68d9c03b17
Model: fpadovani/nld-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.725124943207633,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 9037520831232000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|