68d9c03b17
Model: fpadovani/nld-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.45433893684688775,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1493455435776000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|