|
"mean_token_accuracy": 0.15981971770524978,
|
|
"num_tokens": 5837603.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.585697984695434,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.3799,
|
|
"mean_token_accuracy": 0.1676763653755188,
|
|
"num_tokens": 5847013.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.571235704421997,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.3644,
|
|
"mean_token_accuracy": 0.1653046041727066,
|
|
"num_tokens": 5856783.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.563139486312866,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.3845,
|
|
"mean_token_accuracy": 0.16995761841535567,
|
|
"num_tokens": 5866241.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.516332674026489,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.4277,
|
|
"mean_token_accuracy": 0.16753742694854737,
|
|
"num_tokens": 5876135.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.547815990447998,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.4918,
|
|
"mean_token_accuracy": 0.16449052840471268,
|
|
"num_tokens": 5886841.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.669221067428589,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.4393,
|
|
"mean_token_accuracy": 0.16600144952535628,
|
|
"num_tokens": 5896884.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.572008800506592,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.16310074776411057,
|
|
"num_tokens": 5906194.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.556898689270019,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.3838,
|
|
"mean_token_accuracy": 0.16949001103639602,
|
|
"num_tokens": 5917217.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.570665645599365,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.3143,
|
|
"mean_token_accuracy": 0.16935492008924485,
|
|
"num_tokens": 5926307.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.497607421875,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.4334,
|
|
"mean_token_accuracy": 0.16009773015975953,
|
|
"num_tokens": 5936718.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.519324159622192,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.3936,
|
|
"mean_token_accuracy": 0.1676113560795784,
|
|
"num_tokens": 5947870.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.666380119323731,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.5596,
|
|
"mean_token_accuracy": 0.16087401807308196,
|
|
"num_tokens": 5959090.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.627331924438477,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.542,
|
|
"mean_token_accuracy": 0.15740964114665984,
|
|
"num_tokens": 5970435.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.573572015762329,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.3563,
|
|
"mean_token_accuracy": 0.17464001327753068,
|
|
"num_tokens": 5979732.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.543536520004272,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.4958,
|
|
"mean_token_accuracy": 0.16749019026756287,
|
|
"num_tokens": 5989965.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.652641105651855,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.4815,
|
|
"mean_token_accuracy": 0.17087581902742385,
|
|
"num_tokens": 6000752.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.590217638015747,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.4391,
|
|
"mean_token_accuracy": 0.16385532021522523,
|
|
"num_tokens": 6011318.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.58248119354248,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.4554,
|
|
"mean_token_accuracy": 0.16662627458572388,
|
|
"num_tokens": 6022024.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.547089099884033,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.4085,
|
|
"mean_token_accuracy": 0.1666083201766014,
|
|
"num_tokens": 6032118.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.634429264068603,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.4366,
|
|
"mean_token_accuracy": 0.16996325105428695,
|
|
"num_tokens": 6042607.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.5135270118713375,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.16581401228904724,
|
|
"num_tokens": 6053625.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.57910418510437,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.3887,
|
|
"mean_token_accuracy": 0.17291778028011323,
|
|
"num_tokens": 6063823.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.653094100952148,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.3938,
|
|
"mean_token_accuracy": 0.16758090257644653,
|
|
"num_tokens": 6073914.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.530201768875122,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.3831,
|
|
"mean_token_accuracy": 0.16896505057811737,
|
|
"num_tokens": 6085121.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.509311628341675,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.2966,
|
|
"mean_token_accuracy": 0.17877447456121445,
|
|
"num_tokens": 6095324.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.505864191055298,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.4193,
|
|
"mean_token_accuracy": 0.17353674918413162,
|
|
"num_tokens": 6104704.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.607789134979248,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.53,
|
|
"mean_token_accuracy": 0.16431805193424226,
|
|
"num_tokens": 6116569.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.701474571228028,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.5074,
|
|
"mean_token_accuracy": 0.1631111428141594,
|
|
"num_tokens": 6126166.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.59194164276123,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.3666,
|
|
"mean_token_accuracy": 0.17036575376987456,
|
|
"num_tokens": 6136169.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.575629711151123,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.4518,
|
|
"mean_token_accuracy": 0.17123062908649445,
|
|
"num_tokens": 6146249.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.66314377784729,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.5226,
|
|
"mean_token_accuracy": 0.16588372439146043,
|
|
"num_tokens": 6156892.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.561169719696045,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.3961,
|
|
"mean_token_accuracy": 0.17383137196302414,
|
|
"num_tokens": 6166915.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.521164608001709,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.337,
|
|
"mean_token_accuracy": 0.17068690657615662,
|
|
"num_tokens": 6176837.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.5042627334594725,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.4366,
|
|
"mean_token_accuracy": 0.1676924481987953,
|
|
"num_tokens": 6188038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.578170919418335,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.4656,
|
|
"mean_token_accuracy": 0.16738367676734925,
|
|
"num_tokens": 6199835.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.609300231933593,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.3912,
|
|
"mean_token_accuracy": 0.1685423269867897,
|
|
"num_tokens": 6209122.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.504460890348567,
|
|
"eval_loss": 5.882698059082031,
|
|
"eval_mean_token_accuracy": 0.1543973061234452,
|
|
"eval_num_tokens": 6209122.0,
|
|
"eval_runtime": 2.5784,
|
|
"eval_samples_per_second": 1332.21,
|
|
"eval_steps_per_second": 166.769,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.571905994415284,
|
|
"epoch": 2.729668332576102,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004568624876891796,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.16442082971334457,
|
|
"num_tokens": 6219317.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.588744258880615,
|
|
"epoch": 2.7342117219445705,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00045665416380864587,
|
|
"loss": 5.4786,
|
|
"mean_token_accuracy": 0.1656958982348442,
|
|
"num_tokens": 6229561.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.5960996627807615,
|
|
"epoch": 2.7387551113130395,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004564453917145206,
|
|
"loss": 5.463,
|
|
"mean_token_accuracy": 0.16588701456785201,
|
|
"num_tokens": 6240150.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.568467617034912,
|
|
"epoch": 2.7432985006815085,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004562361719219283,
|
|
"loss": 5.3451,
|
|
"mean_token_accuracy": 0.1669515401124954,
|
|
"num_tokens": 6250641.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.576804828643799,
|
|
"epoch": 2.747841890049977,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045602650494709786,
|
|
"loss": 5.4304,
|
|
"mean_token_accuracy": 0.1632714092731476,
|
|
"num_tokens": 6261239.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.550257635116577,
|
|
"epoch": 2.752385279418446,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00045581639130736173,
|
|
"loss": 5.4443,
|
|
"mean_token_accuracy": 0.1659520834684372,
|
|
"num_tokens": 6270919.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.644656324386597,
|
|
"epoch": 2.756928668786915,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004556058315211546,
|
|
"loss": 5.5159,
|
|
"mean_token_accuracy": 0.16210834681987762,
|
|
"num_tokens": 6283007.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.477072668075562,
|
|
"epoch": 2.761472058155384,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045539482610801156,
|
|
"loss": 5.3631,
|
|
"mean_token_accuracy": 0.17472303807735443,
|
|
"num_tokens": 6293596.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.5631073951721195,
|
|
"epoch": 2.766015447523853,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004551833755885677,
|
|
"loss": 5.4185,
|
|
"mean_token_accuracy": 0.1663610577583313,
|
|
"num_tokens": 6304568.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.587749385833741,
|
|
"epoch": 2.7705588368923215,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004549714804845561,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.16811080425977706,
|
|
"num_tokens": 6315275.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.593580484390259,
|
|
"epoch": 2.7751022262607905,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004547591413188068,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.1686153680086136,
|
|
"num_tokens": 6324986.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.592192840576172,
|
|
"epoch": 2.7796456156292595,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004545463586152456,
|
|
"loss": 5.503,
|
|
"mean_token_accuracy": 0.15756331980228425,
|
|
"num_tokens": 6336318.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.517651748657227,
|
|
"epoch": 2.784189004997728,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004543331328988927,
|
|
"loss": 5.3722,
|
|
"mean_token_accuracy": 0.17039327025413514,
|
|
"num_tokens": 6345986.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.587380504608154,
|
|
"epoch": 2.788732394366197,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004541194646958614,
|
|
"loss": 5.4589,
|
|
"mean_token_accuracy": 0.16623624861240388,
|
|
"num_tokens": 6357025.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.589878416061401,
|
|
"epoch": 2.793275783734666,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045390535453335666,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.16604645699262618,
|
|
"num_tokens": 6367885.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.495131206512451,
|
|
"epoch": 2.797819173103135,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00045369080293967414,
|
|
"loss": 5.3768,
|
|
"mean_token_accuracy": 0.16661519259214402,
|
|
"num_tokens": 6377729.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.524296712875366,
|
|
"epoch": 2.802362562471604,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045347581044419846,
|
|
"loss": 5.3716,
|
|
"mean_token_accuracy": 0.1703293278813362,
|
|
"num_tokens": 6387368.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.558839941024781,
|
|
"epoch": 2.806905951840073,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045326037757740216,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.16629040241241455,
|
|
"num_tokens": 6397771.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.658164215087891,
|
|
"epoch": 2.8114493412085415,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004530445048708447,
|
|
"loss": 5.3575,
|
|
"mean_token_accuracy": 0.1759626403450966,
|
|
"num_tokens": 6408238.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.586536169052124,
|
|
"epoch": 2.8159927305770105,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00045282819285717024,
|
|
"loss": 5.4922,
|
|
"mean_token_accuracy": 0.1615571767091751,
|
|
"num_tokens": 6419196.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.544446897506714,
|
|
"epoch": 2.8205361199454795,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00045261144207010736,
|
|
"loss": 5.4552,
|
|
"mean_token_accuracy": 0.163881815969944,
|
|
"num_tokens": 6430342.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.591366910934449,
|
|
"epoch": 2.825079509313948,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00045239425304446684,
|
|
"loss": 5.3471,
|
|
"mean_token_accuracy": 0.17166430056095122,
|
|
"num_tokens": 6440218.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.515160179138183,
|
|
"epoch": 2.829622898682417,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004521766263161413,
|
|
"loss": 5.3617,
|
|
"mean_token_accuracy": 0.17486351430416108,
|
|
"num_tokens": 6450128.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.507632780075073,
|
|
"epoch": 2.834166288050886,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000451958562422103,
|
|
"loss": 5.3616,
|
|
"mean_token_accuracy": 0.17907665073871612,
|
|
"num_tokens": 6459916.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.606981611251831,
|
|
"epoch": 2.838709677419355,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045174006190040267,
|
|
"loss": 5.4001,
|
|
"mean_token_accuracy": 0.16881245970726014,
|
|
"num_tokens": 6470250.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.613751316070557,
|
|
"epoch": 2.843253066787824,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000451521125290169,
|
|
"loss": 5.4961,
|
|
"mean_token_accuracy": 0.16227143555879592,
|
|
"num_tokens": 6481623.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.592410039901734,
|
|
"epoch": 2.8477964561562925,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000451301753131606,
|
|
"loss": 5.4011,
|
|
"mean_token_accuracy": 0.16916279047727584,
|
|
"num_tokens": 6492143.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.627229928970337,
|
|
"epoch": 2.8523398455247615,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00045108194596599286,
|
|
"loss": 5.4126,
|
|
"mean_token_accuracy": 0.17499956488609314,
|
|
"num_tokens": 6501643.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.470270824432373,
|
|
"epoch": 2.8568832348932305,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004508617043356818,
|
|
"loss": 5.3257,
|
|
"mean_token_accuracy": 0.17122260332107545,
|
|
"num_tokens": 6511442.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.537878561019897,
|
|
"epoch": 2.861426624261699,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 0.0004506410287840972,
|
|
"loss": 5.2755,
|
|
"mean_token_accuracy": 0.17936174422502518,
|
|
"num_tokens": 6521303.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.5049985408782955,
|
|
"epoch": 2.865970013630168,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000450419919855734,
|
|
"loss": 5.4076,
|
|
"mean_token_accuracy": 0.16595449298620224,
|
|
"num_tokens": 6531143.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.557400512695312,
|
|
"epoch": 2.870513402998637,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004501983780961566,
|
|
"loss": 5.4795,
|
|
"mean_token_accuracy": 0.1663627192378044,
|
|
"num_tokens": 6540899.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.693310499191284,
|
|
"epoch": 2.875056792367106,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004499764040519972,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.16179296225309373,
|
|
"num_tokens": 6552131.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.540104341506958,
|
|
"epoch": 2.879600181735575,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00044975399827095467,
|
|
"loss": 5.4369,
|
|
"mean_token_accuracy": 0.16047633737325667,
|
|
"num_tokens": 6563275.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.5416570663452145,
|
|
"epoch": 2.8841435711040435,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.00044953116130179313,
|
|
"loss": 5.3089,
|
|
"mean_token_accuracy": 0.17159310579299927,
|
|
"num_tokens": 6573109.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.482046127319336,
|
|
"epoch": 2.8886869604725125,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004493078936943408,
|
|
"loss": 5.4197,
|
|
"mean_token_accuracy": 0.16622330248355865,
|
|
"num_tokens": 6584978.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.460956192016601,
|
|
"epoch": 2.8932303498409815,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004490841959994882,
|
|
"loss": 5.4178,
|
|
"mean_token_accuracy": 0.16623902171850205,
|
|
"num_tokens": 6595608.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.500478315353393,
|
|
"epoch": 2.89777373920945,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00044886006876918716,
|
|
"loss": 5.3044,
|
|
"mean_token_accuracy": 0.17136270105838775,
|
|
"num_tokens": 6605983.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.6015228748321535,
|
|
"epoch": 2.902317128577919,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004486355125564493,
|
|
"loss": 5.4033,
|
|
"mean_token_accuracy": 0.16127358376979828,
|
|
"num_tokens": 6616682.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.6346742630004885,
|
|
"epoch": 2.906860517946388,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004484105279153449,
|
|
"loss": 5.4555,
|
|
"mean_token_accuracy": 0.1707941025495529,
|
|
"num_tokens": 6628715.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.600561285018921,
|
|
"epoch": 2.911403907314857,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000448185115401001,
|
|
"loss": 5.4415,
|
|
"mean_token_accuracy": 0.16821425557136535,
|
|
"num_tokens": 6639219.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.522152423858643,
|
|
"epoch": 2.915947296683326,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004479592755696008,
|
|
"loss": 5.3459,
|
|
"mean_token_accuracy": 0.1731230065226555,
|
|
"num_tokens": 6649516.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.567919445037842,
|
|
"epoch": 2.9204906860517945,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00044773300897838137,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.17011146247386932,
|
|
"num_tokens": 6659470.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.518619537353516,
|
|
"epoch": 2.9250340754202635,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004475063161856332,
|
|
"loss": 5.4336,
|
|
"mean_token_accuracy": 0.16712405383586884,
|
|
"num_tokens": 6669637.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.499840211868286,
|
|
"epoch": 2.9295774647887325,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004472791977506983,
|
|
"loss": 5.3722,
|
|
"mean_token_accuracy": 0.16638589054346084,
|
|
"num_tokens": 6680450.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.58185043334961,
|
|
"epoch": 2.934120854157201,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00044705165423396855,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.16374873816967012,
|
|
"num_tokens": 6691526.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.5559807300567625,
|
|
"epoch": 2.93866424352567,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00044682368619688523,
|
|
"loss": 5.4044,
|
|
"mean_token_accuracy": 0.17189990878105163,
|
|
"num_tokens": 6702543.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.561328268051147,
|
|
"epoch": 2.943207632894139,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004465952942019367,
|
|
"loss": 5.4958,
|
|
"mean_token_accuracy": 0.1628070890903473,
|
|
"num_tokens": 6712374.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.579527282714844,
|
|
"epoch": 2.947751022262608,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00044636647881265745,
|
|
"loss": 5.3756,
|
|
"mean_token_accuracy": 0.1661432847380638,
|
|
"num_tokens": 6722091.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.539435005187988,
|
|
"epoch": 2.952294411631077,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004461372405936269,
|
|
"loss": 5.3458,
|
|
"mean_token_accuracy": 0.17658119201660155,
|
|
"num_tokens": 6732024.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.5762941360473635,
|
|
"epoch": 2.9568378009995455,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004459075801104673,
|
|
"loss": 5.4798,
|
|
"mean_token_accuracy": 0.1615043580532074,
|
|
"num_tokens": 6741480.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.559581327438354,
|
|
"epoch": 2.9613811903680145,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00044567749792984325,
|
|
"loss": 5.3561,
|
|
"mean_token_accuracy": 0.17535003274679184,
|
|
"num_tokens": 6752504.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.576532649993896,
|
|
"epoch": 2.9659245797364835,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00044544699461945966,
|
|
"loss": 5.4219,
|
|
"mean_token_accuracy": 0.1645856410264969,
|
|
"num_tokens": 6762945.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.555626010894775,
|
|
"epoch": 2.970467969104952,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00044521607074806037,
|
|
"loss": 5.3793,
|
|
"mean_token_accuracy": 0.17216335386037826,
|
|
"num_tokens": 6772862.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.560421752929687,
|
|
"epoch": 2.975011358473421,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004449847268854272,
|
|
"loss": 5.3825,
|
|
"mean_token_accuracy": 0.16928240954875945,
|
|
"num_tokens": 6782912.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.562179040908814,
|
|
"epoch": 2.97955474784189,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004447529636023782,
|
|
"loss": 5.4369,
|
|
"mean_token_accuracy": 0.171390900015831,
|
|
"num_tokens": 6793249.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.544011211395263,
|
|
"epoch": 2.984098137210359,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044452078147076614,
|
|
"loss": 5.4527,
|
|
"mean_token_accuracy": 0.16391034573316574,
|
|
"num_tokens": 6803721.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.660193061828613,
|
|
"epoch": 2.988641526578828,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004442881810634774,
|
|
"loss": 5.4154,
|
|
"mean_token_accuracy": 0.17080647200345994,
|
|
"num_tokens": 6813637.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.570539569854736,
|
|
"epoch": 2.9931849159472965,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004440551629544303,
|
|
"loss": 5.4296,
|
|
"mean_token_accuracy": 0.16353966891765595,
|
|
"num_tokens": 6824171.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.4771934986114506,
|
|
"epoch": 2.9977283053157655,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00044382172771857383,
|
|
"loss": 5.2633,
|
|
"mean_token_accuracy": 0.17647497057914735,
|
|
"num_tokens": 6833668.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.50168153974745,
|
|
"epoch": 3.0018173557473875,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0004435878759318864,
|
|
"loss": 5.2173,
|
|
"mean_token_accuracy": 0.17264852921168009,
|
|
"num_tokens": 6842297.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.582038354873657,
|
|
"epoch": 3.0063607451158565,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004433536081713738,
|
|
"loss": 5.2017,
|
|
"mean_token_accuracy": 0.17235297709703445,
|
|
"num_tokens": 6853124.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.573977899551392,
|
|
"epoch": 3.0109041344843255,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004431189250150686,
|
|
"loss": 5.0484,
|
|
"mean_token_accuracy": 0.18736036419868468,
|
|
"num_tokens": 6861990.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.510259819030762,
|
|
"epoch": 3.015447523852794,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00044288382704202816,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.18592314124107362,
|
|
"num_tokens": 6872024.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.458439064025879,
|
|
"epoch": 3.019990913221263,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044264831483233335,
|
|
"loss": 5.1361,
|
|
"mean_token_accuracy": 0.1793292135000229,
|
|
"num_tokens": 6881205.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.44033989906311,
|
|
"epoch": 3.024534302589732,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044241238896708716,
|
|
"loss": 5.0844,
|
|
"mean_token_accuracy": 0.18597992211580278,
|
|
"num_tokens": 6891741.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.469294691085816,
|
|
"epoch": 3.029077691958201,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004421760500284131,
|
|
"loss": 5.1753,
|
|
"mean_token_accuracy": 0.17299405634403228,
|
|
"num_tokens": 6902122.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.520148515701294,
|
|
"epoch": 3.0336210813266695,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004419392985994541,
|
|
"loss": 5.1402,
|
|
"mean_token_accuracy": 0.1796269476413727,
|
|
"num_tokens": 6912129.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.446143102645874,
|
|
"epoch": 3.0381644706951385,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00044170213526437076,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.18651112914085388,
|
|
"num_tokens": 6921409.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.436380863189697,
|
|
"epoch": 3.0427078600636075,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00044146456060833994,
|
|
"loss": 5.1477,
|
|
"mean_token_accuracy": 0.1810881659388542,
|
|
"num_tokens": 6932346.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.518705892562866,
|
|
"epoch": 3.0472512494320765,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00044122657521755373,
|
|
"loss": 5.1585,
|
|
"mean_token_accuracy": 0.18036397099494933,
|
|
"num_tokens": 6942139.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.500319337844848,
|
|
"epoch": 3.051794638800545,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044098817967921716,
|
|
"loss": 5.1696,
|
|
"mean_token_accuracy": 0.17719566375017165,
|
|
"num_tokens": 6952344.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.535175800323486,
|
|
"epoch": 3.056338028169014,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00044074937458154763,
|
|
"loss": 5.1659,
|
|
"mean_token_accuracy": 0.17821048349142074,
|
|
"num_tokens": 6962430.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.538405227661133,
|
|
"epoch": 3.060881417537483,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00044051016051377303,
|
|
"loss": 5.1949,
|
|
"mean_token_accuracy": 0.17264898866415024,
|
|
"num_tokens": 6973213.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.472163105010987,
|
|
"epoch": 3.065424806905952,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004402705380661302,
|
|
"loss": 5.1387,
|
|
"mean_token_accuracy": 0.1777632713317871,
|
|
"num_tokens": 6983896.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.451051521301269,
|
|
"epoch": 3.0699681962744205,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00044003050782986377,
|
|
"loss": 5.0988,
|
|
"mean_token_accuracy": 0.18774761855602265,
|
|
"num_tokens": 6993549.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.475644063949585,
|
|
"epoch": 3.0745115856428895,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004397900703972245,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.1838320165872574,
|
|
"num_tokens": 7003893.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.550777721405029,
|
|
"epoch": 3.0790549750113585,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004395492263614677,
|
|
"loss": 5.2618,
|
|
"mean_token_accuracy": 0.16851040869951248,
|
|
"num_tokens": 7014826.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.489721155166626,
|
|
"epoch": 3.0835983643798275,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00043930797631685234,
|
|
"loss": 5.1917,
|
|
"mean_token_accuracy": 0.18085327595472336,
|
|
"num_tokens": 7024758.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.452193975448608,
|
|
"epoch": 3.088141753748296,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00043906632085863876,
|
|
"loss": 5.1392,
|
|
"mean_token_accuracy": 0.179022353887558,
|
|
"num_tokens": 7034063.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.433679676055908,
|
|
"epoch": 3.092685143116765,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00043882426058308775,
|
|
"loss": 5.1472,
|
|
"mean_token_accuracy": 0.17723008394241332,
|
|
"num_tokens": 7045137.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.427992486953736,
|
|
"epoch": 3.097228532485234,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00043858179608745915,
|
|
"loss": 5.0818,
|
|
"mean_token_accuracy": 0.17813556045293807,
|
|
"num_tokens": 7056381.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.54514799118042,
|
|
"epoch": 3.101771921853703,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004383389279700098,
|
|
"loss": 5.2195,
|
|
"mean_token_accuracy": 0.1720323771238327,
|
|
"num_tokens": 7066462.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.5610706329345705,
|
|
"epoch": 3.1063153112221715,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00043809565682999274,
|
|
"loss": 5.2475,
|
|
"mean_token_accuracy": 0.17274146527051926,
|
|
"num_tokens": 7078438.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.515238332748413,
|
|
"epoch": 3.1108587005906405,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043785198326765535,
|
|
"loss": 5.1102,
|
|
"mean_token_accuracy": 0.17806554734706878,
|
|
"num_tokens": 7088981.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.485534954071045,
|
|
"epoch": 3.1154020899591095,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043760790788423785,
|
|
"loss": 5.1628,
|
|
"mean_token_accuracy": 0.17726323753595352,
|
|
"num_tokens": 7099407.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.453364896774292,
|
|
"epoch": 3.1199454793275785,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00043736343128197207,
|
|
"loss": 5.0654,
|
|
"mean_token_accuracy": 0.18315556049346923,
|
|
"num_tokens": 7109454.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.357700538635254,
|
|
"epoch": 3.124488868696047,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004371185540640796,
|
|
"loss": 5.148,
|
|
"mean_token_accuracy": 0.18148666322231294,
|
|
"num_tokens": 7119639.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.492453718185425,
|
|
"epoch": 3.129032258064516,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00043687327683477077,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.18315582871437072,
|
|
"num_tokens": 7129872.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.454137706756592,
|
|
"epoch": 3.133575647432985,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004366276001992427,
|
|
"loss": 5.0681,
|
|
"mean_token_accuracy": 0.18447718322277068,
|
|
"num_tokens": 7140456.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.388547563552857,
|
|
"epoch": 3.138119036801454,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.000436381524763678,
|
|
"loss": 5.0905,
|
|
"mean_token_accuracy": 0.1849132224917412,
|
|
"num_tokens": 7149787.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.3914776802062985,
|
|
"epoch": 3.142662426169923,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004361350511352435,
|
|
"loss": 5.0803,
|
|
"mean_token_accuracy": 0.18337713778018952,
|
|
"num_tokens": 7160482.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.464912462234497,
|
|
"epoch": 3.1472058155383915,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004358881799220882,
|
|
"loss": 5.1,
|
|
"mean_token_accuracy": 0.1844940021634102,
|
|
"num_tokens": 7170818.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.431268692016602,
|
|
"epoch": 3.1517492049068605,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00043564091173334236,
|
|
"loss": 5.1542,
|
|
"mean_token_accuracy": 0.18008783161640168,
|
|
"num_tokens": 7180770.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.3913825988769535,
|
|
"epoch": 3.1562925942753295,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00043539324717911573,
|
|
"loss": 5.0619,
|
|
"mean_token_accuracy": 0.18589441627264022,
|
|
"num_tokens": 7190751.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.437940263748169,
|
|
"epoch": 3.1608359836437985,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00043514518687049587,
|
|
"loss": 5.0611,
|
|
"mean_token_accuracy": 0.1921260640025139,
|
|
"num_tokens": 7200093.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.523890018463135,
|
|
"epoch": 3.165379373012267,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000434896731419547,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.18091005682945252,
|
|
"num_tokens": 7210526.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.413506650924683,
|
|
"epoch": 3.169922762380736,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 0.0004346478814393082,
|
|
"loss": 5.1468,
|
|
"mean_token_accuracy": 0.1842249482870102,
|
|
"num_tokens": 7222127.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.382342147827148,
|
|
"epoch": 3.174466151749205,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004343986375437921,
|
|
"loss": 5.1076,
|
|
"mean_token_accuracy": 0.18242321014404297,
|
|
"num_tokens": 7233418.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.490155506134033,
|
|
"epoch": 3.179009541117674,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00043414900034798314,
|
|
"loss": 5.1751,
|
|
"mean_token_accuracy": 0.17678478956222535,
|
|
"num_tokens": 7243743.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.179009541117674,
|
|
"eval_entropy": 5.373775104034778,
|
|
"eval_loss": 5.847410678863525,
|
|
"eval_mean_token_accuracy": 0.15943208842776543,
|
|
"eval_num_tokens": 7243743.0,
|
|
"eval_runtime": 2.5709,
|
|
"eval_samples_per_second": 1336.105,
|
|
"eval_steps_per_second": 167.256,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.455605173110962,
|
|
"epoch": 3.1835529304861425,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004338989704678364,
|
|
"loss": 5.1971,
|
|
"mean_token_accuracy": 0.17006030678749084,
|
|
"num_tokens": 7254906.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.445714902877808,
|
|
"epoch": 3.1880963198546115,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00043364854852027566,
|
|
"loss": 5.1503,
|
|
"mean_token_accuracy": 0.17842386811971664,
|
|
"num_tokens": 7264193.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.539492273330689,
|
|
"epoch": 3.1926397092230805,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004333977351231924,
|
|
"loss": 5.2096,
|
|
"mean_token_accuracy": 0.17794540673494338,
|
|
"num_tokens": 7273723.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.504536581039429,
|
|
"epoch": 3.1971830985915495,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00043314653089544336,
|
|
"loss": 5.2216,
|
|
"mean_token_accuracy": 0.17015519589185715,
|
|
"num_tokens": 7284142.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.434875535964966,
|
|
"epoch": 3.201726487960018,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004328949364568504,
|
|
"loss": 5.1045,
|
|
"mean_token_accuracy": 0.18106790781021118,
|
|
"num_tokens": 7294228.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.430656623840332,
|
|
"epoch": 3.206269877328487,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004326429524281975,
|
|
"loss": 5.1418,
|
|
"mean_token_accuracy": 0.1846308797597885,
|
|
"num_tokens": 7304555.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.501480007171631,
|
|
"epoch": 3.210813266696956,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0004323905794312303,
|
|
"loss": 5.2532,
|
|
"mean_token_accuracy": 0.1675907000899315,
|
|
"num_tokens": 7315955.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.4519788265228275,
|
|
"epoch": 3.215356656065425,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00043213781808865414,
|
|
"loss": 5.1808,
|
|
"mean_token_accuracy": 0.1757256880402565,
|
|
"num_tokens": 7326757.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.469808292388916,
|
|
"epoch": 3.2199000454338935,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004318846690241324,
|
|
"loss": 5.1404,
|
|
"mean_token_accuracy": 0.1841190680861473,
|
|
"num_tokens": 7335969.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.441014528274536,
|
|
"epoch": 3.2244434348023625,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00043163113286228533,
|
|
"loss": 5.1532,
|
|
"mean_token_accuracy": 0.17956622689962387,
|
|
"num_tokens": 7345824.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.510980129241943,
|
|
"epoch": 3.2289868241708315,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004313772102286882,
|
|
"loss": 5.2195,
|
|
"mean_token_accuracy": 0.17319514304399491,
|
|
"num_tokens": 7356292.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.447022247314453,
|
|
"epoch": 3.2335302135393005,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0004311229017498698,
|
|
"loss": 5.2096,
|
|
"mean_token_accuracy": 0.17547392398118972,
|
|
"num_tokens": 7366809.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.4817718982696535,
|
|
"epoch": 3.238073602907769,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000430868208053311,
|
|
"loss": 5.1204,
|
|
"mean_token_accuracy": 0.19000674337148665,
|
|
"num_tokens": 7376076.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.480525779724121,
|
|
"epoch": 3.242616992276238,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00043061312976744326,
|
|
"loss": 5.1408,
|
|
"mean_token_accuracy": 0.17823605388402938,
|
|
"num_tokens": 7386338.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.440802001953125,
|
|
"epoch": 3.247160381644707,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0004303576675216469,
|
|
"loss": 5.117,
|
|
"mean_token_accuracy": 0.1793022036552429,
|
|
"num_tokens": 7395792.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.454340600967408,
|
|
"epoch": 3.251703771013176,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00043010182194624957,
|
|
"loss": 5.1495,
|
|
"mean_token_accuracy": 0.1824739471077919,
|
|
"num_tokens": 7405765.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.447493028640747,
|
|
"epoch": 3.2562471603816445,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004298455936725248,
|
|
"loss": 5.211,
|
|
"mean_token_accuracy": 0.17772205024957657,
|
|
"num_tokens": 7415213.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.496893739700317,
|
|
"epoch": 3.2607905497501135,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004295889833326903,
|
|
"loss": 5.1688,
|
|
"mean_token_accuracy": 0.1726471111178398,
|
|
"num_tokens": 7425722.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.560001516342163,
|
|
"epoch": 3.2653339391185825,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 0.00042933199155990675,
|
|
"loss": 5.2257,
|
|
"mean_token_accuracy": 0.18278490751981735,
|
|
"num_tokens": 7436240.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.522018718719482,
|
|
"epoch": 3.2698773284870515,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004290746189882757,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.1700237810611725,
|
|
"num_tokens": 7447878.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.4914027690887455,
|
|
"epoch": 3.27442071785552,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00042881686625283844,
|
|
"loss": 5.1452,
|
|
"mean_token_accuracy": 0.18039888739585877,
|
|
"num_tokens": 7458635.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.452340078353882,
|
|
"epoch": 3.278964107223989,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004285587339895742,
|
|
"loss": 5.2046,
|
|
"mean_token_accuracy": 0.16521691381931305,
|
|
"num_tokens": 7468740.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.521694993972778,
|
|
"epoch": 3.283507496592458,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00042830022283539863,
|
|
"loss": 5.259,
|
|
"mean_token_accuracy": 0.17513288408517838,
|
|
"num_tokens": 7480977.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.479649019241333,
|
|
"epoch": 3.288050885960927,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004280413334281623,
|
|
"loss": 5.2712,
|
|
"mean_token_accuracy": 0.1735505446791649,
|
|
"num_tokens": 7491758.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.501490783691406,
|
|
"epoch": 3.292594275329396,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00042778206640664915,
|
|
"loss": 5.1959,
|
|
"mean_token_accuracy": 0.17478640675544738,
|
|
"num_tokens": 7500486.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.427266025543213,
|
|
"epoch": 3.2971376646978645,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004275224224105748,
|
|
"loss": 5.1795,
|
|
"mean_token_accuracy": 0.18620938658714295,
|
|
"num_tokens": 7510371.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.487931394577027,
|
|
"epoch": 3.3016810540663335,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00042726240208058484,
|
|
"loss": 5.2056,
|
|
"mean_token_accuracy": 0.1818501189351082,
|
|
"num_tokens": 7520697.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.485713958740234,
|
|
"epoch": 3.3062244434348025,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004270020060582538,
|
|
"loss": 5.1359,
|
|
"mean_token_accuracy": 0.17839961796998977,
|
|
"num_tokens": 7530757.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.476201391220092,
|
|
"epoch": 3.310767832803271,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00042674123498608285,
|
|
"loss": 5.1713,
|
|
"mean_token_accuracy": 0.17760373800992965,
|
|
"num_tokens": 7541810.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.538935661315918,
|
|
"epoch": 3.31531122217174,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00042648008950749877,
|
|
"loss": 5.3716,
|
|
"mean_token_accuracy": 0.16463814228773116,
|
|
"num_tokens": 7552754.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.531646823883056,
|
|
"epoch": 3.319854611540209,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.000426218570266852,
|
|
"loss": 5.1456,
|
|
"mean_token_accuracy": 0.1809786930680275,
|
|
"num_tokens": 7562906.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.365774726867675,
|
|
"epoch": 3.324398000908678,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00042595667790941536,
|
|
"loss": 5.1597,
|
|
"mean_token_accuracy": 0.17980565875768661,
|
|
"num_tokens": 7574149.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.415554189682007,
|
|
"epoch": 3.328941390277147,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00042569441308138204,
|
|
"loss": 5.132,
|
|
"mean_token_accuracy": 0.187614868581295,
|
|
"num_tokens": 7584610.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.441869163513184,
|
|
"epoch": 3.3334847796456155,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004254317764298645,
|
|
"loss": 5.1094,
|
|
"mean_token_accuracy": 0.17508845627307892,
|
|
"num_tokens": 7594954.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.474972534179687,
|
|
"epoch": 3.3380281690140845,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004251687686028927,
|
|
"loss": 5.2323,
|
|
"mean_token_accuracy": 0.16958318650722504,
|
|
"num_tokens": 7606069.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.424328422546386,
|
|
"epoch": 3.3425715583825535,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00042490539024941216,
|
|
"loss": 5.0815,
|
|
"mean_token_accuracy": 0.1843983218073845,
|
|
"num_tokens": 7615671.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.472716522216797,
|
|
"epoch": 3.347114947751022,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.00042464164201928286,
|
|
"loss": 5.2658,
|
|
"mean_token_accuracy": 0.17172005921602249,
|
|
"num_tokens": 7626740.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.497398805618286,
|
|
"epoch": 3.351658337119491,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00042437752456327734,
|
|
"loss": 5.1943,
|
|
"mean_token_accuracy": 0.18103505820035934,
|
|
"num_tokens": 7636459.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.529231214523316,
|
|
"epoch": 3.35620172648796,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00042411303853307907,
|
|
"loss": 5.1426,
|
|
"mean_token_accuracy": 0.17729359716176987,
|
|
"num_tokens": 7646974.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.400937604904175,
|
|
"epoch": 3.360745115856429,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004238481845812811,
|
|
"loss": 5.0969,
|
|
"mean_token_accuracy": 0.18112910240888597,
|
|
"num_tokens": 7657372.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.367641639709473,
|
|
"epoch": 3.365288505224898,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00042358296336138436,
|
|
"loss": 5.1595,
|
|
"mean_token_accuracy": 0.17441086620092391,
|
|
"num_tokens": 7668972.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.481690502166748,
|
|
"epoch": 3.3698318945933665,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004233173755277956,
|
|
"loss": 5.1529,
|
|
"mean_token_accuracy": 0.1946765437722206,
|
|
"num_tokens": 7679742.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.448097991943359,
|
|
"epoch": 3.3743752839618355,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00042305142173582657,
|
|
"loss": 5.1905,
|
|
"mean_token_accuracy": 0.1816701129078865,
|
|
"num_tokens": 7690759.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.415245628356933,
|
|
"epoch": 3.3789186733303045,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00042278510264169165,
|
|
"loss": 5.1681,
|
|
"mean_token_accuracy": 0.17963534593582153,
|
|
"num_tokens": 7700482.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.4490992546081545,
|
|
"epoch": 3.3834620626987735,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00042251841890250696,
|
|
"loss": 5.1514,
|
|
"mean_token_accuracy": 0.17664661705493928,
|
|
"num_tokens": 7710414.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.512318134307861,
|
|
"epoch": 3.388005452067242,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000422251371176288,
|
|
"loss": 5.1792,
|
|
"mean_token_accuracy": 0.1759432762861252,
|
|
"num_tokens": 7721214.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.42315788269043,
|
|
"epoch": 3.392548841435711,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00042198396012194854,
|
|
"loss": 5.1508,
|
|
"mean_token_accuracy": 0.17981941848993302,
|
|
"num_tokens": 7732551.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.51419095993042,
|
|
"epoch": 3.39709223080418,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00042171618639929865,
|
|
"loss": 5.2359,
|
|
"mean_token_accuracy": 0.1766679108142853,
|
|
"num_tokens": 7742266.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.4067340850830075,
|
|
"epoch": 3.401635620172649,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00042144805066904345,
|
|
"loss": 5.0893,
|
|
"mean_token_accuracy": 0.1799272745847702,
|
|
"num_tokens": 7753152.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.377173137664795,
|
|
"epoch": 3.4061790095411175,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00042117955359278125,
|
|
"loss": 5.1775,
|
|
"mean_token_accuracy": 0.17800859957933426,
|
|
"num_tokens": 7764420.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.5033769607543945,
|
|
"epoch": 3.4107223989095865,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00042091069583300177,
|
|
"loss": 5.166,
|
|
"mean_token_accuracy": 0.1811123937368393,
|
|
"num_tokens": 7776044.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.465174865722656,
|
|
"epoch": 3.4152657882780555,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0004206414780530849,
|
|
"loss": 5.2053,
|
|
"mean_token_accuracy": 0.1752263143658638,
|
|
"num_tokens": 7786737.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.390635681152344,
|
|
"epoch": 3.4198091776465245,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 0.0004203719009172987,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.18294474184513093,
|
|
"num_tokens": 7796974.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.473478603363037,
|
|
"epoch": 3.424352567014993,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000420101965090798,
|
|
"loss": 5.2751,
|
|
"mean_token_accuracy": 0.1697794884443283,
|
|
"num_tokens": 7808221.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.416862869262696,
|
|
"epoch": 3.428895956383462,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0004198316712396226,
|
|
"loss": 5.1077,
|
|
"mean_token_accuracy": 0.18563559800386428,
|
|
"num_tokens": 7818422.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.442232418060303,
|
|
"epoch": 3.433439345751931,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00041956102003069573,
|
|
"loss": 5.1277,
|
|
"mean_token_accuracy": 0.18608348816633224,
|
|
"num_tokens": 7828313.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.417854928970337,
|
|
"epoch": 3.4379827351204,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0004192900121318223,
|
|
"loss": 5.174,
|
|
"mean_token_accuracy": 0.18715976923704147,
|
|
"num_tokens": 7839493.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.444969892501831,
|
|
"epoch": 3.4425261244888685,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004190186482116874,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.17483771294355394,
|
|
"num_tokens": 7849796.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.450568056106567,
|
|
"epoch": 3.4470695138573375,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004187469289398545,
|
|
"loss": 5.2278,
|
|
"mean_token_accuracy": 0.17209161967039108,
|
|
"num_tokens": 7860865.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.410475587844848,
|
|
"epoch": 3.4516129032258065,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00041847485498676385,
|
|
"loss": 5.1659,
|
|
"mean_token_accuracy": 0.18606956601142882,
|
|
"num_tokens": 7871144.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.415421819686889,
|
|
"epoch": 3.4561562925942755,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.00041820242702373094,
|
|
"loss": 5.1988,
|
|
"mean_token_accuracy": 0.18072634935379028,
|
|
"num_tokens": 7880916.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.341844940185547,
|
|
"epoch": 3.460699681962744,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0004179296457229448,
|
|
"loss": 5.1339,
|
|
"mean_token_accuracy": 0.18067546039819718,
|
|
"num_tokens": 7891813.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.45639271736145,
|
|
"epoch": 3.465243071331213,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004176565117574659,
|
|
"loss": 5.2055,
|
|
"mean_token_accuracy": 0.17605607509613036,
|
|
"num_tokens": 7902697.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.504477214813233,
|
|
"epoch": 3.469786460699682,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004173830258012255,
|
|
"loss": 5.1563,
|
|
"mean_token_accuracy": 0.1824314370751381,
|
|
"num_tokens": 7912777.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.439791107177735,
|
|
"epoch": 3.474329850068151,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004171091885290226,
|
|
"loss": 5.2085,
|
|
"mean_token_accuracy": 0.17921757996082305,
|
|
"num_tokens": 7923255.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.453814363479614,
|
|
"epoch": 3.4788732394366195,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 0.0004168350006165238,
|
|
"loss": 5.0901,
|
|
"mean_token_accuracy": 0.1806154206395149,
|
|
"num_tokens": 7934268.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.406710243225097,
|
|
"epoch": 3.4834166288050885,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00041656046274026034,
|
|
"loss": 5.11,
|
|
"mean_token_accuracy": 0.18290983140468597,
|
|
"num_tokens": 7944526.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.39798789024353,
|
|
"epoch": 3.4879600181735575,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.0004162855755776272,
|
|
"loss": 5.2053,
|
|
"mean_token_accuracy": 0.1815582647919655,
|
|
"num_tokens": 7954512.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.464317846298218,
|
|
"epoch": 3.4925034075420265,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 0.00041601033980688106,
|
|
"loss": 5.2419,
|
|
"mean_token_accuracy": 0.1661909192800522,
|
|
"num_tokens": 7965703.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.515943717956543,
|
|
"epoch": 3.497046796910495,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0004157347561071389,
|
|
"loss": 5.2156,
|
|
"mean_token_accuracy": 0.17564896941185,
|
|
"num_tokens": 7977822.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.387682771682739,
|
|
"epoch": 3.501590186278964,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.00041545882515837604,
|
|
"loss": 5.1179,
|
|
"mean_token_accuracy": 0.1787295937538147,
|
|
"num_tokens": 7988318.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.454885768890381,
|
|
"epoch": 3.506133575647433,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00041518254764142477,
|
|
"loss": 5.0884,
|
|
"mean_token_accuracy": 0.18690006732940673,
|
|
"num_tokens": 7998076.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.423733711242676,
|
|
"epoch": 3.510676965015902,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00041490592423797226,
|
|
"loss": 5.1817,
|
|
"mean_token_accuracy": 0.182457035779953,
|
|
"num_tokens": 8007913.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.421347284317017,
|
|
"epoch": 3.515220354384371,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004146289556305594,
|
|
"loss": 5.1756,
|
|
"mean_token_accuracy": 0.17665185481309892,
|
|
"num_tokens": 8018340.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.501410627365113,
|
|
"epoch": 3.5197637437528395,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 0.0004143516425025786,
|
|
"loss": 5.2086,
|
|
"mean_token_accuracy": 0.17663790583610534,
|
|
"num_tokens": 8028312.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.509435033798217,
|
|
"epoch": 3.5243071331213085,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0004140739855382724,
|
|
"loss": 5.2163,
|
|
"mean_token_accuracy": 0.18351612985134125,
|
|
"num_tokens": 8038580.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.475415515899658,
|
|
"epoch": 3.5288505224897775,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00041379598542273197,
|
|
"loss": 5.1824,
|
|
"mean_token_accuracy": 0.17666554749011992,
|
|
"num_tokens": 8048365.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.461563968658448,
|
|
"epoch": 3.533393911858246,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00041351764284189464,
|
|
"loss": 5.2247,
|
|
"mean_token_accuracy": 0.17258904576301576,
|
|
"num_tokens": 8059638.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.515468740463257,
|
|
"epoch": 3.537937301226715,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004132389584825433,
|
|
"loss": 5.2939,
|
|
"mean_token_accuracy": 0.16828196346759797,
|
|
"num_tokens": 8070993.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.423805856704712,
|
|
"epoch": 3.542480690595184,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00041295993303230383,
|
|
"loss": 5.13,
|
|
"mean_token_accuracy": 0.17964119464159012,
|
|
"num_tokens": 8081203.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.402665090560913,
|
|
"epoch": 3.547024079963653,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004126805671796438,
|
|
"loss": 5.1459,
|
|
"mean_token_accuracy": 0.1772849217057228,
|
|
"num_tokens": 8091564.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.464258813858033,
|
|
"epoch": 3.551567469332122,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004124008616138707,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.18072547018527985,
|
|
"num_tokens": 8100675.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.431096696853638,
|
|
"epoch": 3.5561108587005905,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 0.00041212081702513015,
|
|
"loss": 5.1807,
|
|
"mean_token_accuracy": 0.18447889685630797,
|
|
"num_tokens": 8110605.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.417476224899292,
|
|
"epoch": 3.5606542480690595,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004118404341044044,
|
|
"loss": 5.2216,
|
|
"mean_token_accuracy": 0.17528523951768876,
|
|
"num_tokens": 8120910.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.464909553527832,
|
|
"epoch": 3.5651976374375285,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00041155971354351034,
|
|
"loss": 5.1792,
|
|
"mean_token_accuracy": 0.17850044071674348,
|
|
"num_tokens": 8131827.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.41238694190979,
|
|
"epoch": 3.569741026805997,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00041127865603509815,
|
|
"loss": 5.1577,
|
|
"mean_token_accuracy": 0.17781717628240584,
|
|
"num_tokens": 8141785.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.467757892608643,
|
|
"epoch": 3.574284416174466,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.0004109972622726493,
|
|
"loss": 5.1658,
|
|
"mean_token_accuracy": 0.1758715733885765,
|
|
"num_tokens": 8151732.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.529815530776977,
|
|
"epoch": 3.578827805542935,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00041071553295047485,
|
|
"loss": 5.2428,
|
|
"mean_token_accuracy": 0.1739981070160866,
|
|
"num_tokens": 8162334.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.429641389846802,
|
|
"epoch": 3.583371194911404,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004104334687637139,
|
|
"loss": 5.1332,
|
|
"mean_token_accuracy": 0.17753940224647521,
|
|
"num_tokens": 8172810.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.389226341247559,
|
|
"epoch": 3.587914584279873,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.00041015107040833187,
|
|
"loss": 5.1829,
|
|
"mean_token_accuracy": 0.17421576231718064,
|
|
"num_tokens": 8183240.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.413247728347779,
|
|
"epoch": 3.5924579736483415,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0004098683385811185,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.1795205220580101,
|
|
"num_tokens": 8193824.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.405896329879761,
|
|
"epoch": 3.5970013630168105,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.00040958527397968666,
|
|
"loss": 5.1131,
|
|
"mean_token_accuracy": 0.18436967581510544,
|
|
"num_tokens": 8203792.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.510301637649536,
|
|
"epoch": 3.6015447523852795,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004093018773024699,
|
|
"loss": 5.2759,
|
|
"mean_token_accuracy": 0.17884058356285096,
|
|
"num_tokens": 8214436.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.466083145141601,
|
|
"epoch": 3.606088141753748,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004090181492487215,
|
|
"loss": 5.1448,
|
|
"mean_token_accuracy": 0.17608668953180312,
|
|
"num_tokens": 8224518.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.490726709365845,
|
|
"epoch": 3.610631531122217,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00040873409051851223,
|
|
"loss": 5.3009,
|
|
"mean_token_accuracy": 0.17026201635599136,
|
|
"num_tokens": 8235234.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.438040828704834,
|
|
"epoch": 3.615174920490686,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004084497018127286,
|
|
"loss": 5.1959,
|
|
"mean_token_accuracy": 0.17208356112241746,
|
|
"num_tokens": 8246612.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.533259868621826,
|
|
"epoch": 3.619718309859155,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.0004081649838330717,
|
|
"loss": 5.1226,
|
|
"mean_token_accuracy": 0.18128403127193451,
|
|
"num_tokens": 8256720.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.458050107955932,
|
|
"epoch": 3.624261699227624,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00040787993728205464,
|
|
"loss": 5.2158,
|
|
"mean_token_accuracy": 0.17919420450925827,
|
|
"num_tokens": 8267666.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.443320989608765,
|
|
"epoch": 3.6288050885960925,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00040759456286300154,
|
|
"loss": 5.1761,
|
|
"mean_token_accuracy": 0.17740476727485657,
|
|
"num_tokens": 8277882.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.458083915710449,
|
|
"epoch": 3.6333484779645615,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00040730886128004544,
|
|
"loss": 5.1167,
|
|
"mean_token_accuracy": 0.18049341440200806,
|
|
"num_tokens": 8286968.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.6333484779645615,
|
|
"eval_entropy": 5.410700931105503,
|
|
"eval_loss": 5.822494029998779,
|
|
"eval_mean_token_accuracy": 0.1598711833877619,
|
|
"eval_num_tokens": 8286968.0,
|
|
"eval_runtime": 2.5663,
|
|
"eval_samples_per_second": 1338.488,
|
|
"eval_steps_per_second": 167.555,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.467584753036499,
|
|
"epoch": 3.6378918673330305,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004070228332381266,
|
|
"loss": 5.1884,
|
|
"mean_token_accuracy": 0.18219147473573685,
|
|
"num_tokens": 8297154.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.456792211532592,
|
|
"epoch": 3.6424352567014995,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0004067364794429908,
|
|
"loss": 5.1601,
|
|
"mean_token_accuracy": 0.17762396335601807,
|
|
"num_tokens": 8307954.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.422146272659302,
|
|
"epoch": 3.646978646069968,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004064498006011874,
|
|
"loss": 5.153,
|
|
"mean_token_accuracy": 0.17737304419279099,
|
|
"num_tokens": 8317714.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.391381931304932,
|
|
"epoch": 3.651522035438437,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00040616279742006813,
|
|
"loss": 5.1677,
|
|
"mean_token_accuracy": 0.18638110160827637,
|
|
"num_tokens": 8327963.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.395629644393921,
|
|
"epoch": 3.656065424806906,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00040587547060778466,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.19039573818445205,
|
|
"num_tokens": 8338190.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.480742740631103,
|
|
"epoch": 3.660608814175375,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004055878208732874,
|
|
"loss": 5.2379,
|
|
"mean_token_accuracy": 0.17104028463363646,
|
|
"num_tokens": 8348564.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.442214012145996,
|
|
"epoch": 3.665152203543844,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00040529984892632363,
|
|
"loss": 5.1411,
|
|
"mean_token_accuracy": 0.18297356218099595,
|
|
"num_tokens": 8359463.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.42068886756897,
|
|
"epoch": 3.6696955929123125,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004050115554774353,
|
|
"loss": 5.2134,
|
|
"mean_token_accuracy": 0.17563298642635344,
|
|
"num_tokens": 8369676.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.414844655990601,
|
|
"epoch": 3.6742389822807815,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004047229412379578,
|
|
"loss": 5.1318,
|
|
"mean_token_accuracy": 0.18917577862739562,
|
|
"num_tokens": 8379579.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.438452863693238,
|
|
"epoch": 3.6787823716492505,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004044340069200183,
|
|
"loss": 5.1947,
|
|
"mean_token_accuracy": 0.18181420415639876,
|
|
"num_tokens": 8390164.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.3900548934936525,
|
|
"epoch": 3.683325761017719,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00040414475323653326,
|
|
"loss": 5.0934,
|
|
"mean_token_accuracy": 0.18446266651153564,
|
|
"num_tokens": 8399220.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.403749084472656,
|
|
"epoch": 3.687869150386188,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0004038551809012076,
|
|
"loss": 5.1566,
|
|
"mean_token_accuracy": 0.17952479124069215,
|
|
"num_tokens": 8409488.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.42854871749878,
|
|
"epoch": 3.692412539754657,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004035652906285322,
|
|
"loss": 5.1116,
|
|
"mean_token_accuracy": 0.1911702811717987,
|
|
"num_tokens": 8418817.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.423796892166138,
|
|
"epoch": 3.696955929123126,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.00040327508313378244,
|
|
"loss": 5.2241,
|
|
"mean_token_accuracy": 0.1786786288022995,
|
|
"num_tokens": 8428742.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.408110904693603,
|
|
"epoch": 3.701499318491595,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0004029845591330164,
|
|
"loss": 5.0896,
|
|
"mean_token_accuracy": 0.19170673787593842,
|
|
"num_tokens": 8438016.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.482151508331299,
|
|
"epoch": 3.7060427078600635,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00040269371934307336,
|
|
"loss": 5.2175,
|
|
"mean_token_accuracy": 0.18434158712625504,
|
|
"num_tokens": 8448428.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.37024621963501,
|
|
"epoch": 3.7105860972285325,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00040240256448157146,
|
|
"loss": 5.0554,
|
|
"mean_token_accuracy": 0.18492696583271026,
|
|
"num_tokens": 8458444.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.396601295471191,
|
|
"epoch": 3.7151294865970015,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004021110952669065,
|
|
"loss": 5.1869,
|
|
"mean_token_accuracy": 0.18095253258943558,
|
|
"num_tokens": 8467721.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.411504220962525,
|
|
"epoch": 3.71967287596547,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.0004018193124182497,
|
|
"loss": 5.1165,
|
|
"mean_token_accuracy": 0.18424260020256042,
|
|
"num_tokens": 8477484.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.42717924118042,
|
|
"epoch": 3.724216265333939,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004015272166555462,
|
|
"loss": 5.1328,
|
|
"mean_token_accuracy": 0.1814862921833992,
|
|
"num_tokens": 8486902.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.386594533920288,
|
|
"epoch": 3.728759654702408,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004012348086995132,
|
|
"loss": 5.0976,
|
|
"mean_token_accuracy": 0.18146499246358871,
|
|
"num_tokens": 8497626.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.418752241134643,
|
|
"epoch": 3.733303044070877,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004009420892716385,
|
|
"loss": 5.197,
|
|
"mean_token_accuracy": 0.17187473326921462,
|
|
"num_tokens": 8507518.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.41076397895813,
|
|
"epoch": 3.737846433439346,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004006490590941781,
|
|
"loss": 5.1744,
|
|
"mean_token_accuracy": 0.17898114621639252,
|
|
"num_tokens": 8517835.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.4592029571533205,
|
|
"epoch": 3.7423898228078145,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004003557188901547,
|
|
"loss": 5.2113,
|
|
"mean_token_accuracy": 0.17653865367174149,
|
|
"num_tokens": 8527552.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.4556434631347654,
|
|
"epoch": 3.7469332121762835,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004000620693833562,
|
|
"loss": 5.1963,
|
|
"mean_token_accuracy": 0.18428903073072433,
|
|
"num_tokens": 8538082.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.438845920562744,
|
|
"epoch": 3.7514766015447525,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.0003997681112983335,
|
|
"loss": 5.1272,
|
|
"mean_token_accuracy": 0.17878906726837157,
|
|
"num_tokens": 8548558.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.378269147872925,
|
|
"epoch": 3.756019990913221,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00039947384536039905,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.18036806732416152,
|
|
"num_tokens": 8559406.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.481081867218018,
|
|
"epoch": 3.76056338028169,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00039917927229562474,
|
|
"loss": 5.1707,
|
|
"mean_token_accuracy": 0.17657368928194045,
|
|
"num_tokens": 8569519.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.429835796356201,
|
|
"epoch": 3.765106769650159,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 0.00039888439283084035,
|
|
"loss": 5.1945,
|
|
"mean_token_accuracy": 0.17817631512880325,
|
|
"num_tokens": 8579658.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.396280908584595,
|
|
"epoch": 3.769650159018628,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00039858920769363164,
|
|
"loss": 5.107,
|
|
"mean_token_accuracy": 0.18225110322237015,
|
|
"num_tokens": 8589681.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.454093933105469,
|
|
"epoch": 3.774193548387097,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.0003982937176123386,
|
|
"loss": 5.185,
|
|
"mean_token_accuracy": 0.1793945088982582,
|
|
"num_tokens": 8600310.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.441811752319336,
|
|
"epoch": 3.7787369377555655,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00039799792331605364,
|
|
"loss": 5.219,
|
|
"mean_token_accuracy": 0.17503565698862075,
|
|
"num_tokens": 8610920.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.403436994552612,
|
|
"epoch": 3.7832803271240345,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00039770182553461965,
|
|
"loss": 5.114,
|
|
"mean_token_accuracy": 0.18590138256549835,
|
|
"num_tokens": 8620536.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.438713788986206,
|
|
"epoch": 3.7878237164925035,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00039740542499862866,
|
|
"loss": 5.2167,
|
|
"mean_token_accuracy": 0.1767923817038536,
|
|
"num_tokens": 8631647.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.422046279907226,
|
|
"epoch": 3.792367105860972,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003971087224394194,
|
|
"loss": 5.1829,
|
|
"mean_token_accuracy": 0.1789577141404152,
|
|
"num_tokens": 8642659.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.409045219421387,
|
|
"epoch": 3.796910495229441,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0003968117185890761,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.18799003064632416,
|
|
"num_tokens": 8651775.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.433942556381226,
|
|
"epoch": 3.80145388459791,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00039651441418042606,
|
|
"loss": 5.1528,
|
|
"mean_token_accuracy": 0.1778918907046318,
|
|
"num_tokens": 8662520.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.4341737747192385,
|
|
"epoch": 3.805997273966379,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003962168099470384,
|
|
"loss": 5.1717,
|
|
"mean_token_accuracy": 0.1820436418056488,
|
|
"num_tokens": 8672630.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.477026653289795,
|
|
"epoch": 3.810540663334848,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000395918906623222,
|
|
"loss": 5.2546,
|
|
"mean_token_accuracy": 0.16686135083436965,
|
|
"num_tokens": 8683348.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.417185258865357,
|
|
"epoch": 3.8150840527033165,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003956207049440238,
|
|
"loss": 5.0898,
|
|
"mean_token_accuracy": 0.1822441875934601,
|
|
"num_tokens": 8693423.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.429653072357178,
|
|
"epoch": 3.8196274420717855,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003953222056452265,
|
|
"loss": 5.164,
|
|
"mean_token_accuracy": 0.1788071408867836,
|
|
"num_tokens": 8703766.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.482513523101806,
|
|
"epoch": 3.8241708314402545,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003950234094633479,
|
|
"loss": 5.2723,
|
|
"mean_token_accuracy": 0.17580373734235763,
|
|
"num_tokens": 8715330.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.4618727684021,
|
|
"epoch": 3.828714220808723,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 0.00039472431713563746,
|
|
"loss": 5.1557,
|
|
"mean_token_accuracy": 0.18469288498163222,
|
|
"num_tokens": 8725732.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.4686198234558105,
|
|
"epoch": 3.833257610177192,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000394424929400076,
|
|
"loss": 5.1239,
|
|
"mean_token_accuracy": 0.1830092117190361,
|
|
"num_tokens": 8736377.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.453915023803711,
|
|
"epoch": 3.837800999545661,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003941252469953729,
|
|
"loss": 5.1589,
|
|
"mean_token_accuracy": 0.17954563945531846,
|
|
"num_tokens": 8746991.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.404045581817627,
|
|
"epoch": 3.84234438891413,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003938252706609647,
|
|
"loss": 5.1583,
|
|
"mean_token_accuracy": 0.17999178618192674,
|
|
"num_tokens": 8756920.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.400370264053345,
|
|
"epoch": 3.846887778282599,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003935250011370132,
|
|
"loss": 5.2289,
|
|
"mean_token_accuracy": 0.16836348325014114,
|
|
"num_tokens": 8768617.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.491107320785522,
|
|
"epoch": 3.851431167651068,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003932244391644037,
|
|
"loss": 5.1648,
|
|
"mean_token_accuracy": 0.18902726769447326,
|
|
"num_tokens": 8778427.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.438788843154907,
|
|
"epoch": 3.8559745570195365,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003929235854847428,
|
|
"loss": 5.2154,
|
|
"mean_token_accuracy": 0.1792486235499382,
|
|
"num_tokens": 8789093.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.372583198547363,
|
|
"epoch": 3.8605179463880055,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00039262244084035724,
|
|
"loss": 5.1695,
|
|
"mean_token_accuracy": 0.17550684213638307,
|
|
"num_tokens": 8799832.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.424095535278321,
|
|
"epoch": 3.8650613357564745,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.00039232100597429143,
|
|
"loss": 5.1726,
|
|
"mean_token_accuracy": 0.18045941442251207,
|
|
"num_tokens": 8810149.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.484337091445923,
|
|
"epoch": 3.869604725124943,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00039201928163030584,
|
|
"loss": 5.1909,
|
|
"mean_token_accuracy": 0.1840778574347496,
|
|
"num_tokens": 8820087.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.433952188491821,
|
|
"epoch": 3.874148114493412,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039171726855287544,
|
|
"loss": 5.2584,
|
|
"mean_token_accuracy": 0.17090077102184295,
|
|
"num_tokens": 8830959.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.470605087280274,
|
|
"epoch": 3.878691503861881,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00039141496748718735,
|
|
"loss": 5.1745,
|
|
"mean_token_accuracy": 0.17811236530542374,
|
|
"num_tokens": 8842522.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.396602582931519,
|
|
"epoch": 3.88323489323035,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003911123791791395,
|
|
"loss": 5.084,
|
|
"mean_token_accuracy": 0.18735769391059875,
|
|
"num_tokens": 8852521.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.460910129547119,
|
|
"epoch": 3.887778282598819,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0003908095043753384,
|
|
"loss": 5.2236,
|
|
"mean_token_accuracy": 0.17589408159255981,
|
|
"num_tokens": 8862156.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.457566165924073,
|
|
"epoch": 3.8923216719672875,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00039050634382309753,
|
|
"loss": 5.0889,
|
|
"mean_token_accuracy": 0.17956527173519135,
|
|
"num_tokens": 8872772.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.46908712387085,
|
|
"epoch": 3.8968650613357565,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003902028982704355,
|
|
"loss": 5.18,
|
|
"mean_token_accuracy": 0.18370604515075684,
|
|
"num_tokens": 8883229.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.383393144607544,
|
|
"epoch": 3.9014084507042255,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003898991684660738,
|
|
"loss": 5.1657,
|
|
"mean_token_accuracy": 0.18166502863168715,
|
|
"num_tokens": 8892403.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.3773016929626465,
|
|
"epoch": 3.905951840072694,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0003895951551594358,
|
|
"loss": 5.1688,
|
|
"mean_token_accuracy": 0.17470889836549758,
|
|
"num_tokens": 8902617.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.494909572601318,
|
|
"epoch": 3.910495229441163,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000389290859100644,
|
|
"loss": 5.2114,
|
|
"mean_token_accuracy": 0.18089583963155748,
|
|
"num_tokens": 8913099.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.382600164413452,
|
|
"epoch": 3.915038618809632,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003889862810405187,
|
|
"loss": 5.1138,
|
|
"mean_token_accuracy": 0.18431212306022643,
|
|
"num_tokens": 8922469.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.413410186767578,
|
|
"epoch": 3.919582008178101,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.000388681421730576,
|
|
"loss": 5.1941,
|
|
"mean_token_accuracy": 0.18283328115940095,
|
|
"num_tokens": 8933261.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.491539764404297,
|
|
"epoch": 3.92412539754657,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0003883762819230259,
|
|
"loss": 5.2133,
|
|
"mean_token_accuracy": 0.18321413695812225,
|
|
"num_tokens": 8943873.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.4187267303466795,
|
|
"epoch": 3.9286687869150385,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00038807086237077057,
|
|
"loss": 5.0895,
|
|
"mean_token_accuracy": 0.1941248133778572,
|
|
"num_tokens": 8953802.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.4645789623260494,
|
|
"epoch": 3.9332121762835075,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00038776516382740237,
|
|
"loss": 5.2824,
|
|
"mean_token_accuracy": 0.1782743066549301,
|
|
"num_tokens": 8964309.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.441825294494629,
|
|
"epoch": 3.9377555656519765,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 0.00038745918704720205,
|
|
"loss": 5.1502,
|
|
"mean_token_accuracy": 0.17935221046209335,
|
|
"num_tokens": 8973611.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.388883352279663,
|
|
"epoch": 3.942298955020445,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00038715293278513693,
|
|
"loss": 5.1949,
|
|
"mean_token_accuracy": 0.17420602440834046,
|
|
"num_tokens": 8985783.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.447239589691162,
|
|
"epoch": 3.946842344388914,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00038684640179685896,
|
|
"loss": 5.225,
|
|
"mean_token_accuracy": 0.17294554710388182,
|
|
"num_tokens": 8996736.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.495539045333862,
|
|
"epoch": 3.951385733757383,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003865395948387029,
|
|
"loss": 5.297,
|
|
"mean_token_accuracy": 0.16954004168510436,
|
|
"num_tokens": 9007635.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.415700817108155,
|
|
"epoch": 3.955929123125852,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00038623251266768437,
|
|
"loss": 5.1337,
|
|
"mean_token_accuracy": 0.1760878399014473,
|
|
"num_tokens": 9017892.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.4047637462615965,
|
|
"epoch": 3.960472512494321,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00038592515604149815,
|
|
"loss": 5.1842,
|
|
"mean_token_accuracy": 0.18052294105291367,
|
|
"num_tokens": 9027594.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.446075391769409,
|
|
"epoch": 3.9650159018627895,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003856175257185162,
|
|
"loss": 5.1993,
|
|
"mean_token_accuracy": 0.17505759596824647,
|
|
"num_tokens": 9037083.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.418712329864502,
|
|
"epoch": 3.9695592912312585,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00038530962245778564,
|
|
"loss": 5.1403,
|
|
"mean_token_accuracy": 0.18622052371501924,
|
|
"num_tokens": 9047176.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.4395936012268065,
|
|
"epoch": 3.9741026805997275,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003850014470190272,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.17966077625751495,
|
|
"num_tokens": 9057810.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.496283149719238,
|
|
"epoch": 3.978646069968196,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003846930001626332,
|
|
"loss": 5.2255,
|
|
"mean_token_accuracy": 0.17851194441318513,
|
|
"num_tokens": 9067765.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.521961498260498,
|
|
"epoch": 3.983189459336665,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003843842826496656,
|
|
"loss": 5.1816,
|
|
"mean_token_accuracy": 0.17836789041757584,
|
|
"num_tokens": 9078796.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.43505859375,
|
|
"epoch": 3.987732848705134,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 0.000384075295241854,
|
|
"loss": 5.1187,
|
|
"mean_token_accuracy": 0.18782972544431686,
|
|
"num_tokens": 9088368.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.38691349029541,
|
|
"epoch": 3.992276238073603,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003837660387015944,
|
|
"loss": 5.096,
|
|
"mean_token_accuracy": 0.18011491298675536,
|
|
"num_tokens": 9099082.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.437783288955688,
|
|
"epoch": 3.996819627442072,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0003834565137919463,
|
|
"loss": 5.2223,
|
|
"mean_token_accuracy": 0.1785713955760002,
|
|
"num_tokens": 9110543.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.455229335361057,
|
|
"epoch": 4.000908677873694,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.0003831467212766318,
|
|
"loss": 5.1589,
|
|
"mean_token_accuracy": 0.17899038063155281,
|
|
"num_tokens": 9119347.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.337488412857056,
|
|
"epoch": 4.005452067242163,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003828366619200331,
|
|
"loss": 4.8884,
|
|
"mean_token_accuracy": 0.19192289263010026,
|
|
"num_tokens": 9129154.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.337045526504516,
|
|
"epoch": 4.009995456610632,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003825263364871908,
|
|
"loss": 4.9124,
|
|
"mean_token_accuracy": 0.19535221457481383,
|
|
"num_tokens": 9139877.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.369376230239868,
|
|
"epoch": 4.0145388459791,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003822157457438022,
|
|
"loss": 4.9906,
|
|
"mean_token_accuracy": 0.18950691521167756,
|
|
"num_tokens": 9150508.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.379881715774536,
|
|
"epoch": 4.019082235347569,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00038190489045621893,
|
|
"loss": 4.9287,
|
|
"mean_token_accuracy": 0.19516898840665817,
|
|
"num_tokens": 9160403.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.474123144149781,
|
|
"epoch": 4.023625624716038,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003815937713914455,
|
|
"loss": 4.9722,
|
|
"mean_token_accuracy": 0.19405981451272963,
|
|
"num_tokens": 9171026.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.4114744663238525,
|
|
"epoch": 4.028169014084507,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00038128238931713743,
|
|
"loss": 4.9565,
|
|
"mean_token_accuracy": 0.19117813408374787,
|
|
"num_tokens": 9181252.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.455380296707153,
|
|
"epoch": 4.032712403452976,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00038097074500159883,
|
|
"loss": 5.0273,
|
|
"mean_token_accuracy": 0.18462603092193602,
|
|
"num_tokens": 9190803.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.445877408981323,
|
|
"epoch": 4.037255792821445,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00038065883921378123,
|
|
"loss": 5.0051,
|
|
"mean_token_accuracy": 0.18780569583177567,
|
|
"num_tokens": 9201077.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.387897109985351,
|
|
"epoch": 4.041799182189914,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003803466727232811,
|
|
"loss": 5.025,
|
|
"mean_token_accuracy": 0.18579367101192473,
|
|
"num_tokens": 9211239.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.400881958007813,
|
|
"epoch": 4.046342571558383,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.0003800342463003383,
|
|
"loss": 4.9846,
|
|
"mean_token_accuracy": 0.19478656500577926,
|
|
"num_tokens": 9222683.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.441873502731323,
|
|
"epoch": 4.050885960926851,
|
|
"grad_norm": 3.8125,
|
|
"learning_rate": 0.00037972156071583386,
|
|
"loss": 4.9435,
|
|
"mean_token_accuracy": 0.19211837649345398,
|
|
"num_tokens": 9232099.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.43688817024231,
|
|
"epoch": 4.05542935029532,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003794086167412884,
|
|
"loss": 4.9936,
|
|
"mean_token_accuracy": 0.19290276169776915,
|
|
"num_tokens": 9242660.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.414752054214477,
|
|
"epoch": 4.059972739663789,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0003790954151488601,
|
|
"loss": 5.0289,
|
|
"mean_token_accuracy": 0.17909463196992875,
|
|
"num_tokens": 9253981.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.402527093887329,
|
|
"epoch": 4.064516129032258,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003787819567113427,
|
|
"loss": 4.9627,
|
|
"mean_token_accuracy": 0.18876929134130477,
|
|
"num_tokens": 9264262.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.364192199707031,
|
|
"epoch": 4.069059518400727,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003784682422021638,
|
|
"loss": 4.9513,
|
|
"mean_token_accuracy": 0.19296810775995255,
|
|
"num_tokens": 9273562.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.382873249053955,
|
|
"epoch": 4.073602907769196,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003781542723953828,
|
|
"loss": 4.9788,
|
|
"mean_token_accuracy": 0.1888687089085579,
|
|
"num_tokens": 9283171.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.376923561096191,
|
|
"epoch": 4.078146297137665,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00037784004806568885,
|
|
"loss": 4.9463,
|
|
"mean_token_accuracy": 0.19253140240907668,
|
|
"num_tokens": 9292624.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.446537971496582,
|
|
"epoch": 4.082689686506134,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00037752556998839934,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.18588905185461044,
|
|
"num_tokens": 9303159.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.399434614181518,
|
|
"epoch": 4.087233075874602,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.0003772108389394576,
|
|
"loss": 4.9809,
|
|
"mean_token_accuracy": 0.1910795971751213,
|
|
"num_tokens": 9314340.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.087233075874602,
|
|
"eval_entropy": 5.309824405714523,
|
|
"eval_loss": 5.789700508117676,
|
|
"eval_mean_token_accuracy": 0.16273059368480083,
|
|
"eval_num_tokens": 9314340.0,
|
|
"eval_runtime": 2.5933,
|
|
"eval_samples_per_second": 1324.573,
|
|
"eval_steps_per_second": 165.813,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.486634063720703,
|
|
"epoch": 4.091776465243071,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00037689585569543137,
|
|
"loss": 5.0616,
|
|
"mean_token_accuracy": 0.1894100159406662,
|
|
"num_tokens": 9325381.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.40195894241333,
|
|
"epoch": 4.09631985461154,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003765806210335103,
|
|
"loss": 5.0318,
|
|
"mean_token_accuracy": 0.18149385005235671,
|
|
"num_tokens": 9336639.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.3298773765563965,
|
|
"epoch": 4.100863243980009,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00037626513573150473,
|
|
"loss": 4.9111,
|
|
"mean_token_accuracy": 0.19296008050441743,
|
|
"num_tokens": 9346794.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.341140031814575,
|
|
"epoch": 4.105406633348478,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00037594940056784315,
|
|
"loss": 4.9863,
|
|
"mean_token_accuracy": 0.19331877529621125,
|
|
"num_tokens": 9356326.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.415403127670288,
|
|
"epoch": 4.109950022716947,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003756334163215708,
|
|
"loss": 4.9809,
|
|
"mean_token_accuracy": 0.19089194834232331,
|
|
"num_tokens": 9366402.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.386852502822876,
|
|
"epoch": 4.114493412085416,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 0.00037531718377234723,
|
|
"loss": 4.9477,
|
|
"mean_token_accuracy": 0.19115636348724366,
|
|
"num_tokens": 9377067.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.416745376586914,
|
|
"epoch": 4.119036801453885,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003750007037004451,
|
|
"loss": 4.9989,
|
|
"mean_token_accuracy": 0.18754029870033265,
|
|
"num_tokens": 9387206.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.408817529678345,
|
|
"epoch": 4.123580190822353,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003746839768867473,
|
|
"loss": 4.9712,
|
|
"mean_token_accuracy": 0.18957843482494355,
|
|
"num_tokens": 9397195.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.3523841381073,
|
|
"epoch": 4.128123580190822,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0003743670041127458,
|
|
"loss": 4.9362,
|
|
"mean_token_accuracy": 0.18932909667491912,
|
|
"num_tokens": 9407611.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.442637300491333,
|
|
"epoch": 4.132666969559291,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003740497861605395,
|
|
"loss": 5.0411,
|
|
"mean_token_accuracy": 0.1857879415154457,
|
|
"num_tokens": 9417897.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.426834297180176,
|
|
"epoch": 4.13721035892776,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000373732323812832,
|
|
"loss": 4.9746,
|
|
"mean_token_accuracy": 0.18273799121379852,
|
|
"num_tokens": 9428896.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.352146768569947,
|
|
"epoch": 4.141753748296229,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00037341461785293025,
|
|
"loss": 4.9094,
|
|
"mean_token_accuracy": 0.19667438566684722,
|
|
"num_tokens": 9438854.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.313916110992432,
|
|
"epoch": 4.146297137664698,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0003730966690647421,
|
|
"loss": 4.9767,
|
|
"mean_token_accuracy": 0.1892751082777977,
|
|
"num_tokens": 9447798.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.3575231552124025,
|
|
"epoch": 4.150840527033167,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00037277847823277444,
|
|
"loss": 4.948,
|
|
"mean_token_accuracy": 0.1953787624835968,
|
|
"num_tokens": 9458930.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.348868656158447,
|
|
"epoch": 4.155383916401636,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00037246004614213164,
|
|
"loss": 4.9246,
|
|
"mean_token_accuracy": 0.19456945806741716,
|
|
"num_tokens": 9469740.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.325779247283935,
|
|
"epoch": 4.159927305770104,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0003721413735785132,
|
|
"loss": 4.935,
|
|
"mean_token_accuracy": 0.19863578379154206,
|
|
"num_tokens": 9479823.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.4029584407806395,
|
|
"epoch": 4.164470695138573,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000371822461328212,
|
|
"loss": 4.9072,
|
|
"mean_token_accuracy": 0.2011960342526436,
|
|
"num_tokens": 9489995.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.389583873748779,
|
|
"epoch": 4.169014084507042,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.00037150331017811233,
|
|
"loss": 4.9911,
|
|
"mean_token_accuracy": 0.19074876755475997,
|
|
"num_tokens": 9501311.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.371024322509766,
|
|
"epoch": 4.173557473875511,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0003711839209156878,
|
|
"loss": 5.0358,
|
|
"mean_token_accuracy": 0.1828678846359253,
|
|
"num_tokens": 9511901.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.41847825050354,
|
|
"epoch": 4.17810086324398,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003708642943289999,
|
|
"loss": 5.0015,
|
|
"mean_token_accuracy": 0.18099919110536575,
|
|
"num_tokens": 9522908.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.468344593048096,
|
|
"epoch": 4.182644252612449,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.0003705444312066953,
|
|
"loss": 5.0827,
|
|
"mean_token_accuracy": 0.1854403421282768,
|
|
"num_tokens": 9533363.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.394726085662842,
|
|
"epoch": 4.187187641980918,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003702243323380045,
|
|
"loss": 4.9268,
|
|
"mean_token_accuracy": 0.1893148422241211,
|
|
"num_tokens": 9543302.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.333201169967651,
|
|
"epoch": 4.191731031349387,
|
|
"grad_norm": 3.3125,
|
|
"learning_rate": 0.00036990399851273965,
|
|
"loss": 4.9766,
|
|
"mean_token_accuracy": 0.18024882227182387,
|
|
"num_tokens": 9553143.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.419759368896484,
|
|
"epoch": 4.196274420717856,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003695834305212925,
|
|
"loss": 4.993,
|
|
"mean_token_accuracy": 0.18760381191968917,
|
|
"num_tokens": 9564236.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.4163806438446045,
|
|
"epoch": 4.200817810086324,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00036926262915463294,
|
|
"loss": 4.9416,
|
|
"mean_token_accuracy": 0.19840674698352814,
|
|
"num_tokens": 9574363.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.371914196014404,
|
|
"epoch": 4.205361199454793,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00036894159520430616,
|
|
"loss": 5.0024,
|
|
"mean_token_accuracy": 0.190459144115448,
|
|
"num_tokens": 9585631.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.36880841255188,
|
|
"epoch": 4.209904588823262,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00036862032946243173,
|
|
"loss": 4.9802,
|
|
"mean_token_accuracy": 0.18715289384126663,
|
|
"num_tokens": 9596826.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.315245771408081,
|
|
"epoch": 4.214447978191731,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00036829883272170087,
|
|
"loss": 4.9296,
|
|
"mean_token_accuracy": 0.1897670403122902,
|
|
"num_tokens": 9608158.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.389384746551514,
|
|
"epoch": 4.2189913675602,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00036797710577537486,
|
|
"loss": 5.0568,
|
|
"mean_token_accuracy": 0.18476884812116623,
|
|
"num_tokens": 9619437.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.440706777572632,
|
|
"epoch": 4.223534756928669,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00036765514941728295,
|
|
"loss": 5.0156,
|
|
"mean_token_accuracy": 0.19564879983663558,
|
|
"num_tokens": 9629913.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.340795469284058,
|
|
"epoch": 4.228078146297138,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003673329644418204,
|
|
"loss": 4.9884,
|
|
"mean_token_accuracy": 0.18982828855514527,
|
|
"num_tokens": 9640033.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.383926868438721,
|
|
"epoch": 4.232621535665607,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00036701055164394664,
|
|
"loss": 5.0494,
|
|
"mean_token_accuracy": 0.18545800298452378,
|
|
"num_tokens": 9651010.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.36355562210083,
|
|
"epoch": 4.237164925034075,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00036668791181918313,
|
|
"loss": 4.9553,
|
|
"mean_token_accuracy": 0.19136697053909302,
|
|
"num_tokens": 9661549.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.371076536178589,
|
|
"epoch": 4.241708314402544,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.0003663650457636117,
|
|
"loss": 4.9884,
|
|
"mean_token_accuracy": 0.18359172940254212,
|
|
"num_tokens": 9671619.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.4365369319915775,
|
|
"epoch": 4.246251703771013,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0003660419542738722,
|
|
"loss": 5.0349,
|
|
"mean_token_accuracy": 0.19231729209423065,
|
|
"num_tokens": 9682582.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.331948757171631,
|
|
"epoch": 4.250795093139482,
|
|
"grad_norm": 2.828125,
|
|
"learning_rate": 0.0003657186381471608,
|
|
"loss": 4.9609,
|
|
"mean_token_accuracy": 0.1934656545519829,
|
|
"num_tokens": 9691997.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.416235256195068,
|
|
"epoch": 4.255338482507951,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00036539509818122765,
|
|
"loss": 5.013,
|
|
"mean_token_accuracy": 0.19295045584440232,
|
|
"num_tokens": 9702105.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.420246601104736,
|
|
"epoch": 4.25988187187642,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003650713351743758,
|
|
"loss": 5.0052,
|
|
"mean_token_accuracy": 0.1947663262486458,
|
|
"num_tokens": 9712347.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.389769506454468,
|
|
"epoch": 4.264425261244889,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0003647473499254581,
|
|
"loss": 5.0042,
|
|
"mean_token_accuracy": 0.1912244439125061,
|
|
"num_tokens": 9723116.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.333562564849854,
|
|
"epoch": 4.268968650613358,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000364423143233876,
|
|
"loss": 4.9374,
|
|
"mean_token_accuracy": 0.19349255561828613,
|
|
"num_tokens": 9732656.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.413692617416382,
|
|
"epoch": 4.273512039981826,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00036409871589957723,
|
|
"loss": 5.0475,
|
|
"mean_token_accuracy": 0.18862596154212952,
|
|
"num_tokens": 9743312.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.34252700805664,
|
|
"epoch": 4.278055429350295,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000363774068723054,
|
|
"loss": 4.9988,
|
|
"mean_token_accuracy": 0.18618683367967606,
|
|
"num_tokens": 9753358.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.366300487518311,
|
|
"epoch": 4.282598818718764,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00036344920250534106,
|
|
"loss": 4.9045,
|
|
"mean_token_accuracy": 0.2023068204522133,
|
|
"num_tokens": 9763472.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.359743118286133,
|
|
"epoch": 4.287142208087233,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0003631241180480133,
|
|
"loss": 4.9232,
|
|
"mean_token_accuracy": 0.1944898396730423,
|
|
"num_tokens": 9773700.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.359337663650512,
|
|
"epoch": 4.291685597455702,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003627988161531845,
|
|
"loss": 4.9424,
|
|
"mean_token_accuracy": 0.1918117567896843,
|
|
"num_tokens": 9784042.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.414016580581665,
|
|
"epoch": 4.296228986824171,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003624732976235046,
|
|
"loss": 5.0774,
|
|
"mean_token_accuracy": 0.1915413796901703,
|
|
"num_tokens": 9794464.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.432794809341431,
|
|
"epoch": 4.30077237619264,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.0003621475632621584,
|
|
"loss": 5.006,
|
|
"mean_token_accuracy": 0.19088054597377777,
|
|
"num_tokens": 9804459.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.345897102355957,
|
|
"epoch": 4.305315765561109,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003618216138728627,
|
|
"loss": 4.9897,
|
|
"mean_token_accuracy": 0.18924088180065154,
|
|
"num_tokens": 9814962.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.423156690597534,
|
|
"epoch": 4.309859154929577,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00036149545025986567,
|
|
"loss": 5.0272,
|
|
"mean_token_accuracy": 0.1889437437057495,
|
|
"num_tokens": 9825184.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.417115545272827,
|
|
"epoch": 4.314402544298046,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00036116907322794335,
|
|
"loss": 5.081,
|
|
"mean_token_accuracy": 0.18567907214164733,
|
|
"num_tokens": 9835184.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.386712312698364,
|
|
"epoch": 4.318945933666515,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00036084248358239857,
|
|
"loss": 4.9871,
|
|
"mean_token_accuracy": 0.18762430995702745,
|
|
"num_tokens": 9845358.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.373134326934815,
|
|
"epoch": 4.323489323034984,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00036051568212905897,
|
|
"loss": 4.9404,
|
|
"mean_token_accuracy": 0.1882898762822151,
|
|
"num_tokens": 9856519.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.379983425140381,
|
|
"epoch": 4.328032712403453,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003601886696742745,
|
|
"loss": 4.9887,
|
|
"mean_token_accuracy": 0.18786398619413375,
|
|
"num_tokens": 9867478.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.335583686828613,
|
|
"epoch": 4.332576101771922,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.000359861447024916,
|
|
"loss": 4.9725,
|
|
"mean_token_accuracy": 0.19376827031373978,
|
|
"num_tokens": 9877709.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.362455177307129,
|
|
"epoch": 4.337119491140391,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00035953401498837285,
|
|
"loss": 4.9372,
|
|
"mean_token_accuracy": 0.1963377982378006,
|
|
"num_tokens": 9888134.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.3191088199615475,
|
|
"epoch": 4.34166288050886,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00035920637437255096,
|
|
"loss": 4.9875,
|
|
"mean_token_accuracy": 0.190143358707428,
|
|
"num_tokens": 9899150.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.402696514129639,
|
|
"epoch": 4.346206269877328,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.000358878525985871,
|
|
"loss": 4.9962,
|
|
"mean_token_accuracy": 0.19074421525001525,
|
|
"num_tokens": 9909599.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.356594514846802,
|
|
"epoch": 4.350749659245797,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003585504706372663,
|
|
"loss": 4.9368,
|
|
"mean_token_accuracy": 0.19795213937759398,
|
|
"num_tokens": 9919607.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.3573136806488035,
|
|
"epoch": 4.355293048614266,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003582222091361808,
|
|
"loss": 5.0007,
|
|
"mean_token_accuracy": 0.1899087592959404,
|
|
"num_tokens": 9930431.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.36541748046875,
|
|
"epoch": 4.359836437982735,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003578937422925671,
|
|
"loss": 4.977,
|
|
"mean_token_accuracy": 0.188054358959198,
|
|
"num_tokens": 9940647.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.352145624160767,
|
|
"epoch": 4.364379827351204,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 0.0003575650709168846,
|
|
"loss": 5.0108,
|
|
"mean_token_accuracy": 0.18627723753452302,
|
|
"num_tokens": 9951110.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.361512517929077,
|
|
"epoch": 4.368923216719673,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00035723619582009715,
|
|
"loss": 4.9855,
|
|
"mean_token_accuracy": 0.18833843767642974,
|
|
"num_tokens": 9961158.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.3800272941589355,
|
|
"epoch": 4.373466606088142,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00035690711781367137,
|
|
"loss": 5.0,
|
|
"mean_token_accuracy": 0.19137757420539855,
|
|
"num_tokens": 9971051.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.31362476348877,
|
|
"epoch": 4.378009995456611,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003565778377095744,
|
|
"loss": 4.9921,
|
|
"mean_token_accuracy": 0.18342406302690506,
|
|
"num_tokens": 9981990.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.392795467376709,
|
|
"epoch": 4.38255338482508,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.0003562483563202724,
|
|
"loss": 4.9468,
|
|
"mean_token_accuracy": 0.19157821536064149,
|
|
"num_tokens": 9991668.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.339076185226441,
|
|
"epoch": 4.387096774193548,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00035591867445872784,
|
|
"loss": 4.8651,
|
|
"mean_token_accuracy": 0.1973013013601303,
|
|
"num_tokens": 10001519.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.290941286087036,
|
|
"epoch": 4.391640163562017,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000355588792938398,
|
|
"loss": 4.927,
|
|
"mean_token_accuracy": 0.19318805187940596,
|
|
"num_tokens": 10010856.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.2846324920654295,
|
|
"epoch": 4.396183552930486,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003552587125732326,
|
|
"loss": 4.9529,
|
|
"mean_token_accuracy": 0.1943074271082878,
|
|
"num_tokens": 10020647.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.293172121047974,
|
|
"epoch": 4.400726942298955,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00035492843417767224,
|
|
"loss": 4.8916,
|
|
"mean_token_accuracy": 0.199043570458889,
|
|
"num_tokens": 10030114.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.36516752243042,
|
|
"epoch": 4.405270331667424,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000354597958566646,
|
|
"loss": 4.9535,
|
|
"mean_token_accuracy": 0.19390641748905182,
|
|
"num_tokens": 10039863.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.306846284866333,
|
|
"epoch": 4.409813721035893,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00035426728655556954,
|
|
"loss": 4.9253,
|
|
"mean_token_accuracy": 0.18597330152988434,
|
|
"num_tokens": 10049111.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.349527215957641,
|
|
"epoch": 4.414357110404362,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003539364189603433,
|
|
"loss": 5.0032,
|
|
"mean_token_accuracy": 0.19109094142913818,
|
|
"num_tokens": 10058762.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.379121780395508,
|
|
"epoch": 4.418900499772831,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00035360535659735006,
|
|
"loss": 4.9936,
|
|
"mean_token_accuracy": 0.188161438703537,
|
|
"num_tokens": 10068587.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.377999639511108,
|
|
"epoch": 4.423443889141299,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003532741002834534,
|
|
"loss": 4.9821,
|
|
"mean_token_accuracy": 0.1890724152326584,
|
|
"num_tokens": 10079386.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.326123046875,
|
|
"epoch": 4.427987278509768,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035294265083599526,
|
|
"loss": 4.9158,
|
|
"mean_token_accuracy": 0.20002813041210174,
|
|
"num_tokens": 10089490.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.323672771453857,
|
|
"epoch": 4.432530667878237,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003526110090727942,
|
|
"loss": 5.0,
|
|
"mean_token_accuracy": 0.19350990951061248,
|
|
"num_tokens": 10100465.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.351464939117432,
|
|
"epoch": 4.437074057246706,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003522791758121434,
|
|
"loss": 4.9345,
|
|
"mean_token_accuracy": 0.20031216293573378,
|
|
"num_tokens": 10110919.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.350898027420044,
|
|
"epoch": 4.441617446615175,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00035194715187280835,
|
|
"loss": 4.9699,
|
|
"mean_token_accuracy": 0.19299251288175584,
|
|
"num_tokens": 10121116.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.470292615890503,
|
|
"epoch": 4.446160835983644,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003516149380740252,
|
|
"loss": 5.101,
|
|
"mean_token_accuracy": 0.18566372543573378,
|
|
"num_tokens": 10132269.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.359087800979614,
|
|
"epoch": 4.450704225352113,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003512825352354983,
|
|
"loss": 4.9694,
|
|
"mean_token_accuracy": 0.19472431242465973,
|
|
"num_tokens": 10142856.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.319437742233276,
|
|
"epoch": 4.455247614720582,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00035094994417739884,
|
|
"loss": 4.9299,
|
|
"mean_token_accuracy": 0.18753089457750322,
|
|
"num_tokens": 10152377.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.382904863357544,
|
|
"epoch": 4.45979100408905,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 0.000350617165720362,
|
|
"loss": 4.9989,
|
|
"mean_token_accuracy": 0.18992173671722412,
|
|
"num_tokens": 10161961.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.393486356735229,
|
|
"epoch": 4.464334393457519,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00035028420068548567,
|
|
"loss": 4.9637,
|
|
"mean_token_accuracy": 0.19053333699703218,
|
|
"num_tokens": 10171624.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.419872379302978,
|
|
"epoch": 4.468877782825988,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0003499510498943279,
|
|
"loss": 5.0708,
|
|
"mean_token_accuracy": 0.18718637973070146,
|
|
"num_tokens": 10182642.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.355088996887207,
|
|
"epoch": 4.473421172194457,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00034961771416890526,
|
|
"loss": 5.0235,
|
|
"mean_token_accuracy": 0.19284038245677948,
|
|
"num_tokens": 10193307.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.4032947540283205,
|
|
"epoch": 4.477964561562926,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003492841943316903,
|
|
"loss": 5.0126,
|
|
"mean_token_accuracy": 0.1874317318201065,
|
|
"num_tokens": 10203506.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.330928468704224,
|
|
"epoch": 4.482507950931395,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003489504912056103,
|
|
"loss": 4.9086,
|
|
"mean_token_accuracy": 0.1925170138478279,
|
|
"num_tokens": 10214266.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.290315866470337,
|
|
"epoch": 4.487051340299864,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003486166056140445,
|
|
"loss": 4.9079,
|
|
"mean_token_accuracy": 0.1968353122472763,
|
|
"num_tokens": 10224953.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.323562431335449,
|
|
"epoch": 4.491594729668333,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00034828253838082223,
|
|
"loss": 4.9639,
|
|
"mean_token_accuracy": 0.1855916365981102,
|
|
"num_tokens": 10235435.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.275987005233764,
|
|
"epoch": 4.496138119036801,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00034794829033022134,
|
|
"loss": 4.9524,
|
|
"mean_token_accuracy": 0.19701596796512605,
|
|
"num_tokens": 10245677.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.272549438476562,
|
|
"epoch": 4.50068150840527,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003476138622869657,
|
|
"loss": 4.9172,
|
|
"mean_token_accuracy": 0.19351920187473298,
|
|
"num_tokens": 10256135.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.3953694820404055,
|
|
"epoch": 4.505224897773739,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003472792550762232,
|
|
"loss": 5.0141,
|
|
"mean_token_accuracy": 0.19104439318180083,
|
|
"num_tokens": 10266465.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.432288694381714,
|
|
"epoch": 4.509768287142208,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003469444695236038,
|
|
"loss": 5.0266,
|
|
"mean_token_accuracy": 0.18884656429290772,
|
|
"num_tokens": 10278123.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.334843683242798,
|
|
"epoch": 4.514311676510677,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00034660950645515765,
|
|
"loss": 4.9883,
|
|
"mean_token_accuracy": 0.19115446656942367,
|
|
"num_tokens": 10287486.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.373358964920044,
|
|
"epoch": 4.518855065879146,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003462743666973728,
|
|
"loss": 4.9566,
|
|
"mean_token_accuracy": 0.1944430500268936,
|
|
"num_tokens": 10298100.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.3807793140411375,
|
|
"epoch": 4.523398455247615,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 0.0003459390510771734,
|
|
"loss": 5.0316,
|
|
"mean_token_accuracy": 0.19140505641698838,
|
|
"num_tokens": 10308549.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.363754463195801,
|
|
"epoch": 4.527941844616084,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003456035604219172,
|
|
"loss": 4.9882,
|
|
"mean_token_accuracy": 0.19088154137134553,
|
|
"num_tokens": 10319030.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.286427211761475,
|
|
"epoch": 4.532485233984552,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003452678955593942,
|
|
"loss": 5.0215,
|
|
"mean_token_accuracy": 0.19178841710090638,
|
|
"num_tokens": 10329613.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.367597246170044,
|
|
"epoch": 4.537028623353021,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000344932057317824,
|
|
"loss": 5.03,
|
|
"mean_token_accuracy": 0.18699211180210112,
|
|
"num_tokens": 10339844.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.413221788406372,
|
|
"epoch": 4.54157201272149,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003445960465258542,
|
|
"loss": 5.0548,
|
|
"mean_token_accuracy": 0.18222031891345977,
|
|
"num_tokens": 10350906.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.54157201272149,
|
|
"eval_entropy": 5.263978087624838,
|
|
"eval_loss": 5.76738977432251,
|
|
"eval_mean_token_accuracy": 0.16468495205044748,
|
|
"eval_num_tokens": 10350906.0,
|
|
"eval_runtime": 2.5875,
|
|
"eval_samples_per_second": 1327.511,
|
|
"eval_steps_per_second": 166.18,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.283872413635254,
|
|
"epoch": 4.546115402089959,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003442598640125578,
|
|
"loss": 4.9864,
|
|
"mean_token_accuracy": 0.18617178052663802,
|
|
"num_tokens": 10361574.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.411957550048828,
|
|
"epoch": 4.550658791458428,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000343923510607432,
|
|
"loss": 5.013,
|
|
"mean_token_accuracy": 0.18984230160713195,
|
|
"num_tokens": 10372714.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.320171785354614,
|
|
"epoch": 4.555202180826897,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.0003435869871403952,
|
|
"loss": 4.9041,
|
|
"mean_token_accuracy": 0.1972029447555542,
|
|
"num_tokens": 10382143.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.299863195419311,
|
|
"epoch": 4.559745570195366,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00034325029444178565,
|
|
"loss": 4.9635,
|
|
"mean_token_accuracy": 0.1890595704317093,
|
|
"num_tokens": 10392040.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.34261155128479,
|
|
"epoch": 4.564288959563835,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00034291343334235915,
|
|
"loss": 5.0197,
|
|
"mean_token_accuracy": 0.1796267405152321,
|
|
"num_tokens": 10403230.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.260759878158569,
|
|
"epoch": 4.568832348932304,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003425764046732869,
|
|
"loss": 4.9347,
|
|
"mean_token_accuracy": 0.19403728246688842,
|
|
"num_tokens": 10413466.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.361887407302857,
|
|
"epoch": 4.573375738300772,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00034223920926615376,
|
|
"loss": 4.9363,
|
|
"mean_token_accuracy": 0.19552510529756545,
|
|
"num_tokens": 10422559.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.327618026733399,
|
|
"epoch": 4.577919127669241,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00034190184795295577,
|
|
"loss": 4.9261,
|
|
"mean_token_accuracy": 0.19957540035247803,
|
|
"num_tokens": 10432544.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.329244375228882,
|
|
"epoch": 4.58246251703771,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00034156432156609844,
|
|
"loss": 5.0099,
|
|
"mean_token_accuracy": 0.19274436235427855,
|
|
"num_tokens": 10442014.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.25601167678833,
|
|
"epoch": 4.587005906406179,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00034122663093839466,
|
|
"loss": 4.8672,
|
|
"mean_token_accuracy": 0.20006236881017686,
|
|
"num_tokens": 10451093.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.31429533958435,
|
|
"epoch": 4.591549295774648,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003408887769030624,
|
|
"loss": 4.8999,
|
|
"mean_token_accuracy": 0.1960371434688568,
|
|
"num_tokens": 10461678.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.292968702316284,
|
|
"epoch": 4.596092685143117,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.000340550760293723,
|
|
"loss": 4.9497,
|
|
"mean_token_accuracy": 0.19498579800128937,
|
|
"num_tokens": 10471607.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.334319114685059,
|
|
"epoch": 4.600636074511586,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00034021258194439876,
|
|
"loss": 4.9928,
|
|
"mean_token_accuracy": 0.18937742710113525,
|
|
"num_tokens": 10482453.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.217832756042481,
|
|
"epoch": 4.605179463880054,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00033987424268951117,
|
|
"loss": 4.8837,
|
|
"mean_token_accuracy": 0.19866590499877929,
|
|
"num_tokens": 10492424.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.295894145965576,
|
|
"epoch": 4.609722853248523,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00033953574336387867,
|
|
"loss": 4.9313,
|
|
"mean_token_accuracy": 0.19271864742040634,
|
|
"num_tokens": 10500958.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.357057285308838,
|
|
"epoch": 4.614266242616992,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0003391970848027148,
|
|
"loss": 5.0057,
|
|
"mean_token_accuracy": 0.1968857914209366,
|
|
"num_tokens": 10510906.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.362762832641602,
|
|
"epoch": 4.618809631985461,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003388582678416257,
|
|
"loss": 4.9279,
|
|
"mean_token_accuracy": 0.19629640132188797,
|
|
"num_tokens": 10521832.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.335441541671753,
|
|
"epoch": 4.62335302135393,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00033851929331660876,
|
|
"loss": 4.9134,
|
|
"mean_token_accuracy": 0.19557331800460814,
|
|
"num_tokens": 10532310.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.297559213638306,
|
|
"epoch": 4.627896410722399,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033818016206404983,
|
|
"loss": 4.9731,
|
|
"mean_token_accuracy": 0.18789515048265457,
|
|
"num_tokens": 10543979.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.433013248443603,
|
|
"epoch": 4.632439800090868,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003378408749207215,
|
|
"loss": 5.0589,
|
|
"mean_token_accuracy": 0.18233964145183562,
|
|
"num_tokens": 10553757.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.38985857963562,
|
|
"epoch": 4.636983189459337,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003375014327237812,
|
|
"loss": 4.9884,
|
|
"mean_token_accuracy": 0.1842988282442093,
|
|
"num_tokens": 10563575.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.376491260528565,
|
|
"epoch": 4.641526578827806,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003371618363107686,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.18659371733665467,
|
|
"num_tokens": 10572828.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.253942489624023,
|
|
"epoch": 4.646069968196274,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00033682208651960444,
|
|
"loss": 4.9814,
|
|
"mean_token_accuracy": 0.1850336715579033,
|
|
"num_tokens": 10582567.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.282211828231811,
|
|
"epoch": 4.650613357564743,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.00033648218418858716,
|
|
"loss": 4.9656,
|
|
"mean_token_accuracy": 0.1979486510157585,
|
|
"num_tokens": 10594046.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.383515405654907,
|
|
"epoch": 4.655156746933212,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0003361421301563922,
|
|
"loss": 4.9399,
|
|
"mean_token_accuracy": 0.19595108032226563,
|
|
"num_tokens": 10604343.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.322145938873291,
|
|
"epoch": 4.659700136301681,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003358019252620691,
|
|
"loss": 5.0234,
|
|
"mean_token_accuracy": 0.18899079859256745,
|
|
"num_tokens": 10615140.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.330113649368286,
|
|
"epoch": 4.66424352567015,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00033546157034503954,
|
|
"loss": 4.9949,
|
|
"mean_token_accuracy": 0.19599897414445877,
|
|
"num_tokens": 10624768.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.328687572479248,
|
|
"epoch": 4.668786915038619,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003351210662450955,
|
|
"loss": 4.9033,
|
|
"mean_token_accuracy": 0.1956720009446144,
|
|
"num_tokens": 10635246.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.277680587768555,
|
|
"epoch": 4.673330304407088,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.0003347804138023968,
|
|
"loss": 4.8964,
|
|
"mean_token_accuracy": 0.1992383375763893,
|
|
"num_tokens": 10645118.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.337792015075683,
|
|
"epoch": 4.677873693775557,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.0003344396138574698,
|
|
"loss": 4.924,
|
|
"mean_token_accuracy": 0.19318317025899887,
|
|
"num_tokens": 10655846.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.37283616065979,
|
|
"epoch": 4.682417083144026,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.0003340986672512042,
|
|
"loss": 5.0198,
|
|
"mean_token_accuracy": 0.1900196611881256,
|
|
"num_tokens": 10665741.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.411409091949463,
|
|
"epoch": 4.686960472512494,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003337575748248521,
|
|
"loss": 4.9736,
|
|
"mean_token_accuracy": 0.18948383033275604,
|
|
"num_tokens": 10676120.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.360564184188843,
|
|
"epoch": 4.691503861880963,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00033341633742002503,
|
|
"loss": 4.9503,
|
|
"mean_token_accuracy": 0.18896364122629167,
|
|
"num_tokens": 10686195.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.296868848800659,
|
|
"epoch": 4.696047251249432,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003330749558786924,
|
|
"loss": 4.9658,
|
|
"mean_token_accuracy": 0.19361142218112945,
|
|
"num_tokens": 10697278.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.364632558822632,
|
|
"epoch": 4.700590640617901,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003327334310431791,
|
|
"loss": 5.0023,
|
|
"mean_token_accuracy": 0.19373507797718048,
|
|
"num_tokens": 10707937.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.342183923721313,
|
|
"epoch": 4.70513402998637,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003323917637561638,
|
|
"loss": 4.8778,
|
|
"mean_token_accuracy": 0.190831957757473,
|
|
"num_tokens": 10717963.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.3353965282440186,
|
|
"epoch": 4.709677419354839,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 0.00033204995486067664,
|
|
"loss": 4.971,
|
|
"mean_token_accuracy": 0.19090821593999863,
|
|
"num_tokens": 10728294.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.415113115310669,
|
|
"epoch": 4.714220808723308,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00033170800520009703,
|
|
"loss": 5.0264,
|
|
"mean_token_accuracy": 0.1850395530462265,
|
|
"num_tokens": 10739617.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.36203932762146,
|
|
"epoch": 4.718764198091776,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003313659156181518,
|
|
"loss": 5.048,
|
|
"mean_token_accuracy": 0.18760942667722702,
|
|
"num_tokens": 10750704.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.34307074546814,
|
|
"epoch": 4.723307587460245,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00033102368695891283,
|
|
"loss": 4.9407,
|
|
"mean_token_accuracy": 0.19550167471170427,
|
|
"num_tokens": 10760106.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.436530685424804,
|
|
"epoch": 4.727850976828714,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00033068132006679554,
|
|
"loss": 5.1117,
|
|
"mean_token_accuracy": 0.18790678530931473,
|
|
"num_tokens": 10770963.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.385390567779541,
|
|
"epoch": 4.732394366197183,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00033033881578655605,
|
|
"loss": 5.0405,
|
|
"mean_token_accuracy": 0.18734891414642335,
|
|
"num_tokens": 10781441.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.431859731674194,
|
|
"epoch": 4.736937755565652,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003299961749632898,
|
|
"loss": 5.0226,
|
|
"mean_token_accuracy": 0.19410906881093978,
|
|
"num_tokens": 10791548.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.350888776779175,
|
|
"epoch": 4.741481144934121,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00032965339844242864,
|
|
"loss": 4.9729,
|
|
"mean_token_accuracy": 0.19307624101638793,
|
|
"num_tokens": 10801525.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.261187410354614,
|
|
"epoch": 4.74602453430259,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003293104870697399,
|
|
"loss": 4.9458,
|
|
"mean_token_accuracy": 0.18961290568113326,
|
|
"num_tokens": 10811907.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.308267593383789,
|
|
"epoch": 4.750567923671059,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00032896744169132305,
|
|
"loss": 4.934,
|
|
"mean_token_accuracy": 0.20096516758203506,
|
|
"num_tokens": 10822996.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.339084339141846,
|
|
"epoch": 4.755111313039528,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003286242631536087,
|
|
"loss": 4.8919,
|
|
"mean_token_accuracy": 0.19203527867794037,
|
|
"num_tokens": 10832561.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.3334156513214115,
|
|
"epoch": 4.759654702407996,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00032828095230335555,
|
|
"loss": 4.9727,
|
|
"mean_token_accuracy": 0.18872866034507751,
|
|
"num_tokens": 10842591.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.310991525650024,
|
|
"epoch": 4.764198091776465,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003279375099876491,
|
|
"loss": 4.9918,
|
|
"mean_token_accuracy": 0.19160071462392808,
|
|
"num_tokens": 10852888.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.376796579360962,
|
|
"epoch": 4.768741481144934,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0003275939370538992,
|
|
"loss": 5.0117,
|
|
"mean_token_accuracy": 0.19392362385988235,
|
|
"num_tokens": 10863792.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.290318584442138,
|
|
"epoch": 4.773284870513403,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.00032725023434983767,
|
|
"loss": 4.886,
|
|
"mean_token_accuracy": 0.19952467679977418,
|
|
"num_tokens": 10873206.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.369866991043091,
|
|
"epoch": 4.777828259881872,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00032690640272351715,
|
|
"loss": 5.0513,
|
|
"mean_token_accuracy": 0.187387515604496,
|
|
"num_tokens": 10883686.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.243828296661377,
|
|
"epoch": 4.782371649250341,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003265624430233076,
|
|
"loss": 4.944,
|
|
"mean_token_accuracy": 0.20173004865646363,
|
|
"num_tokens": 10894416.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.408784484863281,
|
|
"epoch": 4.78691503861881,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003262183560978956,
|
|
"loss": 5.0479,
|
|
"mean_token_accuracy": 0.19413791298866273,
|
|
"num_tokens": 10904405.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.436379051208496,
|
|
"epoch": 4.791458427987278,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 0.0003258741427962814,
|
|
"loss": 4.9606,
|
|
"mean_token_accuracy": 0.1968202605843544,
|
|
"num_tokens": 10915339.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.350435829162597,
|
|
"epoch": 4.796001817355747,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00032552980396777714,
|
|
"loss": 4.9759,
|
|
"mean_token_accuracy": 0.19249845445156097,
|
|
"num_tokens": 10926373.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.346661949157715,
|
|
"epoch": 4.800545206724216,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032518534046200464,
|
|
"loss": 5.0341,
|
|
"mean_token_accuracy": 0.18927804082632066,
|
|
"num_tokens": 10936695.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.447178792953491,
|
|
"epoch": 4.805088596092685,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003248407531288932,
|
|
"loss": 5.0866,
|
|
"mean_token_accuracy": 0.18202512413263322,
|
|
"num_tokens": 10949741.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.380243873596191,
|
|
"epoch": 4.809631985461154,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000324496042818678,
|
|
"loss": 4.994,
|
|
"mean_token_accuracy": 0.19422316700220107,
|
|
"num_tokens": 10960786.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.350778818130493,
|
|
"epoch": 4.814175374829623,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00032415121038189746,
|
|
"loss": 4.9549,
|
|
"mean_token_accuracy": 0.1960028663277626,
|
|
"num_tokens": 10970423.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.330668020248413,
|
|
"epoch": 4.818718764198092,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00032380625666939125,
|
|
"loss": 4.9495,
|
|
"mean_token_accuracy": 0.1955626428127289,
|
|
"num_tokens": 10980196.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.288684272766114,
|
|
"epoch": 4.823262153566561,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00032346118253229834,
|
|
"loss": 4.9828,
|
|
"mean_token_accuracy": 0.19362895786762238,
|
|
"num_tokens": 10989857.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.353769445419312,
|
|
"epoch": 4.82780554293503,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00032311598882205496,
|
|
"loss": 4.9877,
|
|
"mean_token_accuracy": 0.18278684914112092,
|
|
"num_tokens": 11000276.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.356082820892334,
|
|
"epoch": 4.832348932303498,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00032277067639039216,
|
|
"loss": 5.0347,
|
|
"mean_token_accuracy": 0.18239014446735383,
|
|
"num_tokens": 11012502.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.310602807998658,
|
|
"epoch": 4.836892321671967,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.00032242524608933423,
|
|
"loss": 4.9029,
|
|
"mean_token_accuracy": 0.19386888146400452,
|
|
"num_tokens": 11023042.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.323847770690918,
|
|
"epoch": 4.841435711040436,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00032207969877119587,
|
|
"loss": 4.9445,
|
|
"mean_token_accuracy": 0.2011082574725151,
|
|
"num_tokens": 11032259.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.423577356338501,
|
|
"epoch": 4.845979100408905,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00032173403528858093,
|
|
"loss": 5.0822,
|
|
"mean_token_accuracy": 0.190763221681118,
|
|
"num_tokens": 11042324.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.304279232025147,
|
|
"epoch": 4.850522489777374,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003213882564943796,
|
|
"loss": 4.9973,
|
|
"mean_token_accuracy": 0.19040728658437728,
|
|
"num_tokens": 11053519.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.350554895401001,
|
|
"epoch": 4.855065879145843,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003210423632417666,
|
|
"loss": 4.9652,
|
|
"mean_token_accuracy": 0.1956503450870514,
|
|
"num_tokens": 11063413.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.290701198577881,
|
|
"epoch": 4.859609268514312,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003206963563841993,
|
|
"loss": 4.9402,
|
|
"mean_token_accuracy": 0.19059386104345322,
|
|
"num_tokens": 11073232.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.302380609512329,
|
|
"epoch": 4.86415265788278,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00032035023677541515,
|
|
"loss": 4.9062,
|
|
"mean_token_accuracy": 0.18717576414346696,
|
|
"num_tokens": 11083408.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.357814025878906,
|
|
"epoch": 4.868696047251249,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003200040052694299,
|
|
"loss": 5.0258,
|
|
"mean_token_accuracy": 0.18858310878276824,
|
|
"num_tokens": 11094125.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.404850673675537,
|
|
"epoch": 4.873239436619718,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00031965766272053534,
|
|
"loss": 4.9953,
|
|
"mean_token_accuracy": 0.18857405483722686,
|
|
"num_tokens": 11106293.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.290663433074951,
|
|
"epoch": 4.877782825988187,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00031931120998329737,
|
|
"loss": 4.8903,
|
|
"mean_token_accuracy": 0.2038312003016472,
|
|
"num_tokens": 11116485.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.329820775985718,
|
|
"epoch": 4.882326215356656,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003189646479125536,
|
|
"loss": 5.0539,
|
|
"mean_token_accuracy": 0.185985966026783,
|
|
"num_tokens": 11127876.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.397577905654908,
|
|
"epoch": 4.886869604725125,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003186179773634115,
|
|
"loss": 5.0937,
|
|
"mean_token_accuracy": 0.18836888968944548,
|
|
"num_tokens": 11138871.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.392291164398193,
|
|
"epoch": 4.891412994093594,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00031827119919124614,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.1885474741458893,
|
|
"num_tokens": 11148872.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.266137266159058,
|
|
"epoch": 4.895956383462063,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00031792431425169826,
|
|
"loss": 4.865,
|
|
"mean_token_accuracy": 0.2075335755944252,
|
|
"num_tokens": 11159716.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.339373445510864,
|
|
"epoch": 4.900499772830532,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00031757732340067195,
|
|
"loss": 4.9532,
|
|
"mean_token_accuracy": 0.19534672051668167,
|
|
"num_tokens": 11169187.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.359593152999878,
|
|
"epoch": 4.905043162199,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00031723022749433267,
|
|
"loss": 4.9271,
|
|
"mean_token_accuracy": 0.19506461918354034,
|
|
"num_tokens": 11179889.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.397574138641358,
|
|
"epoch": 4.909586551567469,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000316883027389105,
|
|
"loss": 5.0897,
|
|
"mean_token_accuracy": 0.18759266287088394,
|
|
"num_tokens": 11189439.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.378802967071533,
|
|
"epoch": 4.914129940935938,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003165357239416706,
|
|
"loss": 5.0258,
|
|
"mean_token_accuracy": 0.19039431661367417,
|
|
"num_tokens": 11200834.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.318107652664184,
|
|
"epoch": 4.918673330304407,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003161883180089665,
|
|
"loss": 4.9259,
|
|
"mean_token_accuracy": 0.19801055192947387,
|
|
"num_tokens": 11210475.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.311107635498047,
|
|
"epoch": 4.923216719672876,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0003158408104481821,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.18994834572076796,
|
|
"num_tokens": 11220744.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.397562265396118,
|
|
"epoch": 4.927760109041345,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00031549320211675776,
|
|
"loss": 5.0706,
|
|
"mean_token_accuracy": 0.18086918145418168,
|
|
"num_tokens": 11231108.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.364572668075562,
|
|
"epoch": 4.932303498409814,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00031514549387238246,
|
|
"loss": 5.0173,
|
|
"mean_token_accuracy": 0.1937827005982399,
|
|
"num_tokens": 11241160.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.346155500411987,
|
|
"epoch": 4.936846887778283,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003147976865729918,
|
|
"loss": 5.0465,
|
|
"mean_token_accuracy": 0.18490591943264006,
|
|
"num_tokens": 11251711.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.348287916183471,
|
|
"epoch": 4.941390277146752,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00031444978107676567,
|
|
"loss": 5.0007,
|
|
"mean_token_accuracy": 0.19004596620798112,
|
|
"num_tokens": 11263025.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.367938232421875,
|
|
"epoch": 4.94593366651522,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00031410177824212635,
|
|
"loss": 4.9872,
|
|
"mean_token_accuracy": 0.1904970735311508,
|
|
"num_tokens": 11273927.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.3646440505981445,
|
|
"epoch": 4.950477055883689,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003137536789277362,
|
|
"loss": 4.96,
|
|
"mean_token_accuracy": 0.19044254273176192,
|
|
"num_tokens": 11284108.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.36102991104126,
|
|
"epoch": 4.955020445252158,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003134054839924957,
|
|
"loss": 4.8971,
|
|
"mean_token_accuracy": 0.19603829830884933,
|
|
"num_tokens": 11294530.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.262496852874756,
|
|
"epoch": 4.959563834620627,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00031305719429554124,
|
|
"loss": 4.9292,
|
|
"mean_token_accuracy": 0.1915763720870018,
|
|
"num_tokens": 11304803.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.3387470722198485,
|
|
"epoch": 4.964107223989096,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003127088106962429,
|
|
"loss": 5.0303,
|
|
"mean_token_accuracy": 0.19123712480068206,
|
|
"num_tokens": 11314567.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.293166446685791,
|
|
"epoch": 4.968650613357565,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031236033405420284,
|
|
"loss": 4.9787,
|
|
"mean_token_accuracy": 0.19235899299383163,
|
|
"num_tokens": 11324452.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.348763370513916,
|
|
"epoch": 4.973194002726034,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00031201176522925236,
|
|
"loss": 4.9887,
|
|
"mean_token_accuracy": 0.19334562867879868,
|
|
"num_tokens": 11334930.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.323974180221557,
|
|
"epoch": 4.977737392094502,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003116631050814503,
|
|
"loss": 4.8977,
|
|
"mean_token_accuracy": 0.2002630889415741,
|
|
"num_tokens": 11345487.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.33988881111145,
|
|
"epoch": 4.982280781462971,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00031131435447108116,
|
|
"loss": 4.9778,
|
|
"mean_token_accuracy": 0.1899196058511734,
|
|
"num_tokens": 11355150.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.335041332244873,
|
|
"epoch": 4.98682417083144,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003109655142586521,
|
|
"loss": 4.9571,
|
|
"mean_token_accuracy": 0.19761783629655838,
|
|
"num_tokens": 11366642.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.317358779907226,
|
|
"epoch": 4.991367560199909,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00031061658530489195,
|
|
"loss": 4.9889,
|
|
"mean_token_accuracy": 0.18887935876846312,
|
|
"num_tokens": 11377229.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.313145780563355,
|
|
"epoch": 4.995910949568378,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00031026756847074804,
|
|
"loss": 5.0131,
|
|
"mean_token_accuracy": 0.1929445117712021,
|
|
"num_tokens": 11387444.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.995910949568378,
|
|
"eval_entropy": 5.257368088877478,
|
|
"eval_loss": 5.723143100738525,
|
|
"eval_mean_token_accuracy": 0.16668621275660603,
|
|
"eval_num_tokens": 11387444.0,
|
|
"eval_runtime": 2.5795,
|
|
"eval_samples_per_second": 1331.634,
|
|
"eval_steps_per_second": 166.697,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.355203469594319,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 0.0003099184646173849,
|
|
"loss": 4.989,
|
|
"mean_token_accuracy": 0.19893405006991494,
|
|
"num_tokens": 11396985.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.314251232147217,
|
|
"epoch": 5.004543389368469,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00030956927460618126,
|
|
"loss": 4.7898,
|
|
"mean_token_accuracy": 0.1985960438847542,
|
|
"num_tokens": 11408747.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.274362182617187,
|
|
"epoch": 5.009086778736938,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003092199992987291,
|
|
"loss": 4.7218,
|
|
"mean_token_accuracy": 0.20721909254789353,
|
|
"num_tokens": 11418695.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.305945777893067,
|
|
"epoch": 5.013630168105407,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00030887063955683044,
|
|
"loss": 4.7945,
|
|
"mean_token_accuracy": 0.2055928274989128,
|
|
"num_tokens": 11428143.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.342715787887573,
|
|
"epoch": 5.018173557473876,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00030852119624249553,
|
|
"loss": 4.8471,
|
|
"mean_token_accuracy": 0.20211713314056395,
|
|
"num_tokens": 11439610.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.361601734161377,
|
|
"epoch": 5.022716946842344,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003081716702179412,
|
|
"loss": 4.7992,
|
|
"mean_token_accuracy": 0.2032486915588379,
|
|
"num_tokens": 11450035.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.2582355499267575,
|
|
"epoch": 5.027260336210813,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000307822062345588,
|
|
"loss": 4.7665,
|
|
"mean_token_accuracy": 0.21366539150476455,
|
|
"num_tokens": 11460736.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.311851501464844,
|
|
"epoch": 5.031803725579282,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003074723734880588,
|
|
"loss": 4.7613,
|
|
"mean_token_accuracy": 0.20563632994890213,
|
|
"num_tokens": 11471319.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.268526887893676,
|
|
"epoch": 5.036347114947751,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000307122604508176,
|
|
"loss": 4.7547,
|
|
"mean_token_accuracy": 0.20495051443576812,
|
|
"num_tokens": 11481052.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.260550594329834,
|
|
"epoch": 5.04089050431622,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003067727562689598,
|
|
"loss": 4.7731,
|
|
"mean_token_accuracy": 0.20005600154399872,
|
|
"num_tokens": 11491166.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.317758226394654,
|
|
"epoch": 5.045433893684689,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003064228296336259,
|
|
"loss": 4.7449,
|
|
"mean_token_accuracy": 0.2092271104454994,
|
|
"num_tokens": 11501007.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.335279273986816,
|
|
"epoch": 5.049977283053158,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003060728254655836,
|
|
"loss": 4.835,
|
|
"mean_token_accuracy": 0.2002784788608551,
|
|
"num_tokens": 11512197.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.310590219497681,
|
|
"epoch": 5.054520672421627,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00030572274462843327,
|
|
"loss": 4.8057,
|
|
"mean_token_accuracy": 0.19899851232767105,
|
|
"num_tokens": 11522396.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.350581789016724,
|
|
"epoch": 5.059064061790095,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00030537258798596465,
|
|
"loss": 4.8287,
|
|
"mean_token_accuracy": 0.20290527045726775,
|
|
"num_tokens": 11533034.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.379289245605468,
|
|
"epoch": 5.063607451158564,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 0.0003050223564021543,
|
|
"loss": 4.8456,
|
|
"mean_token_accuracy": 0.20130965113639832,
|
|
"num_tokens": 11543281.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.305382204055786,
|
|
"epoch": 5.068150840527033,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003046720507411641,
|
|
"loss": 4.7302,
|
|
"mean_token_accuracy": 0.21038950383663177,
|
|
"num_tokens": 11552905.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.266563129425049,
|
|
"epoch": 5.072694229895502,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0003043216718673383,
|
|
"loss": 4.7813,
|
|
"mean_token_accuracy": 0.20856122076511383,
|
|
"num_tokens": 11562397.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.3473944664001465,
|
|
"epoch": 5.077237619263971,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030397122064520186,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.1952403262257576,
|
|
"num_tokens": 11573375.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.3259439945220945,
|
|
"epoch": 5.08178100863244,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00030362069793945856,
|
|
"loss": 4.7666,
|
|
"mean_token_accuracy": 0.20514967888593674,
|
|
"num_tokens": 11584155.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.327657890319824,
|
|
"epoch": 5.086324398000909,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.0003032701046149881,
|
|
"loss": 4.7675,
|
|
"mean_token_accuracy": 0.20547526180744172,
|
|
"num_tokens": 11593929.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.327135801315308,
|
|
"epoch": 5.090867787369378,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003029194415368448,
|
|
"loss": 4.8116,
|
|
"mean_token_accuracy": 0.20603902041912078,
|
|
"num_tokens": 11604891.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.343165445327759,
|
|
"epoch": 5.095411176737846,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0003025687095702548,
|
|
"loss": 4.8681,
|
|
"mean_token_accuracy": 0.19003528207540513,
|
|
"num_tokens": 11616377.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.222209024429321,
|
|
"epoch": 5.099954566106315,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003022179095806146,
|
|
"loss": 4.8097,
|
|
"mean_token_accuracy": 0.19964264184236527,
|
|
"num_tokens": 11627079.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.303332328796387,
|
|
"epoch": 5.104497955474784,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.000301867042433488,
|
|
"loss": 4.7831,
|
|
"mean_token_accuracy": 0.20562289357185365,
|
|
"num_tokens": 11638014.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.39164171218872,
|
|
"epoch": 5.109041344843253,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030151610899460505,
|
|
"loss": 4.8662,
|
|
"mean_token_accuracy": 0.2063317656517029,
|
|
"num_tokens": 11648759.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.316876077651978,
|
|
"epoch": 5.113584734211722,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00030116511012985916,
|
|
"loss": 4.8052,
|
|
"mean_token_accuracy": 0.20284639447927474,
|
|
"num_tokens": 11659050.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.328123569488525,
|
|
"epoch": 5.118128123580191,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003008140467053048,
|
|
"loss": 4.8405,
|
|
"mean_token_accuracy": 0.1961578443646431,
|
|
"num_tokens": 11670131.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.298899221420288,
|
|
"epoch": 5.12267151294866,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 0.0003004629195871565,
|
|
"loss": 4.7248,
|
|
"mean_token_accuracy": 0.20844633281230926,
|
|
"num_tokens": 11680564.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.322790622711182,
|
|
"epoch": 5.127214902317129,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 0.0003001117296417852,
|
|
"loss": 4.8632,
|
|
"mean_token_accuracy": 0.196859447658062,
|
|
"num_tokens": 11691910.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.264201354980469,
|
|
"epoch": 5.131758291685597,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002997604777357173,
|
|
"loss": 4.8093,
|
|
"mean_token_accuracy": 0.20051469504833222,
|
|
"num_tokens": 11702174.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.381486463546753,
|
|
"epoch": 5.136301681054066,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000299409164735632,
|
|
"loss": 4.9349,
|
|
"mean_token_accuracy": 0.1959015130996704,
|
|
"num_tokens": 11712841.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.371714353561401,
|
|
"epoch": 5.140845070422535,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002990577915083591,
|
|
"loss": 4.8174,
|
|
"mean_token_accuracy": 0.20348872393369674,
|
|
"num_tokens": 11722781.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.274111413955689,
|
|
"epoch": 5.145388459791004,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.00029870635892087716,
|
|
"loss": 4.6998,
|
|
"mean_token_accuracy": 0.21555744111537933,
|
|
"num_tokens": 11732774.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.246600580215454,
|
|
"epoch": 5.149931849159473,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0002983548678403112,
|
|
"loss": 4.8174,
|
|
"mean_token_accuracy": 0.19977861046791076,
|
|
"num_tokens": 11743308.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.289332914352417,
|
|
"epoch": 5.154475238527942,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.0002980033191339305,
|
|
"loss": 4.8781,
|
|
"mean_token_accuracy": 0.19565684497356414,
|
|
"num_tokens": 11754572.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.281759452819824,
|
|
"epoch": 5.159018627896411,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0002976517136691465,
|
|
"loss": 4.8036,
|
|
"mean_token_accuracy": 0.2020352303981781,
|
|
"num_tokens": 11764406.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.300207471847534,
|
|
"epoch": 5.16356201726488,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00029730005231351074,
|
|
"loss": 4.7793,
|
|
"mean_token_accuracy": 0.2002014324069023,
|
|
"num_tokens": 11774121.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.3081433296203615,
|
|
"epoch": 5.168105406633348,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.0002969483359347127,
|
|
"loss": 4.829,
|
|
"mean_token_accuracy": 0.20155582278966905,
|
|
"num_tokens": 11784820.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.274715948104858,
|
|
"epoch": 5.172648796001817,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0002965965654005776,
|
|
"loss": 4.7886,
|
|
"mean_token_accuracy": 0.2085181087255478,
|
|
"num_tokens": 11794891.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.337316179275513,
|
|
"epoch": 5.177192185370286,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002962447415790642,
|
|
"loss": 4.775,
|
|
"mean_token_accuracy": 0.20240263342857362,
|
|
"num_tokens": 11804842.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.3387978076934814,
|
|
"epoch": 5.181735574738755,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00029589286533826276,
|
|
"loss": 4.8285,
|
|
"mean_token_accuracy": 0.1999972939491272,
|
|
"num_tokens": 11815893.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.236787176132202,
|
|
"epoch": 5.186278964107224,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0002955409375463931,
|
|
"loss": 4.7328,
|
|
"mean_token_accuracy": 0.21103683710098267,
|
|
"num_tokens": 11826332.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.312748289108276,
|
|
"epoch": 5.190822353475693,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.0002951889590718019,
|
|
"loss": 4.8626,
|
|
"mean_token_accuracy": 0.19659258276224137,
|
|
"num_tokens": 11836977.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.332038545608521,
|
|
"epoch": 5.195365742844162,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002948369307829612,
|
|
"loss": 4.7727,
|
|
"mean_token_accuracy": 0.2045678064227104,
|
|
"num_tokens": 11846483.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.32895336151123,
|
|
"epoch": 5.199909132212631,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0002944848535484656,
|
|
"loss": 4.8332,
|
|
"mean_token_accuracy": 0.20245538055896758,
|
|
"num_tokens": 11857055.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.221473264694214,
|
|
"epoch": 5.2044525215811,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00029413272823703085,
|
|
"loss": 4.6909,
|
|
"mean_token_accuracy": 0.21462486386299134,
|
|
"num_tokens": 11866929.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.295747184753418,
|
|
"epoch": 5.208995910949568,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0002937805557174911,
|
|
"loss": 4.8318,
|
|
"mean_token_accuracy": 0.19675479382276534,
|
|
"num_tokens": 11875973.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.313238668441772,
|
|
"epoch": 5.213539300318037,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00029342833685879713,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.20764228254556655,
|
|
"num_tokens": 11886273.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.37005877494812,
|
|
"epoch": 5.218082689686506,
|
|
"grad_norm": 3.015625,
|
|
"learning_rate": 0.0002930760725300137,
|
|
"loss": 4.8639,
|
|
"mean_token_accuracy": 0.20370941311120988,
|
|
"num_tokens": 11897462.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.325822877883911,
|
|
"epoch": 5.222626079054975,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00029272376360031833,
|
|
"loss": 4.8248,
|
|
"mean_token_accuracy": 0.19871455579996108,
|
|
"num_tokens": 11908116.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.322547388076782,
|
|
"epoch": 5.227169468423444,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00029237141093899804,
|
|
"loss": 4.8812,
|
|
"mean_token_accuracy": 0.19539507031440734,
|
|
"num_tokens": 11918990.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.348210239410401,
|
|
"epoch": 5.231712857791913,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00029201901541544824,
|
|
"loss": 4.8629,
|
|
"mean_token_accuracy": 0.20139847546815873,
|
|
"num_tokens": 11929201.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.300044107437134,
|
|
"epoch": 5.236256247160382,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0002916665778991696,
|
|
"loss": 4.8455,
|
|
"mean_token_accuracy": 0.19100299179553987,
|
|
"num_tokens": 11939002.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.278189849853516,
|
|
"epoch": 5.240799636528851,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002913140992597668,
|
|
"loss": 4.7143,
|
|
"mean_token_accuracy": 0.20845458954572677,
|
|
"num_tokens": 11949255.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.274648380279541,
|
|
"epoch": 5.245343025897319,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.0002909615803669458,
|
|
"loss": 4.7752,
|
|
"mean_token_accuracy": 0.2118154436349869,
|
|
"num_tokens": 11959108.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.205221223831177,
|
|
"epoch": 5.249886415265788,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0002906090220905119,
|
|
"loss": 4.7105,
|
|
"mean_token_accuracy": 0.2064744457602501,
|
|
"num_tokens": 11969196.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.267990922927856,
|
|
"epoch": 5.254429804634257,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002902564253003677,
|
|
"loss": 4.7472,
|
|
"mean_token_accuracy": 0.20631950497627258,
|
|
"num_tokens": 11979101.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.287361907958984,
|
|
"epoch": 5.258973194002726,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002899037908665106,
|
|
"loss": 4.8116,
|
|
"mean_token_accuracy": 0.20188073217868804,
|
|
"num_tokens": 11988396.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.278177881240845,
|
|
"epoch": 5.263516583371195,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002895511196590312,
|
|
"loss": 4.7971,
|
|
"mean_token_accuracy": 0.20554658025503159,
|
|
"num_tokens": 11998355.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.284485483169556,
|
|
"epoch": 5.268059972739664,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002891984125481104,
|
|
"loss": 4.7811,
|
|
"mean_token_accuracy": 0.20703572034835815,
|
|
"num_tokens": 12008938.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.324875545501709,
|
|
"epoch": 5.272603362108133,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00028884567040401814,
|
|
"loss": 4.8402,
|
|
"mean_token_accuracy": 0.19388228058815002,
|
|
"num_tokens": 12019875.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.280922555923462,
|
|
"epoch": 5.277146751476602,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002884928940971105,
|
|
"loss": 4.8108,
|
|
"mean_token_accuracy": 0.20480692088603974,
|
|
"num_tokens": 12029368.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.316416072845459,
|
|
"epoch": 5.28169014084507,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000288140084497828,
|
|
"loss": 4.7765,
|
|
"mean_token_accuracy": 0.2029802307486534,
|
|
"num_tokens": 12040191.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.287885665893555,
|
|
"epoch": 5.286233530213539,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002877872424766932,
|
|
"loss": 4.7704,
|
|
"mean_token_accuracy": 0.21332027465105058,
|
|
"num_tokens": 12050087.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.2464258670806885,
|
|
"epoch": 5.290776919582008,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002874343689043086,
|
|
"loss": 4.8076,
|
|
"mean_token_accuracy": 0.20980110317468642,
|
|
"num_tokens": 12061014.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.273265743255616,
|
|
"epoch": 5.295320308950477,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000287081464651355,
|
|
"loss": 4.8231,
|
|
"mean_token_accuracy": 0.20693545341491698,
|
|
"num_tokens": 12071055.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.310388612747192,
|
|
"epoch": 5.299863698318946,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002867285305885882,
|
|
"loss": 4.8075,
|
|
"mean_token_accuracy": 0.2070198431611061,
|
|
"num_tokens": 12081323.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.270347976684571,
|
|
"epoch": 5.304407087687415,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00028637556758683816,
|
|
"loss": 4.7567,
|
|
"mean_token_accuracy": 0.2071988210082054,
|
|
"num_tokens": 12091482.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.233462715148926,
|
|
"epoch": 5.308950477055884,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002860225765170058,
|
|
"loss": 4.7219,
|
|
"mean_token_accuracy": 0.20832815021276474,
|
|
"num_tokens": 12101960.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.2578874111175535,
|
|
"epoch": 5.313493866424353,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00028566955825006174,
|
|
"loss": 4.7951,
|
|
"mean_token_accuracy": 0.20612651854753494,
|
|
"num_tokens": 12111654.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.3511645793914795,
|
|
"epoch": 5.318037255792821,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00028531651365704325,
|
|
"loss": 4.8862,
|
|
"mean_token_accuracy": 0.20034325420856475,
|
|
"num_tokens": 12123096.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.274366998672486,
|
|
"epoch": 5.32258064516129,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00028496344360905287,
|
|
"loss": 4.7724,
|
|
"mean_token_accuracy": 0.20729893893003465,
|
|
"num_tokens": 12133064.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.253252077102661,
|
|
"epoch": 5.327124034529759,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00028461034897725587,
|
|
"loss": 4.8096,
|
|
"mean_token_accuracy": 0.20582525581121444,
|
|
"num_tokens": 12144386.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.305161714553833,
|
|
"epoch": 5.331667423898228,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00028425723063287803,
|
|
"loss": 4.8566,
|
|
"mean_token_accuracy": 0.19901687055826187,
|
|
"num_tokens": 12156111.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.389003324508667,
|
|
"epoch": 5.336210813266697,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00028390408944720385,
|
|
"loss": 4.8939,
|
|
"mean_token_accuracy": 0.19477416276931764,
|
|
"num_tokens": 12166462.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.288669586181641,
|
|
"epoch": 5.340754202635166,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002835509262915741,
|
|
"loss": 4.9173,
|
|
"mean_token_accuracy": 0.19453637599945067,
|
|
"num_tokens": 12178635.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.251773977279663,
|
|
"epoch": 5.345297592003635,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00028319774203738375,
|
|
"loss": 4.7794,
|
|
"mean_token_accuracy": 0.20567141026258468,
|
|
"num_tokens": 12188880.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.226381015777588,
|
|
"epoch": 5.349840981372104,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002828445375560798,
|
|
"loss": 4.7625,
|
|
"mean_token_accuracy": 0.2034046858549118,
|
|
"num_tokens": 12199733.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.21810302734375,
|
|
"epoch": 5.354384370740572,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028249131371915927,
|
|
"loss": 4.8313,
|
|
"mean_token_accuracy": 0.20026933699846267,
|
|
"num_tokens": 12209925.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.378061151504516,
|
|
"epoch": 5.358927760109041,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002821380713981668,
|
|
"loss": 4.8746,
|
|
"mean_token_accuracy": 0.1986991986632347,
|
|
"num_tokens": 12219870.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.282768106460571,
|
|
"epoch": 5.36347114947751,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00028178481146469273,
|
|
"loss": 4.7892,
|
|
"mean_token_accuracy": 0.2109663501381874,
|
|
"num_tokens": 12230295.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.324818658828735,
|
|
"epoch": 5.368014538845979,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0002814315347903709,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.20832245647907258,
|
|
"num_tokens": 12239700.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.284470129013061,
|
|
"epoch": 5.372557928214448,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002810782422468763,
|
|
"loss": 4.8746,
|
|
"mean_token_accuracy": 0.20065496116876602,
|
|
"num_tokens": 12249529.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.284618330001831,
|
|
"epoch": 5.377101317582917,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00028072493470592317,
|
|
"loss": 4.7517,
|
|
"mean_token_accuracy": 0.2071136236190796,
|
|
"num_tokens": 12259774.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.2980506896972654,
|
|
"epoch": 5.381644706951386,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00028037161303926265,
|
|
"loss": 4.8048,
|
|
"mean_token_accuracy": 0.2049766957759857,
|
|
"num_tokens": 12270484.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.2718267917633055,
|
|
"epoch": 5.386188096319855,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000280018278118681,
|
|
"loss": 4.7859,
|
|
"mean_token_accuracy": 0.20488373935222626,
|
|
"num_tokens": 12281195.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.249173879623413,
|
|
"epoch": 5.390731485688324,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000279664930815997,
|
|
"loss": 4.7706,
|
|
"mean_token_accuracy": 0.20318662226200104,
|
|
"num_tokens": 12292052.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.240638303756714,
|
|
"epoch": 5.395274875056792,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002793115720030599,
|
|
"loss": 4.8496,
|
|
"mean_token_accuracy": 0.20583048462867737,
|
|
"num_tokens": 12302562.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.297917604446411,
|
|
"epoch": 5.399818264425261,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002789582025517474,
|
|
"loss": 4.8374,
|
|
"mean_token_accuracy": 0.20446386486291884,
|
|
"num_tokens": 12313875.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.316703987121582,
|
|
"epoch": 5.40436165379373,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 0.0002786048233339637,
|
|
"loss": 4.8398,
|
|
"mean_token_accuracy": 0.1948122888803482,
|
|
"num_tokens": 12324500.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.245751905441284,
|
|
"epoch": 5.408905043162199,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00027825143522163675,
|
|
"loss": 4.8199,
|
|
"mean_token_accuracy": 0.20545164346694947,
|
|
"num_tokens": 12335474.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.236779499053955,
|
|
"epoch": 5.413448432530668,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002778980390867165,
|
|
"loss": 4.7712,
|
|
"mean_token_accuracy": 0.20655177980661393,
|
|
"num_tokens": 12346174.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.292743253707886,
|
|
"epoch": 5.417991821899137,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027754463580117306,
|
|
"loss": 4.7953,
|
|
"mean_token_accuracy": 0.21172325164079667,
|
|
"num_tokens": 12356108.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.3042552947998045,
|
|
"epoch": 5.422535211267606,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027719122623699353,
|
|
"loss": 4.8397,
|
|
"mean_token_accuracy": 0.20886390060186386,
|
|
"num_tokens": 12366699.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.366331958770752,
|
|
"epoch": 5.427078600636074,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002768378112661813,
|
|
"loss": 4.8452,
|
|
"mean_token_accuracy": 0.19330077469348908,
|
|
"num_tokens": 12377225.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.3250799655914305,
|
|
"epoch": 5.431621990004543,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002764843917607523,
|
|
"loss": 4.8688,
|
|
"mean_token_accuracy": 0.20934968590736389,
|
|
"num_tokens": 12388936.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.263410758972168,
|
|
"epoch": 5.436165379373012,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002761309685927343,
|
|
"loss": 4.7876,
|
|
"mean_token_accuracy": 0.2011903703212738,
|
|
"num_tokens": 12399130.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.238879728317261,
|
|
"epoch": 5.440708768741481,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002757775426341636,
|
|
"loss": 4.8062,
|
|
"mean_token_accuracy": 0.212798972427845,
|
|
"num_tokens": 12410181.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.300264358520508,
|
|
"epoch": 5.44525215810995,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027542411475708385,
|
|
"loss": 4.8966,
|
|
"mean_token_accuracy": 0.19284610748291015,
|
|
"num_tokens": 12421173.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.390559816360474,
|
|
"epoch": 5.449795547478419,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027507068583354304,
|
|
"loss": 4.895,
|
|
"mean_token_accuracy": 0.18880967199802398,
|
|
"num_tokens": 12433195.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.449795547478419,
|
|
"eval_entropy": 5.2714127185732815,
|
|
"eval_loss": 5.717171669006348,
|
|
"eval_mean_token_accuracy": 0.16813963939284168,
|
|
"eval_num_tokens": 12433195.0,
|
|
"eval_runtime": 2.5865,
|
|
"eval_samples_per_second": 1328.025,
|
|
"eval_steps_per_second": 166.245,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.286482048034668,
|
|
"epoch": 5.454338936846888,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00027471725673559205,
|
|
"loss": 4.8009,
|
|
"mean_token_accuracy": 0.20356534719467162,
|
|
"num_tokens": 12442944.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.243269586563111,
|
|
"epoch": 5.458882326215357,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027436382833528184,
|
|
"loss": 4.7564,
|
|
"mean_token_accuracy": 0.21502138674259186,
|
|
"num_tokens": 12453133.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.287388849258423,
|
|
"epoch": 5.463425715583826,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0002740104015046621,
|
|
"loss": 4.8816,
|
|
"mean_token_accuracy": 0.20068138241767883,
|
|
"num_tokens": 12464606.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.275341939926148,
|
|
"epoch": 5.467969104952294,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00027365697711577815,
|
|
"loss": 4.7948,
|
|
"mean_token_accuracy": 0.19945428073406218,
|
|
"num_tokens": 12474490.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.2260754108428955,
|
|
"epoch": 5.472512494320763,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002733035560406697,
|
|
"loss": 4.7847,
|
|
"mean_token_accuracy": 0.20645896196365357,
|
|
"num_tokens": 12485158.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.23867130279541,
|
|
"epoch": 5.477055883689232,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00027295013915136805,
|
|
"loss": 4.852,
|
|
"mean_token_accuracy": 0.19600539207458495,
|
|
"num_tokens": 12496265.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.25579047203064,
|
|
"epoch": 5.481599273057701,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00027259672731989436,
|
|
"loss": 4.7225,
|
|
"mean_token_accuracy": 0.2093784913420677,
|
|
"num_tokens": 12506719.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.274996328353882,
|
|
"epoch": 5.48614266242617,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000272243321418257,
|
|
"loss": 4.8086,
|
|
"mean_token_accuracy": 0.20485190451145172,
|
|
"num_tokens": 12515990.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.25696816444397,
|
|
"epoch": 5.490686051794639,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002718899223184501,
|
|
"loss": 4.8249,
|
|
"mean_token_accuracy": 0.19955365508794784,
|
|
"num_tokens": 12526557.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.216456413269043,
|
|
"epoch": 5.495229441163108,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.00027153653089245067,
|
|
"loss": 4.7597,
|
|
"mean_token_accuracy": 0.211626099050045,
|
|
"num_tokens": 12536657.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.25298810005188,
|
|
"epoch": 5.499772830531577,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000271183148012217,
|
|
"loss": 4.7749,
|
|
"mean_token_accuracy": 0.2068590596318245,
|
|
"num_tokens": 12546958.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.263179588317871,
|
|
"epoch": 5.504316219900045,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002708297745496861,
|
|
"loss": 4.7904,
|
|
"mean_token_accuracy": 0.20097048878669738,
|
|
"num_tokens": 12556978.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.266425466537475,
|
|
"epoch": 5.508859609268514,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027047641137677213,
|
|
"loss": 4.773,
|
|
"mean_token_accuracy": 0.20670468658208846,
|
|
"num_tokens": 12566954.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.283720397949219,
|
|
"epoch": 5.513402998636983,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002701230593653632,
|
|
"loss": 4.8001,
|
|
"mean_token_accuracy": 0.20849338322877883,
|
|
"num_tokens": 12577129.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.295004272460938,
|
|
"epoch": 5.517946388005452,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0002697697193873207,
|
|
"loss": 4.8077,
|
|
"mean_token_accuracy": 0.1972612902522087,
|
|
"num_tokens": 12586802.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.313274240493774,
|
|
"epoch": 5.522489777373921,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00026941639231447547,
|
|
"loss": 4.8522,
|
|
"mean_token_accuracy": 0.2003218039870262,
|
|
"num_tokens": 12597732.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.272136449813843,
|
|
"epoch": 5.52703316674239,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0002690630790186273,
|
|
"loss": 4.7267,
|
|
"mean_token_accuracy": 0.20478794276714324,
|
|
"num_tokens": 12607185.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.294095849990844,
|
|
"epoch": 5.531576556110859,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026870978037154137,
|
|
"loss": 4.8311,
|
|
"mean_token_accuracy": 0.20364048182964326,
|
|
"num_tokens": 12616821.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.309929895401001,
|
|
"epoch": 5.536119945479328,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002683564972449472,
|
|
"loss": 4.8482,
|
|
"mean_token_accuracy": 0.19911444336175918,
|
|
"num_tokens": 12626698.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.325423049926758,
|
|
"epoch": 5.540663334847796,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002680032305105355,
|
|
"loss": 4.7937,
|
|
"mean_token_accuracy": 0.2105136349797249,
|
|
"num_tokens": 12636588.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.341064691543579,
|
|
"epoch": 5.545206724216265,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002676499810399572,
|
|
"loss": 4.8451,
|
|
"mean_token_accuracy": 0.19782129526138306,
|
|
"num_tokens": 12647196.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.268216371536255,
|
|
"epoch": 5.549750113584734,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00026729674970481985,
|
|
"loss": 4.8246,
|
|
"mean_token_accuracy": 0.20448949933052063,
|
|
"num_tokens": 12658631.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.242969131469726,
|
|
"epoch": 5.554293502953203,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00026694353737668705,
|
|
"loss": 4.8378,
|
|
"mean_token_accuracy": 0.20355365574359893,
|
|
"num_tokens": 12668905.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.203650569915771,
|
|
"epoch": 5.558836892321672,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00026659034492707473,
|
|
"loss": 4.7694,
|
|
"mean_token_accuracy": 0.2106305480003357,
|
|
"num_tokens": 12678270.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.2606648921966555,
|
|
"epoch": 5.563380281690141,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002662371732274504,
|
|
"loss": 4.7966,
|
|
"mean_token_accuracy": 0.20512018501758575,
|
|
"num_tokens": 12688248.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.34873399734497,
|
|
"epoch": 5.56792367105861,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00026588402314923,
|
|
"loss": 4.8945,
|
|
"mean_token_accuracy": 0.19796316921710969,
|
|
"num_tokens": 12698418.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.210602807998657,
|
|
"epoch": 5.572467060427079,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002655308955637763,
|
|
"loss": 4.7295,
|
|
"mean_token_accuracy": 0.20859063267707825,
|
|
"num_tokens": 12708578.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.2473972797393795,
|
|
"epoch": 5.577010449795548,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0002651777913423965,
|
|
"loss": 4.8381,
|
|
"mean_token_accuracy": 0.1976973980665207,
|
|
"num_tokens": 12718899.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.205964231491089,
|
|
"epoch": 5.581553839164016,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002648247113563402,
|
|
"loss": 4.7584,
|
|
"mean_token_accuracy": 0.1992357239127159,
|
|
"num_tokens": 12728342.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.310240459442139,
|
|
"epoch": 5.586097228532485,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00026447165647679703,
|
|
"loss": 4.8123,
|
|
"mean_token_accuracy": 0.21126005351543425,
|
|
"num_tokens": 12739337.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.296712636947632,
|
|
"epoch": 5.590640617900954,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000264118627574895,
|
|
"loss": 4.8068,
|
|
"mean_token_accuracy": 0.20234532356262208,
|
|
"num_tokens": 12749650.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.307099866867065,
|
|
"epoch": 5.595184007269423,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00026376562552169776,
|
|
"loss": 4.8673,
|
|
"mean_token_accuracy": 0.19362275600433348,
|
|
"num_tokens": 12760573.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.358535432815552,
|
|
"epoch": 5.599727396637892,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00026341265118820284,
|
|
"loss": 4.8923,
|
|
"mean_token_accuracy": 0.19580060839653016,
|
|
"num_tokens": 12771070.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.352646446228027,
|
|
"epoch": 5.604270786006361,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00026305970544533903,
|
|
"loss": 4.7996,
|
|
"mean_token_accuracy": 0.20734547525644303,
|
|
"num_tokens": 12781896.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.24811110496521,
|
|
"epoch": 5.60881417537483,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00026270678916396527,
|
|
"loss": 4.8498,
|
|
"mean_token_accuracy": 0.20107945501804353,
|
|
"num_tokens": 12792675.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.210130071640014,
|
|
"epoch": 5.613357564743298,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00026235390321486705,
|
|
"loss": 4.747,
|
|
"mean_token_accuracy": 0.2096434786915779,
|
|
"num_tokens": 12802371.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.292365503311157,
|
|
"epoch": 5.617900954111767,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026200104846875574,
|
|
"loss": 4.8172,
|
|
"mean_token_accuracy": 0.20616569072008134,
|
|
"num_tokens": 12813281.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.260284090042115,
|
|
"epoch": 5.622444343480236,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000261648225796265,
|
|
"loss": 4.8576,
|
|
"mean_token_accuracy": 0.19886635690927507,
|
|
"num_tokens": 12823859.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.278480291366577,
|
|
"epoch": 5.626987732848705,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002612954360679499,
|
|
"loss": 4.739,
|
|
"mean_token_accuracy": 0.20730520337820052,
|
|
"num_tokens": 12833288.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.269114923477173,
|
|
"epoch": 5.631531122217174,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.000260942680154284,
|
|
"loss": 4.7835,
|
|
"mean_token_accuracy": 0.2049938127398491,
|
|
"num_tokens": 12843698.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.209796285629272,
|
|
"epoch": 5.636074511585643,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002605899589256574,
|
|
"loss": 4.7161,
|
|
"mean_token_accuracy": 0.21461195051670073,
|
|
"num_tokens": 12853956.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.297564506530762,
|
|
"epoch": 5.640617900954112,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00026023727325237463,
|
|
"loss": 4.8449,
|
|
"mean_token_accuracy": 0.20040213614702224,
|
|
"num_tokens": 12864680.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.276399087905884,
|
|
"epoch": 5.645161290322581,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0002598846240046527,
|
|
"loss": 4.7222,
|
|
"mean_token_accuracy": 0.20671052485704422,
|
|
"num_tokens": 12873790.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.28139066696167,
|
|
"epoch": 5.64970467969105,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002595320120526183,
|
|
"loss": 4.7957,
|
|
"mean_token_accuracy": 0.2057262986898422,
|
|
"num_tokens": 12884559.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.253578042984008,
|
|
"epoch": 5.654248069059518,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025917943826630666,
|
|
"loss": 4.827,
|
|
"mean_token_accuracy": 0.2039135828614235,
|
|
"num_tokens": 12896176.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.285609006881714,
|
|
"epoch": 5.658791458427987,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002588269035156584,
|
|
"loss": 4.7629,
|
|
"mean_token_accuracy": 0.20175277590751647,
|
|
"num_tokens": 12906381.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.265052223205567,
|
|
"epoch": 5.663334847796456,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00025847440867051796,
|
|
"loss": 4.7729,
|
|
"mean_token_accuracy": 0.20036162286996842,
|
|
"num_tokens": 12915738.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.240390396118164,
|
|
"epoch": 5.667878237164925,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002581219546006314,
|
|
"loss": 4.7897,
|
|
"mean_token_accuracy": 0.2057756155729294,
|
|
"num_tokens": 12926059.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.260892486572265,
|
|
"epoch": 5.672421626533394,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0002577695421756441,
|
|
"loss": 4.8255,
|
|
"mean_token_accuracy": 0.19921373426914216,
|
|
"num_tokens": 12936730.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.308874654769897,
|
|
"epoch": 5.676965015901863,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.0002574171722650986,
|
|
"loss": 4.8313,
|
|
"mean_token_accuracy": 0.1984620586037636,
|
|
"num_tokens": 12945687.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.369046783447265,
|
|
"epoch": 5.681508405270332,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00025706484573843273,
|
|
"loss": 4.8276,
|
|
"mean_token_accuracy": 0.2080907016992569,
|
|
"num_tokens": 12955087.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.252665662765503,
|
|
"epoch": 5.6860517946388,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000256712563464977,
|
|
"loss": 4.7892,
|
|
"mean_token_accuracy": 0.19885947555303574,
|
|
"num_tokens": 12965497.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.337397527694702,
|
|
"epoch": 5.690595184007269,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002563603263139532,
|
|
"loss": 4.957,
|
|
"mean_token_accuracy": 0.191968634724617,
|
|
"num_tokens": 12976520.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.306397104263306,
|
|
"epoch": 5.695138573375738,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002560081351544711,
|
|
"loss": 4.7795,
|
|
"mean_token_accuracy": 0.20773693919181824,
|
|
"num_tokens": 12987271.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.278060722351074,
|
|
"epoch": 5.699681962744207,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025565599085552776,
|
|
"loss": 4.8263,
|
|
"mean_token_accuracy": 0.19958744794130326,
|
|
"num_tokens": 12998170.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.257146215438842,
|
|
"epoch": 5.704225352112676,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00025530389428600394,
|
|
"loss": 4.8211,
|
|
"mean_token_accuracy": 0.20067884773015976,
|
|
"num_tokens": 13008059.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.2592016696929935,
|
|
"epoch": 5.708768741481145,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002549518463146631,
|
|
"loss": 4.7712,
|
|
"mean_token_accuracy": 0.2051811471581459,
|
|
"num_tokens": 13017904.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.316100645065307,
|
|
"epoch": 5.713312130849614,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002545998478101485,
|
|
"loss": 4.8346,
|
|
"mean_token_accuracy": 0.20483010411262512,
|
|
"num_tokens": 13028399.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.281266498565674,
|
|
"epoch": 5.717855520218083,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00025424789964098154,
|
|
"loss": 4.8137,
|
|
"mean_token_accuracy": 0.20255436897277831,
|
|
"num_tokens": 13038420.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.313267469406128,
|
|
"epoch": 5.722398909586552,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002538960026755593,
|
|
"loss": 4.8058,
|
|
"mean_token_accuracy": 0.2029904156923294,
|
|
"num_tokens": 13048514.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.317473840713501,
|
|
"epoch": 5.72694229895502,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00025354415778215255,
|
|
"loss": 4.8993,
|
|
"mean_token_accuracy": 0.20023373812437056,
|
|
"num_tokens": 13059728.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.28926043510437,
|
|
"epoch": 5.731485688323489,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00025319236582890364,
|
|
"loss": 4.8937,
|
|
"mean_token_accuracy": 0.19693628549575806,
|
|
"num_tokens": 13069535.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.235255193710327,
|
|
"epoch": 5.736029077691958,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025284062768382424,
|
|
"loss": 4.7655,
|
|
"mean_token_accuracy": 0.2049593985080719,
|
|
"num_tokens": 13079012.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.320984411239624,
|
|
"epoch": 5.740572467060427,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002524889442147932,
|
|
"loss": 4.7782,
|
|
"mean_token_accuracy": 0.21228712648153306,
|
|
"num_tokens": 13089067.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.314213991165161,
|
|
"epoch": 5.745115856428896,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025213731628955454,
|
|
"loss": 4.8261,
|
|
"mean_token_accuracy": 0.20250903367996215,
|
|
"num_tokens": 13100576.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.321662950515747,
|
|
"epoch": 5.749659245797365,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025178574477571525,
|
|
"loss": 4.9751,
|
|
"mean_token_accuracy": 0.19074491411447525,
|
|
"num_tokens": 13111926.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.232034969329834,
|
|
"epoch": 5.754202635165834,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025143423054074305,
|
|
"loss": 4.7453,
|
|
"mean_token_accuracy": 0.2057296559214592,
|
|
"num_tokens": 13122362.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.238938426971435,
|
|
"epoch": 5.758746024534303,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002510827744519644,
|
|
"loss": 4.7974,
|
|
"mean_token_accuracy": 0.20645043402910232,
|
|
"num_tokens": 13133372.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.2460309028625485,
|
|
"epoch": 5.763289413902772,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00025073137737656226,
|
|
"loss": 4.7606,
|
|
"mean_token_accuracy": 0.2044596865773201,
|
|
"num_tokens": 13143440.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.302167081832886,
|
|
"epoch": 5.76783280327124,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002503800401815739,
|
|
"loss": 4.8191,
|
|
"mean_token_accuracy": 0.20314935892820357,
|
|
"num_tokens": 13153564.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.2795275211334225,
|
|
"epoch": 5.772376192639709,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00025002876373388913,
|
|
"loss": 4.8996,
|
|
"mean_token_accuracy": 0.19302715808153154,
|
|
"num_tokens": 13163679.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.29702615737915,
|
|
"epoch": 5.776919582008178,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024967754890024737,
|
|
"loss": 4.8103,
|
|
"mean_token_accuracy": 0.20786179900169371,
|
|
"num_tokens": 13173930.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.288027000427246,
|
|
"epoch": 5.781462971376647,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002493263965472366,
|
|
"loss": 4.7979,
|
|
"mean_token_accuracy": 0.20700994729995728,
|
|
"num_tokens": 13184805.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.245089054107666,
|
|
"epoch": 5.786006360745116,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002489753075412901,
|
|
"loss": 4.8551,
|
|
"mean_token_accuracy": 0.20153616964817048,
|
|
"num_tokens": 13196229.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.274089622497558,
|
|
"epoch": 5.790549750113585,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00024862428274868533,
|
|
"loss": 4.7342,
|
|
"mean_token_accuracy": 0.2128339245915413,
|
|
"num_tokens": 13205906.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.267417669296265,
|
|
"epoch": 5.795093139482054,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0002482733230355408,
|
|
"loss": 4.7549,
|
|
"mean_token_accuracy": 0.203622704744339,
|
|
"num_tokens": 13215076.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.311213064193725,
|
|
"epoch": 5.799636528850522,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00024792242926781495,
|
|
"loss": 4.8884,
|
|
"mean_token_accuracy": 0.19969306141138077,
|
|
"num_tokens": 13225761.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.328445529937744,
|
|
"epoch": 5.804179918218991,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00024757160231130313,
|
|
"loss": 4.888,
|
|
"mean_token_accuracy": 0.19889026880264282,
|
|
"num_tokens": 13236342.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.256750154495239,
|
|
"epoch": 5.80872330758746,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00024722084303163607,
|
|
"loss": 4.7666,
|
|
"mean_token_accuracy": 0.21052155196666716,
|
|
"num_tokens": 13244989.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.288610029220581,
|
|
"epoch": 5.813266696955929,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002468701522942774,
|
|
"loss": 4.8033,
|
|
"mean_token_accuracy": 0.20309991240501404,
|
|
"num_tokens": 13255495.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.328080177307129,
|
|
"epoch": 5.817810086324398,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024651953096452165,
|
|
"loss": 4.8622,
|
|
"mean_token_accuracy": 0.19808589667081833,
|
|
"num_tokens": 13266189.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.274410152435303,
|
|
"epoch": 5.822353475692867,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024616897990749196,
|
|
"loss": 4.8115,
|
|
"mean_token_accuracy": 0.20253337919712067,
|
|
"num_tokens": 13276194.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.25294189453125,
|
|
"epoch": 5.826896865061336,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002458184999881385,
|
|
"loss": 4.8305,
|
|
"mean_token_accuracy": 0.19954493790864944,
|
|
"num_tokens": 13286188.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.324479961395264,
|
|
"epoch": 5.831440254429805,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00024546809207123534,
|
|
"loss": 4.7745,
|
|
"mean_token_accuracy": 0.20648476779460906,
|
|
"num_tokens": 13295926.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.335936546325684,
|
|
"epoch": 5.835983643798274,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002451177570213794,
|
|
"loss": 4.873,
|
|
"mean_token_accuracy": 0.19872704595327378,
|
|
"num_tokens": 13307373.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.2688384532928465,
|
|
"epoch": 5.840527033166742,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024476749570298746,
|
|
"loss": 4.7947,
|
|
"mean_token_accuracy": 0.1991908609867096,
|
|
"num_tokens": 13316984.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.26218695640564,
|
|
"epoch": 5.845070422535211,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024441730898029456,
|
|
"loss": 4.7662,
|
|
"mean_token_accuracy": 0.20748126804828643,
|
|
"num_tokens": 13327188.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.360145330429077,
|
|
"epoch": 5.84961381190368,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00024406719771735165,
|
|
"loss": 4.8406,
|
|
"mean_token_accuracy": 0.20328497439622878,
|
|
"num_tokens": 13338108.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.231265878677368,
|
|
"epoch": 5.854157201272149,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002437171627780234,
|
|
"loss": 4.7717,
|
|
"mean_token_accuracy": 0.2056691437959671,
|
|
"num_tokens": 13347898.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.266537189483643,
|
|
"epoch": 5.858700590640618,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024336720502598624,
|
|
"loss": 4.8478,
|
|
"mean_token_accuracy": 0.2016562581062317,
|
|
"num_tokens": 13358649.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.273508834838867,
|
|
"epoch": 5.863243980009087,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002430173253247262,
|
|
"loss": 4.8309,
|
|
"mean_token_accuracy": 0.2023981347680092,
|
|
"num_tokens": 13368447.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.264497709274292,
|
|
"epoch": 5.867787369377556,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00024266752453753655,
|
|
"loss": 4.8071,
|
|
"mean_token_accuracy": 0.20245419144630433,
|
|
"num_tokens": 13377809.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.304609775543213,
|
|
"epoch": 5.872330758746024,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024231780352751615,
|
|
"loss": 4.9076,
|
|
"mean_token_accuracy": 0.198614664375782,
|
|
"num_tokens": 13389352.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.279650020599365,
|
|
"epoch": 5.876874148114493,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00024196816315756665,
|
|
"loss": 4.7973,
|
|
"mean_token_accuracy": 0.20462385565042496,
|
|
"num_tokens": 13399929.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.297680282592774,
|
|
"epoch": 5.881417537482962,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002416186042903909,
|
|
"loss": 4.7003,
|
|
"mean_token_accuracy": 0.21073048561811447,
|
|
"num_tokens": 13410096.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.297827482223511,
|
|
"epoch": 5.885960926851431,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00024126912778849074,
|
|
"loss": 4.9049,
|
|
"mean_token_accuracy": 0.20015881955623627,
|
|
"num_tokens": 13420384.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.326197862625122,
|
|
"epoch": 5.8905043162199,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024091973451416472,
|
|
"loss": 4.8103,
|
|
"mean_token_accuracy": 0.200961734354496,
|
|
"num_tokens": 13430887.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.267475032806397,
|
|
"epoch": 5.895047705588369,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 0.00024057042532950597,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.20270390659570695,
|
|
"num_tokens": 13440914.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.264398860931396,
|
|
"epoch": 5.899591094956838,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002402212010964001,
|
|
"loss": 4.8662,
|
|
"mean_token_accuracy": 0.19793045818805693,
|
|
"num_tokens": 13452476.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.284561824798584,
|
|
"epoch": 5.904134484325307,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00023987206267652334,
|
|
"loss": 4.7903,
|
|
"mean_token_accuracy": 0.20297755002975465,
|
|
"num_tokens": 13463779.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.904134484325307,
|
|
"eval_entropy": 5.2187281996704815,
|
|
"eval_loss": 5.691013813018799,
|
|
"eval_mean_token_accuracy": 0.1701197989111723,
|
|
"eval_num_tokens": 13463779.0,
|
|
"eval_runtime": 2.5806,
|
|
"eval_samples_per_second": 1331.092,
|
|
"eval_steps_per_second": 166.629,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.319182586669922,
|
|
"epoch": 5.908677873693776,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023952301093133993,
|
|
"loss": 4.9125,
|
|
"mean_token_accuracy": 0.19798723608255386,
|
|
"num_tokens": 13475467.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.261123323440552,
|
|
"epoch": 5.913221263062244,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023917404672210046,
|
|
"loss": 4.7578,
|
|
"mean_token_accuracy": 0.20501208454370498,
|
|
"num_tokens": 13485484.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.266103839874267,
|
|
"epoch": 5.917764652430713,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002388251709098392,
|
|
"loss": 4.7625,
|
|
"mean_token_accuracy": 0.20922042727470397,
|
|
"num_tokens": 13494941.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.326749086380005,
|
|
"epoch": 5.922308041799182,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023847638435537283,
|
|
"loss": 4.8452,
|
|
"mean_token_accuracy": 0.20820602029561996,
|
|
"num_tokens": 13504762.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.300112676620484,
|
|
"epoch": 5.926851431167651,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023812768791929724,
|
|
"loss": 4.7966,
|
|
"mean_token_accuracy": 0.20499637126922607,
|
|
"num_tokens": 13515436.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.284777498245239,
|
|
"epoch": 5.93139482053612,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023777908246198643,
|
|
"loss": 4.8149,
|
|
"mean_token_accuracy": 0.20660513639450073,
|
|
"num_tokens": 13525211.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.278512859344483,
|
|
"epoch": 5.935938209904589,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023743056884358954,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.20704597383737564,
|
|
"num_tokens": 13534954.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.297599220275879,
|
|
"epoch": 5.940481599273058,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00023708214792402942,
|
|
"loss": 4.825,
|
|
"mean_token_accuracy": 0.2093295380473137,
|
|
"num_tokens": 13545459.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.2701623916625975,
|
|
"epoch": 5.945024988641527,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023673382056299997,
|
|
"loss": 4.8217,
|
|
"mean_token_accuracy": 0.20492678582668306,
|
|
"num_tokens": 13554541.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.303726387023926,
|
|
"epoch": 5.949568378009996,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023638558761996436,
|
|
"loss": 4.8081,
|
|
"mean_token_accuracy": 0.20359983742237092,
|
|
"num_tokens": 13564810.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.272811985015869,
|
|
"epoch": 5.954111767378464,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00023603744995415277,
|
|
"loss": 4.7656,
|
|
"mean_token_accuracy": 0.20557126849889756,
|
|
"num_tokens": 13574756.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.34186863899231,
|
|
"epoch": 5.958655156746933,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023568940842456026,
|
|
"loss": 4.8408,
|
|
"mean_token_accuracy": 0.20334313213825225,
|
|
"num_tokens": 13584346.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.367609167098999,
|
|
"epoch": 5.963198546115402,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002353414638899447,
|
|
"loss": 4.8327,
|
|
"mean_token_accuracy": 0.20456442832946778,
|
|
"num_tokens": 13595713.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.190651988983154,
|
|
"epoch": 5.967741935483871,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00023499361720882468,
|
|
"loss": 4.7095,
|
|
"mean_token_accuracy": 0.2171497941017151,
|
|
"num_tokens": 13605065.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.2678783416748045,
|
|
"epoch": 5.97228532485234,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00023464586923947718,
|
|
"loss": 4.8801,
|
|
"mean_token_accuracy": 0.2037242203950882,
|
|
"num_tokens": 13615028.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.252307319641114,
|
|
"epoch": 5.976828714220809,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00023429822083993596,
|
|
"loss": 4.7897,
|
|
"mean_token_accuracy": 0.20354368686676025,
|
|
"num_tokens": 13625752.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.305717277526855,
|
|
"epoch": 5.981372103589278,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023395067286798872,
|
|
"loss": 4.7642,
|
|
"mean_token_accuracy": 0.21082998365163802,
|
|
"num_tokens": 13635094.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.270577001571655,
|
|
"epoch": 5.985915492957746,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00023360322618117563,
|
|
"loss": 4.7993,
|
|
"mean_token_accuracy": 0.21192242950201035,
|
|
"num_tokens": 13645007.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.225088977813721,
|
|
"epoch": 5.990458882326215,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00023325588163678674,
|
|
"loss": 4.7025,
|
|
"mean_token_accuracy": 0.21045738011598586,
|
|
"num_tokens": 13655101.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.218331289291382,
|
|
"epoch": 5.995002271694684,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00023290864009186036,
|
|
"loss": 4.7732,
|
|
"mean_token_accuracy": 0.20872749537229537,
|
|
"num_tokens": 13665289.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.298364162445068,
|
|
"epoch": 5.999545661063153,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00023256150240318025,
|
|
"loss": 4.8856,
|
|
"mean_token_accuracy": 0.20231215208768843,
|
|
"num_tokens": 13675384.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.339073446061876,
|
|
"epoch": 6.003634711494775,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023221446942727438,
|
|
"loss": 4.7852,
|
|
"mean_token_accuracy": 0.21538383430904812,
|
|
"num_tokens": 13683918.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.313893270492554,
|
|
"epoch": 6.008178100863244,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002318675420204119,
|
|
"loss": 4.7049,
|
|
"mean_token_accuracy": 0.20984520465135575,
|
|
"num_tokens": 13693999.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.333902597427368,
|
|
"epoch": 6.012721490231713,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023152072103860188,
|
|
"loss": 4.7729,
|
|
"mean_token_accuracy": 0.20661194622516632,
|
|
"num_tokens": 13705465.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.270451498031616,
|
|
"epoch": 6.017264879600182,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023117400733759053,
|
|
"loss": 4.6314,
|
|
"mean_token_accuracy": 0.22027163505554198,
|
|
"num_tokens": 13715858.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.307543659210205,
|
|
"epoch": 6.021808268968651,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002308274017728595,
|
|
"loss": 4.6807,
|
|
"mean_token_accuracy": 0.21739052087068558,
|
|
"num_tokens": 13725639.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.173598527908325,
|
|
"epoch": 6.026351658337119,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002304809051996235,
|
|
"loss": 4.5742,
|
|
"mean_token_accuracy": 0.21951912492513656,
|
|
"num_tokens": 13736015.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.108713006973266,
|
|
"epoch": 6.030895047705588,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002301345184728284,
|
|
"loss": 4.5491,
|
|
"mean_token_accuracy": 0.2174864873290062,
|
|
"num_tokens": 13745433.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.303092288970947,
|
|
"epoch": 6.035438437074057,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00022978824244714903,
|
|
"loss": 4.6459,
|
|
"mean_token_accuracy": 0.22149067968130112,
|
|
"num_tokens": 13754970.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.262698841094971,
|
|
"epoch": 6.039981826442526,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00022944207797698707,
|
|
"loss": 4.7386,
|
|
"mean_token_accuracy": 0.20984985381364823,
|
|
"num_tokens": 13766356.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.2533163070678714,
|
|
"epoch": 6.044525215810995,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002290960259164689,
|
|
"loss": 4.7134,
|
|
"mean_token_accuracy": 0.21134636402130128,
|
|
"num_tokens": 13776503.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.285929489135742,
|
|
"epoch": 6.049068605179464,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022875008711944355,
|
|
"loss": 4.7162,
|
|
"mean_token_accuracy": 0.20675788521766664,
|
|
"num_tokens": 13787918.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.204592180252075,
|
|
"epoch": 6.053611994547933,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00022840426243948077,
|
|
"loss": 4.6621,
|
|
"mean_token_accuracy": 0.20984945595264434,
|
|
"num_tokens": 13798305.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.269285821914673,
|
|
"epoch": 6.058155383916402,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022805855272986842,
|
|
"loss": 4.6677,
|
|
"mean_token_accuracy": 0.21044817119836806,
|
|
"num_tokens": 13808867.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.254004669189453,
|
|
"epoch": 6.062698773284871,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00022771295884361088,
|
|
"loss": 4.5561,
|
|
"mean_token_accuracy": 0.22348132729530334,
|
|
"num_tokens": 13820073.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.310808897018433,
|
|
"epoch": 6.067242162653339,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002273674816334268,
|
|
"loss": 4.7141,
|
|
"mean_token_accuracy": 0.21549597084522248,
|
|
"num_tokens": 13830318.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.167801570892334,
|
|
"epoch": 6.071785552021808,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022702212195174664,
|
|
"loss": 4.6344,
|
|
"mean_token_accuracy": 0.21551823168992995,
|
|
"num_tokens": 13840847.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.248471641540528,
|
|
"epoch": 6.076328941390277,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00022667688065071146,
|
|
"loss": 4.652,
|
|
"mean_token_accuracy": 0.21194809228181838,
|
|
"num_tokens": 13850478.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.2485480308532715,
|
|
"epoch": 6.080872330758746,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00022633175858216936,
|
|
"loss": 4.7274,
|
|
"mean_token_accuracy": 0.20632485002279283,
|
|
"num_tokens": 13860749.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.290609550476074,
|
|
"epoch": 6.085415720127215,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022598675659767515,
|
|
"loss": 4.712,
|
|
"mean_token_accuracy": 0.2090269148349762,
|
|
"num_tokens": 13869909.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.281312084197998,
|
|
"epoch": 6.089959109495684,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022564187554848663,
|
|
"loss": 4.6461,
|
|
"mean_token_accuracy": 0.2172894686460495,
|
|
"num_tokens": 13880235.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.159547662734985,
|
|
"epoch": 6.094502498864153,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00022529711628556387,
|
|
"loss": 4.5637,
|
|
"mean_token_accuracy": 0.22269825637340546,
|
|
"num_tokens": 13890324.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.268808269500733,
|
|
"epoch": 6.099045888232622,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022495247965956567,
|
|
"loss": 4.6873,
|
|
"mean_token_accuracy": 0.21398976892232896,
|
|
"num_tokens": 13900250.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.258668041229248,
|
|
"epoch": 6.10358927760109,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002246079665208489,
|
|
"loss": 4.6328,
|
|
"mean_token_accuracy": 0.21042170077562333,
|
|
"num_tokens": 13909736.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.2520112037658695,
|
|
"epoch": 6.108132666969559,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00022426357771946544,
|
|
"loss": 4.6702,
|
|
"mean_token_accuracy": 0.21557483971118926,
|
|
"num_tokens": 13920187.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.232370090484619,
|
|
"epoch": 6.112676056338028,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00022391931410516036,
|
|
"loss": 4.6367,
|
|
"mean_token_accuracy": 0.2071380525827408,
|
|
"num_tokens": 13930858.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.175109767913819,
|
|
"epoch": 6.117219445706497,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022357517652736992,
|
|
"loss": 4.5789,
|
|
"mean_token_accuracy": 0.225791597366333,
|
|
"num_tokens": 13940771.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.264018726348877,
|
|
"epoch": 6.121762835074966,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00022323116583521948,
|
|
"loss": 4.7532,
|
|
"mean_token_accuracy": 0.2039385199546814,
|
|
"num_tokens": 13951312.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.311155366897583,
|
|
"epoch": 6.126306224443435,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022288728287752109,
|
|
"loss": 4.7152,
|
|
"mean_token_accuracy": 0.21024034172296524,
|
|
"num_tokens": 13960862.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.250777673721314,
|
|
"epoch": 6.130849613811904,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002225435285027719,
|
|
"loss": 4.6625,
|
|
"mean_token_accuracy": 0.21636074483394624,
|
|
"num_tokens": 13972228.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.218579959869385,
|
|
"epoch": 6.135393003180373,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00022219990355915148,
|
|
"loss": 4.6224,
|
|
"mean_token_accuracy": 0.22177349030971527,
|
|
"num_tokens": 13982977.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.283806180953979,
|
|
"epoch": 6.139936392548841,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022185640889452036,
|
|
"loss": 4.627,
|
|
"mean_token_accuracy": 0.2218010038137436,
|
|
"num_tokens": 13991645.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.288509654998779,
|
|
"epoch": 6.14447978191731,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022151304535641747,
|
|
"loss": 4.6814,
|
|
"mean_token_accuracy": 0.20856865048408507,
|
|
"num_tokens": 14001313.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.2308062553405765,
|
|
"epoch": 6.149023171285779,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002211698137920582,
|
|
"loss": 4.6774,
|
|
"mean_token_accuracy": 0.21466973423957825,
|
|
"num_tokens": 14011248.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.232339000701904,
|
|
"epoch": 6.153566560654248,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00022082671504833223,
|
|
"loss": 4.6653,
|
|
"mean_token_accuracy": 0.2121764063835144,
|
|
"num_tokens": 14022290.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.23998761177063,
|
|
"epoch": 6.158109950022717,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022048374997180175,
|
|
"loss": 4.6568,
|
|
"mean_token_accuracy": 0.21196332722902297,
|
|
"num_tokens": 14032899.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.262250757217407,
|
|
"epoch": 6.162653339391186,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022014091940869885,
|
|
"loss": 4.6593,
|
|
"mean_token_accuracy": 0.219775952398777,
|
|
"num_tokens": 14043065.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.234580659866333,
|
|
"epoch": 6.167196728759655,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021979822420492402,
|
|
"loss": 4.7103,
|
|
"mean_token_accuracy": 0.2067109152674675,
|
|
"num_tokens": 14054594.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.374506711959839,
|
|
"epoch": 6.171740118128124,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021945566520604337,
|
|
"loss": 4.7767,
|
|
"mean_token_accuracy": 0.20854289531707765,
|
|
"num_tokens": 14064211.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.2676472663879395,
|
|
"epoch": 6.176283507496592,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002191132432572875,
|
|
"loss": 4.6764,
|
|
"mean_token_accuracy": 0.22211036086082458,
|
|
"num_tokens": 14073946.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.239787435531616,
|
|
"epoch": 6.180826896865061,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021877095920354818,
|
|
"loss": 4.5807,
|
|
"mean_token_accuracy": 0.2224801003932953,
|
|
"num_tokens": 14083303.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.24082989692688,
|
|
"epoch": 6.18537028623353,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00021842881388937764,
|
|
"loss": 4.5956,
|
|
"mean_token_accuracy": 0.22200829088687896,
|
|
"num_tokens": 14093475.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.30450668334961,
|
|
"epoch": 6.189913675601999,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00021808680815898508,
|
|
"loss": 4.7462,
|
|
"mean_token_accuracy": 0.20560940206050873,
|
|
"num_tokens": 14104117.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.165324354171753,
|
|
"epoch": 6.194457064970468,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021774494285623598,
|
|
"loss": 4.6202,
|
|
"mean_token_accuracy": 0.21722320169210435,
|
|
"num_tokens": 14115480.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.173626947402954,
|
|
"epoch": 6.199000454338937,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002174032188246487,
|
|
"loss": 4.6178,
|
|
"mean_token_accuracy": 0.21470047831535338,
|
|
"num_tokens": 14125827.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.210063123703003,
|
|
"epoch": 6.203543843707406,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.00021706163690739344,
|
|
"loss": 4.6693,
|
|
"mean_token_accuracy": 0.22398434579372406,
|
|
"num_tokens": 14136424.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.3059875011444095,
|
|
"epoch": 6.208087233075875,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021672019794728963,
|
|
"loss": 4.7468,
|
|
"mean_token_accuracy": 0.20568601191043853,
|
|
"num_tokens": 14147478.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.3313232421875,
|
|
"epoch": 6.212630622444343,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00021637890278680406,
|
|
"loss": 4.739,
|
|
"mean_token_accuracy": 0.21284224689006806,
|
|
"num_tokens": 14159044.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.245356130599975,
|
|
"epoch": 6.217174011812812,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00021603775226804845,
|
|
"loss": 4.6901,
|
|
"mean_token_accuracy": 0.21195679754018784,
|
|
"num_tokens": 14169370.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.248649549484253,
|
|
"epoch": 6.221717401181281,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00021569674723277783,
|
|
"loss": 4.6943,
|
|
"mean_token_accuracy": 0.21183323115110397,
|
|
"num_tokens": 14179774.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.245226764678955,
|
|
"epoch": 6.22626079054975,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00021535588852238823,
|
|
"loss": 4.6709,
|
|
"mean_token_accuracy": 0.21890810281038284,
|
|
"num_tokens": 14190182.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.263800096511841,
|
|
"epoch": 6.230804179918219,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021501517697791479,
|
|
"loss": 4.5828,
|
|
"mean_token_accuracy": 0.2222355753183365,
|
|
"num_tokens": 14199505.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.290008211135865,
|
|
"epoch": 6.235347569286688,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002146746134400291,
|
|
"loss": 4.7122,
|
|
"mean_token_accuracy": 0.20517320334911346,
|
|
"num_tokens": 14209677.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.203843832015991,
|
|
"epoch": 6.239890958655157,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021433419874903803,
|
|
"loss": 4.6399,
|
|
"mean_token_accuracy": 0.2134006768465042,
|
|
"num_tokens": 14219356.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.188125324249268,
|
|
"epoch": 6.244434348023626,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00021399393374488097,
|
|
"loss": 4.5858,
|
|
"mean_token_accuracy": 0.22675073146820068,
|
|
"num_tokens": 14227992.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.215114402770996,
|
|
"epoch": 6.248977737392094,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021365381926712797,
|
|
"loss": 4.6411,
|
|
"mean_token_accuracy": 0.2184091776609421,
|
|
"num_tokens": 14238354.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.26951117515564,
|
|
"epoch": 6.253521126760563,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021331385615497756,
|
|
"loss": 4.6934,
|
|
"mean_token_accuracy": 0.21244042217731476,
|
|
"num_tokens": 14249560.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.26614146232605,
|
|
"epoch": 6.258064516129032,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002129740452472552,
|
|
"loss": 4.7096,
|
|
"mean_token_accuracy": 0.21101521998643874,
|
|
"num_tokens": 14259372.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.233666706085205,
|
|
"epoch": 6.262607905497501,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021263438738241026,
|
|
"loss": 4.6195,
|
|
"mean_token_accuracy": 0.21667539030313493,
|
|
"num_tokens": 14270837.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.228752565383911,
|
|
"epoch": 6.26715129486597,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021229488339851498,
|
|
"loss": 4.62,
|
|
"mean_token_accuracy": 0.2218576490879059,
|
|
"num_tokens": 14280138.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.2644312381744385,
|
|
"epoch": 6.271694684234439,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00021195553413326146,
|
|
"loss": 4.761,
|
|
"mean_token_accuracy": 0.20626553893089294,
|
|
"num_tokens": 14290131.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.219017028808594,
|
|
"epoch": 6.276238073602908,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002116163404239605,
|
|
"loss": 4.6784,
|
|
"mean_token_accuracy": 0.21211068332195282,
|
|
"num_tokens": 14299961.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.245612144470215,
|
|
"epoch": 6.280781462971377,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021127730310753866,
|
|
"loss": 4.6245,
|
|
"mean_token_accuracy": 0.22132833898067475,
|
|
"num_tokens": 14310529.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.220850849151612,
|
|
"epoch": 6.285324852339846,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00021093842302053706,
|
|
"loss": 4.5716,
|
|
"mean_token_accuracy": 0.2271314024925232,
|
|
"num_tokens": 14320220.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.322644662857056,
|
|
"epoch": 6.289868241708314,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021059970099910835,
|
|
"loss": 4.7025,
|
|
"mean_token_accuracy": 0.21694063097238542,
|
|
"num_tokens": 14331700.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.291501808166504,
|
|
"epoch": 6.294411631076783,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.00021026113787901568,
|
|
"loss": 4.6553,
|
|
"mean_token_accuracy": 0.2109191581606865,
|
|
"num_tokens": 14341163.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.209377336502075,
|
|
"epoch": 6.298955020445252,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020992273449562966,
|
|
"loss": 4.6456,
|
|
"mean_token_accuracy": 0.2186720162630081,
|
|
"num_tokens": 14351398.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.238816118240356,
|
|
"epoch": 6.303498409813721,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00020958449168392723,
|
|
"loss": 4.6491,
|
|
"mean_token_accuracy": 0.21624014675617217,
|
|
"num_tokens": 14361920.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.228332042694092,
|
|
"epoch": 6.30804179918219,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002092464102784888,
|
|
"loss": 4.6463,
|
|
"mean_token_accuracy": 0.22152708023786544,
|
|
"num_tokens": 14371980.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.253279066085815,
|
|
"epoch": 6.312585188550659,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020890849111349663,
|
|
"loss": 4.6844,
|
|
"mean_token_accuracy": 0.21151918172836304,
|
|
"num_tokens": 14381397.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.245932769775391,
|
|
"epoch": 6.317128577919128,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020857073502273266,
|
|
"loss": 4.7083,
|
|
"mean_token_accuracy": 0.21285642683506012,
|
|
"num_tokens": 14393476.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.237629652023315,
|
|
"epoch": 6.321671967287597,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002082331428395765,
|
|
"loss": 4.6854,
|
|
"mean_token_accuracy": 0.21789849996566774,
|
|
"num_tokens": 14404801.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.255968475341797,
|
|
"epoch": 6.326215356656065,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020789571539700326,
|
|
"loss": 4.6787,
|
|
"mean_token_accuracy": 0.21786294430494307,
|
|
"num_tokens": 14415921.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.2207787990570065,
|
|
"epoch": 6.330758746024534,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020755845352758175,
|
|
"loss": 4.7119,
|
|
"mean_token_accuracy": 0.2130756139755249,
|
|
"num_tokens": 14426455.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.253996896743774,
|
|
"epoch": 6.335302135393003,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 0.0002072213580634718,
|
|
"loss": 4.6825,
|
|
"mean_token_accuracy": 0.21790721565485,
|
|
"num_tokens": 14438720.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.265485048294067,
|
|
"epoch": 6.339845524761472,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002068844298364233,
|
|
"loss": 4.7005,
|
|
"mean_token_accuracy": 0.21332386136054993,
|
|
"num_tokens": 14449644.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.2956037521362305,
|
|
"epoch": 6.344388914129941,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020654766967777288,
|
|
"loss": 4.7244,
|
|
"mean_token_accuracy": 0.20778272897005082,
|
|
"num_tokens": 14459712.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.252825736999512,
|
|
"epoch": 6.34893230349841,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002062110784184431,
|
|
"loss": 4.6571,
|
|
"mean_token_accuracy": 0.21298161298036575,
|
|
"num_tokens": 14469440.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.247277688980103,
|
|
"epoch": 6.353475692866879,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020587465688893904,
|
|
"loss": 4.7507,
|
|
"mean_token_accuracy": 0.20674612373113632,
|
|
"num_tokens": 14480587.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.346015405654907,
|
|
"epoch": 6.358019082235348,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020553840591934765,
|
|
"loss": 4.7357,
|
|
"mean_token_accuracy": 0.21161360293626785,
|
|
"num_tokens": 14490360.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.358019082235348,
|
|
"eval_entropy": 5.224280507065529,
|
|
"eval_loss": 5.686985015869141,
|
|
"eval_mean_token_accuracy": 0.1713664277001869,
|
|
"eval_num_tokens": 14490360.0,
|
|
"eval_runtime": 2.5834,
|
|
"eval_samples_per_second": 1329.628,
|
|
"eval_steps_per_second": 166.445,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.254722309112549,
|
|
"epoch": 6.362562471603816,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002052023263393346,
|
|
"loss": 4.6667,
|
|
"mean_token_accuracy": 0.21602638363838195,
|
|
"num_tokens": 14501135.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.263490295410156,
|
|
"epoch": 6.367105860972285,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020486641897814295,
|
|
"loss": 4.668,
|
|
"mean_token_accuracy": 0.2096094161272049,
|
|
"num_tokens": 14511828.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.242901754379273,
|
|
"epoch": 6.371649250340754,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020453068466459067,
|
|
"loss": 4.6683,
|
|
"mean_token_accuracy": 0.21172192096710205,
|
|
"num_tokens": 14521973.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.280440616607666,
|
|
"epoch": 6.376192639709223,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00020419512422706893,
|
|
"loss": 4.7365,
|
|
"mean_token_accuracy": 0.21045666933059692,
|
|
"num_tokens": 14532515.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.2366979122161865,
|
|
"epoch": 6.380736029077692,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020385973849353957,
|
|
"loss": 4.6537,
|
|
"mean_token_accuracy": 0.21246603429317473,
|
|
"num_tokens": 14542577.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.215287208557129,
|
|
"epoch": 6.385279418446161,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020352452829153369,
|
|
"loss": 4.6054,
|
|
"mean_token_accuracy": 0.2167500615119934,
|
|
"num_tokens": 14552472.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.21115779876709,
|
|
"epoch": 6.38982280781463,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000203189494448149,
|
|
"loss": 4.6365,
|
|
"mean_token_accuracy": 0.21864844858646393,
|
|
"num_tokens": 14562615.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.251078844070435,
|
|
"epoch": 6.394366197183099,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00020285463779004837,
|
|
"loss": 4.6516,
|
|
"mean_token_accuracy": 0.21984951943159103,
|
|
"num_tokens": 14573200.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.312428283691406,
|
|
"epoch": 6.398909586551568,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020251995914345722,
|
|
"loss": 4.6916,
|
|
"mean_token_accuracy": 0.21577997803688048,
|
|
"num_tokens": 14583042.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.298718643188477,
|
|
"epoch": 6.403452975920036,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020218545933416181,
|
|
"loss": 4.6998,
|
|
"mean_token_accuracy": 0.2159578949213028,
|
|
"num_tokens": 14594409.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.27925672531128,
|
|
"epoch": 6.407996365288505,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020185113918750724,
|
|
"loss": 4.7473,
|
|
"mean_token_accuracy": 0.20996573865413665,
|
|
"num_tokens": 14605370.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.282946252822876,
|
|
"epoch": 6.412539754656974,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002015169995283952,
|
|
"loss": 4.6909,
|
|
"mean_token_accuracy": 0.20985879600048066,
|
|
"num_tokens": 14615056.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.388379859924316,
|
|
"epoch": 6.417083144025443,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020118304118128211,
|
|
"loss": 4.763,
|
|
"mean_token_accuracy": 0.20694683492183685,
|
|
"num_tokens": 14624959.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.268580150604248,
|
|
"epoch": 6.421626533393912,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00020084926497017703,
|
|
"loss": 4.6254,
|
|
"mean_token_accuracy": 0.21907588988542556,
|
|
"num_tokens": 14635348.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.220173120498657,
|
|
"epoch": 6.426169922762381,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020051567171863944,
|
|
"loss": 4.6844,
|
|
"mean_token_accuracy": 0.21127890944480895,
|
|
"num_tokens": 14645232.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.2530828475952145,
|
|
"epoch": 6.43071331213085,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0002001822622497777,
|
|
"loss": 4.6528,
|
|
"mean_token_accuracy": 0.22057541757822036,
|
|
"num_tokens": 14656194.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.261120939254761,
|
|
"epoch": 6.435256701499318,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019984903738624632,
|
|
"loss": 4.6848,
|
|
"mean_token_accuracy": 0.2164546713232994,
|
|
"num_tokens": 14667019.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.2882670879364015,
|
|
"epoch": 6.439800090867787,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00019951599795024478,
|
|
"loss": 4.6748,
|
|
"mean_token_accuracy": 0.21532880365848542,
|
|
"num_tokens": 14676339.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.274350166320801,
|
|
"epoch": 6.444343480236256,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019918314476351462,
|
|
"loss": 4.6912,
|
|
"mean_token_accuracy": 0.21541735678911209,
|
|
"num_tokens": 14688206.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.213504219055176,
|
|
"epoch": 6.448886869604725,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019885047864733803,
|
|
"loss": 4.6653,
|
|
"mean_token_accuracy": 0.22355851382017136,
|
|
"num_tokens": 14698397.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.250280714035034,
|
|
"epoch": 6.453430258973194,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019851800042253553,
|
|
"loss": 4.6074,
|
|
"mean_token_accuracy": 0.22305436283349991,
|
|
"num_tokens": 14708140.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.278135061264038,
|
|
"epoch": 6.457973648341663,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001981857109094642,
|
|
"loss": 4.6858,
|
|
"mean_token_accuracy": 0.2197541892528534,
|
|
"num_tokens": 14718710.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.25558910369873,
|
|
"epoch": 6.462517037710132,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019785361092801524,
|
|
"loss": 4.7163,
|
|
"mean_token_accuracy": 0.20874851942062378,
|
|
"num_tokens": 14729143.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.260909605026245,
|
|
"epoch": 6.467060427078601,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019752170129761244,
|
|
"loss": 4.6886,
|
|
"mean_token_accuracy": 0.2161649599671364,
|
|
"num_tokens": 14739386.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.232439804077148,
|
|
"epoch": 6.47160381644707,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001971899828372098,
|
|
"loss": 4.7053,
|
|
"mean_token_accuracy": 0.21201059222221375,
|
|
"num_tokens": 14749209.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.281756591796875,
|
|
"epoch": 6.476147205815538,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019685845636528972,
|
|
"loss": 4.6798,
|
|
"mean_token_accuracy": 0.22088144570589066,
|
|
"num_tokens": 14759351.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.276137590408325,
|
|
"epoch": 6.480690595184007,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019652712269986074,
|
|
"loss": 4.6723,
|
|
"mean_token_accuracy": 0.21214569509029388,
|
|
"num_tokens": 14769069.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.310868215560913,
|
|
"epoch": 6.485233984552476,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019619598265845556,
|
|
"loss": 4.6878,
|
|
"mean_token_accuracy": 0.2185814216732979,
|
|
"num_tokens": 14778563.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.221215057373047,
|
|
"epoch": 6.489777373920945,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00019586503705812969,
|
|
"loss": 4.6556,
|
|
"mean_token_accuracy": 0.2195277214050293,
|
|
"num_tokens": 14788305.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.257626533508301,
|
|
"epoch": 6.494320763289414,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001955342867154582,
|
|
"loss": 4.721,
|
|
"mean_token_accuracy": 0.20737510323524475,
|
|
"num_tokens": 14799463.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.20029730796814,
|
|
"epoch": 6.498864152657883,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00019520373244653484,
|
|
"loss": 4.6279,
|
|
"mean_token_accuracy": 0.2254675269126892,
|
|
"num_tokens": 14809516.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.27101697921753,
|
|
"epoch": 6.503407542026352,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0001948733750669693,
|
|
"loss": 4.7291,
|
|
"mean_token_accuracy": 0.20683671087026595,
|
|
"num_tokens": 14819605.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.246508169174194,
|
|
"epoch": 6.50795093139482,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019454321539188568,
|
|
"loss": 4.6808,
|
|
"mean_token_accuracy": 0.21297200620174409,
|
|
"num_tokens": 14830241.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.274373388290405,
|
|
"epoch": 6.512494320763289,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019421325423592007,
|
|
"loss": 4.7163,
|
|
"mean_token_accuracy": 0.21184756904840468,
|
|
"num_tokens": 14840294.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.319458055496216,
|
|
"epoch": 6.517037710131758,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001938834924132189,
|
|
"loss": 4.7422,
|
|
"mean_token_accuracy": 0.21136158108711242,
|
|
"num_tokens": 14850156.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.22994589805603,
|
|
"epoch": 6.521581099500227,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001935539307374366,
|
|
"loss": 4.6231,
|
|
"mean_token_accuracy": 0.22024643123149873,
|
|
"num_tokens": 14859889.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.293275356292725,
|
|
"epoch": 6.526124488868696,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019322457002173395,
|
|
"loss": 4.6976,
|
|
"mean_token_accuracy": 0.2173983484506607,
|
|
"num_tokens": 14871659.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.254808616638184,
|
|
"epoch": 6.530667878237165,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019289541107877562,
|
|
"loss": 4.6902,
|
|
"mean_token_accuracy": 0.21533323377370833,
|
|
"num_tokens": 14882041.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.240185689926148,
|
|
"epoch": 6.535211267605634,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00019256645472072876,
|
|
"loss": 4.6303,
|
|
"mean_token_accuracy": 0.22474135011434554,
|
|
"num_tokens": 14891709.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.264193916320801,
|
|
"epoch": 6.539754656974103,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019223770175926032,
|
|
"loss": 4.6907,
|
|
"mean_token_accuracy": 0.2157044753432274,
|
|
"num_tokens": 14901898.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.225185203552246,
|
|
"epoch": 6.544298046342572,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019190915300553574,
|
|
"loss": 4.6149,
|
|
"mean_token_accuracy": 0.2208397403359413,
|
|
"num_tokens": 14911910.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.258855295181275,
|
|
"epoch": 6.54884143571104,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001915808092702162,
|
|
"loss": 4.6484,
|
|
"mean_token_accuracy": 0.21097374707460403,
|
|
"num_tokens": 14922055.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.209173393249512,
|
|
"epoch": 6.553384825079509,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001912526713634574,
|
|
"loss": 4.6355,
|
|
"mean_token_accuracy": 0.21914678663015366,
|
|
"num_tokens": 14931841.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.261049890518189,
|
|
"epoch": 6.557928214447978,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001909247400949069,
|
|
"loss": 4.6661,
|
|
"mean_token_accuracy": 0.2152576357126236,
|
|
"num_tokens": 14942356.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.217480659484863,
|
|
"epoch": 6.562471603816447,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00019059701627370263,
|
|
"loss": 4.6512,
|
|
"mean_token_accuracy": 0.212629996240139,
|
|
"num_tokens": 14952552.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.16615138053894,
|
|
"epoch": 6.567014993184916,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019026950070847025,
|
|
"loss": 4.6498,
|
|
"mean_token_accuracy": 0.22111873030662538,
|
|
"num_tokens": 14962728.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.213552379608155,
|
|
"epoch": 6.571558382553385,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018994219420732228,
|
|
"loss": 4.6378,
|
|
"mean_token_accuracy": 0.2224404036998749,
|
|
"num_tokens": 14974071.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.309884977340698,
|
|
"epoch": 6.576101771921854,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018961509757785462,
|
|
"loss": 4.7427,
|
|
"mean_token_accuracy": 0.21085004806518554,
|
|
"num_tokens": 14983567.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.216222524642944,
|
|
"epoch": 6.580645161290323,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018928821162714587,
|
|
"loss": 4.6627,
|
|
"mean_token_accuracy": 0.2165367752313614,
|
|
"num_tokens": 14994172.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.227495098114014,
|
|
"epoch": 6.585188550658792,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018896153716175458,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.2192710742354393,
|
|
"num_tokens": 15004094.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.23339614868164,
|
|
"epoch": 6.58973194002726,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0001886350749877175,
|
|
"loss": 4.6545,
|
|
"mean_token_accuracy": 0.21624967455863953,
|
|
"num_tokens": 15013986.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.26328673362732,
|
|
"epoch": 6.594275329395729,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001883088259105476,
|
|
"loss": 4.6753,
|
|
"mean_token_accuracy": 0.21718561202287673,
|
|
"num_tokens": 15024764.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.283977508544922,
|
|
"epoch": 6.598818718764198,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018798279073523204,
|
|
"loss": 4.6926,
|
|
"mean_token_accuracy": 0.2114184617996216,
|
|
"num_tokens": 15034971.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.2769446849823,
|
|
"epoch": 6.603362108132667,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00018765697026623017,
|
|
"loss": 4.7527,
|
|
"mean_token_accuracy": 0.20878670364618301,
|
|
"num_tokens": 15045750.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.293875360488892,
|
|
"epoch": 6.607905497501136,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001873313653074718,
|
|
"loss": 4.7872,
|
|
"mean_token_accuracy": 0.20239365845918655,
|
|
"num_tokens": 15056840.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.2063071727752686,
|
|
"epoch": 6.612448886869605,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00018700597666235452,
|
|
"loss": 4.6526,
|
|
"mean_token_accuracy": 0.2172047808766365,
|
|
"num_tokens": 15067451.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.247951412200928,
|
|
"epoch": 6.616992276238074,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018668080513374268,
|
|
"loss": 4.6728,
|
|
"mean_token_accuracy": 0.21665652245283126,
|
|
"num_tokens": 15077780.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.250485563278199,
|
|
"epoch": 6.621535665606542,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018635585152396457,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.2036831259727478,
|
|
"num_tokens": 15089100.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.191010189056397,
|
|
"epoch": 6.626079054975011,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000186031116634811,
|
|
"loss": 4.6887,
|
|
"mean_token_accuracy": 0.21216764748096467,
|
|
"num_tokens": 15099420.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.264903593063354,
|
|
"epoch": 6.63062244434348,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018570660126753291,
|
|
"loss": 4.6908,
|
|
"mean_token_accuracy": 0.2093848779797554,
|
|
"num_tokens": 15109884.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.261797475814819,
|
|
"epoch": 6.635165833711949,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001853823062228399,
|
|
"loss": 4.631,
|
|
"mean_token_accuracy": 0.22413872927427292,
|
|
"num_tokens": 15119761.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.226210832595825,
|
|
"epoch": 6.639709223080418,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018505823230089758,
|
|
"loss": 4.6013,
|
|
"mean_token_accuracy": 0.230231973528862,
|
|
"num_tokens": 15129159.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.2506327629089355,
|
|
"epoch": 6.644252612448887,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018473438030132633,
|
|
"loss": 4.612,
|
|
"mean_token_accuracy": 0.21688394397497177,
|
|
"num_tokens": 15138855.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.216858148574829,
|
|
"epoch": 6.648796001817356,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001844107510231985,
|
|
"loss": 4.6539,
|
|
"mean_token_accuracy": 0.21414976119995116,
|
|
"num_tokens": 15149957.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.229018688201904,
|
|
"epoch": 6.653339391185825,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00018408734526503744,
|
|
"loss": 4.678,
|
|
"mean_token_accuracy": 0.2202347531914711,
|
|
"num_tokens": 15159110.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.27342529296875,
|
|
"epoch": 6.657882780554294,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018376416382481442,
|
|
"loss": 4.7362,
|
|
"mean_token_accuracy": 0.2085210233926773,
|
|
"num_tokens": 15169444.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.386235523223877,
|
|
"epoch": 6.662426169922762,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018344120749994787,
|
|
"loss": 4.88,
|
|
"mean_token_accuracy": 0.19211344420909882,
|
|
"num_tokens": 15180246.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.180595016479492,
|
|
"epoch": 6.666969559291231,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018311847708730005,
|
|
"loss": 4.6516,
|
|
"mean_token_accuracy": 0.2150442659854889,
|
|
"num_tokens": 15191578.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.200475835800171,
|
|
"epoch": 6.6715129486597,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001827959733831765,
|
|
"loss": 4.6898,
|
|
"mean_token_accuracy": 0.20946430265903473,
|
|
"num_tokens": 15202380.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.188045072555542,
|
|
"epoch": 6.676056338028169,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00018247369718332274,
|
|
"loss": 4.674,
|
|
"mean_token_accuracy": 0.21930219680070878,
|
|
"num_tokens": 15214073.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.231937456130981,
|
|
"epoch": 6.680599727396638,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018215164928292354,
|
|
"loss": 4.5996,
|
|
"mean_token_accuracy": 0.21992871463298796,
|
|
"num_tokens": 15223655.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.293695497512817,
|
|
"epoch": 6.685143116765107,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00018182983047659984,
|
|
"loss": 4.7813,
|
|
"mean_token_accuracy": 0.20364303439855574,
|
|
"num_tokens": 15235567.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.248773908615112,
|
|
"epoch": 6.689686506133576,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018150824155840778,
|
|
"loss": 4.6622,
|
|
"mean_token_accuracy": 0.21577077656984328,
|
|
"num_tokens": 15246349.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.298004531860352,
|
|
"epoch": 6.694229895502044,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001811868833218358,
|
|
"loss": 4.6976,
|
|
"mean_token_accuracy": 0.22175818532705308,
|
|
"num_tokens": 15256443.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.246001482009888,
|
|
"epoch": 6.698773284870513,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018086575655980376,
|
|
"loss": 4.6857,
|
|
"mean_token_accuracy": 0.21930325031280518,
|
|
"num_tokens": 15267392.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.231525230407715,
|
|
"epoch": 6.703316674238982,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018054486206465952,
|
|
"loss": 4.6132,
|
|
"mean_token_accuracy": 0.22538460046052933,
|
|
"num_tokens": 15277371.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.258160877227783,
|
|
"epoch": 6.707860063607451,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001802242006281788,
|
|
"loss": 4.6227,
|
|
"mean_token_accuracy": 0.22003697901964187,
|
|
"num_tokens": 15287490.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.181690216064453,
|
|
"epoch": 6.71240345297592,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017990377304156165,
|
|
"loss": 4.6387,
|
|
"mean_token_accuracy": 0.21493539065122605,
|
|
"num_tokens": 15298061.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.201219987869263,
|
|
"epoch": 6.716946842344389,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 0.0001795835800954312,
|
|
"loss": 4.6396,
|
|
"mean_token_accuracy": 0.2185338705778122,
|
|
"num_tokens": 15307862.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.171203994750977,
|
|
"epoch": 6.721490231712858,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017926362257983174,
|
|
"loss": 4.6924,
|
|
"mean_token_accuracy": 0.21825126111507415,
|
|
"num_tokens": 15317693.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.2441455841064455,
|
|
"epoch": 6.726033621081327,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0001789439012842269,
|
|
"loss": 4.7375,
|
|
"mean_token_accuracy": 0.21135578453540801,
|
|
"num_tokens": 15329000.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.304090595245361,
|
|
"epoch": 6.730577010449796,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017862441699749687,
|
|
"loss": 4.7155,
|
|
"mean_token_accuracy": 0.22038295418024062,
|
|
"num_tokens": 15339967.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.300751972198486,
|
|
"epoch": 6.735120399818264,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001783051705079375,
|
|
"loss": 4.7136,
|
|
"mean_token_accuracy": 0.2104785069823265,
|
|
"num_tokens": 15350478.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.291823101043701,
|
|
"epoch": 6.739663789186733,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017798616260325767,
|
|
"loss": 4.6775,
|
|
"mean_token_accuracy": 0.21263580471277238,
|
|
"num_tokens": 15360728.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.258435010910034,
|
|
"epoch": 6.744207178555202,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001776673940705778,
|
|
"loss": 4.6973,
|
|
"mean_token_accuracy": 0.21202701479196548,
|
|
"num_tokens": 15370689.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.215921545028687,
|
|
"epoch": 6.748750567923671,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001773488656964276,
|
|
"loss": 4.5866,
|
|
"mean_token_accuracy": 0.22349108010530472,
|
|
"num_tokens": 15380987.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.294268941879272,
|
|
"epoch": 6.75329395729214,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000177030578266744,
|
|
"loss": 4.6926,
|
|
"mean_token_accuracy": 0.2189079150557518,
|
|
"num_tokens": 15391989.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.294797945022583,
|
|
"epoch": 6.757837346660609,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017671253256686963,
|
|
"loss": 4.7247,
|
|
"mean_token_accuracy": 0.2101029247045517,
|
|
"num_tokens": 15402376.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.232749605178833,
|
|
"epoch": 6.762380736029078,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017639472938155075,
|
|
"loss": 4.6593,
|
|
"mean_token_accuracy": 0.21454359591007233,
|
|
"num_tokens": 15413034.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.195667457580567,
|
|
"epoch": 6.766924125397547,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017607716949493514,
|
|
"loss": 4.6229,
|
|
"mean_token_accuracy": 0.21489844024181365,
|
|
"num_tokens": 15423378.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.161777067184448,
|
|
"epoch": 6.771467514766016,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017575985369057008,
|
|
"loss": 4.6189,
|
|
"mean_token_accuracy": 0.21496208161115646,
|
|
"num_tokens": 15434092.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.1847950458526615,
|
|
"epoch": 6.776010904134484,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001754427827514008,
|
|
"loss": 4.6443,
|
|
"mean_token_accuracy": 0.21475446820259095,
|
|
"num_tokens": 15445178.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.275688648223877,
|
|
"epoch": 6.780554293502953,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017512595745976855,
|
|
"loss": 4.6835,
|
|
"mean_token_accuracy": 0.21544518023729325,
|
|
"num_tokens": 15455456.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.230772876739502,
|
|
"epoch": 6.785097682871422,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017480937859740793,
|
|
"loss": 4.6582,
|
|
"mean_token_accuracy": 0.22523362338542938,
|
|
"num_tokens": 15465805.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.273218584060669,
|
|
"epoch": 6.789641072239891,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017449304694544608,
|
|
"loss": 4.7122,
|
|
"mean_token_accuracy": 0.21086454391479492,
|
|
"num_tokens": 15476590.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.224421167373658,
|
|
"epoch": 6.79418446160836,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017417696328439974,
|
|
"loss": 4.6314,
|
|
"mean_token_accuracy": 0.21585592478513718,
|
|
"num_tokens": 15486791.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.25825138092041,
|
|
"epoch": 6.798727850976829,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00017386112839417402,
|
|
"loss": 4.6414,
|
|
"mean_token_accuracy": 0.22312817722558975,
|
|
"num_tokens": 15497116.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.252547264099121,
|
|
"epoch": 6.803271240345298,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017354554305405994,
|
|
"loss": 4.6712,
|
|
"mean_token_accuracy": 0.22069894075393676,
|
|
"num_tokens": 15506731.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.201738977432251,
|
|
"epoch": 6.807814629713766,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001732302080427332,
|
|
"loss": 4.712,
|
|
"mean_token_accuracy": 0.21428730189800263,
|
|
"num_tokens": 15517333.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.188584089279175,
|
|
"epoch": 6.812358019082235,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017291512413825148,
|
|
"loss": 4.6774,
|
|
"mean_token_accuracy": 0.21530684232711791,
|
|
"num_tokens": 15526995.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.812358019082235,
|
|
"eval_entropy": 5.172897092686143,
|
|
"eval_loss": 5.663166522979736,
|
|
"eval_mean_token_accuracy": 0.17367457184334134,
|
|
"eval_num_tokens": 15526995.0,
|
|
"eval_runtime": 2.5894,
|
|
"eval_samples_per_second": 1326.579,
|
|
"eval_steps_per_second": 166.064,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.251057434082031,
|
|
"epoch": 6.816901408450704,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00017260029211805297,
|
|
"loss": 4.6423,
|
|
"mean_token_accuracy": 0.21061441600322722,
|
|
"num_tokens": 15536686.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.284023475646973,
|
|
"epoch": 6.821444797819173,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00017228571275895434,
|
|
"loss": 4.7759,
|
|
"mean_token_accuracy": 0.2050710752606392,
|
|
"num_tokens": 15547565.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.196148490905761,
|
|
"epoch": 6.825988187187642,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017197138683714902,
|
|
"loss": 4.6298,
|
|
"mean_token_accuracy": 0.21200600117444993,
|
|
"num_tokens": 15557962.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.2387810230255125,
|
|
"epoch": 6.830531576556111,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001716573151282049,
|
|
"loss": 4.6493,
|
|
"mean_token_accuracy": 0.21730404645204543,
|
|
"num_tokens": 15567164.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.2139935970306395,
|
|
"epoch": 6.83507496592458,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001713434984070627,
|
|
"loss": 4.637,
|
|
"mean_token_accuracy": 0.214555087685585,
|
|
"num_tokens": 15577917.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.2825695991516115,
|
|
"epoch": 6.839618355293049,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00017102993744803392,
|
|
"loss": 4.7328,
|
|
"mean_token_accuracy": 0.20602056682109832,
|
|
"num_tokens": 15589106.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.258858680725098,
|
|
"epoch": 6.844161744661518,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00017071663302479923,
|
|
"loss": 4.714,
|
|
"mean_token_accuracy": 0.2074183017015457,
|
|
"num_tokens": 15600368.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.250766277313232,
|
|
"epoch": 6.848705134029986,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00017040358591040606,
|
|
"loss": 4.6403,
|
|
"mean_token_accuracy": 0.21749922931194304,
|
|
"num_tokens": 15610671.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.151281070709229,
|
|
"epoch": 6.853248523398455,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000170090796877267,
|
|
"loss": 4.5998,
|
|
"mean_token_accuracy": 0.21988233476877211,
|
|
"num_tokens": 15620403.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.259158039093018,
|
|
"epoch": 6.857791912766924,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016977826669715783,
|
|
"loss": 4.6647,
|
|
"mean_token_accuracy": 0.21959518045186996,
|
|
"num_tokens": 15631260.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.234355545043945,
|
|
"epoch": 6.862335302135393,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000169465996141216,
|
|
"loss": 4.6797,
|
|
"mean_token_accuracy": 0.213654924929142,
|
|
"num_tokens": 15641604.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.2651139259338375,
|
|
"epoch": 6.866878691503862,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001691539859799377,
|
|
"loss": 4.6601,
|
|
"mean_token_accuracy": 0.2158714711666107,
|
|
"num_tokens": 15652469.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.221493482589722,
|
|
"epoch": 6.871422080872331,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00016884223698317722,
|
|
"loss": 4.6844,
|
|
"mean_token_accuracy": 0.21246726363897322,
|
|
"num_tokens": 15662615.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.2297389030456545,
|
|
"epoch": 6.8759654702408,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001685307499201441,
|
|
"loss": 4.6436,
|
|
"mean_token_accuracy": 0.20592823028564453,
|
|
"num_tokens": 15674868.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.235596227645874,
|
|
"epoch": 6.880508859609268,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001682195255594019,
|
|
"loss": 4.7138,
|
|
"mean_token_accuracy": 0.2193004310131073,
|
|
"num_tokens": 15686246.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.240719318389893,
|
|
"epoch": 6.885052248977737,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016790856466886544,
|
|
"loss": 4.6509,
|
|
"mean_token_accuracy": 0.2151799514889717,
|
|
"num_tokens": 15697025.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.238690042495728,
|
|
"epoch": 6.889595638346206,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016759786801580018,
|
|
"loss": 4.7013,
|
|
"mean_token_accuracy": 0.2151371121406555,
|
|
"num_tokens": 15706636.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.208232355117798,
|
|
"epoch": 6.894139027714675,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016728743636681897,
|
|
"loss": 4.6606,
|
|
"mean_token_accuracy": 0.21908408850431443,
|
|
"num_tokens": 15717319.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.21625657081604,
|
|
"epoch": 6.898682417083144,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00016697727048788118,
|
|
"loss": 4.5988,
|
|
"mean_token_accuracy": 0.22070427536964415,
|
|
"num_tokens": 15726546.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.266551208496094,
|
|
"epoch": 6.903225806451613,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016666737114429017,
|
|
"loss": 4.6954,
|
|
"mean_token_accuracy": 0.21175629198551177,
|
|
"num_tokens": 15737176.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.250291919708252,
|
|
"epoch": 6.907769195820082,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001663577391006919,
|
|
"loss": 4.6748,
|
|
"mean_token_accuracy": 0.22392287999391555,
|
|
"num_tokens": 15747422.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.252000570297241,
|
|
"epoch": 6.912312585188551,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016604837512107263,
|
|
"loss": 4.6903,
|
|
"mean_token_accuracy": 0.21330186277627944,
|
|
"num_tokens": 15757591.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.226497507095337,
|
|
"epoch": 6.91685597455702,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016573927996875714,
|
|
"loss": 4.7343,
|
|
"mean_token_accuracy": 0.20969702303409576,
|
|
"num_tokens": 15768375.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.1968639373779295,
|
|
"epoch": 6.921399363925488,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016543045440640697,
|
|
"loss": 4.6073,
|
|
"mean_token_accuracy": 0.22442987859249114,
|
|
"num_tokens": 15778935.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.252734851837158,
|
|
"epoch": 6.925942753293957,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016512189919601865,
|
|
"loss": 4.6374,
|
|
"mean_token_accuracy": 0.22416206449270248,
|
|
"num_tokens": 15788955.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.2405541896820065,
|
|
"epoch": 6.930486142662426,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001648136150989214,
|
|
"loss": 4.6151,
|
|
"mean_token_accuracy": 0.22632551938295364,
|
|
"num_tokens": 15798795.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.201674509048462,
|
|
"epoch": 6.935029532030895,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001645056028757756,
|
|
"loss": 4.6949,
|
|
"mean_token_accuracy": 0.21300850808620453,
|
|
"num_tokens": 15809251.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.184117078781128,
|
|
"epoch": 6.939572921399364,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.00016419786328657066,
|
|
"loss": 4.5989,
|
|
"mean_token_accuracy": 0.22483847737312318,
|
|
"num_tokens": 15819418.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.264193296432495,
|
|
"epoch": 6.944116310767833,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00016389039709062376,
|
|
"loss": 4.6668,
|
|
"mean_token_accuracy": 0.207116562128067,
|
|
"num_tokens": 15829933.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.212735366821289,
|
|
"epoch": 6.948659700136302,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016358320504657677,
|
|
"loss": 4.6826,
|
|
"mean_token_accuracy": 0.21755619943141938,
|
|
"num_tokens": 15841032.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.250994634628296,
|
|
"epoch": 6.95320308950477,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016327628791239576,
|
|
"loss": 4.7016,
|
|
"mean_token_accuracy": 0.20894907116889955,
|
|
"num_tokens": 15851126.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.270348978042603,
|
|
"epoch": 6.957746478873239,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016296964644536816,
|
|
"loss": 4.6838,
|
|
"mean_token_accuracy": 0.21706764698028563,
|
|
"num_tokens": 15861832.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.293090105056763,
|
|
"epoch": 6.962289868241708,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001626632814021015,
|
|
"loss": 4.6697,
|
|
"mean_token_accuracy": 0.2130077674984932,
|
|
"num_tokens": 15872295.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.263749980926514,
|
|
"epoch": 6.966833257610177,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016235719353852079,
|
|
"loss": 4.671,
|
|
"mean_token_accuracy": 0.21949903815984725,
|
|
"num_tokens": 15882453.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.216774415969849,
|
|
"epoch": 6.971376646978646,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00016205138360986753,
|
|
"loss": 4.7068,
|
|
"mean_token_accuracy": 0.2129571706056595,
|
|
"num_tokens": 15891962.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.194906234741211,
|
|
"epoch": 6.975920036347115,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016174585237069738,
|
|
"loss": 4.6013,
|
|
"mean_token_accuracy": 0.22052549719810485,
|
|
"num_tokens": 15902309.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.194825744628906,
|
|
"epoch": 6.980463425715584,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016144060057487824,
|
|
"loss": 4.6269,
|
|
"mean_token_accuracy": 0.2139191746711731,
|
|
"num_tokens": 15912789.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.259652614593506,
|
|
"epoch": 6.985006815084053,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016113562897558849,
|
|
"loss": 4.7114,
|
|
"mean_token_accuracy": 0.21551052778959273,
|
|
"num_tokens": 15922722.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.195772552490235,
|
|
"epoch": 6.989550204452522,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001608309383253154,
|
|
"loss": 4.5903,
|
|
"mean_token_accuracy": 0.22103402018547058,
|
|
"num_tokens": 15933069.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.310867643356323,
|
|
"epoch": 6.99409359382099,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001605265293758528,
|
|
"loss": 4.6863,
|
|
"mean_token_accuracy": 0.21484410315752028,
|
|
"num_tokens": 15942427.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.250721883773804,
|
|
"epoch": 6.998636983189459,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016022240287829947,
|
|
"loss": 4.6695,
|
|
"mean_token_accuracy": 0.21429374665021897,
|
|
"num_tokens": 15952765.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.1640735732184515,
|
|
"epoch": 7.002726033621081,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015991855958305724,
|
|
"loss": 4.6468,
|
|
"mean_token_accuracy": 0.22277632852395376,
|
|
"num_tokens": 15963025.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.20646390914917,
|
|
"epoch": 7.00726942298955,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00015961500023982945,
|
|
"loss": 4.5681,
|
|
"mean_token_accuracy": 0.22533840984106063,
|
|
"num_tokens": 15973425.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.270082902908325,
|
|
"epoch": 7.011812812358019,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015931172559761848,
|
|
"loss": 4.5845,
|
|
"mean_token_accuracy": 0.2288178563117981,
|
|
"num_tokens": 15984064.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.237667369842529,
|
|
"epoch": 7.016356201726488,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001590087364047243,
|
|
"loss": 4.5703,
|
|
"mean_token_accuracy": 0.22651014029979705,
|
|
"num_tokens": 15994512.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.241020154953003,
|
|
"epoch": 7.020899591094957,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015870603340874267,
|
|
"loss": 4.5018,
|
|
"mean_token_accuracy": 0.22735404819250107,
|
|
"num_tokens": 16004140.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.175865554809571,
|
|
"epoch": 7.025442980463426,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015840361735656333,
|
|
"loss": 4.5062,
|
|
"mean_token_accuracy": 0.23281169533729554,
|
|
"num_tokens": 16013710.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.287718868255615,
|
|
"epoch": 7.029986369831895,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015810148899436747,
|
|
"loss": 4.563,
|
|
"mean_token_accuracy": 0.23018514662981032,
|
|
"num_tokens": 16023846.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.211607646942139,
|
|
"epoch": 7.034529759200363,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015779964906762706,
|
|
"loss": 4.5234,
|
|
"mean_token_accuracy": 0.2311800464987755,
|
|
"num_tokens": 16034327.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.225309991836548,
|
|
"epoch": 7.039073148568832,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015749809832110197,
|
|
"loss": 4.57,
|
|
"mean_token_accuracy": 0.22699210792779922,
|
|
"num_tokens": 16045055.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.169312334060669,
|
|
"epoch": 7.043616537937301,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001571968374988389,
|
|
"loss": 4.5556,
|
|
"mean_token_accuracy": 0.22047570198774338,
|
|
"num_tokens": 16055198.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.186239528656006,
|
|
"epoch": 7.04815992730577,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00015689586734416876,
|
|
"loss": 4.5195,
|
|
"mean_token_accuracy": 0.23176031112670897,
|
|
"num_tokens": 16065030.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.244939947128296,
|
|
"epoch": 7.052703316674239,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015659518859970545,
|
|
"loss": 4.597,
|
|
"mean_token_accuracy": 0.22669316530227662,
|
|
"num_tokens": 16075974.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.25231819152832,
|
|
"epoch": 7.057246706042708,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001562948020073441,
|
|
"loss": 4.6079,
|
|
"mean_token_accuracy": 0.225293330848217,
|
|
"num_tokens": 16086835.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.255233287811279,
|
|
"epoch": 7.061790095411177,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00015599470830825868,
|
|
"loss": 4.585,
|
|
"mean_token_accuracy": 0.22374805361032485,
|
|
"num_tokens": 16098116.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.267121934890747,
|
|
"epoch": 7.066333484779646,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015569490824290054,
|
|
"loss": 4.5913,
|
|
"mean_token_accuracy": 0.212683367729187,
|
|
"num_tokens": 16109185.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.206568574905395,
|
|
"epoch": 7.070876874148114,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015539540255099643,
|
|
"loss": 4.5546,
|
|
"mean_token_accuracy": 0.2236502707004547,
|
|
"num_tokens": 16120025.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.220963191986084,
|
|
"epoch": 7.075420263516583,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00015509619197154705,
|
|
"loss": 4.5319,
|
|
"mean_token_accuracy": 0.23020597249269487,
|
|
"num_tokens": 16131145.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.27653956413269,
|
|
"epoch": 7.079963652885052,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00015479727724282476,
|
|
"loss": 4.594,
|
|
"mean_token_accuracy": 0.21907325237989425,
|
|
"num_tokens": 16141063.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.224508666992188,
|
|
"epoch": 7.084507042253521,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00015449865910237184,
|
|
"loss": 4.5443,
|
|
"mean_token_accuracy": 0.22381241172552108,
|
|
"num_tokens": 16151059.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.222726488113404,
|
|
"epoch": 7.08905043162199,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015420033828699882,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.2381626158952713,
|
|
"num_tokens": 16160679.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.2308954238891605,
|
|
"epoch": 7.093593820990459,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001539023155327829,
|
|
"loss": 4.5788,
|
|
"mean_token_accuracy": 0.21982610374689102,
|
|
"num_tokens": 16170269.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.218662738800049,
|
|
"epoch": 7.098137210358928,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015360459157506525,
|
|
"loss": 4.5412,
|
|
"mean_token_accuracy": 0.2233443483710289,
|
|
"num_tokens": 16180496.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.230857229232788,
|
|
"epoch": 7.102680599727397,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015330716714845048,
|
|
"loss": 4.5609,
|
|
"mean_token_accuracy": 0.2203302040696144,
|
|
"num_tokens": 16191043.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.278061866760254,
|
|
"epoch": 7.107223989095866,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001530100429868035,
|
|
"loss": 4.7141,
|
|
"mean_token_accuracy": 0.2107528731226921,
|
|
"num_tokens": 16201551.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.2396149158477785,
|
|
"epoch": 7.111767378464334,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015271321982324892,
|
|
"loss": 4.5908,
|
|
"mean_token_accuracy": 0.21543675363063813,
|
|
"num_tokens": 16212928.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.223546743392944,
|
|
"epoch": 7.116310767832803,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015241669839016818,
|
|
"loss": 4.5494,
|
|
"mean_token_accuracy": 0.22915434688329697,
|
|
"num_tokens": 16223233.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.224034261703491,
|
|
"epoch": 7.120854157201272,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00015212047941919854,
|
|
"loss": 4.6515,
|
|
"mean_token_accuracy": 0.21428634375333785,
|
|
"num_tokens": 16234902.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.229031610488891,
|
|
"epoch": 7.125397546569741,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.00015182456364123086,
|
|
"loss": 4.6012,
|
|
"mean_token_accuracy": 0.22238935977220536,
|
|
"num_tokens": 16244846.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.210952568054199,
|
|
"epoch": 7.12994093593821,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015152895178640788,
|
|
"loss": 4.5616,
|
|
"mean_token_accuracy": 0.22535636872053147,
|
|
"num_tokens": 16255067.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.245619058609009,
|
|
"epoch": 7.134484325306679,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001512336445841224,
|
|
"loss": 4.5919,
|
|
"mean_token_accuracy": 0.2193811684846878,
|
|
"num_tokens": 16265482.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.282940864562988,
|
|
"epoch": 7.139027714675148,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015093864276301573,
|
|
"loss": 4.5909,
|
|
"mean_token_accuracy": 0.22275354713201523,
|
|
"num_tokens": 16275319.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.219920969009399,
|
|
"epoch": 7.143571104043617,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015064394705097546,
|
|
"loss": 4.5747,
|
|
"mean_token_accuracy": 0.2198408678174019,
|
|
"num_tokens": 16284889.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.284860372543335,
|
|
"epoch": 7.148114493412085,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015034955817513397,
|
|
"loss": 4.5914,
|
|
"mean_token_accuracy": 0.2239375814795494,
|
|
"num_tokens": 16295259.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.183380603790283,
|
|
"epoch": 7.152657882780554,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001500554768618665,
|
|
"loss": 4.5501,
|
|
"mean_token_accuracy": 0.22388984113931656,
|
|
"num_tokens": 16305489.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.196062612533569,
|
|
"epoch": 7.157201272149023,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014976170383678954,
|
|
"loss": 4.5257,
|
|
"mean_token_accuracy": 0.23448060154914857,
|
|
"num_tokens": 16315167.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.222267293930054,
|
|
"epoch": 7.161744661517492,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014946823982475883,
|
|
"loss": 4.5198,
|
|
"mean_token_accuracy": 0.23643112778663636,
|
|
"num_tokens": 16325059.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.29891414642334,
|
|
"epoch": 7.166288050885961,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014917508554986754,
|
|
"loss": 4.6239,
|
|
"mean_token_accuracy": 0.22289974391460418,
|
|
"num_tokens": 16336647.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.22138032913208,
|
|
"epoch": 7.17083144025443,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00014888224173544476,
|
|
"loss": 4.5318,
|
|
"mean_token_accuracy": 0.21693261712789536,
|
|
"num_tokens": 16346146.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.267798852920532,
|
|
"epoch": 7.175374829622899,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014858970910405368,
|
|
"loss": 4.6387,
|
|
"mean_token_accuracy": 0.22112399935722352,
|
|
"num_tokens": 16356124.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.243106174468994,
|
|
"epoch": 7.179918218991368,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014829748837748906,
|
|
"loss": 4.6202,
|
|
"mean_token_accuracy": 0.2143951803445816,
|
|
"num_tokens": 16367197.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.238667964935303,
|
|
"epoch": 7.184461608359836,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001480055802767769,
|
|
"loss": 4.5537,
|
|
"mean_token_accuracy": 0.223260298371315,
|
|
"num_tokens": 16376757.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.197329664230347,
|
|
"epoch": 7.189004997728305,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014771398552217114,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.23132639974355698,
|
|
"num_tokens": 16387454.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.231648063659668,
|
|
"epoch": 7.193548387096774,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014742270483315319,
|
|
"loss": 4.572,
|
|
"mean_token_accuracy": 0.2244747444987297,
|
|
"num_tokens": 16397852.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.189863395690918,
|
|
"epoch": 7.198091776465243,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014713173892842878,
|
|
"loss": 4.5495,
|
|
"mean_token_accuracy": 0.22543570548295974,
|
|
"num_tokens": 16407752.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.195296573638916,
|
|
"epoch": 7.202635165833712,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014684108852592765,
|
|
"loss": 4.5873,
|
|
"mean_token_accuracy": 0.215851229429245,
|
|
"num_tokens": 16418500.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.232487630844116,
|
|
"epoch": 7.207178555202181,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00014655075434280064,
|
|
"loss": 4.59,
|
|
"mean_token_accuracy": 0.22034069299697875,
|
|
"num_tokens": 16428706.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.26331729888916,
|
|
"epoch": 7.21172194457065,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014626073709541853,
|
|
"loss": 4.538,
|
|
"mean_token_accuracy": 0.2287825748324394,
|
|
"num_tokens": 16438184.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.199267339706421,
|
|
"epoch": 7.216265333939119,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00014597103749937,
|
|
"loss": 4.5528,
|
|
"mean_token_accuracy": 0.22676606327295304,
|
|
"num_tokens": 16449340.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.231654071807862,
|
|
"epoch": 7.220808723307587,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00014568165626946012,
|
|
"loss": 4.5354,
|
|
"mean_token_accuracy": 0.2211272582411766,
|
|
"num_tokens": 16459604.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.296084594726563,
|
|
"epoch": 7.225352112676056,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014539259411970823,
|
|
"loss": 4.6728,
|
|
"mean_token_accuracy": 0.21753816306591034,
|
|
"num_tokens": 16470150.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.25042495727539,
|
|
"epoch": 7.229895502044525,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001451038517633465,
|
|
"loss": 4.6004,
|
|
"mean_token_accuracy": 0.22982532232999803,
|
|
"num_tokens": 16480797.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.280168724060059,
|
|
"epoch": 7.234438891412994,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00014481542991281782,
|
|
"loss": 4.6476,
|
|
"mean_token_accuracy": 0.21795613765716554,
|
|
"num_tokens": 16492629.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.228167915344239,
|
|
"epoch": 7.238982280781463,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014452732927977465,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.22551799714565277,
|
|
"num_tokens": 16503225.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.2671427726745605,
|
|
"epoch": 7.243525670149932,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001442395505750766,
|
|
"loss": 4.507,
|
|
"mean_token_accuracy": 0.23313942551612854,
|
|
"num_tokens": 16512446.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.228613948822021,
|
|
"epoch": 7.248069059518401,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00014395209450878899,
|
|
"loss": 4.5544,
|
|
"mean_token_accuracy": 0.22680300921201707,
|
|
"num_tokens": 16522944.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.237939691543579,
|
|
"epoch": 7.25261244888687,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000143664961790181,
|
|
"loss": 4.5585,
|
|
"mean_token_accuracy": 0.227813458442688,
|
|
"num_tokens": 16533100.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.258956432342529,
|
|
"epoch": 7.257155838255338,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001433781531277243,
|
|
"loss": 4.6487,
|
|
"mean_token_accuracy": 0.21317924559116364,
|
|
"num_tokens": 16542893.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.215850448608398,
|
|
"epoch": 7.261699227623807,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001430916692290905,
|
|
"loss": 4.5769,
|
|
"mean_token_accuracy": 0.22542574405670165,
|
|
"num_tokens": 16553209.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.2517335414886475,
|
|
"epoch": 7.266242616992276,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00014280551080115032,
|
|
"loss": 4.5774,
|
|
"mean_token_accuracy": 0.22562101632356643,
|
|
"num_tokens": 16564574.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 7.266242616992276,
|
|
"eval_entropy": 5.152003928117974,
|
|
"eval_loss": 5.657458782196045,
|
|
"eval_mean_token_accuracy": 0.17438650120829427,
|
|
"eval_num_tokens": 16564574.0,
|
|
"eval_runtime": 2.5858,
|
|
"eval_samples_per_second": 1328.425,
|
|
"eval_steps_per_second": 166.295,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.25633978843689,
|
|
"epoch": 7.270786006360745,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014251967854997106,
|
|
"loss": 4.5902,
|
|
"mean_token_accuracy": 0.22231813222169877,
|
|
"num_tokens": 16574517.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.253251552581787,
|
|
"epoch": 7.275329395729214,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014223417318081572,
|
|
"loss": 4.5964,
|
|
"mean_token_accuracy": 0.22077925503253937,
|
|
"num_tokens": 16585311.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.201621103286743,
|
|
"epoch": 7.279872785097683,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014194899539814,
|
|
"loss": 4.5958,
|
|
"mean_token_accuracy": 0.2198190599679947,
|
|
"num_tokens": 16596055.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.169779300689697,
|
|
"epoch": 7.284416174466152,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014166414590559197,
|
|
"loss": 4.5284,
|
|
"mean_token_accuracy": 0.22371095716953276,
|
|
"num_tokens": 16606076.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.254501438140869,
|
|
"epoch": 7.288959563834621,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014137962540600944,
|
|
"loss": 4.5961,
|
|
"mean_token_accuracy": 0.22435845285654069,
|
|
"num_tokens": 16616740.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.229991388320923,
|
|
"epoch": 7.29350295320309,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00014109543460141837,
|
|
"loss": 4.5351,
|
|
"mean_token_accuracy": 0.23002929240465164,
|
|
"num_tokens": 16627412.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.214050722122193,
|
|
"epoch": 7.298046342571558,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001408115741930313,
|
|
"loss": 4.532,
|
|
"mean_token_accuracy": 0.23218650817871095,
|
|
"num_tokens": 16637032.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.17647123336792,
|
|
"epoch": 7.302589731940027,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014052804488124575,
|
|
"loss": 4.5415,
|
|
"mean_token_accuracy": 0.2246716096997261,
|
|
"num_tokens": 16646596.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.208134555816651,
|
|
"epoch": 7.307133121308496,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014024484736564196,
|
|
"loss": 4.5017,
|
|
"mean_token_accuracy": 0.23275112807750703,
|
|
"num_tokens": 16656404.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.23730993270874,
|
|
"epoch": 7.311676510676965,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001399619823449817,
|
|
"loss": 4.5612,
|
|
"mean_token_accuracy": 0.22251387536525727,
|
|
"num_tokens": 16666372.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.189354753494262,
|
|
"epoch": 7.316219900045434,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013967945051720614,
|
|
"loss": 4.552,
|
|
"mean_token_accuracy": 0.22754601389169693,
|
|
"num_tokens": 16677150.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.215758657455444,
|
|
"epoch": 7.320763289413903,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013939725257943475,
|
|
"loss": 4.6346,
|
|
"mean_token_accuracy": 0.22098578810691832,
|
|
"num_tokens": 16688712.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.216437196731567,
|
|
"epoch": 7.325306678782372,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00013911538922796275,
|
|
"loss": 4.5476,
|
|
"mean_token_accuracy": 0.22005417197942734,
|
|
"num_tokens": 16699214.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.227729415893554,
|
|
"epoch": 7.32985006815084,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013883386115825984,
|
|
"loss": 4.5826,
|
|
"mean_token_accuracy": 0.2259720116853714,
|
|
"num_tokens": 16709294.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.226204681396484,
|
|
"epoch": 7.334393457519309,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013855266906496867,
|
|
"loss": 4.5225,
|
|
"mean_token_accuracy": 0.22755974531173706,
|
|
"num_tokens": 16718761.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.2700108051300045,
|
|
"epoch": 7.338936846887778,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013827181364190288,
|
|
"loss": 4.6445,
|
|
"mean_token_accuracy": 0.21511484384536744,
|
|
"num_tokens": 16729075.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.2207972526550295,
|
|
"epoch": 7.343480236256247,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00013799129558204505,
|
|
"loss": 4.517,
|
|
"mean_token_accuracy": 0.22982999980449675,
|
|
"num_tokens": 16739293.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.2190478324890135,
|
|
"epoch": 7.348023625624716,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00013771111557754577,
|
|
"loss": 4.5765,
|
|
"mean_token_accuracy": 0.22116925269365312,
|
|
"num_tokens": 16749214.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.184898519515992,
|
|
"epoch": 7.352567014993185,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013743127431972124,
|
|
"loss": 4.5226,
|
|
"mean_token_accuracy": 0.2283172994852066,
|
|
"num_tokens": 16758715.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.224375152587891,
|
|
"epoch": 7.357110404361654,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001371517724990522,
|
|
"loss": 4.6131,
|
|
"mean_token_accuracy": 0.22402420938014983,
|
|
"num_tokens": 16768965.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.203349828720093,
|
|
"epoch": 7.361653793730123,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001368726108051813,
|
|
"loss": 4.4727,
|
|
"mean_token_accuracy": 0.23579326868057252,
|
|
"num_tokens": 16778368.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.256468439102173,
|
|
"epoch": 7.366197183098592,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013659378992691244,
|
|
"loss": 4.6042,
|
|
"mean_token_accuracy": 0.21908179074525833,
|
|
"num_tokens": 16789396.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.25421986579895,
|
|
"epoch": 7.37074057246706,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013631531055220835,
|
|
"loss": 4.5893,
|
|
"mean_token_accuracy": 0.22143297493457795,
|
|
"num_tokens": 16800063.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.227025890350342,
|
|
"epoch": 7.375283961835529,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013603717336818934,
|
|
"loss": 4.5575,
|
|
"mean_token_accuracy": 0.22685863673686982,
|
|
"num_tokens": 16810344.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.1635030746459964,
|
|
"epoch": 7.379827351203998,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00013575937906113095,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.22610011100769042,
|
|
"num_tokens": 16820233.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.25134310722351,
|
|
"epoch": 7.384370740572467,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0001354819283164634,
|
|
"loss": 4.6133,
|
|
"mean_token_accuracy": 0.2223345920443535,
|
|
"num_tokens": 16829761.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.301743030548096,
|
|
"epoch": 7.388914129940936,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001352048218187686,
|
|
"loss": 4.6664,
|
|
"mean_token_accuracy": 0.22189030200242996,
|
|
"num_tokens": 16841068.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.243192338943482,
|
|
"epoch": 7.393457519309405,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00013492806025177935,
|
|
"loss": 4.5355,
|
|
"mean_token_accuracy": 0.22680718302726746,
|
|
"num_tokens": 16850733.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.20181097984314,
|
|
"epoch": 7.398000908677874,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013465164429837716,
|
|
"loss": 4.5362,
|
|
"mean_token_accuracy": 0.22188206017017365,
|
|
"num_tokens": 16860416.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.2357536315917965,
|
|
"epoch": 7.402544298046343,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001343755746405912,
|
|
"loss": 4.5718,
|
|
"mean_token_accuracy": 0.23030321896076203,
|
|
"num_tokens": 16870662.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.19888219833374,
|
|
"epoch": 7.407087687414811,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013409985195959574,
|
|
"loss": 4.6438,
|
|
"mean_token_accuracy": 0.21326191872358322,
|
|
"num_tokens": 16881758.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.252466106414795,
|
|
"epoch": 7.41163107678328,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013382447693570908,
|
|
"loss": 4.6011,
|
|
"mean_token_accuracy": 0.21805257201194764,
|
|
"num_tokens": 16892399.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.225783348083496,
|
|
"epoch": 7.416174466151749,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00013354945024839166,
|
|
"loss": 4.6398,
|
|
"mean_token_accuracy": 0.21656263917684554,
|
|
"num_tokens": 16903696.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.253020858764648,
|
|
"epoch": 7.420717855520218,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00013327477257624466,
|
|
"loss": 4.5718,
|
|
"mean_token_accuracy": 0.22651004940271377,
|
|
"num_tokens": 16914326.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.213275051116943,
|
|
"epoch": 7.425261244888687,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001330004445970078,
|
|
"loss": 4.4911,
|
|
"mean_token_accuracy": 0.2399069771170616,
|
|
"num_tokens": 16924533.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.216855096817016,
|
|
"epoch": 7.429804634257156,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00013272646698755816,
|
|
"loss": 4.5121,
|
|
"mean_token_accuracy": 0.23128149211406707,
|
|
"num_tokens": 16934157.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.150432682037353,
|
|
"epoch": 7.434348023625625,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00013245284042390814,
|
|
"loss": 4.503,
|
|
"mean_token_accuracy": 0.23006280213594438,
|
|
"num_tokens": 16944148.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.218782186508179,
|
|
"epoch": 7.438891412994094,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013217956558120436,
|
|
"loss": 4.6561,
|
|
"mean_token_accuracy": 0.21561591178178788,
|
|
"num_tokens": 16956388.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.211469459533691,
|
|
"epoch": 7.443434802362562,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000131906643133725,
|
|
"loss": 4.6037,
|
|
"mean_token_accuracy": 0.2261236935853958,
|
|
"num_tokens": 16965216.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.230607748031616,
|
|
"epoch": 7.447978191731031,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013163407375487926,
|
|
"loss": 4.5739,
|
|
"mean_token_accuracy": 0.22329719960689545,
|
|
"num_tokens": 16976250.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.199106884002686,
|
|
"epoch": 7.4525215810995,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.00013136185811720496,
|
|
"loss": 4.6256,
|
|
"mean_token_accuracy": 0.21579130738973618,
|
|
"num_tokens": 16987375.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.182454204559326,
|
|
"epoch": 7.457064970467969,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001310899968923671,
|
|
"loss": 4.6062,
|
|
"mean_token_accuracy": 0.22765926122665406,
|
|
"num_tokens": 16997526.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.2586416721344,
|
|
"epoch": 7.461608359836438,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001308184907511562,
|
|
"loss": 4.6101,
|
|
"mean_token_accuracy": 0.21906366646289827,
|
|
"num_tokens": 17008342.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.237203168869018,
|
|
"epoch": 7.466151749204907,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013054734036348675,
|
|
"loss": 4.6055,
|
|
"mean_token_accuracy": 0.22194729298353194,
|
|
"num_tokens": 17019555.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.258501291275024,
|
|
"epoch": 7.470695138573376,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013027654639839536,
|
|
"loss": 4.63,
|
|
"mean_token_accuracy": 0.21807322949171065,
|
|
"num_tokens": 17030686.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.263095045089722,
|
|
"epoch": 7.475238527941845,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00013000610952403917,
|
|
"loss": 4.543,
|
|
"mean_token_accuracy": 0.23368019014596939,
|
|
"num_tokens": 17041089.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.1596235752105715,
|
|
"epoch": 7.479781917310314,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012973603040769427,
|
|
"loss": 4.5215,
|
|
"mean_token_accuracy": 0.22630759030580522,
|
|
"num_tokens": 17050194.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.236288070678711,
|
|
"epoch": 7.484325306678782,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 0.00012946630971575414,
|
|
"loss": 4.5687,
|
|
"mean_token_accuracy": 0.22006708830595018,
|
|
"num_tokens": 17060596.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.243351411819458,
|
|
"epoch": 7.488868696047251,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012919694811372768,
|
|
"loss": 4.6134,
|
|
"mean_token_accuracy": 0.22112150490283966,
|
|
"num_tokens": 17069763.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.175983095169068,
|
|
"epoch": 7.49341208541572,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012892794626623793,
|
|
"loss": 4.5358,
|
|
"mean_token_accuracy": 0.22512242048978806,
|
|
"num_tokens": 17080031.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.223270893096924,
|
|
"epoch": 7.497955474784189,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001286593048370201,
|
|
"loss": 4.5746,
|
|
"mean_token_accuracy": 0.22631481438875198,
|
|
"num_tokens": 17091339.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.185304117202759,
|
|
"epoch": 7.502498864152658,
|
|
"grad_norm": 2.953125,
|
|
"learning_rate": 0.00012839102448892033,
|
|
"loss": 4.5214,
|
|
"mean_token_accuracy": 0.22839743047952651,
|
|
"num_tokens": 17102264.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.228479099273682,
|
|
"epoch": 7.507042253521127,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012812310588389367,
|
|
"loss": 4.5644,
|
|
"mean_token_accuracy": 0.21607417166233062,
|
|
"num_tokens": 17112493.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.218078088760376,
|
|
"epoch": 7.511585642889596,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00012785554968300266,
|
|
"loss": 4.5387,
|
|
"mean_token_accuracy": 0.23243934512138367,
|
|
"num_tokens": 17122037.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.2124560356140135,
|
|
"epoch": 7.516129032258064,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00012758835654641548,
|
|
"loss": 4.5734,
|
|
"mean_token_accuracy": 0.2228848949074745,
|
|
"num_tokens": 17133403.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.170206689834595,
|
|
"epoch": 7.520672421626533,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000127321527133405,
|
|
"loss": 4.5226,
|
|
"mean_token_accuracy": 0.2237117260694504,
|
|
"num_tokens": 17143619.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.244374322891235,
|
|
"epoch": 7.525215810995002,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00012705506210234583,
|
|
"loss": 4.6309,
|
|
"mean_token_accuracy": 0.21674645096063613,
|
|
"num_tokens": 17154811.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.1738382339477536,
|
|
"epoch": 7.529759200363471,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012678896211071422,
|
|
"loss": 4.5432,
|
|
"mean_token_accuracy": 0.22319968193769454,
|
|
"num_tokens": 17166039.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.240162801742554,
|
|
"epoch": 7.53430258973194,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001265232278150854,
|
|
"loss": 4.5632,
|
|
"mean_token_accuracy": 0.22503907829523087,
|
|
"num_tokens": 17176555.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.207495307922363,
|
|
"epoch": 7.538845979100409,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001262578598711323,
|
|
"loss": 4.61,
|
|
"mean_token_accuracy": 0.21605585068464278,
|
|
"num_tokens": 17188220.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.238838720321655,
|
|
"epoch": 7.543389368468878,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001259928589336239,
|
|
"loss": 4.5778,
|
|
"mean_token_accuracy": 0.22842023968696595,
|
|
"num_tokens": 17199528.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.235700798034668,
|
|
"epoch": 7.547932757837347,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012572822565642378,
|
|
"loss": 4.53,
|
|
"mean_token_accuracy": 0.2256921038031578,
|
|
"num_tokens": 17210030.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.247352933883667,
|
|
"epoch": 7.552476147205816,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012546396069248813,
|
|
"loss": 4.6164,
|
|
"mean_token_accuracy": 0.22206749469041825,
|
|
"num_tokens": 17220231.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.215401029586792,
|
|
"epoch": 7.557019536574284,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00012520006469386458,
|
|
"loss": 4.5733,
|
|
"mean_token_accuracy": 0.22866043597459793,
|
|
"num_tokens": 17229966.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.210735845565796,
|
|
"epoch": 7.561562925942753,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00012493653831169014,
|
|
"loss": 4.6061,
|
|
"mean_token_accuracy": 0.2198980525135994,
|
|
"num_tokens": 17240654.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.154702043533325,
|
|
"epoch": 7.566106315311222,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00012467338219619015,
|
|
"loss": 4.4963,
|
|
"mean_token_accuracy": 0.23870454281568526,
|
|
"num_tokens": 17250653.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.2529905319213865,
|
|
"epoch": 7.570649704679691,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000124410596996676,
|
|
"loss": 4.5882,
|
|
"mean_token_accuracy": 0.2246700644493103,
|
|
"num_tokens": 17260223.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.166658878326416,
|
|
"epoch": 7.57519309404816,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001241481833615442,
|
|
"loss": 4.5468,
|
|
"mean_token_accuracy": 0.2252572402358055,
|
|
"num_tokens": 17269998.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.199185132980347,
|
|
"epoch": 7.579736483416629,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00012388614193827407,
|
|
"loss": 4.5263,
|
|
"mean_token_accuracy": 0.23009968996047975,
|
|
"num_tokens": 17279824.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.181958913803101,
|
|
"epoch": 7.584279872785098,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.000123624473373427,
|
|
"loss": 4.5276,
|
|
"mean_token_accuracy": 0.22702550441026687,
|
|
"num_tokens": 17290242.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.156328535079956,
|
|
"epoch": 7.588823262153566,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012336317831264411,
|
|
"loss": 4.5207,
|
|
"mean_token_accuracy": 0.23420267850160598,
|
|
"num_tokens": 17299515.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.207658910751343,
|
|
"epoch": 7.593366651522036,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012310225740064498,
|
|
"loss": 4.5023,
|
|
"mean_token_accuracy": 0.2274177610874176,
|
|
"num_tokens": 17308627.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.172543859481811,
|
|
"epoch": 7.597910040890504,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001228417112812259,
|
|
"loss": 4.4881,
|
|
"mean_token_accuracy": 0.22478106170892714,
|
|
"num_tokens": 17319465.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.1818266868591305,
|
|
"epoch": 7.602453430258973,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012258154059725877,
|
|
"loss": 4.528,
|
|
"mean_token_accuracy": 0.2263115108013153,
|
|
"num_tokens": 17329202.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.183373737335205,
|
|
"epoch": 7.606996819627442,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001223217459906886,
|
|
"loss": 4.4666,
|
|
"mean_token_accuracy": 0.2379268229007721,
|
|
"num_tokens": 17338956.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.24989423751831,
|
|
"epoch": 7.611540208995911,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00012206232810253297,
|
|
"loss": 4.6387,
|
|
"mean_token_accuracy": 0.21884799897670745,
|
|
"num_tokens": 17349380.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.245648813247681,
|
|
"epoch": 7.61608359836438,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00012180328757287967,
|
|
"loss": 4.7073,
|
|
"mean_token_accuracy": 0.21554791331291198,
|
|
"num_tokens": 17359770.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.2012087345123295,
|
|
"epoch": 7.620626987732849,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012154462504088545,
|
|
"loss": 4.5527,
|
|
"mean_token_accuracy": 0.23017272353172302,
|
|
"num_tokens": 17369209.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.190491247177124,
|
|
"epoch": 7.625170377101318,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001212863411447744,
|
|
"loss": 4.5566,
|
|
"mean_token_accuracy": 0.22460397332906723,
|
|
"num_tokens": 17379221.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.242994928359986,
|
|
"epoch": 7.629713766469786,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012102843652183628,
|
|
"loss": 4.4796,
|
|
"mean_token_accuracy": 0.22753357887268066,
|
|
"num_tokens": 17389164.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.280991172790527,
|
|
"epoch": 7.634257155838255,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012077091180842523,
|
|
"loss": 4.6015,
|
|
"mean_token_accuracy": 0.22099053114652634,
|
|
"num_tokens": 17399392.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.234323883056641,
|
|
"epoch": 7.638800545206724,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00012051376763995784,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.2207357257604599,
|
|
"num_tokens": 17408933.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.215939903259278,
|
|
"epoch": 7.643343934575193,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012025700465091177,
|
|
"loss": 4.5468,
|
|
"mean_token_accuracy": 0.2209831565618515,
|
|
"num_tokens": 17419391.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.20884337425232,
|
|
"epoch": 7.647887323943662,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00012000062347482413,
|
|
"loss": 4.5799,
|
|
"mean_token_accuracy": 0.22439638823270797,
|
|
"num_tokens": 17430562.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.196576499938965,
|
|
"epoch": 7.652430713312131,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011974462474429015,
|
|
"loss": 4.5506,
|
|
"mean_token_accuracy": 0.22183343172073364,
|
|
"num_tokens": 17441345.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.200856304168701,
|
|
"epoch": 7.6569741026806,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011948900909096101,
|
|
"loss": 4.5578,
|
|
"mean_token_accuracy": 0.22090665102005005,
|
|
"num_tokens": 17450972.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.208098888397217,
|
|
"epoch": 7.661517492049069,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001192337771455431,
|
|
"loss": 4.5642,
|
|
"mean_token_accuracy": 0.22315150052309035,
|
|
"num_tokens": 17460930.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.231241130828858,
|
|
"epoch": 7.666060881417538,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011897892953779576,
|
|
"loss": 4.5707,
|
|
"mean_token_accuracy": 0.21993254274129867,
|
|
"num_tokens": 17470865.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.1978106021881105,
|
|
"epoch": 7.670604270786006,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011872446689653032,
|
|
"loss": 4.4808,
|
|
"mean_token_accuracy": 0.2307516112923622,
|
|
"num_tokens": 17481379.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.303891181945801,
|
|
"epoch": 7.675147660154475,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001184703898496078,
|
|
"loss": 4.6158,
|
|
"mean_token_accuracy": 0.2263522058725357,
|
|
"num_tokens": 17492741.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.23414831161499,
|
|
"epoch": 7.679691049522944,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011821669902393823,
|
|
"loss": 4.6309,
|
|
"mean_token_accuracy": 0.21490714997053145,
|
|
"num_tokens": 17504179.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.179196071624756,
|
|
"epoch": 7.684234438891413,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011796339504547846,
|
|
"loss": 4.5586,
|
|
"mean_token_accuracy": 0.22981976866722106,
|
|
"num_tokens": 17514299.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.260849094390869,
|
|
"epoch": 7.688777828259882,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011771047853923086,
|
|
"loss": 4.5629,
|
|
"mean_token_accuracy": 0.2271700009703636,
|
|
"num_tokens": 17524454.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.224223899841308,
|
|
"epoch": 7.693321217628351,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011745795012924169,
|
|
"loss": 4.5112,
|
|
"mean_token_accuracy": 0.2315181940793991,
|
|
"num_tokens": 17534887.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.244089651107788,
|
|
"epoch": 7.69786460699682,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011720581043859988,
|
|
"loss": 4.6053,
|
|
"mean_token_accuracy": 0.22575889825820922,
|
|
"num_tokens": 17544944.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.2166492462158205,
|
|
"epoch": 7.702407996365288,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011695406008943495,
|
|
"loss": 4.539,
|
|
"mean_token_accuracy": 0.2320433408021927,
|
|
"num_tokens": 17556481.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.255328512191772,
|
|
"epoch": 7.706951385733757,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011670269970291584,
|
|
"loss": 4.6029,
|
|
"mean_token_accuracy": 0.22217052280902863,
|
|
"num_tokens": 17566776.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.155510807037354,
|
|
"epoch": 7.711494775102226,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001164517298992493,
|
|
"loss": 4.5194,
|
|
"mean_token_accuracy": 0.21987317502498627,
|
|
"num_tokens": 17575441.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.248950862884522,
|
|
"epoch": 7.716038164470695,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011620115129767848,
|
|
"loss": 4.6425,
|
|
"mean_token_accuracy": 0.21865012794733046,
|
|
"num_tokens": 17585792.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.196743440628052,
|
|
"epoch": 7.720581553839164,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011595096451648111,
|
|
"loss": 4.5582,
|
|
"mean_token_accuracy": 0.21970709264278412,
|
|
"num_tokens": 17595820.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.720581553839164,
|
|
"eval_entropy": 5.138251077297122,
|
|
"eval_loss": 5.651764392852783,
|
|
"eval_mean_token_accuracy": 0.17516402335360992,
|
|
"eval_num_tokens": 17595820.0,
|
|
"eval_runtime": 2.582,
|
|
"eval_samples_per_second": 1330.347,
|
|
"eval_steps_per_second": 166.535,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.178137254714966,
|
|
"epoch": 7.725124943207633,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011570117017296819,
|
|
"loss": 4.5114,
|
|
"mean_token_accuracy": 0.22983548194169998,
|
|
"num_tokens": 17605791.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.209797668457031,
|
|
"epoch": 7.729668332576102,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011545176888348237,
|
|
"loss": 4.5734,
|
|
"mean_token_accuracy": 0.22523413598537445,
|
|
"num_tokens": 17616159.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.158917331695557,
|
|
"epoch": 7.734211721944571,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011520276126339667,
|
|
"loss": 4.5465,
|
|
"mean_token_accuracy": 0.22151664048433303,
|
|
"num_tokens": 17626612.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.219311141967774,
|
|
"epoch": 7.73875511131304,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011495414792711259,
|
|
"loss": 4.5637,
|
|
"mean_token_accuracy": 0.22384814769029618,
|
|
"num_tokens": 17636254.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.289140892028809,
|
|
"epoch": 7.743298500681508,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011470592948805882,
|
|
"loss": 4.6582,
|
|
"mean_token_accuracy": 0.21537579149007796,
|
|
"num_tokens": 17648097.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.207537841796875,
|
|
"epoch": 7.747841890049977,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011445810655868957,
|
|
"loss": 4.528,
|
|
"mean_token_accuracy": 0.22534339278936386,
|
|
"num_tokens": 17658361.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.169511127471924,
|
|
"epoch": 7.752385279418446,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00011421067975048355,
|
|
"loss": 4.4997,
|
|
"mean_token_accuracy": 0.2316489189863205,
|
|
"num_tokens": 17668640.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.154721117019653,
|
|
"epoch": 7.756928668786915,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011396364967394152,
|
|
"loss": 4.4508,
|
|
"mean_token_accuracy": 0.23602576553821564,
|
|
"num_tokens": 17678257.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.155816221237183,
|
|
"epoch": 7.761472058155384,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011371701693858582,
|
|
"loss": 4.5604,
|
|
"mean_token_accuracy": 0.2244935989379883,
|
|
"num_tokens": 17688635.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.181255531311035,
|
|
"epoch": 7.766015447523853,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011347078215295817,
|
|
"loss": 4.6111,
|
|
"mean_token_accuracy": 0.2230816066265106,
|
|
"num_tokens": 17699917.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.232228469848633,
|
|
"epoch": 7.770558836892322,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011322494592461839,
|
|
"loss": 4.5746,
|
|
"mean_token_accuracy": 0.21799688786268234,
|
|
"num_tokens": 17710278.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.201872730255127,
|
|
"epoch": 7.77510222626079,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011297950886014292,
|
|
"loss": 4.536,
|
|
"mean_token_accuracy": 0.23191049098968505,
|
|
"num_tokens": 17720375.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.2076600074768065,
|
|
"epoch": 7.779645615629259,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001127344715651234,
|
|
"loss": 4.5631,
|
|
"mean_token_accuracy": 0.22134747356176376,
|
|
"num_tokens": 17730299.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.244504737854004,
|
|
"epoch": 7.784189004997728,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011248983464416497,
|
|
"loss": 4.5892,
|
|
"mean_token_accuracy": 0.22587901204824448,
|
|
"num_tokens": 17741096.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.207184648513794,
|
|
"epoch": 7.788732394366197,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011224559870088488,
|
|
"loss": 4.5758,
|
|
"mean_token_accuracy": 0.22582703977823257,
|
|
"num_tokens": 17751502.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.206006193161011,
|
|
"epoch": 7.793275783734666,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011200176433791101,
|
|
"loss": 4.561,
|
|
"mean_token_accuracy": 0.22687047272920607,
|
|
"num_tokens": 17761955.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.161537599563599,
|
|
"epoch": 7.797819173103135,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001117583321568805,
|
|
"loss": 4.5098,
|
|
"mean_token_accuracy": 0.23175639361143113,
|
|
"num_tokens": 17772330.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.205416536331176,
|
|
"epoch": 7.802362562471604,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011151530275843805,
|
|
"loss": 4.4673,
|
|
"mean_token_accuracy": 0.2326839178800583,
|
|
"num_tokens": 17781688.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.14605941772461,
|
|
"epoch": 7.806905951840073,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011127267674223449,
|
|
"loss": 4.5324,
|
|
"mean_token_accuracy": 0.22843084931373597,
|
|
"num_tokens": 17792037.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.224978590011597,
|
|
"epoch": 7.811449341208542,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011103045470692532,
|
|
"loss": 4.6145,
|
|
"mean_token_accuracy": 0.22449600398540498,
|
|
"num_tokens": 17802705.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.272964572906494,
|
|
"epoch": 7.81599273057701,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011078863725016955,
|
|
"loss": 4.659,
|
|
"mean_token_accuracy": 0.21228035390377045,
|
|
"num_tokens": 17813758.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.1934449672698975,
|
|
"epoch": 7.820536119945479,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011054722496862759,
|
|
"loss": 4.5246,
|
|
"mean_token_accuracy": 0.23345967531204223,
|
|
"num_tokens": 17823465.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.235465717315674,
|
|
"epoch": 7.825079509313948,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011030621845796021,
|
|
"loss": 4.562,
|
|
"mean_token_accuracy": 0.22745550274848939,
|
|
"num_tokens": 17833719.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.185354566574096,
|
|
"epoch": 7.829622898682417,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011006561831282703,
|
|
"loss": 4.5208,
|
|
"mean_token_accuracy": 0.23617078363895416,
|
|
"num_tokens": 17844519.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.1785801410675045,
|
|
"epoch": 7.834166288050886,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010982542512688515,
|
|
"loss": 4.5108,
|
|
"mean_token_accuracy": 0.232280895113945,
|
|
"num_tokens": 17856463.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.163903141021729,
|
|
"epoch": 7.838709677419355,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010958563949278711,
|
|
"loss": 4.5066,
|
|
"mean_token_accuracy": 0.22818759381771087,
|
|
"num_tokens": 17867449.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.1462122917175295,
|
|
"epoch": 7.843253066787824,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010934626200218032,
|
|
"loss": 4.5427,
|
|
"mean_token_accuracy": 0.22524810433387757,
|
|
"num_tokens": 17878522.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.236309766769409,
|
|
"epoch": 7.847796456156293,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010910729324570488,
|
|
"loss": 4.5907,
|
|
"mean_token_accuracy": 0.22669459581375123,
|
|
"num_tokens": 17888544.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.261119651794433,
|
|
"epoch": 7.852339845524762,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010886873381299248,
|
|
"loss": 4.6383,
|
|
"mean_token_accuracy": 0.2187759980559349,
|
|
"num_tokens": 17897432.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.237616539001465,
|
|
"epoch": 7.85688323489323,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010863058429266469,
|
|
"loss": 4.5598,
|
|
"mean_token_accuracy": 0.23596204370260238,
|
|
"num_tokens": 17907868.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.235971832275391,
|
|
"epoch": 7.861426624261699,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010839284527233192,
|
|
"loss": 4.5068,
|
|
"mean_token_accuracy": 0.22331051230430604,
|
|
"num_tokens": 17917887.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.1687110424041744,
|
|
"epoch": 7.865970013630168,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010815551733859153,
|
|
"loss": 4.5096,
|
|
"mean_token_accuracy": 0.22566710114479066,
|
|
"num_tokens": 17928030.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.206781387329102,
|
|
"epoch": 7.870513402998637,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 0.00010791860107702658,
|
|
"loss": 4.6212,
|
|
"mean_token_accuracy": 0.2137369304895401,
|
|
"num_tokens": 17937990.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.188587474822998,
|
|
"epoch": 7.875056792367106,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010768209707220435,
|
|
"loss": 4.6147,
|
|
"mean_token_accuracy": 0.22039198130369186,
|
|
"num_tokens": 17948262.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.225181436538696,
|
|
"epoch": 7.879600181735575,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010744600590767504,
|
|
"loss": 4.5784,
|
|
"mean_token_accuracy": 0.220207642018795,
|
|
"num_tokens": 17958942.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.197871685028076,
|
|
"epoch": 7.884143571104044,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010721032816597012,
|
|
"loss": 4.5691,
|
|
"mean_token_accuracy": 0.22591895014047622,
|
|
"num_tokens": 17969362.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.240658617019653,
|
|
"epoch": 7.888686960472512,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000106975064428601,
|
|
"loss": 4.5361,
|
|
"mean_token_accuracy": 0.23047971725463867,
|
|
"num_tokens": 17979009.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.294917869567871,
|
|
"epoch": 7.893230349840981,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00010674021527605742,
|
|
"loss": 4.5979,
|
|
"mean_token_accuracy": 0.22703611403703688,
|
|
"num_tokens": 17989064.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.2024085521698,
|
|
"epoch": 7.89777373920945,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010650578128780656,
|
|
"loss": 4.514,
|
|
"mean_token_accuracy": 0.2265704795718193,
|
|
"num_tokens": 17998574.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.294367170333862,
|
|
"epoch": 7.902317128577919,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001062717630422908,
|
|
"loss": 4.7253,
|
|
"mean_token_accuracy": 0.21670208275318145,
|
|
"num_tokens": 18010735.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.209484243392945,
|
|
"epoch": 7.906860517946388,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010603816111692699,
|
|
"loss": 4.5582,
|
|
"mean_token_accuracy": 0.2280753493309021,
|
|
"num_tokens": 18020316.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.245192909240723,
|
|
"epoch": 7.911403907314857,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 0.00010580497608810455,
|
|
"loss": 4.5618,
|
|
"mean_token_accuracy": 0.22282399535179137,
|
|
"num_tokens": 18031058.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.202867269515991,
|
|
"epoch": 7.915947296683326,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010557220853118456,
|
|
"loss": 4.538,
|
|
"mean_token_accuracy": 0.2260320395231247,
|
|
"num_tokens": 18040603.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.27614917755127,
|
|
"epoch": 7.920490686051795,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010533985902049755,
|
|
"loss": 4.6246,
|
|
"mean_token_accuracy": 0.2238526999950409,
|
|
"num_tokens": 18051369.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.195500087738037,
|
|
"epoch": 7.925034075420264,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00010510792812934307,
|
|
"loss": 4.515,
|
|
"mean_token_accuracy": 0.2281131401658058,
|
|
"num_tokens": 18062283.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.234424638748169,
|
|
"epoch": 7.929577464788732,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010487641642998746,
|
|
"loss": 4.595,
|
|
"mean_token_accuracy": 0.22423552572727204,
|
|
"num_tokens": 18072125.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.203469514846802,
|
|
"epoch": 7.934120854157201,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00010464532449366281,
|
|
"loss": 4.6275,
|
|
"mean_token_accuracy": 0.22039141058921813,
|
|
"num_tokens": 18083812.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.247008800506592,
|
|
"epoch": 7.93866424352567,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00010441465289056548,
|
|
"loss": 4.5834,
|
|
"mean_token_accuracy": 0.2302800625562668,
|
|
"num_tokens": 18095399.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.238218021392822,
|
|
"epoch": 7.943207632894139,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00010418440218985482,
|
|
"loss": 4.5711,
|
|
"mean_token_accuracy": 0.23411377072334288,
|
|
"num_tokens": 18106234.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.172398042678833,
|
|
"epoch": 7.947751022262608,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010395457295965143,
|
|
"loss": 4.4746,
|
|
"mean_token_accuracy": 0.23601692914962769,
|
|
"num_tokens": 18116129.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.2564130306243895,
|
|
"epoch": 7.952294411631077,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00010372516576703617,
|
|
"loss": 4.5971,
|
|
"mean_token_accuracy": 0.220488104224205,
|
|
"num_tokens": 18126757.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.2084815979003904,
|
|
"epoch": 7.956837800999546,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0001034961811780484,
|
|
"loss": 4.5628,
|
|
"mean_token_accuracy": 0.22312868237495423,
|
|
"num_tokens": 18136645.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.272539806365967,
|
|
"epoch": 7.961381190368014,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010326761975768492,
|
|
"loss": 4.618,
|
|
"mean_token_accuracy": 0.22238818556070328,
|
|
"num_tokens": 18146458.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.197665500640869,
|
|
"epoch": 7.965924579736483,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010303948206989827,
|
|
"loss": 4.5532,
|
|
"mean_token_accuracy": 0.2271757483482361,
|
|
"num_tokens": 18156151.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.22778000831604,
|
|
"epoch": 7.970467969104952,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010281176867759556,
|
|
"loss": 4.5843,
|
|
"mean_token_accuracy": 0.22606052160263063,
|
|
"num_tokens": 18166695.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.193287706375122,
|
|
"epoch": 7.975011358473421,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010258448014263682,
|
|
"loss": 4.532,
|
|
"mean_token_accuracy": 0.2257544979453087,
|
|
"num_tokens": 18177011.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.268028450012207,
|
|
"epoch": 7.97955474784189,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010235761702583405,
|
|
"loss": 4.626,
|
|
"mean_token_accuracy": 0.21656066328287124,
|
|
"num_tokens": 18187179.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.2173055648803714,
|
|
"epoch": 7.984098137210359,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010213117988694944,
|
|
"loss": 4.5791,
|
|
"mean_token_accuracy": 0.22019704431295395,
|
|
"num_tokens": 18197106.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.156585693359375,
|
|
"epoch": 7.988641526578828,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00010190516928469407,
|
|
"loss": 4.5535,
|
|
"mean_token_accuracy": 0.22639950811862947,
|
|
"num_tokens": 18207729.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.185166740417481,
|
|
"epoch": 7.993184915947297,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010167958577672654,
|
|
"loss": 4.4857,
|
|
"mean_token_accuracy": 0.22903526127338408,
|
|
"num_tokens": 18217885.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.218082523345947,
|
|
"epoch": 7.997728305315766,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010145442991965198,
|
|
"loss": 4.5954,
|
|
"mean_token_accuracy": 0.22043910324573518,
|
|
"num_tokens": 18229633.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.180166191524929,
|
|
"epoch": 8.001817355747388,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010122970226901981,
|
|
"loss": 4.579,
|
|
"mean_token_accuracy": 0.22380098203818002,
|
|
"num_tokens": 18239194.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.260691976547241,
|
|
"epoch": 8.006360745115856,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00010100540337932339,
|
|
"loss": 4.5275,
|
|
"mean_token_accuracy": 0.22550746500492097,
|
|
"num_tokens": 18250061.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.21523118019104,
|
|
"epoch": 8.010904134484326,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00010078153380399773,
|
|
"loss": 4.452,
|
|
"mean_token_accuracy": 0.23921645283699036,
|
|
"num_tokens": 18260977.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.211930990219116,
|
|
"epoch": 8.015447523852794,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010055809409541902,
|
|
"loss": 4.481,
|
|
"mean_token_accuracy": 0.23739960938692092,
|
|
"num_tokens": 18271866.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.213113737106323,
|
|
"epoch": 8.019990913221264,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010033508480490222,
|
|
"loss": 4.5645,
|
|
"mean_token_accuracy": 0.2260512262582779,
|
|
"num_tokens": 18283353.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.135661602020264,
|
|
"epoch": 8.024534302589732,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010011250648270081,
|
|
"loss": 4.4775,
|
|
"mean_token_accuracy": 0.2323247417807579,
|
|
"num_tokens": 18293619.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.2273499965667725,
|
|
"epoch": 8.0290776919582,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.98903596780046e-05,
|
|
"loss": 4.4931,
|
|
"mean_token_accuracy": 0.2379540368914604,
|
|
"num_tokens": 18303411.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.213728284835815,
|
|
"epoch": 8.03362108132667,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.966864493893876e-05,
|
|
"loss": 4.438,
|
|
"mean_token_accuracy": 0.2458496555685997,
|
|
"num_tokens": 18312798.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.195835208892822,
|
|
"epoch": 8.038164470695138,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.94473628125623e-05,
|
|
"loss": 4.4977,
|
|
"mean_token_accuracy": 0.23027886897325517,
|
|
"num_tokens": 18324111.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.190727233886719,
|
|
"epoch": 8.042707860063608,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.922651384486695e-05,
|
|
"loss": 4.4647,
|
|
"mean_token_accuracy": 0.22865334153175354,
|
|
"num_tokens": 18335197.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.20472822189331,
|
|
"epoch": 8.047251249432076,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.900609858077562e-05,
|
|
"loss": 4.5148,
|
|
"mean_token_accuracy": 0.23035187572240828,
|
|
"num_tokens": 18345896.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.247004699707031,
|
|
"epoch": 8.051794638800546,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 9.878611756414094e-05,
|
|
"loss": 4.5354,
|
|
"mean_token_accuracy": 0.22985993772745134,
|
|
"num_tokens": 18356941.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.200949716567993,
|
|
"epoch": 8.056338028169014,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.856657133774424e-05,
|
|
"loss": 4.421,
|
|
"mean_token_accuracy": 0.2374473586678505,
|
|
"num_tokens": 18366767.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.245094966888428,
|
|
"epoch": 8.060881417537482,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.834746044329411e-05,
|
|
"loss": 4.5702,
|
|
"mean_token_accuracy": 0.22773967385292054,
|
|
"num_tokens": 18376698.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.204232835769654,
|
|
"epoch": 8.065424806905952,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.81287854214248e-05,
|
|
"loss": 4.4438,
|
|
"mean_token_accuracy": 0.2325809821486473,
|
|
"num_tokens": 18386678.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.228427791595459,
|
|
"epoch": 8.06996819627442,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.791054681169522e-05,
|
|
"loss": 4.5018,
|
|
"mean_token_accuracy": 0.23008545339107514,
|
|
"num_tokens": 18397020.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.1644023895263675,
|
|
"epoch": 8.07451158564289,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.769274515258739e-05,
|
|
"loss": 4.4859,
|
|
"mean_token_accuracy": 0.22655516415834426,
|
|
"num_tokens": 18407330.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.1832507133483885,
|
|
"epoch": 8.079054975011358,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.747538098150539e-05,
|
|
"loss": 4.4591,
|
|
"mean_token_accuracy": 0.23385779261589051,
|
|
"num_tokens": 18417046.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.204177045822144,
|
|
"epoch": 8.083598364379828,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 9.725845483477352e-05,
|
|
"loss": 4.5254,
|
|
"mean_token_accuracy": 0.2312478691339493,
|
|
"num_tokens": 18427907.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.211804437637329,
|
|
"epoch": 8.088141753748296,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.704196724763565e-05,
|
|
"loss": 4.4938,
|
|
"mean_token_accuracy": 0.2265399619936943,
|
|
"num_tokens": 18437972.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.178389692306519,
|
|
"epoch": 8.092685143116766,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 9.682591875425323e-05,
|
|
"loss": 4.51,
|
|
"mean_token_accuracy": 0.2334580972790718,
|
|
"num_tokens": 18448188.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.2195384979248045,
|
|
"epoch": 8.097228532485234,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.661030988770463e-05,
|
|
"loss": 4.5173,
|
|
"mean_token_accuracy": 0.22832322716712952,
|
|
"num_tokens": 18458130.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.204656887054443,
|
|
"epoch": 8.101771921853702,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.63951411799831e-05,
|
|
"loss": 4.4796,
|
|
"mean_token_accuracy": 0.23702031522989273,
|
|
"num_tokens": 18468257.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.158678770065308,
|
|
"epoch": 8.106315311222172,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.618041316199617e-05,
|
|
"loss": 4.5096,
|
|
"mean_token_accuracy": 0.2260836735367775,
|
|
"num_tokens": 18478720.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.2167058944702145,
|
|
"epoch": 8.11085870059064,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.596612636356383e-05,
|
|
"loss": 4.4718,
|
|
"mean_token_accuracy": 0.22713564783334733,
|
|
"num_tokens": 18489543.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.2357782363891605,
|
|
"epoch": 8.11540208995911,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.575228131341745e-05,
|
|
"loss": 4.531,
|
|
"mean_token_accuracy": 0.2251293733716011,
|
|
"num_tokens": 18499505.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.281885719299316,
|
|
"epoch": 8.119945479327578,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 9.55388785391984e-05,
|
|
"loss": 4.5885,
|
|
"mean_token_accuracy": 0.22472924292087554,
|
|
"num_tokens": 18510044.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.217501497268676,
|
|
"epoch": 8.124488868696048,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.532591856745689e-05,
|
|
"loss": 4.5159,
|
|
"mean_token_accuracy": 0.22912381291389466,
|
|
"num_tokens": 18520537.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.207627677917481,
|
|
"epoch": 8.129032258064516,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 9.511340192365044e-05,
|
|
"loss": 4.561,
|
|
"mean_token_accuracy": 0.22138841599225997,
|
|
"num_tokens": 18531840.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.200653839111328,
|
|
"epoch": 8.133575647432984,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.490132913214277e-05,
|
|
"loss": 4.4757,
|
|
"mean_token_accuracy": 0.22833633422851562,
|
|
"num_tokens": 18542097.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.232006072998047,
|
|
"epoch": 8.138119036801454,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.468970071620231e-05,
|
|
"loss": 4.5321,
|
|
"mean_token_accuracy": 0.22704195529222487,
|
|
"num_tokens": 18551629.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.1937511444091795,
|
|
"epoch": 8.142662426169922,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.447851719800126e-05,
|
|
"loss": 4.4673,
|
|
"mean_token_accuracy": 0.2283680871129036,
|
|
"num_tokens": 18562248.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.1845756530761715,
|
|
"epoch": 8.147205815538392,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.426777909861395e-05,
|
|
"loss": 4.4588,
|
|
"mean_token_accuracy": 0.23345676809549332,
|
|
"num_tokens": 18572107.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.205668497085571,
|
|
"epoch": 8.15174920490686,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.405748693801562e-05,
|
|
"loss": 4.5228,
|
|
"mean_token_accuracy": 0.22562928944826127,
|
|
"num_tokens": 18582665.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.2646876811981205,
|
|
"epoch": 8.15629259427533,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.384764123508124e-05,
|
|
"loss": 4.5398,
|
|
"mean_token_accuracy": 0.22920592576265336,
|
|
"num_tokens": 18593737.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.2327563762664795,
|
|
"epoch": 8.160835983643798,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 9.363824250758443e-05,
|
|
"loss": 4.5034,
|
|
"mean_token_accuracy": 0.23391259461641312,
|
|
"num_tokens": 18604180.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.169515895843506,
|
|
"epoch": 8.165379373012268,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.34292912721955e-05,
|
|
"loss": 4.4253,
|
|
"mean_token_accuracy": 0.23076368123292923,
|
|
"num_tokens": 18614328.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.216651868820191,
|
|
"epoch": 8.169922762380736,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.322078804448095e-05,
|
|
"loss": 4.505,
|
|
"mean_token_accuracy": 0.24133139997720718,
|
|
"num_tokens": 18625357.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.146533060073852,
|
|
"epoch": 8.174466151749204,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.301273333890171e-05,
|
|
"loss": 4.4572,
|
|
"mean_token_accuracy": 0.2365337133407593,
|
|
"num_tokens": 18636306.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 8.174466151749204,
|
|
"eval_entropy": 5.1300362808759825,
|
|
"eval_loss": 5.644595623016357,
|
|
"eval_mean_token_accuracy": 0.17571929266979527,
|
|
"eval_num_tokens": 18636306.0,
|
|
"eval_runtime": 2.5832,
|
|
"eval_samples_per_second": 1329.768,
|
|
"eval_steps_per_second": 166.463,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.240426397323608,
|
|
"epoch": 8.179009541117674,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.280512766881227e-05,
|
|
"loss": 4.5199,
|
|
"mean_token_accuracy": 0.2355484962463379,
|
|
"num_tokens": 18647134.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.201512432098388,
|
|
"epoch": 8.183552930486142,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.25979715464588e-05,
|
|
"loss": 4.4565,
|
|
"mean_token_accuracy": 0.23898531049489974,
|
|
"num_tokens": 18656948.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.206055498123169,
|
|
"epoch": 8.188096319854612,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.23912654829784e-05,
|
|
"loss": 4.488,
|
|
"mean_token_accuracy": 0.22067562788724898,
|
|
"num_tokens": 18667682.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.246758842468262,
|
|
"epoch": 8.19263970922308,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.218500998839794e-05,
|
|
"loss": 4.5516,
|
|
"mean_token_accuracy": 0.2266480430960655,
|
|
"num_tokens": 18678124.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.233915662765503,
|
|
"epoch": 8.19718309859155,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.197920557163215e-05,
|
|
"loss": 4.514,
|
|
"mean_token_accuracy": 0.23063594102859497,
|
|
"num_tokens": 18689063.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.228711462020874,
|
|
"epoch": 8.201726487960018,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 9.177385274048312e-05,
|
|
"loss": 4.5465,
|
|
"mean_token_accuracy": 0.2295500561594963,
|
|
"num_tokens": 18700118.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.231200265884399,
|
|
"epoch": 8.206269877328488,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.156895200163837e-05,
|
|
"loss": 4.5482,
|
|
"mean_token_accuracy": 0.22415401935577392,
|
|
"num_tokens": 18711683.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.2213465690612795,
|
|
"epoch": 8.210813266696956,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.13645038606703e-05,
|
|
"loss": 4.5126,
|
|
"mean_token_accuracy": 0.22989322394132614,
|
|
"num_tokens": 18722033.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.261202383041382,
|
|
"epoch": 8.215356656065424,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.116050882203427e-05,
|
|
"loss": 4.5417,
|
|
"mean_token_accuracy": 0.22430808395147322,
|
|
"num_tokens": 18732387.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.223684883117675,
|
|
"epoch": 8.219900045433894,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.09569673890678e-05,
|
|
"loss": 4.5324,
|
|
"mean_token_accuracy": 0.2341057389974594,
|
|
"num_tokens": 18743573.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.149905443191528,
|
|
"epoch": 8.224443434802362,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.075388006398907e-05,
|
|
"loss": 4.4267,
|
|
"mean_token_accuracy": 0.23581122010946273,
|
|
"num_tokens": 18753897.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.183023881912232,
|
|
"epoch": 8.228986824170832,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.0551247347896e-05,
|
|
"loss": 4.5608,
|
|
"mean_token_accuracy": 0.2164709284901619,
|
|
"num_tokens": 18764849.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.158408832550049,
|
|
"epoch": 8.2335302135393,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.034906974076449e-05,
|
|
"loss": 4.4294,
|
|
"mean_token_accuracy": 0.23355284333229065,
|
|
"num_tokens": 18776404.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.219209671020508,
|
|
"epoch": 8.23807360290777,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.014734774144784e-05,
|
|
"loss": 4.5101,
|
|
"mean_token_accuracy": 0.2292255014181137,
|
|
"num_tokens": 18786803.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.243441772460938,
|
|
"epoch": 8.242616992276238,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.994608184767491e-05,
|
|
"loss": 4.5513,
|
|
"mean_token_accuracy": 0.2252140685915947,
|
|
"num_tokens": 18797135.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.234616231918335,
|
|
"epoch": 8.247160381644706,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.97452725560495e-05,
|
|
"loss": 4.4334,
|
|
"mean_token_accuracy": 0.237919645011425,
|
|
"num_tokens": 18806495.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.225677871704102,
|
|
"epoch": 8.251703771013176,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.954492036204829e-05,
|
|
"loss": 4.5099,
|
|
"mean_token_accuracy": 0.22612342238426208,
|
|
"num_tokens": 18816925.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.203834390640258,
|
|
"epoch": 8.256247160381644,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.934502576002059e-05,
|
|
"loss": 4.4455,
|
|
"mean_token_accuracy": 0.23522180020809175,
|
|
"num_tokens": 18826604.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.238829278945923,
|
|
"epoch": 8.260790549750114,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.914558924318643e-05,
|
|
"loss": 4.5436,
|
|
"mean_token_accuracy": 0.2314549967646599,
|
|
"num_tokens": 18837278.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.204307937622071,
|
|
"epoch": 8.265333939118582,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.894661130363557e-05,
|
|
"loss": 4.5477,
|
|
"mean_token_accuracy": 0.22338440865278245,
|
|
"num_tokens": 18846588.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.161233949661255,
|
|
"epoch": 8.269877328487052,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.874809243232624e-05,
|
|
"loss": 4.4946,
|
|
"mean_token_accuracy": 0.24197443723678588,
|
|
"num_tokens": 18856345.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.205551338195801,
|
|
"epoch": 8.27442071785552,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.85500331190841e-05,
|
|
"loss": 4.5621,
|
|
"mean_token_accuracy": 0.22074697613716127,
|
|
"num_tokens": 18867000.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.1958507061004635,
|
|
"epoch": 8.27896410722399,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.835243385260083e-05,
|
|
"loss": 4.5554,
|
|
"mean_token_accuracy": 0.22012001574039458,
|
|
"num_tokens": 18876925.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.2248554706573485,
|
|
"epoch": 8.283507496592458,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.815529512043293e-05,
|
|
"loss": 4.4777,
|
|
"mean_token_accuracy": 0.23585545271635056,
|
|
"num_tokens": 18886376.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.194869947433472,
|
|
"epoch": 8.288050885960926,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.795861740900061e-05,
|
|
"loss": 4.4847,
|
|
"mean_token_accuracy": 0.2343673437833786,
|
|
"num_tokens": 18897258.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.2668108463287355,
|
|
"epoch": 8.292594275329396,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.77624012035866e-05,
|
|
"loss": 4.4531,
|
|
"mean_token_accuracy": 0.24057314097881316,
|
|
"num_tokens": 18906959.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.255810499191284,
|
|
"epoch": 8.297137664697864,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.756664698833492e-05,
|
|
"loss": 4.5988,
|
|
"mean_token_accuracy": 0.21885324120521546,
|
|
"num_tokens": 18917930.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.238526821136475,
|
|
"epoch": 8.301681054066334,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.737135524624959e-05,
|
|
"loss": 4.5877,
|
|
"mean_token_accuracy": 0.22492407709360124,
|
|
"num_tokens": 18928506.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.224182891845703,
|
|
"epoch": 8.306224443434802,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.717652645919352e-05,
|
|
"loss": 4.5021,
|
|
"mean_token_accuracy": 0.2363664597272873,
|
|
"num_tokens": 18938811.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.184027147293091,
|
|
"epoch": 8.310767832803272,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.698216110788753e-05,
|
|
"loss": 4.4707,
|
|
"mean_token_accuracy": 0.2326377108693123,
|
|
"num_tokens": 18948487.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.194627714157105,
|
|
"epoch": 8.31531122217174,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.67882596719086e-05,
|
|
"loss": 4.4758,
|
|
"mean_token_accuracy": 0.2332702696323395,
|
|
"num_tokens": 18958021.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.1884771347045895,
|
|
"epoch": 8.319854611540208,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.659482262968943e-05,
|
|
"loss": 4.4989,
|
|
"mean_token_accuracy": 0.2324329599738121,
|
|
"num_tokens": 18967937.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.245235300064087,
|
|
"epoch": 8.324398000908678,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.640185045851653e-05,
|
|
"loss": 4.518,
|
|
"mean_token_accuracy": 0.22789670825004577,
|
|
"num_tokens": 18978119.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.17412486076355,
|
|
"epoch": 8.328941390277146,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.620934363452973e-05,
|
|
"loss": 4.4929,
|
|
"mean_token_accuracy": 0.22842726111412048,
|
|
"num_tokens": 18989036.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.221337985992432,
|
|
"epoch": 8.333484779645616,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 8.601730263272034e-05,
|
|
"loss": 4.5492,
|
|
"mean_token_accuracy": 0.2307686597108841,
|
|
"num_tokens": 18999561.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.162964487075806,
|
|
"epoch": 8.338028169014084,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.58257279269305e-05,
|
|
"loss": 4.4446,
|
|
"mean_token_accuracy": 0.2322291061282158,
|
|
"num_tokens": 19010086.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.132960510253906,
|
|
"epoch": 8.342571558382554,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.563461998985172e-05,
|
|
"loss": 4.4053,
|
|
"mean_token_accuracy": 0.23884500414133072,
|
|
"num_tokens": 19019697.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.24687933921814,
|
|
"epoch": 8.347114947751022,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.544397929302389e-05,
|
|
"loss": 4.5409,
|
|
"mean_token_accuracy": 0.23310733437538148,
|
|
"num_tokens": 19030429.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.231929302215576,
|
|
"epoch": 8.351658337119492,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.525380630683389e-05,
|
|
"loss": 4.5337,
|
|
"mean_token_accuracy": 0.22881446033716202,
|
|
"num_tokens": 19041052.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.157002353668213,
|
|
"epoch": 8.35620172648796,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.506410150051478e-05,
|
|
"loss": 4.4646,
|
|
"mean_token_accuracy": 0.23568211048841475,
|
|
"num_tokens": 19050936.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.170928621292115,
|
|
"epoch": 8.360745115856428,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.487486534214423e-05,
|
|
"loss": 4.4515,
|
|
"mean_token_accuracy": 0.23189760595560074,
|
|
"num_tokens": 19060884.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.195332336425781,
|
|
"epoch": 8.365288505224898,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 8.468609829864366e-05,
|
|
"loss": 4.5056,
|
|
"mean_token_accuracy": 0.23216475695371627,
|
|
"num_tokens": 19071405.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.227158164978027,
|
|
"epoch": 8.369831894593366,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.449780083577697e-05,
|
|
"loss": 4.4867,
|
|
"mean_token_accuracy": 0.2265643373131752,
|
|
"num_tokens": 19081186.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.2565271854400635,
|
|
"epoch": 8.374375283961836,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.430997341814947e-05,
|
|
"loss": 4.5307,
|
|
"mean_token_accuracy": 0.23243416845798492,
|
|
"num_tokens": 19091042.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.21032485961914,
|
|
"epoch": 8.378918673330304,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.41226165092066e-05,
|
|
"loss": 4.4926,
|
|
"mean_token_accuracy": 0.22909659892320633,
|
|
"num_tokens": 19101408.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.18392767906189,
|
|
"epoch": 8.383462062698774,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.393573057123299e-05,
|
|
"loss": 4.483,
|
|
"mean_token_accuracy": 0.23234323859214784,
|
|
"num_tokens": 19111773.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.206689357757568,
|
|
"epoch": 8.388005452067242,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 8.374931606535099e-05,
|
|
"loss": 4.5415,
|
|
"mean_token_accuracy": 0.23048438727855683,
|
|
"num_tokens": 19121532.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.200222730636597,
|
|
"epoch": 8.392548841435712,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.356337345152002e-05,
|
|
"loss": 4.5388,
|
|
"mean_token_accuracy": 0.22230129092931747,
|
|
"num_tokens": 19131389.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.159313344955445,
|
|
"epoch": 8.39709223080418,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.33779031885348e-05,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.23093403428792952,
|
|
"num_tokens": 19141304.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.146255779266357,
|
|
"epoch": 8.401635620172648,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 8.319290573402494e-05,
|
|
"loss": 4.4441,
|
|
"mean_token_accuracy": 0.23340053260326385,
|
|
"num_tokens": 19151039.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.181550025939941,
|
|
"epoch": 8.406179009541118,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.300838154445314e-05,
|
|
"loss": 4.4631,
|
|
"mean_token_accuracy": 0.23341148048639299,
|
|
"num_tokens": 19160909.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.239724969863891,
|
|
"epoch": 8.410722398909586,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.282433107511466e-05,
|
|
"loss": 4.4941,
|
|
"mean_token_accuracy": 0.22765978425741196,
|
|
"num_tokens": 19171906.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.211581134796143,
|
|
"epoch": 8.415265788278056,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.264075478013552e-05,
|
|
"loss": 4.5121,
|
|
"mean_token_accuracy": 0.23408535718917847,
|
|
"num_tokens": 19182688.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.185322427749634,
|
|
"epoch": 8.419809177646524,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.245765311247216e-05,
|
|
"loss": 4.4364,
|
|
"mean_token_accuracy": 0.23226522356271745,
|
|
"num_tokens": 19192429.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.242787790298462,
|
|
"epoch": 8.424352567014994,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.227502652390968e-05,
|
|
"loss": 4.6,
|
|
"mean_token_accuracy": 0.21734050363302232,
|
|
"num_tokens": 19202454.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.3026049613952635,
|
|
"epoch": 8.428895956383462,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.2092875465061e-05,
|
|
"loss": 4.5791,
|
|
"mean_token_accuracy": 0.22171479165554048,
|
|
"num_tokens": 19213057.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.148788070678711,
|
|
"epoch": 8.43343934575193,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 8.191120038536575e-05,
|
|
"loss": 4.4096,
|
|
"mean_token_accuracy": 0.2332531452178955,
|
|
"num_tokens": 19222417.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.228498935699463,
|
|
"epoch": 8.4379827351204,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.173000173308921e-05,
|
|
"loss": 4.5256,
|
|
"mean_token_accuracy": 0.22073906511068345,
|
|
"num_tokens": 19233743.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.246389961242675,
|
|
"epoch": 8.442526124488868,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.154927995532098e-05,
|
|
"loss": 4.64,
|
|
"mean_token_accuracy": 0.2226926028728485,
|
|
"num_tokens": 19245000.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.2616184711456295,
|
|
"epoch": 8.447069513857338,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.136903549797408e-05,
|
|
"loss": 4.5009,
|
|
"mean_token_accuracy": 0.22887718677520752,
|
|
"num_tokens": 19255738.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.244961309432983,
|
|
"epoch": 8.451612903225806,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.118926880578378e-05,
|
|
"loss": 4.5378,
|
|
"mean_token_accuracy": 0.23211107701063155,
|
|
"num_tokens": 19266268.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.343849039077758,
|
|
"epoch": 8.456156292594276,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.100998032230656e-05,
|
|
"loss": 4.5846,
|
|
"mean_token_accuracy": 0.2259268045425415,
|
|
"num_tokens": 19276721.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.214672803878784,
|
|
"epoch": 8.460699681962744,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.083117048991895e-05,
|
|
"loss": 4.4721,
|
|
"mean_token_accuracy": 0.22675121128559111,
|
|
"num_tokens": 19287241.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.240182876586914,
|
|
"epoch": 8.465243071331214,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.065283974981638e-05,
|
|
"loss": 4.5893,
|
|
"mean_token_accuracy": 0.22070799618959427,
|
|
"num_tokens": 19298065.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.275504636764526,
|
|
"epoch": 8.469786460699682,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.047498854201222e-05,
|
|
"loss": 4.568,
|
|
"mean_token_accuracy": 0.22215922474861144,
|
|
"num_tokens": 19308108.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.1969116687774655,
|
|
"epoch": 8.47432985006815,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.029761730533678e-05,
|
|
"loss": 4.5383,
|
|
"mean_token_accuracy": 0.2309856817126274,
|
|
"num_tokens": 19317909.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.24124321937561,
|
|
"epoch": 8.47887323943662,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.012072647743569e-05,
|
|
"loss": 4.5706,
|
|
"mean_token_accuracy": 0.23151241540908812,
|
|
"num_tokens": 19328278.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.163302707672119,
|
|
"epoch": 8.483416628805088,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.994431649476974e-05,
|
|
"loss": 4.4952,
|
|
"mean_token_accuracy": 0.2334277391433716,
|
|
"num_tokens": 19338995.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.2264142513275145,
|
|
"epoch": 8.487960018173558,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.976838779261285e-05,
|
|
"loss": 4.5278,
|
|
"mean_token_accuracy": 0.23228322118520736,
|
|
"num_tokens": 19349177.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.307556867599487,
|
|
"epoch": 8.492503407542026,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.959294080505179e-05,
|
|
"loss": 4.5396,
|
|
"mean_token_accuracy": 0.2259279891848564,
|
|
"num_tokens": 19358871.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.245112037658691,
|
|
"epoch": 8.497046796910496,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.941797596498435e-05,
|
|
"loss": 4.5623,
|
|
"mean_token_accuracy": 0.23470604568719863,
|
|
"num_tokens": 19368594.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.187015151977539,
|
|
"epoch": 8.501590186278964,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.924349370411904e-05,
|
|
"loss": 4.4618,
|
|
"mean_token_accuracy": 0.23584866821765899,
|
|
"num_tokens": 19379973.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.222504949569702,
|
|
"epoch": 8.506133575647432,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.906949445297342e-05,
|
|
"loss": 4.4834,
|
|
"mean_token_accuracy": 0.23530194759368897,
|
|
"num_tokens": 19390762.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.226660680770874,
|
|
"epoch": 8.510676965015902,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.889597864087336e-05,
|
|
"loss": 4.5553,
|
|
"mean_token_accuracy": 0.2319432780146599,
|
|
"num_tokens": 19401918.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.171429920196533,
|
|
"epoch": 8.51522035438437,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 7.872294669595185e-05,
|
|
"loss": 4.4917,
|
|
"mean_token_accuracy": 0.23433414250612258,
|
|
"num_tokens": 19411973.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.194205093383789,
|
|
"epoch": 8.51976374375284,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.855039904514808e-05,
|
|
"loss": 4.4709,
|
|
"mean_token_accuracy": 0.2240083783864975,
|
|
"num_tokens": 19422888.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.300940370559692,
|
|
"epoch": 8.524307133121308,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.837833611420618e-05,
|
|
"loss": 4.5496,
|
|
"mean_token_accuracy": 0.23186347931623458,
|
|
"num_tokens": 19433611.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.147952747344971,
|
|
"epoch": 8.528850522489778,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.820675832767435e-05,
|
|
"loss": 4.4765,
|
|
"mean_token_accuracy": 0.23543111532926558,
|
|
"num_tokens": 19443712.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.172258567810059,
|
|
"epoch": 8.533393911858246,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.803566610890365e-05,
|
|
"loss": 4.3874,
|
|
"mean_token_accuracy": 0.23786530643701553,
|
|
"num_tokens": 19453410.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.168212604522705,
|
|
"epoch": 8.537937301226716,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.786505988004722e-05,
|
|
"loss": 4.4709,
|
|
"mean_token_accuracy": 0.23948812782764434,
|
|
"num_tokens": 19463306.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.23259916305542,
|
|
"epoch": 8.542480690595184,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.769494006205894e-05,
|
|
"loss": 4.5672,
|
|
"mean_token_accuracy": 0.2287536531686783,
|
|
"num_tokens": 19474127.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.2641921043396,
|
|
"epoch": 8.547024079963652,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.752530707469252e-05,
|
|
"loss": 4.5954,
|
|
"mean_token_accuracy": 0.22618156522512436,
|
|
"num_tokens": 19486472.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.210583162307739,
|
|
"epoch": 8.551567469332122,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.73561613365005e-05,
|
|
"loss": 4.5207,
|
|
"mean_token_accuracy": 0.2329728499054909,
|
|
"num_tokens": 19497130.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.197551965713501,
|
|
"epoch": 8.55611085870059,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.718750326483321e-05,
|
|
"loss": 4.4375,
|
|
"mean_token_accuracy": 0.23253639936447143,
|
|
"num_tokens": 19507321.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.261976289749145,
|
|
"epoch": 8.56065424806906,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.701933327583766e-05,
|
|
"loss": 4.5672,
|
|
"mean_token_accuracy": 0.22778965830802916,
|
|
"num_tokens": 19516244.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.193121385574341,
|
|
"epoch": 8.565197637437528,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.685165178445658e-05,
|
|
"loss": 4.457,
|
|
"mean_token_accuracy": 0.24018448144197463,
|
|
"num_tokens": 19526594.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.174620008468628,
|
|
"epoch": 8.569741026805998,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.668445920442732e-05,
|
|
"loss": 4.4753,
|
|
"mean_token_accuracy": 0.22565681487321854,
|
|
"num_tokens": 19536835.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.171035194396973,
|
|
"epoch": 8.574284416174466,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.651775594828112e-05,
|
|
"loss": 4.5746,
|
|
"mean_token_accuracy": 0.21909554451704025,
|
|
"num_tokens": 19546474.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.26349196434021,
|
|
"epoch": 8.578827805542936,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.635154242734146e-05,
|
|
"loss": 4.587,
|
|
"mean_token_accuracy": 0.22266013026237488,
|
|
"num_tokens": 19556604.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.139011478424072,
|
|
"epoch": 8.583371194911404,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.618581905172385e-05,
|
|
"loss": 4.4521,
|
|
"mean_token_accuracy": 0.22887528836727142,
|
|
"num_tokens": 19566824.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.192534446716309,
|
|
"epoch": 8.587914584279872,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.60205862303342e-05,
|
|
"loss": 4.4477,
|
|
"mean_token_accuracy": 0.23357955664396285,
|
|
"num_tokens": 19576917.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.212857723236084,
|
|
"epoch": 8.592457973648342,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.585584437086806e-05,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.22591808140277864,
|
|
"num_tokens": 19587651.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.2172308444976805,
|
|
"epoch": 8.59700136301681,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.569159387980958e-05,
|
|
"loss": 4.5048,
|
|
"mean_token_accuracy": 0.23360133916139603,
|
|
"num_tokens": 19598421.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.157787275314331,
|
|
"epoch": 8.60154475238528,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.552783516243056e-05,
|
|
"loss": 4.458,
|
|
"mean_token_accuracy": 0.23750558346509934,
|
|
"num_tokens": 19608715.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.199044609069825,
|
|
"epoch": 8.606088141753748,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.536456862278936e-05,
|
|
"loss": 4.5173,
|
|
"mean_token_accuracy": 0.22793461829423906,
|
|
"num_tokens": 19619676.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.2325035572052006,
|
|
"epoch": 8.610631531122218,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.520179466372988e-05,
|
|
"loss": 4.5519,
|
|
"mean_token_accuracy": 0.23241727501153947,
|
|
"num_tokens": 19630837.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.153389930725098,
|
|
"epoch": 8.615174920490686,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.503951368688068e-05,
|
|
"loss": 4.3714,
|
|
"mean_token_accuracy": 0.23952197283506393,
|
|
"num_tokens": 19640607.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.2273376941680905,
|
|
"epoch": 8.619718309859154,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.487772609265401e-05,
|
|
"loss": 4.5753,
|
|
"mean_token_accuracy": 0.2292728081345558,
|
|
"num_tokens": 19651354.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.170546293258667,
|
|
"epoch": 8.624261699227624,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.471643228024466e-05,
|
|
"loss": 4.4505,
|
|
"mean_token_accuracy": 0.23841069340705873,
|
|
"num_tokens": 19661266.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.219140672683716,
|
|
"epoch": 8.628805088596092,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.455563264762903e-05,
|
|
"loss": 4.5114,
|
|
"mean_token_accuracy": 0.22709287256002425,
|
|
"num_tokens": 19671275.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 8.628805088596092,
|
|
"eval_entropy": 5.1491661947827,
|
|
"eval_loss": 5.642927646636963,
|
|
"eval_mean_token_accuracy": 0.17594369525133177,
|
|
"eval_num_tokens": 19671275.0,
|
|
"eval_runtime": 2.5764,
|
|
"eval_samples_per_second": 1333.262,
|
|
"eval_steps_per_second": 166.9,
|
|
"step": 9500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"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": 2.8544463752832e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|