6aa7ea38aa
Model: fpadovani/nld-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.8169014084507042,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.856400108337402,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 21.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.2106,
|
|
"mean_token_accuracy": 0.0006149356486275792,
|
|
"num_tokens": 10156.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.860177421569825,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 21.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.2151,
|
|
"mean_token_accuracy": 0.0003687591990455985,
|
|
"num_tokens": 20669.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.876665878295898,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 22.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.0374,
|
|
"mean_token_accuracy": 0.000468046945752576,
|
|
"num_tokens": 30952.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.957574272155762,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 27.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.7887,
|
|
"mean_token_accuracy": 0.0006278944842051714,
|
|
"num_tokens": 41429.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.298445510864258,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 35.0,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.1575,
|
|
"mean_token_accuracy": 0.0006482779106590896,
|
|
"num_tokens": 52166.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.667004203796386,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 36.25,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 12.0099,
|
|
"mean_token_accuracy": 0.00022935778833925723,
|
|
"num_tokens": 61670.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.085978126525879,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 4.0,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.8036,
|
|
"mean_token_accuracy": 0.006481384776998311,
|
|
"num_tokens": 71702.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.696837425231934,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 3.5625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.629,
|
|
"mean_token_accuracy": 0.014822260290384293,
|
|
"num_tokens": 82840.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.674850368499756,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 3.09375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.4551,
|
|
"mean_token_accuracy": 0.017661335971206428,
|
|
"num_tokens": 92290.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.59873275756836,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.2805,
|
|
"mean_token_accuracy": 0.017723128758370878,
|
|
"num_tokens": 101842.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.64423999786377,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.2044,
|
|
"mean_token_accuracy": 0.013444735389202832,
|
|
"num_tokens": 114030.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.725485229492188,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.0243,
|
|
"mean_token_accuracy": 0.02563867699354887,
|
|
"num_tokens": 124349.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.725618076324462,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8587,
|
|
"mean_token_accuracy": 0.03280576281249523,
|
|
"num_tokens": 133534.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.69280939102173,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8074,
|
|
"mean_token_accuracy": 0.019261055253446104,
|
|
"num_tokens": 144383.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.707190322875977,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7725,
|
|
"mean_token_accuracy": 0.028312317933887242,
|
|
"num_tokens": 155467.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.719540405273438,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6826,
|
|
"mean_token_accuracy": 0.05074650049209595,
|
|
"num_tokens": 165903.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.707062530517579,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5739,
|
|
"mean_token_accuracy": 0.04917050153017044,
|
|
"num_tokens": 175812.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.685445880889892,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4801,
|
|
"mean_token_accuracy": 0.04583706185221672,
|
|
"num_tokens": 186357.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.684220504760741,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3766,
|
|
"mean_token_accuracy": 0.04345137663185596,
|
|
"num_tokens": 196124.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.67547779083252,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2652,
|
|
"mean_token_accuracy": 0.04514526352286339,
|
|
"num_tokens": 205655.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.653054904937743,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1821,
|
|
"mean_token_accuracy": 0.050565244257450105,
|
|
"num_tokens": 215870.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.627846431732177,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0638,
|
|
"mean_token_accuracy": 0.05385896116495133,
|
|
"num_tokens": 225027.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.588682270050048,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9803,
|
|
"mean_token_accuracy": 0.05429651960730553,
|
|
"num_tokens": 236352.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.560179710388184,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8435,
|
|
"mean_token_accuracy": 0.0539657898247242,
|
|
"num_tokens": 246713.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.498886871337891,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6823,
|
|
"mean_token_accuracy": 0.05030508264899254,
|
|
"num_tokens": 256950.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.440382480621338,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6331,
|
|
"mean_token_accuracy": 0.05017514526844025,
|
|
"num_tokens": 267345.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.351335620880127,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4422,
|
|
"mean_token_accuracy": 0.054022491723299024,
|
|
"num_tokens": 276782.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.266972255706786,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.3194,
|
|
"mean_token_accuracy": 0.05632990375161171,
|
|
"num_tokens": 287024.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.132027339935302,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2954,
|
|
"mean_token_accuracy": 0.05156242474913597,
|
|
"num_tokens": 296597.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.018408966064452,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1037,
|
|
"mean_token_accuracy": 0.05491906628012657,
|
|
"num_tokens": 307709.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.880233860015869,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.0694,
|
|
"mean_token_accuracy": 0.05425432473421097,
|
|
"num_tokens": 317976.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.71201810836792,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8984,
|
|
"mean_token_accuracy": 0.054642430320382115,
|
|
"num_tokens": 327319.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.520874309539796,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8208,
|
|
"mean_token_accuracy": 0.05670260861515999,
|
|
"num_tokens": 337337.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.264504623413085,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7862,
|
|
"mean_token_accuracy": 0.0514026153832674,
|
|
"num_tokens": 347779.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.049597454071044,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7529,
|
|
"mean_token_accuracy": 0.06217045783996582,
|
|
"num_tokens": 358088.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.819415187835693,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.6422,
|
|
"mean_token_accuracy": 0.059666012227535245,
|
|
"num_tokens": 368314.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.560069179534912,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5159,
|
|
"mean_token_accuracy": 0.07182336524128914,
|
|
"num_tokens": 378406.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.396781253814698,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5248,
|
|
"mean_token_accuracy": 0.07036102451384067,
|
|
"num_tokens": 388912.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.250075912475586,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.5042,
|
|
"mean_token_accuracy": 0.07041607536375523,
|
|
"num_tokens": 399290.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.157392597198486,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4089,
|
|
"mean_token_accuracy": 0.07651857286691666,
|
|
"num_tokens": 409757.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.099771499633789,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5407,
|
|
"mean_token_accuracy": 0.07969405055046082,
|
|
"num_tokens": 420182.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.01094584465027,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.4483,
|
|
"mean_token_accuracy": 0.07581862695515156,
|
|
"num_tokens": 429897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.986596918106079,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3481,
|
|
"mean_token_accuracy": 0.07829344719648361,
|
|
"num_tokens": 441131.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.942036485671997,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3161,
|
|
"mean_token_accuracy": 0.0775233581662178,
|
|
"num_tokens": 451894.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.821108770370484,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2593,
|
|
"mean_token_accuracy": 0.07877892255783081,
|
|
"num_tokens": 462613.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.904732894897461,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3536,
|
|
"mean_token_accuracy": 0.07972405850887299,
|
|
"num_tokens": 471606.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.824862718582153,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2458,
|
|
"mean_token_accuracy": 0.08337627723813057,
|
|
"num_tokens": 481559.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.791320323944092,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2611,
|
|
"mean_token_accuracy": 0.08033445328474045,
|
|
"num_tokens": 492766.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.805171489715576,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3194,
|
|
"mean_token_accuracy": 0.08032750971615314,
|
|
"num_tokens": 503357.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.816235494613648,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1556,
|
|
"mean_token_accuracy": 0.08248411491513252,
|
|
"num_tokens": 513098.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.719041204452514,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2404,
|
|
"mean_token_accuracy": 0.08086299896240234,
|
|
"num_tokens": 522807.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.781749677658081,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.259,
|
|
"mean_token_accuracy": 0.08047396913170815,
|
|
"num_tokens": 532670.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.758613204956054,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2064,
|
|
"mean_token_accuracy": 0.07659858912229538,
|
|
"num_tokens": 543647.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.7667632579803465,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2166,
|
|
"mean_token_accuracy": 0.08752154409885407,
|
|
"num_tokens": 553047.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.720886993408203,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2571,
|
|
"mean_token_accuracy": 0.0864359587430954,
|
|
"num_tokens": 563413.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.675430965423584,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2182,
|
|
"mean_token_accuracy": 0.08417232409119606,
|
|
"num_tokens": 573790.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.737106895446777,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1452,
|
|
"mean_token_accuracy": 0.08157877773046493,
|
|
"num_tokens": 582958.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.618109798431396,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0887,
|
|
"mean_token_accuracy": 0.09229333996772766,
|
|
"num_tokens": 592886.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.615639877319336,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1444,
|
|
"mean_token_accuracy": 0.08253698348999024,
|
|
"num_tokens": 603692.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.630686283111572,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1779,
|
|
"mean_token_accuracy": 0.08081393018364906,
|
|
"num_tokens": 613534.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.684085559844971,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1265,
|
|
"mean_token_accuracy": 0.0869300678372383,
|
|
"num_tokens": 623862.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.708980464935303,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2041,
|
|
"mean_token_accuracy": 0.0886037103831768,
|
|
"num_tokens": 633759.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.514765357971191,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0843,
|
|
"mean_token_accuracy": 0.0886026717722416,
|
|
"num_tokens": 643553.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.49160099029541,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0034,
|
|
"mean_token_accuracy": 0.08772724494338036,
|
|
"num_tokens": 653714.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.63023567199707,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.033,
|
|
"mean_token_accuracy": 0.09123715907335281,
|
|
"num_tokens": 662817.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.539647436141967,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0547,
|
|
"mean_token_accuracy": 0.0879720188677311,
|
|
"num_tokens": 672381.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.440394639968872,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9355,
|
|
"mean_token_accuracy": 0.08899887949228287,
|
|
"num_tokens": 682525.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.532028865814209,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0283,
|
|
"mean_token_accuracy": 0.08837513625621796,
|
|
"num_tokens": 691826.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.473609542846679,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.007,
|
|
"mean_token_accuracy": 0.0950367122888565,
|
|
"num_tokens": 701319.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.459249973297119,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0947,
|
|
"mean_token_accuracy": 0.08535790592432022,
|
|
"num_tokens": 710888.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.56165566444397,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0342,
|
|
"mean_token_accuracy": 0.09012823775410653,
|
|
"num_tokens": 722516.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.527815628051758,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.08682614788413048,
|
|
"num_tokens": 733058.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.45044322013855,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0592,
|
|
"mean_token_accuracy": 0.09242936447262765,
|
|
"num_tokens": 744832.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.463720369338989,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0049,
|
|
"mean_token_accuracy": 0.09356781840324402,
|
|
"num_tokens": 755387.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.43138747215271,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.9355,
|
|
"mean_token_accuracy": 0.09147335067391396,
|
|
"num_tokens": 764455.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.4005152702331545,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.98,
|
|
"mean_token_accuracy": 0.09236079826951027,
|
|
"num_tokens": 774513.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.440145206451416,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0819,
|
|
"mean_token_accuracy": 0.08978344351053238,
|
|
"num_tokens": 785540.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.4839293479919435,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.947,
|
|
"mean_token_accuracy": 0.09177564308047295,
|
|
"num_tokens": 795233.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.431085395812988,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9969,
|
|
"mean_token_accuracy": 0.09167078733444214,
|
|
"num_tokens": 804855.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.446485424041748,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9593,
|
|
"mean_token_accuracy": 0.0934673197567463,
|
|
"num_tokens": 814818.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.352738714218139,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.9751,
|
|
"mean_token_accuracy": 0.09469850584864617,
|
|
"num_tokens": 824622.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.417457246780396,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8664,
|
|
"mean_token_accuracy": 0.09812097549438477,
|
|
"num_tokens": 834813.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.318056726455689,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.9516,
|
|
"mean_token_accuracy": 0.09538300186395646,
|
|
"num_tokens": 844887.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.351348257064819,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9529,
|
|
"mean_token_accuracy": 0.0924244724214077,
|
|
"num_tokens": 855430.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.276928663253784,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.794,
|
|
"mean_token_accuracy": 0.09646925926208497,
|
|
"num_tokens": 866052.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.296294832229615,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.9297,
|
|
"mean_token_accuracy": 0.09176526665687561,
|
|
"num_tokens": 877961.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.363173007965088,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8984,
|
|
"mean_token_accuracy": 0.1007479526102543,
|
|
"num_tokens": 888386.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.295883321762085,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9439,
|
|
"mean_token_accuracy": 0.09971338957548141,
|
|
"num_tokens": 898272.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.353458261489868,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9055,
|
|
"mean_token_accuracy": 0.10149330794811248,
|
|
"num_tokens": 908317.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.281500577926636,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.867,
|
|
"mean_token_accuracy": 0.09293231442570686,
|
|
"num_tokens": 917970.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.282706642150879,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9115,
|
|
"mean_token_accuracy": 0.0950246959924698,
|
|
"num_tokens": 928132.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.30927529335022,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8424,
|
|
"mean_token_accuracy": 0.09944264888763428,
|
|
"num_tokens": 938092.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.37966022491455,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9599,
|
|
"mean_token_accuracy": 0.09508244097232818,
|
|
"num_tokens": 948650.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.15450644493103,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.868,
|
|
"mean_token_accuracy": 0.10101077854633331,
|
|
"num_tokens": 960195.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.254466104507446,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.8956,
|
|
"mean_token_accuracy": 0.10054716616868972,
|
|
"num_tokens": 970608.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.369555997848511,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8689,
|
|
"mean_token_accuracy": 0.09615805298089981,
|
|
"num_tokens": 981053.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.163026285171509,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7827,
|
|
"mean_token_accuracy": 0.10568929314613343,
|
|
"num_tokens": 991560.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.18547511100769,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8472,
|
|
"mean_token_accuracy": 0.09352557733654976,
|
|
"num_tokens": 1002008.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.173881673812867,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8306,
|
|
"mean_token_accuracy": 0.10517893880605697,
|
|
"num_tokens": 1012349.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.266806602478027,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8159,
|
|
"mean_token_accuracy": 0.10154995843768119,
|
|
"num_tokens": 1022886.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 7.165732025545696,
|
|
"eval_loss": 6.923176288604736,
|
|
"eval_mean_token_accuracy": 0.10160443299385004,
|
|
"eval_num_tokens": 1022886.0,
|
|
"eval_runtime": 2.056,
|
|
"eval_samples_per_second": 1670.696,
|
|
"eval_steps_per_second": 209.141,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.297193765640259,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8619,
|
|
"mean_token_accuracy": 0.09589311704039574,
|
|
"num_tokens": 1033129.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.214840221405029,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7604,
|
|
"mean_token_accuracy": 0.10675142779946327,
|
|
"num_tokens": 1043069.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.178305196762085,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.785,
|
|
"mean_token_accuracy": 0.10303612574934959,
|
|
"num_tokens": 1054049.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.15740852355957,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.772,
|
|
"mean_token_accuracy": 0.09831079840660095,
|
|
"num_tokens": 1065155.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.140080070495605,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7699,
|
|
"mean_token_accuracy": 0.1037091501057148,
|
|
"num_tokens": 1075510.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.111771297454834,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7653,
|
|
"mean_token_accuracy": 0.10416601449251175,
|
|
"num_tokens": 1086494.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.242225551605225,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8098,
|
|
"mean_token_accuracy": 0.10401513352990151,
|
|
"num_tokens": 1097311.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.118653678894043,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7668,
|
|
"mean_token_accuracy": 0.10299587845802308,
|
|
"num_tokens": 1107757.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.0831122398376465,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6761,
|
|
"mean_token_accuracy": 0.10623887702822685,
|
|
"num_tokens": 1118889.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.184380054473877,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.8628,
|
|
"mean_token_accuracy": 0.0998254045844078,
|
|
"num_tokens": 1128951.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.096487092971802,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6669,
|
|
"mean_token_accuracy": 0.10797309428453446,
|
|
"num_tokens": 1138583.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.1161802291870115,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7456,
|
|
"mean_token_accuracy": 0.103419478982687,
|
|
"num_tokens": 1149572.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.167421865463257,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7063,
|
|
"mean_token_accuracy": 0.10864525139331818,
|
|
"num_tokens": 1160532.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.055079984664917,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7034,
|
|
"mean_token_accuracy": 0.10719505548477173,
|
|
"num_tokens": 1170915.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.016624212265015,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6622,
|
|
"mean_token_accuracy": 0.1151500552892685,
|
|
"num_tokens": 1180822.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.104376125335693,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7169,
|
|
"mean_token_accuracy": 0.10622723251581193,
|
|
"num_tokens": 1190447.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.050690603256226,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.745,
|
|
"mean_token_accuracy": 0.10602018609642982,
|
|
"num_tokens": 1200951.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.107309436798095,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6341,
|
|
"mean_token_accuracy": 0.11055795922875404,
|
|
"num_tokens": 1211846.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.894921445846558,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5876,
|
|
"mean_token_accuracy": 0.1088622160255909,
|
|
"num_tokens": 1222074.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.963123083114624,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6548,
|
|
"mean_token_accuracy": 0.10943065285682678,
|
|
"num_tokens": 1231741.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.993402624130249,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.5906,
|
|
"mean_token_accuracy": 0.10523484498262406,
|
|
"num_tokens": 1241078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.046900510787964,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7859,
|
|
"mean_token_accuracy": 0.10277008563280106,
|
|
"num_tokens": 1251685.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.997891521453857,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6293,
|
|
"mean_token_accuracy": 0.11614568382501603,
|
|
"num_tokens": 1262044.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.896301555633545,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6362,
|
|
"mean_token_accuracy": 0.10905065536499023,
|
|
"num_tokens": 1273647.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.149247694015503,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6994,
|
|
"mean_token_accuracy": 0.10467459112405778,
|
|
"num_tokens": 1284396.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.961115026473999,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6689,
|
|
"mean_token_accuracy": 0.10689816027879714,
|
|
"num_tokens": 1294339.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.002821254730224,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.597,
|
|
"mean_token_accuracy": 0.11088685020804405,
|
|
"num_tokens": 1304572.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.956146478652954,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.6421,
|
|
"mean_token_accuracy": 0.11130892485380173,
|
|
"num_tokens": 1314780.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.921744728088379,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.64,
|
|
"mean_token_accuracy": 0.10691023096442223,
|
|
"num_tokens": 1325215.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.905644273757934,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5648,
|
|
"mean_token_accuracy": 0.11088442802429199,
|
|
"num_tokens": 1334745.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.0158445835113525,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6651,
|
|
"mean_token_accuracy": 0.11198936477303505,
|
|
"num_tokens": 1345150.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.971886682510376,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6206,
|
|
"mean_token_accuracy": 0.10981507077813149,
|
|
"num_tokens": 1355619.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.895886754989624,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5486,
|
|
"mean_token_accuracy": 0.11562549024820327,
|
|
"num_tokens": 1365933.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.931118965148926,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5762,
|
|
"mean_token_accuracy": 0.11562452912330627,
|
|
"num_tokens": 1376867.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.992132759094238,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.7267,
|
|
"mean_token_accuracy": 0.11105038151144982,
|
|
"num_tokens": 1386739.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.933910131454468,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5837,
|
|
"mean_token_accuracy": 0.11368328630924225,
|
|
"num_tokens": 1397138.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.001461124420166,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6873,
|
|
"mean_token_accuracy": 0.10977592095732688,
|
|
"num_tokens": 1408171.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.838484525680542,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6101,
|
|
"mean_token_accuracy": 0.10415276065468788,
|
|
"num_tokens": 1418259.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.87859959602356,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6169,
|
|
"mean_token_accuracy": 0.11740297675132752,
|
|
"num_tokens": 1428080.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.9925439834594725,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6374,
|
|
"mean_token_accuracy": 0.11313816457986832,
|
|
"num_tokens": 1438502.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.828886651992798,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5904,
|
|
"mean_token_accuracy": 0.11328546851873397,
|
|
"num_tokens": 1448396.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.961853837966919,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6356,
|
|
"mean_token_accuracy": 0.10769107192754745,
|
|
"num_tokens": 1459728.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.863912916183471,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5312,
|
|
"mean_token_accuracy": 0.1191827192902565,
|
|
"num_tokens": 1469733.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.790990257263184,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5308,
|
|
"mean_token_accuracy": 0.11443287208676338,
|
|
"num_tokens": 1479004.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.854830598831176,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4956,
|
|
"mean_token_accuracy": 0.11455681398510933,
|
|
"num_tokens": 1489848.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.813479566574097,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4824,
|
|
"mean_token_accuracy": 0.11726864874362945,
|
|
"num_tokens": 1500877.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.935261297225952,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6316,
|
|
"mean_token_accuracy": 0.11274069175124168,
|
|
"num_tokens": 1511840.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.835966539382935,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5425,
|
|
"mean_token_accuracy": 0.11053763255476952,
|
|
"num_tokens": 1521839.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.827030801773072,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4601,
|
|
"mean_token_accuracy": 0.11425546035170556,
|
|
"num_tokens": 1531285.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.73953218460083,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4915,
|
|
"mean_token_accuracy": 0.11177220493555069,
|
|
"num_tokens": 1542027.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.826250886917114,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5947,
|
|
"mean_token_accuracy": 0.11794949546456338,
|
|
"num_tokens": 1552527.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.8967725276947025,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.639,
|
|
"mean_token_accuracy": 0.11244681403040886,
|
|
"num_tokens": 1562605.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.803361415863037,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5235,
|
|
"mean_token_accuracy": 0.11243330240249634,
|
|
"num_tokens": 1572128.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.795600032806396,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4591,
|
|
"mean_token_accuracy": 0.11609070152044296,
|
|
"num_tokens": 1582352.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.812645721435547,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5547,
|
|
"mean_token_accuracy": 0.11612385958433151,
|
|
"num_tokens": 1592887.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.746767950057984,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.11659797877073289,
|
|
"num_tokens": 1603564.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.6716784000396725,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3529,
|
|
"mean_token_accuracy": 0.12544001638889313,
|
|
"num_tokens": 1613270.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.728840160369873,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4293,
|
|
"mean_token_accuracy": 0.1263292007148266,
|
|
"num_tokens": 1623009.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.746360015869141,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4635,
|
|
"mean_token_accuracy": 0.11441954895853997,
|
|
"num_tokens": 1633209.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.779658365249634,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4894,
|
|
"mean_token_accuracy": 0.1203268215060234,
|
|
"num_tokens": 1643708.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.690767478942871,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5492,
|
|
"mean_token_accuracy": 0.11524337530136108,
|
|
"num_tokens": 1653973.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.796014213562012,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5308,
|
|
"mean_token_accuracy": 0.11834607645869255,
|
|
"num_tokens": 1664856.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.693472051620484,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4518,
|
|
"mean_token_accuracy": 0.11877465546131134,
|
|
"num_tokens": 1675935.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.736241674423217,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.488,
|
|
"mean_token_accuracy": 0.11770976707339287,
|
|
"num_tokens": 1686597.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.7434159278869625,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4418,
|
|
"mean_token_accuracy": 0.11995128318667411,
|
|
"num_tokens": 1697385.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.87018084526062,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4871,
|
|
"mean_token_accuracy": 0.11747646629810334,
|
|
"num_tokens": 1706680.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.659475994110108,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4874,
|
|
"mean_token_accuracy": 0.1250874675810337,
|
|
"num_tokens": 1716130.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.664006137847901,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4137,
|
|
"mean_token_accuracy": 0.11410242542624474,
|
|
"num_tokens": 1726835.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.7605010032653805,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4559,
|
|
"mean_token_accuracy": 0.12729908376932145,
|
|
"num_tokens": 1737708.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.726518821716309,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4323,
|
|
"mean_token_accuracy": 0.12231917381286621,
|
|
"num_tokens": 1747649.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.653599405288697,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4593,
|
|
"mean_token_accuracy": 0.11891063153743744,
|
|
"num_tokens": 1758327.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.721684741973877,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4156,
|
|
"mean_token_accuracy": 0.11827657297253609,
|
|
"num_tokens": 1769767.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.6785807609558105,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5214,
|
|
"mean_token_accuracy": 0.12091457024216652,
|
|
"num_tokens": 1780139.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.760772800445556,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4172,
|
|
"mean_token_accuracy": 0.12346908524632454,
|
|
"num_tokens": 1790783.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.653431367874146,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4131,
|
|
"mean_token_accuracy": 0.11938240155577659,
|
|
"num_tokens": 1801349.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.589143133163452,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.404,
|
|
"mean_token_accuracy": 0.11662679985165596,
|
|
"num_tokens": 1811990.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.688130950927734,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4161,
|
|
"mean_token_accuracy": 0.12564157620072364,
|
|
"num_tokens": 1821725.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.628420114517212,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3812,
|
|
"mean_token_accuracy": 0.11830286234617234,
|
|
"num_tokens": 1831794.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.630347061157226,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4406,
|
|
"mean_token_accuracy": 0.11875352710485458,
|
|
"num_tokens": 1842002.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.713060092926026,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.428,
|
|
"mean_token_accuracy": 0.1267362855374813,
|
|
"num_tokens": 1852282.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.650756549835205,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4456,
|
|
"mean_token_accuracy": 0.11507766097784042,
|
|
"num_tokens": 1862830.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.5850566864013675,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3817,
|
|
"mean_token_accuracy": 0.12454709187150001,
|
|
"num_tokens": 1872746.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.640098190307617,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4152,
|
|
"mean_token_accuracy": 0.1162450760602951,
|
|
"num_tokens": 1884353.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.677035713195801,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4494,
|
|
"mean_token_accuracy": 0.1194159023463726,
|
|
"num_tokens": 1895719.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.643181371688843,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3932,
|
|
"mean_token_accuracy": 0.1279523029923439,
|
|
"num_tokens": 1905905.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.710006141662598,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4895,
|
|
"mean_token_accuracy": 0.1186818890273571,
|
|
"num_tokens": 1916010.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.604852104187012,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3789,
|
|
"mean_token_accuracy": 0.12054645344614982,
|
|
"num_tokens": 1926429.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.486386346817016,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3112,
|
|
"mean_token_accuracy": 0.12320371642708779,
|
|
"num_tokens": 1936322.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.614945411682129,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4767,
|
|
"mean_token_accuracy": 0.12188357040286064,
|
|
"num_tokens": 1947885.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.704386138916016,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4325,
|
|
"mean_token_accuracy": 0.11737854704260826,
|
|
"num_tokens": 1957876.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.615909004211426,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4402,
|
|
"mean_token_accuracy": 0.12112006917595863,
|
|
"num_tokens": 1969531.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.572828912734986,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2983,
|
|
"mean_token_accuracy": 0.12579266652464866,
|
|
"num_tokens": 1980787.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.5395854949951175,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3532,
|
|
"mean_token_accuracy": 0.12586135566234588,
|
|
"num_tokens": 1991645.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.570622396469116,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4259,
|
|
"mean_token_accuracy": 0.12444038391113281,
|
|
"num_tokens": 2002544.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.674864959716797,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4778,
|
|
"mean_token_accuracy": 0.12012744024395942,
|
|
"num_tokens": 2012813.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.624161052703857,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3342,
|
|
"mean_token_accuracy": 0.1272656500339508,
|
|
"num_tokens": 2023756.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.531716966629029,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.415,
|
|
"mean_token_accuracy": 0.12654544711112975,
|
|
"num_tokens": 2034245.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.6605147361755375,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4622,
|
|
"mean_token_accuracy": 0.12089950144290924,
|
|
"num_tokens": 2045635.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.577081489562988,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.4137,
|
|
"mean_token_accuracy": 0.12277993112802506,
|
|
"num_tokens": 2056501.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.601724433898926,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4161,
|
|
"mean_token_accuracy": 0.1239933967590332,
|
|
"num_tokens": 2066540.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.457386301839074,
|
|
"eval_loss": 6.457565784454346,
|
|
"eval_mean_token_accuracy": 0.12266246589810349,
|
|
"eval_num_tokens": 2066540.0,
|
|
"eval_runtime": 2.0261,
|
|
"eval_samples_per_second": 1695.353,
|
|
"eval_steps_per_second": 212.228,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.642101049423218,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.3046,
|
|
"mean_token_accuracy": 0.12406818643212318,
|
|
"num_tokens": 2077008.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.447614336013794,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.2211,
|
|
"mean_token_accuracy": 0.12996889874339104,
|
|
"num_tokens": 2087314.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.564070749282837,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.3952,
|
|
"mean_token_accuracy": 0.12220894247293472,
|
|
"num_tokens": 2097928.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.498920154571533,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.3161,
|
|
"mean_token_accuracy": 0.12189262807369232,
|
|
"num_tokens": 2108662.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.502299547195435,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.3559,
|
|
"mean_token_accuracy": 0.12635889351367952,
|
|
"num_tokens": 2118546.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.61690583229065,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.3872,
|
|
"mean_token_accuracy": 0.1270056627690792,
|
|
"num_tokens": 2128012.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.446464252471924,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.2277,
|
|
"mean_token_accuracy": 0.12683860957622528,
|
|
"num_tokens": 2138241.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.643786096572876,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.3889,
|
|
"mean_token_accuracy": 0.12322892174124718,
|
|
"num_tokens": 2150780.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.422167158126831,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.2808,
|
|
"mean_token_accuracy": 0.12567662224173545,
|
|
"num_tokens": 2161918.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.502702140808106,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.3161,
|
|
"mean_token_accuracy": 0.12407866790890694,
|
|
"num_tokens": 2172742.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.470433187484741,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.2963,
|
|
"mean_token_accuracy": 0.13002401813864709,
|
|
"num_tokens": 2182122.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.475074577331543,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.3398,
|
|
"mean_token_accuracy": 0.12201474159955979,
|
|
"num_tokens": 2193703.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.485757064819336,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.2524,
|
|
"mean_token_accuracy": 0.1314729295670986,
|
|
"num_tokens": 2204041.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.5328117370605465,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.2924,
|
|
"mean_token_accuracy": 0.12847484424710273,
|
|
"num_tokens": 2214276.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.443771505355835,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.2555,
|
|
"mean_token_accuracy": 0.12922132089734079,
|
|
"num_tokens": 2225390.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.469732618331909,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.236,
|
|
"mean_token_accuracy": 0.12894286066293717,
|
|
"num_tokens": 2235708.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.518327808380127,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.13040516525506973,
|
|
"num_tokens": 2246284.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.445136451721192,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.2634,
|
|
"mean_token_accuracy": 0.1280519634485245,
|
|
"num_tokens": 2257219.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.365058374404907,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.2679,
|
|
"mean_token_accuracy": 0.12958364114165305,
|
|
"num_tokens": 2268019.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.465979433059692,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.2853,
|
|
"mean_token_accuracy": 0.12592302560806273,
|
|
"num_tokens": 2278382.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.542138364579943,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 6.1386,
|
|
"mean_token_accuracy": 0.12785813957452774,
|
|
"num_tokens": 2288515.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.407947540283203,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 5.9955,
|
|
"mean_token_accuracy": 0.13868918940424918,
|
|
"num_tokens": 2298396.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.337148284912109,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 6.1121,
|
|
"mean_token_accuracy": 0.13123534470796586,
|
|
"num_tokens": 2308707.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.413614797592163,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 6.0338,
|
|
"mean_token_accuracy": 0.13907074108719825,
|
|
"num_tokens": 2318981.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.406648731231689,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 6.132,
|
|
"mean_token_accuracy": 0.13878810107707978,
|
|
"num_tokens": 2329647.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.378375291824341,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 6.0628,
|
|
"mean_token_accuracy": 0.13353429287672042,
|
|
"num_tokens": 2339449.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.402873182296753,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 6.1569,
|
|
"mean_token_accuracy": 0.13958344906568526,
|
|
"num_tokens": 2349904.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.362394571304321,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 6.0681,
|
|
"mean_token_accuracy": 0.1426156133413315,
|
|
"num_tokens": 2359833.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.438542270660401,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 6.0931,
|
|
"mean_token_accuracy": 0.13478946760296823,
|
|
"num_tokens": 2369857.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.419106435775757,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 6.1102,
|
|
"mean_token_accuracy": 0.13969969153404235,
|
|
"num_tokens": 2380372.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.359906387329102,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 6.0892,
|
|
"mean_token_accuracy": 0.13606216683983802,
|
|
"num_tokens": 2391239.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.311654329299927,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 6.0798,
|
|
"mean_token_accuracy": 0.14339477866888045,
|
|
"num_tokens": 2401178.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.429500484466553,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 6.1488,
|
|
"mean_token_accuracy": 0.13889346718788148,
|
|
"num_tokens": 2412496.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.420183420181274,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 6.1371,
|
|
"mean_token_accuracy": 0.13705583438277244,
|
|
"num_tokens": 2423281.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.307903480529785,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 6.0821,
|
|
"mean_token_accuracy": 0.13973685950040818,
|
|
"num_tokens": 2434078.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.311843490600586,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 6.0074,
|
|
"mean_token_accuracy": 0.14173415526747704,
|
|
"num_tokens": 2443808.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.209487628936768,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.14506323635578156,
|
|
"num_tokens": 2453397.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.438489770889282,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 6.1409,
|
|
"mean_token_accuracy": 0.135777448117733,
|
|
"num_tokens": 2464295.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.37608585357666,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 6.1874,
|
|
"mean_token_accuracy": 0.13546083718538285,
|
|
"num_tokens": 2475311.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.473006963729858,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 6.1026,
|
|
"mean_token_accuracy": 0.13419248089194297,
|
|
"num_tokens": 2485968.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.257586097717285,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 6.1261,
|
|
"mean_token_accuracy": 0.13627575635910033,
|
|
"num_tokens": 2496183.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.336587190628052,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 6.0836,
|
|
"mean_token_accuracy": 0.13671565651893616,
|
|
"num_tokens": 2506242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.3969768524169925,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 6.0852,
|
|
"mean_token_accuracy": 0.1399049386382103,
|
|
"num_tokens": 2516443.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.255747842788696,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 6.0569,
|
|
"mean_token_accuracy": 0.1428102508187294,
|
|
"num_tokens": 2525795.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.344145202636719,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 6.0782,
|
|
"mean_token_accuracy": 0.13670306727290155,
|
|
"num_tokens": 2535046.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.36448302268982,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 6.0756,
|
|
"mean_token_accuracy": 0.1401789128780365,
|
|
"num_tokens": 2546489.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.078287601470947,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 6.0231,
|
|
"mean_token_accuracy": 0.1417386919260025,
|
|
"num_tokens": 2556335.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.46276741027832,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 6.0399,
|
|
"mean_token_accuracy": 0.1342715211212635,
|
|
"num_tokens": 2567647.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.412200593948365,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 6.2511,
|
|
"mean_token_accuracy": 0.12803458347916602,
|
|
"num_tokens": 2578618.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.206548547744751,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 6.0558,
|
|
"mean_token_accuracy": 0.14603335335850715,
|
|
"num_tokens": 2587797.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.319081544876099,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 6.0984,
|
|
"mean_token_accuracy": 0.133018708974123,
|
|
"num_tokens": 2598403.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.366528797149658,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 6.0943,
|
|
"mean_token_accuracy": 0.13550746738910674,
|
|
"num_tokens": 2608296.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.3718188285827635,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 6.1071,
|
|
"mean_token_accuracy": 0.1415480449795723,
|
|
"num_tokens": 2617906.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.257988023757934,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 6.0511,
|
|
"mean_token_accuracy": 0.13821252733469008,
|
|
"num_tokens": 2627699.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.330170059204102,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 6.027,
|
|
"mean_token_accuracy": 0.1415513798594475,
|
|
"num_tokens": 2638384.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.220354127883911,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 5.9593,
|
|
"mean_token_accuracy": 0.14534688517451286,
|
|
"num_tokens": 2648279.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.248266410827637,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 6.0408,
|
|
"mean_token_accuracy": 0.12930047884583473,
|
|
"num_tokens": 2657000.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.254126119613647,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 6.0696,
|
|
"mean_token_accuracy": 0.14027479737997056,
|
|
"num_tokens": 2667760.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.35225100517273,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 6.0009,
|
|
"mean_token_accuracy": 0.14144083932042123,
|
|
"num_tokens": 2678450.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.288917922973633,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 6.1168,
|
|
"mean_token_accuracy": 0.13410752713680268,
|
|
"num_tokens": 2689695.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.342508125305176,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 6.1282,
|
|
"mean_token_accuracy": 0.14176513105630875,
|
|
"num_tokens": 2700414.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.242630481719971,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 5.9502,
|
|
"mean_token_accuracy": 0.15300626754760743,
|
|
"num_tokens": 2709985.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.248297691345215,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 6.0924,
|
|
"mean_token_accuracy": 0.14182935208082198,
|
|
"num_tokens": 2720990.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.269750690460205,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 5.9948,
|
|
"mean_token_accuracy": 0.14390970543026924,
|
|
"num_tokens": 2731285.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.23701663017273,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 6.0861,
|
|
"mean_token_accuracy": 0.14006828367710114,
|
|
"num_tokens": 2741024.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.250706720352173,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 6.0089,
|
|
"mean_token_accuracy": 0.14118139818310738,
|
|
"num_tokens": 2751447.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.288669204711914,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 6.0333,
|
|
"mean_token_accuracy": 0.14222527295351028,
|
|
"num_tokens": 2761283.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.160416793823242,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.9472,
|
|
"mean_token_accuracy": 0.14869927018880844,
|
|
"num_tokens": 2771354.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.327591037750244,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 6.1407,
|
|
"mean_token_accuracy": 0.13928610384464263,
|
|
"num_tokens": 2781063.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.322598695755005,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 6.0377,
|
|
"mean_token_accuracy": 0.14065276831388474,
|
|
"num_tokens": 2791103.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.268867301940918,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 6.1553,
|
|
"mean_token_accuracy": 0.1384136810898781,
|
|
"num_tokens": 2802382.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.20656189918518,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 6.0044,
|
|
"mean_token_accuracy": 0.1394697204232216,
|
|
"num_tokens": 2813740.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.308270502090454,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 6.0168,
|
|
"mean_token_accuracy": 0.13902125507593155,
|
|
"num_tokens": 2824121.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.242033958435059,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 6.0452,
|
|
"mean_token_accuracy": 0.14119891151785852,
|
|
"num_tokens": 2833356.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.288197469711304,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 6.0387,
|
|
"mean_token_accuracy": 0.13836880177259445,
|
|
"num_tokens": 2843839.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.218136358261108,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.13987476527690887,
|
|
"num_tokens": 2853523.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.194594192504883,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 6.0492,
|
|
"mean_token_accuracy": 0.13922794684767723,
|
|
"num_tokens": 2863226.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.40929217338562,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 6.0971,
|
|
"mean_token_accuracy": 0.14521313607692718,
|
|
"num_tokens": 2873822.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.130873823165894,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 5.9524,
|
|
"mean_token_accuracy": 0.14649356976151467,
|
|
"num_tokens": 2883645.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.290006494522094,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 6.071,
|
|
"mean_token_accuracy": 0.13487468957901,
|
|
"num_tokens": 2894397.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.285066986083985,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 6.0467,
|
|
"mean_token_accuracy": 0.14313390702009202,
|
|
"num_tokens": 2904947.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.171815013885498,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.9978,
|
|
"mean_token_accuracy": 0.14585176184773446,
|
|
"num_tokens": 2916183.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.206707763671875,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.9117,
|
|
"mean_token_accuracy": 0.14459551572799684,
|
|
"num_tokens": 2925429.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.226998424530029,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 5.9839,
|
|
"mean_token_accuracy": 0.14731092602014542,
|
|
"num_tokens": 2935142.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.235570096969605,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 6.051,
|
|
"mean_token_accuracy": 0.1416834071278572,
|
|
"num_tokens": 2945151.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.298262405395508,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 6.1376,
|
|
"mean_token_accuracy": 0.13880389630794526,
|
|
"num_tokens": 2956015.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.257140445709228,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 5.9659,
|
|
"mean_token_accuracy": 0.14583468139171601,
|
|
"num_tokens": 2965879.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.238783836364746,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 6.0532,
|
|
"mean_token_accuracy": 0.13990017473697663,
|
|
"num_tokens": 2977021.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.180285978317261,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.9404,
|
|
"mean_token_accuracy": 0.14709279239177703,
|
|
"num_tokens": 2987169.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.32270450592041,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 6.0525,
|
|
"mean_token_accuracy": 0.1423275850713253,
|
|
"num_tokens": 2997666.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.153949642181397,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.9168,
|
|
"mean_token_accuracy": 0.14165920540690422,
|
|
"num_tokens": 3007021.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.16786413192749,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 5.9954,
|
|
"mean_token_accuracy": 0.14898994714021682,
|
|
"num_tokens": 3016557.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.173955011367798,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 5.9605,
|
|
"mean_token_accuracy": 0.14649975001811982,
|
|
"num_tokens": 3026146.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.278796911239624,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 6.0085,
|
|
"mean_token_accuracy": 0.14478534683585167,
|
|
"num_tokens": 3037234.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.171536302566528,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.948,
|
|
"mean_token_accuracy": 0.14746320694684983,
|
|
"num_tokens": 3047282.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.13257532119751,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.9735,
|
|
"mean_token_accuracy": 0.14172942489385604,
|
|
"num_tokens": 3058648.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.366324806213379,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 6.1551,
|
|
"mean_token_accuracy": 0.14091512486338614,
|
|
"num_tokens": 3068996.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.178349304199219,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.9532,
|
|
"mean_token_accuracy": 0.1532169207930565,
|
|
"num_tokens": 3078991.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.131071615219116,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.9267,
|
|
"mean_token_accuracy": 0.14385127127170563,
|
|
"num_tokens": 3090200.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.145272159576416,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 5.9836,
|
|
"mean_token_accuracy": 0.14638550356030464,
|
|
"num_tokens": 3101971.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3625624716038165,
|
|
"eval_entropy": 6.003304544714994,
|
|
"eval_loss": 6.187674522399902,
|
|
"eval_mean_token_accuracy": 0.13976414828279685,
|
|
"eval_num_tokens": 3101971.0,
|
|
"eval_runtime": 2.0245,
|
|
"eval_samples_per_second": 1696.731,
|
|
"eval_steps_per_second": 212.4,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.193512916564941,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 6.05,
|
|
"mean_token_accuracy": 0.14252812564373016,
|
|
"num_tokens": 3113381.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.290583896636963,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 6.0279,
|
|
"mean_token_accuracy": 0.1442758098244667,
|
|
"num_tokens": 3125099.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.086841201782226,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 5.9198,
|
|
"mean_token_accuracy": 0.14686398953199387,
|
|
"num_tokens": 3135007.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.208672046661377,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.9621,
|
|
"mean_token_accuracy": 0.14713747948408126,
|
|
"num_tokens": 3145031.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.197163724899292,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 6.0023,
|
|
"mean_token_accuracy": 0.14530939906835555,
|
|
"num_tokens": 3156044.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.12416296005249,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.9125,
|
|
"mean_token_accuracy": 0.14605395942926408,
|
|
"num_tokens": 3165124.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.086002159118652,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.8588,
|
|
"mean_token_accuracy": 0.15170793682336808,
|
|
"num_tokens": 3175246.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.2570037841796875,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.15199698209762574,
|
|
"num_tokens": 3184817.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.093594932556153,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 5.9871,
|
|
"mean_token_accuracy": 0.13670103028416633,
|
|
"num_tokens": 3195142.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.152914714813233,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.14729512929916383,
|
|
"num_tokens": 3205724.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.198964929580688,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 6.0814,
|
|
"mean_token_accuracy": 0.14055647775530816,
|
|
"num_tokens": 3216919.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.316028022766114,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 6.0752,
|
|
"mean_token_accuracy": 0.14144429787993432,
|
|
"num_tokens": 3228929.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.152643489837646,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 6.0427,
|
|
"mean_token_accuracy": 0.1418978050351143,
|
|
"num_tokens": 3239541.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.203517246246338,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.9156,
|
|
"mean_token_accuracy": 0.14633660465478898,
|
|
"num_tokens": 3249527.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.083546543121338,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.8145,
|
|
"mean_token_accuracy": 0.14704043418169022,
|
|
"num_tokens": 3260497.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.252065896987915,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.14513912349939345,
|
|
"num_tokens": 3270865.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.0464317321777346,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.9389,
|
|
"mean_token_accuracy": 0.14239763393998145,
|
|
"num_tokens": 3281531.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.243140935897827,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 5.9092,
|
|
"mean_token_accuracy": 0.14480504766106606,
|
|
"num_tokens": 3291970.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.072083616256714,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.9548,
|
|
"mean_token_accuracy": 0.14921991974115373,
|
|
"num_tokens": 3301638.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.242140626907348,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 6.0293,
|
|
"mean_token_accuracy": 0.14663881212472915,
|
|
"num_tokens": 3311780.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.103953409194946,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.1474029079079628,
|
|
"num_tokens": 3322096.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.278363752365112,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 6.0609,
|
|
"mean_token_accuracy": 0.1401104636490345,
|
|
"num_tokens": 3332324.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.039387226104736,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 5.8519,
|
|
"mean_token_accuracy": 0.15410613864660264,
|
|
"num_tokens": 3342037.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.2045708179473875,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.14948561191558837,
|
|
"num_tokens": 3353047.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.0782692432403564,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.14826880693435668,
|
|
"num_tokens": 3363841.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.134917449951172,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.9461,
|
|
"mean_token_accuracy": 0.1493829995393753,
|
|
"num_tokens": 3374086.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.147575044631958,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.9277,
|
|
"mean_token_accuracy": 0.1457385838031769,
|
|
"num_tokens": 3384303.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.106037664413452,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.9065,
|
|
"mean_token_accuracy": 0.14968472644686698,
|
|
"num_tokens": 3394858.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.126568031311035,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.921,
|
|
"mean_token_accuracy": 0.1508701652288437,
|
|
"num_tokens": 3405252.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.235639429092407,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.9737,
|
|
"mean_token_accuracy": 0.1500774383544922,
|
|
"num_tokens": 3416007.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.198228073120117,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 6.0023,
|
|
"mean_token_accuracy": 0.145437902957201,
|
|
"num_tokens": 3426932.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.046504831314087,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 5.8834,
|
|
"mean_token_accuracy": 0.15331984907388688,
|
|
"num_tokens": 3437262.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.109807920455933,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.8692,
|
|
"mean_token_accuracy": 0.14938399195671082,
|
|
"num_tokens": 3447525.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.1399143695831295,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 5.9769,
|
|
"mean_token_accuracy": 0.1492663398385048,
|
|
"num_tokens": 3458198.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.119841623306274,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.8413,
|
|
"mean_token_accuracy": 0.15245268642902374,
|
|
"num_tokens": 3468149.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.060994291305542,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.831,
|
|
"mean_token_accuracy": 0.14690444469451905,
|
|
"num_tokens": 3478355.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.075498294830322,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.1500585861504078,
|
|
"num_tokens": 3488726.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.078313875198364,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.8442,
|
|
"mean_token_accuracy": 0.15003334283828734,
|
|
"num_tokens": 3498346.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.047679615020752,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.8935,
|
|
"mean_token_accuracy": 0.15071313083171844,
|
|
"num_tokens": 3508213.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.230852746963501,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 6.0305,
|
|
"mean_token_accuracy": 0.14461363554000856,
|
|
"num_tokens": 3518978.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.110667943954468,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.9616,
|
|
"mean_token_accuracy": 0.14984720945358276,
|
|
"num_tokens": 3529328.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.103526782989502,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.885,
|
|
"mean_token_accuracy": 0.14589395821094514,
|
|
"num_tokens": 3539347.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.139566421508789,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.9158,
|
|
"mean_token_accuracy": 0.13954660817980766,
|
|
"num_tokens": 3549798.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.10515341758728,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 6.0097,
|
|
"mean_token_accuracy": 0.13724498748779296,
|
|
"num_tokens": 3560363.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.176522636413575,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.9223,
|
|
"mean_token_accuracy": 0.14831044971942903,
|
|
"num_tokens": 3570080.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.032936000823975,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.9293,
|
|
"mean_token_accuracy": 0.14602912664413453,
|
|
"num_tokens": 3580053.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.149723958969116,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 6.0176,
|
|
"mean_token_accuracy": 0.14310100600123404,
|
|
"num_tokens": 3590119.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.179568910598755,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.9898,
|
|
"mean_token_accuracy": 0.13816896080970764,
|
|
"num_tokens": 3601809.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.254966831207275,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.8641,
|
|
"mean_token_accuracy": 0.1515677347779274,
|
|
"num_tokens": 3611752.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.020392608642578,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.9126,
|
|
"mean_token_accuracy": 0.14693209081888198,
|
|
"num_tokens": 3621137.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.10863356590271,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.8777,
|
|
"mean_token_accuracy": 0.14968368485569955,
|
|
"num_tokens": 3631567.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.066102790832519,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.845,
|
|
"mean_token_accuracy": 0.14791938215494155,
|
|
"num_tokens": 3641145.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.12293586730957,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.8942,
|
|
"mean_token_accuracy": 0.1458176776766777,
|
|
"num_tokens": 3651716.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.099333572387695,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 5.8407,
|
|
"mean_token_accuracy": 0.15360514372587203,
|
|
"num_tokens": 3662577.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.039719963073731,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.8231,
|
|
"mean_token_accuracy": 0.1538111835718155,
|
|
"num_tokens": 3672984.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.089582061767578,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.9108,
|
|
"mean_token_accuracy": 0.15286104679107665,
|
|
"num_tokens": 3682634.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.070204401016236,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 5.8804,
|
|
"mean_token_accuracy": 0.15201435387134551,
|
|
"num_tokens": 3693022.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.984148263931274,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.757,
|
|
"mean_token_accuracy": 0.15402248799800872,
|
|
"num_tokens": 3702234.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.066392850875855,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.9303,
|
|
"mean_token_accuracy": 0.14461359679698943,
|
|
"num_tokens": 3712235.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.167467546463013,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.8695,
|
|
"mean_token_accuracy": 0.15563700944185258,
|
|
"num_tokens": 3722745.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.976653289794922,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.9161,
|
|
"mean_token_accuracy": 0.14776645004749298,
|
|
"num_tokens": 3732942.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.151669120788574,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.9,
|
|
"mean_token_accuracy": 0.14856889694929123,
|
|
"num_tokens": 3743806.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.032671546936035,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.9028,
|
|
"mean_token_accuracy": 0.15380265712738037,
|
|
"num_tokens": 3755115.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.0663093566894535,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.8501,
|
|
"mean_token_accuracy": 0.15821865797042847,
|
|
"num_tokens": 3765867.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.987074756622315,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.8326,
|
|
"mean_token_accuracy": 0.14925233870744706,
|
|
"num_tokens": 3777071.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.090896701812744,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.8382,
|
|
"mean_token_accuracy": 0.15105062425136567,
|
|
"num_tokens": 3786554.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.056898021697998,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.8785,
|
|
"mean_token_accuracy": 0.15210662931203842,
|
|
"num_tokens": 3797627.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.024951410293579,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.8648,
|
|
"mean_token_accuracy": 0.14908481538295745,
|
|
"num_tokens": 3808841.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.096540880203247,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.8912,
|
|
"mean_token_accuracy": 0.14719861075282098,
|
|
"num_tokens": 3818995.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.090984201431274,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.9448,
|
|
"mean_token_accuracy": 0.14551446288824083,
|
|
"num_tokens": 3828762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.118815183639526,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.14934536665678025,
|
|
"num_tokens": 3838752.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.968673419952393,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.1585403174161911,
|
|
"num_tokens": 3849664.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.095662546157837,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.8264,
|
|
"mean_token_accuracy": 0.15558928102254868,
|
|
"num_tokens": 3859053.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.985531139373779,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.8751,
|
|
"mean_token_accuracy": 0.14425294250249862,
|
|
"num_tokens": 3869654.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.092062377929688,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.8641,
|
|
"mean_token_accuracy": 0.14817842692136765,
|
|
"num_tokens": 3879678.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.999444675445557,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.7796,
|
|
"mean_token_accuracy": 0.15495456159114837,
|
|
"num_tokens": 3889950.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.019551515579224,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.8649,
|
|
"mean_token_accuracy": 0.15243524610996245,
|
|
"num_tokens": 3900095.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.1658735275268555,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.9303,
|
|
"mean_token_accuracy": 0.1504468709230423,
|
|
"num_tokens": 3911034.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.986667251586914,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 5.8907,
|
|
"mean_token_accuracy": 0.14699459969997405,
|
|
"num_tokens": 3922219.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.05580701828003,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.862,
|
|
"mean_token_accuracy": 0.15054864585399627,
|
|
"num_tokens": 3932235.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.985044717788696,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.8085,
|
|
"mean_token_accuracy": 0.15058288127183914,
|
|
"num_tokens": 3941845.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.070745754241943,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.8807,
|
|
"mean_token_accuracy": 0.1437092550098896,
|
|
"num_tokens": 3953392.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.0344319343566895,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.745,
|
|
"mean_token_accuracy": 0.1542268216609955,
|
|
"num_tokens": 3963294.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.939865493774414,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.14705657958984375,
|
|
"num_tokens": 3973660.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.044660520553589,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.8148,
|
|
"mean_token_accuracy": 0.15370839834213257,
|
|
"num_tokens": 3983587.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.041128301620484,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.8705,
|
|
"mean_token_accuracy": 0.1499823346734047,
|
|
"num_tokens": 3994346.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.030775451660157,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.8531,
|
|
"mean_token_accuracy": 0.15244520008563994,
|
|
"num_tokens": 4005234.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.948519134521485,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.7664,
|
|
"mean_token_accuracy": 0.15548717826604844,
|
|
"num_tokens": 4015542.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.020337200164795,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.7732,
|
|
"mean_token_accuracy": 0.15857137143611907,
|
|
"num_tokens": 4024874.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.0677588939666744,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.9227,
|
|
"mean_token_accuracy": 0.15153265595436097,
|
|
"num_tokens": 4035880.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.039951133728027,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.8129,
|
|
"mean_token_accuracy": 0.15864262878894805,
|
|
"num_tokens": 4046484.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.101760339736939,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.8685,
|
|
"mean_token_accuracy": 0.15120580941438674,
|
|
"num_tokens": 4057037.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.0240397453308105,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.9492,
|
|
"mean_token_accuracy": 0.14803495183587073,
|
|
"num_tokens": 4066169.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.03475775718689,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.8525,
|
|
"mean_token_accuracy": 0.14776365533471109,
|
|
"num_tokens": 4076697.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.9979242324829105,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.7369,
|
|
"mean_token_accuracy": 0.15955950319766998,
|
|
"num_tokens": 4087372.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.044879007339477,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.8513,
|
|
"mean_token_accuracy": 0.14841177314519882,
|
|
"num_tokens": 4097653.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.096712970733643,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.789,
|
|
"mean_token_accuracy": 0.15217158794403077,
|
|
"num_tokens": 4107381.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.004293537139892,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.879,
|
|
"mean_token_accuracy": 0.14941695034503938,
|
|
"num_tokens": 4118762.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.994526815414429,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.7885,
|
|
"mean_token_accuracy": 0.1529120072722435,
|
|
"num_tokens": 4129011.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.1039690494537355,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.8606,
|
|
"mean_token_accuracy": 0.14621007740497588,
|
|
"num_tokens": 4138975.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.77301181748856,
|
|
"eval_loss": 5.996906757354736,
|
|
"eval_mean_token_accuracy": 0.15077170481862023,
|
|
"eval_num_tokens": 4138975.0,
|
|
"eval_runtime": 2.0213,
|
|
"eval_samples_per_second": 1699.416,
|
|
"eval_steps_per_second": 212.736,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"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": 891616958576640.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|