6aa7ea38aa
Model: fpadovani/nld-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.6333484779645615,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.0030182838439945,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.8817,
|
|
"mean_token_accuracy": 0.1444647118449211,
|
|
"num_tokens": 4150164.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.977216529846191,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.7721,
|
|
"mean_token_accuracy": 0.15824113339185714,
|
|
"num_tokens": 4159872.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.98731369972229,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.8545,
|
|
"mean_token_accuracy": 0.14761832132935523,
|
|
"num_tokens": 4170665.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.060095405578613,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.7386,
|
|
"mean_token_accuracy": 0.16259402483701707,
|
|
"num_tokens": 4180476.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.02204909324646,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.8705,
|
|
"mean_token_accuracy": 0.15071415081620215,
|
|
"num_tokens": 4190740.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.967234563827515,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.7552,
|
|
"mean_token_accuracy": 0.15657350271940232,
|
|
"num_tokens": 4201504.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.858735990524292,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.7043,
|
|
"mean_token_accuracy": 0.1618433803319931,
|
|
"num_tokens": 4211614.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.028314018249512,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.8137,
|
|
"mean_token_accuracy": 0.15237652361392975,
|
|
"num_tokens": 4221141.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.973033046722412,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.8692,
|
|
"mean_token_accuracy": 0.1551216259598732,
|
|
"num_tokens": 4231877.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.900773811340332,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.6775,
|
|
"mean_token_accuracy": 0.1587761864066124,
|
|
"num_tokens": 4242331.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.9782391548156735,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.6977,
|
|
"mean_token_accuracy": 0.154694864153862,
|
|
"num_tokens": 4252406.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.941512489318848,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.7876,
|
|
"mean_token_accuracy": 0.15253498554229736,
|
|
"num_tokens": 4263036.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.984394598007202,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.7579,
|
|
"mean_token_accuracy": 0.15677025616168977,
|
|
"num_tokens": 4273032.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.966323184967041,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.8364,
|
|
"mean_token_accuracy": 0.15176032930612565,
|
|
"num_tokens": 4283360.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.9785511016845705,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.7799,
|
|
"mean_token_accuracy": 0.15718763768672944,
|
|
"num_tokens": 4292623.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.977948713302612,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.7919,
|
|
"mean_token_accuracy": 0.158698408305645,
|
|
"num_tokens": 4302817.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.035198163986206,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.7915,
|
|
"mean_token_accuracy": 0.1563709110021591,
|
|
"num_tokens": 4313049.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.93516058921814,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.6693,
|
|
"mean_token_accuracy": 0.16123159378767013,
|
|
"num_tokens": 4322873.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.913444948196411,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.8258,
|
|
"mean_token_accuracy": 0.14879332184791566,
|
|
"num_tokens": 4333339.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.007962846755982,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.7947,
|
|
"mean_token_accuracy": 0.15515013188123702,
|
|
"num_tokens": 4344917.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.8405719757080075,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.65,
|
|
"mean_token_accuracy": 0.16141374558210372,
|
|
"num_tokens": 4354140.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.947803544998169,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.8904,
|
|
"mean_token_accuracy": 0.14953781366348268,
|
|
"num_tokens": 4364976.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.978219223022461,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.6633,
|
|
"mean_token_accuracy": 0.1581929862499237,
|
|
"num_tokens": 4376423.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.968920946121216,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.7977,
|
|
"mean_token_accuracy": 0.15216519683599472,
|
|
"num_tokens": 4387865.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.940340948104859,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.7982,
|
|
"mean_token_accuracy": 0.15568334609270096,
|
|
"num_tokens": 4398275.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.0297112464904785,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.7144,
|
|
"mean_token_accuracy": 0.15438816994428634,
|
|
"num_tokens": 4407396.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.145289564132691,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 5.9707,
|
|
"mean_token_accuracy": 0.14780526086688042,
|
|
"num_tokens": 4419682.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.941552352905274,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.8372,
|
|
"mean_token_accuracy": 0.15454522371292115,
|
|
"num_tokens": 4430590.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.9843121528625485,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.7445,
|
|
"mean_token_accuracy": 0.1609554573893547,
|
|
"num_tokens": 4441979.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.954929447174072,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.7625,
|
|
"mean_token_accuracy": 0.15590231418609618,
|
|
"num_tokens": 4451975.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.888317966461182,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.15523761063814162,
|
|
"num_tokens": 4462778.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.031228065490723,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.7566,
|
|
"mean_token_accuracy": 0.16247936636209487,
|
|
"num_tokens": 4472461.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.948527526855469,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.7118,
|
|
"mean_token_accuracy": 0.1612512245774269,
|
|
"num_tokens": 4482507.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.965464496612549,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.7382,
|
|
"mean_token_accuracy": 0.16152099072933196,
|
|
"num_tokens": 4492601.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.95245327949524,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.8439,
|
|
"mean_token_accuracy": 0.15757981091737747,
|
|
"num_tokens": 4503696.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.935918807983398,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.7342,
|
|
"mean_token_accuracy": 0.15598483383655548,
|
|
"num_tokens": 4513649.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.9791747570037845,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.7571,
|
|
"mean_token_accuracy": 0.15890643149614334,
|
|
"num_tokens": 4524054.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.00588173866272,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.7411,
|
|
"mean_token_accuracy": 0.14926089346408844,
|
|
"num_tokens": 4535549.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.926804304122925,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.7689,
|
|
"mean_token_accuracy": 0.15297144651412964,
|
|
"num_tokens": 4545654.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.972972822189331,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.7879,
|
|
"mean_token_accuracy": 0.15464622229337693,
|
|
"num_tokens": 4556230.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.915124787224664,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.5392,
|
|
"mean_token_accuracy": 0.16189659469657475,
|
|
"num_tokens": 4564486.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.868302679061889,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.4421,
|
|
"mean_token_accuracy": 0.16875875890254974,
|
|
"num_tokens": 4575026.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.845345067977905,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.4781,
|
|
"mean_token_accuracy": 0.1636953830718994,
|
|
"num_tokens": 4585902.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.082960891723633,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.5773,
|
|
"mean_token_accuracy": 0.16727212071418762,
|
|
"num_tokens": 4597609.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.8468077182769775,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.4492,
|
|
"mean_token_accuracy": 0.17395626306533812,
|
|
"num_tokens": 4607531.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.864391136169433,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.4745,
|
|
"mean_token_accuracy": 0.16630882173776626,
|
|
"num_tokens": 4618826.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.952245712280273,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.4981,
|
|
"mean_token_accuracy": 0.17101365178823472,
|
|
"num_tokens": 4629526.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.907699346542358,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.52,
|
|
"mean_token_accuracy": 0.1672904148697853,
|
|
"num_tokens": 4639930.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.886648368835449,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.4154,
|
|
"mean_token_accuracy": 0.17574702948331833,
|
|
"num_tokens": 4649661.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.764517688751221,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.4134,
|
|
"mean_token_accuracy": 0.1753751739859581,
|
|
"num_tokens": 4660178.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.907838964462281,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.471,
|
|
"mean_token_accuracy": 0.16770458370447158,
|
|
"num_tokens": 4670417.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.821921634674072,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.1805216059088707,
|
|
"num_tokens": 4680221.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.863324069976807,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.473,
|
|
"mean_token_accuracy": 0.16433386206626893,
|
|
"num_tokens": 4691136.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.720443773269653,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.3165,
|
|
"mean_token_accuracy": 0.18293814212083817,
|
|
"num_tokens": 4700879.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.850162220001221,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.4953,
|
|
"mean_token_accuracy": 0.16432842761278152,
|
|
"num_tokens": 4711477.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.857503032684326,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.17294807881116867,
|
|
"num_tokens": 4721973.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.782174730300904,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.1705305427312851,
|
|
"num_tokens": 4732425.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.843368148803711,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.4453,
|
|
"mean_token_accuracy": 0.17461275309324265,
|
|
"num_tokens": 4743377.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.879863929748535,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.5012,
|
|
"mean_token_accuracy": 0.16959404051303864,
|
|
"num_tokens": 4753534.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.800211668014526,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.4677,
|
|
"mean_token_accuracy": 0.16852423399686814,
|
|
"num_tokens": 4763159.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.798802137374878,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.4238,
|
|
"mean_token_accuracy": 0.17590796947479248,
|
|
"num_tokens": 4773497.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.817171812057495,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.5236,
|
|
"mean_token_accuracy": 0.17162022292613982,
|
|
"num_tokens": 4784291.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.813042688369751,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.4691,
|
|
"mean_token_accuracy": 0.17058789283037185,
|
|
"num_tokens": 4794762.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.961149787902832,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.5994,
|
|
"mean_token_accuracy": 0.1690542832016945,
|
|
"num_tokens": 4804449.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.824402856826782,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.4483,
|
|
"mean_token_accuracy": 0.17529074400663375,
|
|
"num_tokens": 4814172.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.839179754257202,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.5368,
|
|
"mean_token_accuracy": 0.16765836775302886,
|
|
"num_tokens": 4824909.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.90317006111145,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.581,
|
|
"mean_token_accuracy": 0.17135870456695557,
|
|
"num_tokens": 4834546.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.81043848991394,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.4821,
|
|
"mean_token_accuracy": 0.17135401517152787,
|
|
"num_tokens": 4844809.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.8183534145355225,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.4689,
|
|
"mean_token_accuracy": 0.1713588982820511,
|
|
"num_tokens": 4855883.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.844845867156982,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.5044,
|
|
"mean_token_accuracy": 0.173297718167305,
|
|
"num_tokens": 4865936.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.786552810668946,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.17181466668844222,
|
|
"num_tokens": 4876621.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.680826997756958,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.4063,
|
|
"mean_token_accuracy": 0.16907930374145508,
|
|
"num_tokens": 4886315.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.817247343063355,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.4871,
|
|
"mean_token_accuracy": 0.1724187269806862,
|
|
"num_tokens": 4897214.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.777086544036865,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4859,
|
|
"mean_token_accuracy": 0.17181222438812255,
|
|
"num_tokens": 4907360.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.832090520858765,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.4438,
|
|
"mean_token_accuracy": 0.17732187807559968,
|
|
"num_tokens": 4917591.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.823357343673706,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.1624151811003685,
|
|
"num_tokens": 4928075.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.799165534973144,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.4252,
|
|
"mean_token_accuracy": 0.17624630630016327,
|
|
"num_tokens": 4937898.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.806081628799438,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.4877,
|
|
"mean_token_accuracy": 0.16705317199230194,
|
|
"num_tokens": 4948152.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.839706134796143,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.4654,
|
|
"mean_token_accuracy": 0.17072801142930985,
|
|
"num_tokens": 4957973.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.655703830718994,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.4288,
|
|
"mean_token_accuracy": 0.1745622530579567,
|
|
"num_tokens": 4967375.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.7824067115783695,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.433,
|
|
"mean_token_accuracy": 0.175946144759655,
|
|
"num_tokens": 4977286.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.73899655342102,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.4962,
|
|
"mean_token_accuracy": 0.1643753618001938,
|
|
"num_tokens": 4987930.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.910732936859131,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.5566,
|
|
"mean_token_accuracy": 0.1674664482474327,
|
|
"num_tokens": 4998707.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.692483139038086,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.4128,
|
|
"mean_token_accuracy": 0.1799857273697853,
|
|
"num_tokens": 5008540.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.769286012649536,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.4721,
|
|
"mean_token_accuracy": 0.17131308019161223,
|
|
"num_tokens": 5018199.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.860927391052246,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.479,
|
|
"mean_token_accuracy": 0.1737558886408806,
|
|
"num_tokens": 5028251.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.736874771118164,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.4791,
|
|
"mean_token_accuracy": 0.1656083032488823,
|
|
"num_tokens": 5038290.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.770257186889649,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.4409,
|
|
"mean_token_accuracy": 0.1719879612326622,
|
|
"num_tokens": 5048337.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.858885717391968,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.4713,
|
|
"mean_token_accuracy": 0.16860313415527345,
|
|
"num_tokens": 5058533.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.788866710662842,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.4809,
|
|
"mean_token_accuracy": 0.17258588522672652,
|
|
"num_tokens": 5068556.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.765106821060181,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.5676,
|
|
"mean_token_accuracy": 0.16873901337385178,
|
|
"num_tokens": 5079317.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.8879783153533936,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.542,
|
|
"mean_token_accuracy": 0.16907548457384108,
|
|
"num_tokens": 5089544.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.746640348434449,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.5254,
|
|
"mean_token_accuracy": 0.16796262562274933,
|
|
"num_tokens": 5099491.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.87215256690979,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.5314,
|
|
"mean_token_accuracy": 0.16680036932229997,
|
|
"num_tokens": 5109523.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.846458387374878,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.4624,
|
|
"mean_token_accuracy": 0.17200492322444916,
|
|
"num_tokens": 5119940.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.731270360946655,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.4543,
|
|
"mean_token_accuracy": 0.17587835788726808,
|
|
"num_tokens": 5129789.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.773299503326416,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.4736,
|
|
"mean_token_accuracy": 0.16850591599941253,
|
|
"num_tokens": 5139570.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.750662326812744,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.4433,
|
|
"mean_token_accuracy": 0.16742100864648818,
|
|
"num_tokens": 5150526.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.732232856750488,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.3944,
|
|
"mean_token_accuracy": 0.16691581904888153,
|
|
"num_tokens": 5160895.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.711820888519287,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.3854,
|
|
"mean_token_accuracy": 0.17315618991851806,
|
|
"num_tokens": 5171054.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.270786006360745,
|
|
"eval_entropy": 5.636157930729001,
|
|
"eval_loss": 5.90526819229126,
|
|
"eval_mean_token_accuracy": 0.15578120942552423,
|
|
"eval_num_tokens": 5171054.0,
|
|
"eval_runtime": 2.024,
|
|
"eval_samples_per_second": 1697.137,
|
|
"eval_steps_per_second": 212.451,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.849636268615723,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.5413,
|
|
"mean_token_accuracy": 0.15833096504211425,
|
|
"num_tokens": 5180870.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.7907532215118405,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.4966,
|
|
"mean_token_accuracy": 0.1661762699484825,
|
|
"num_tokens": 5190388.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.8149584293365475,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.4615,
|
|
"mean_token_accuracy": 0.16411382257938384,
|
|
"num_tokens": 5200582.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.760423040390014,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.4342,
|
|
"mean_token_accuracy": 0.16865904480218888,
|
|
"num_tokens": 5211079.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.801602220535278,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.5714,
|
|
"mean_token_accuracy": 0.16202916502952575,
|
|
"num_tokens": 5221481.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.815705060958862,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.447,
|
|
"mean_token_accuracy": 0.1744874358177185,
|
|
"num_tokens": 5231387.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.666153478622436,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.3901,
|
|
"mean_token_accuracy": 0.17715386897325516,
|
|
"num_tokens": 5240758.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.777317094802856,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.4235,
|
|
"mean_token_accuracy": 0.17053455710411072,
|
|
"num_tokens": 5250839.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.750581836700439,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.5259,
|
|
"mean_token_accuracy": 0.16632520109415055,
|
|
"num_tokens": 5261554.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.765299463272095,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.4849,
|
|
"mean_token_accuracy": 0.17075166702270508,
|
|
"num_tokens": 5272422.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.768901777267456,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.1710539698600769,
|
|
"num_tokens": 5282522.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.795893335342408,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.17355609238147734,
|
|
"num_tokens": 5293100.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.766564846038818,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.5089,
|
|
"mean_token_accuracy": 0.16715673357248306,
|
|
"num_tokens": 5303103.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.784523344039917,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.4945,
|
|
"mean_token_accuracy": 0.17657329887151718,
|
|
"num_tokens": 5313765.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.85287389755249,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.5624,
|
|
"mean_token_accuracy": 0.16382403075695037,
|
|
"num_tokens": 5324816.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.752615022659302,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.4588,
|
|
"mean_token_accuracy": 0.1738062635064125,
|
|
"num_tokens": 5334882.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.724595737457276,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.4589,
|
|
"mean_token_accuracy": 0.16862281113862992,
|
|
"num_tokens": 5344623.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.8034827709198,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.4573,
|
|
"mean_token_accuracy": 0.17020931392908095,
|
|
"num_tokens": 5355786.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.6630857467651365,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.4676,
|
|
"mean_token_accuracy": 0.16971983313560485,
|
|
"num_tokens": 5366139.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.8910150051116945,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.5624,
|
|
"mean_token_accuracy": 0.1644546926021576,
|
|
"num_tokens": 5376890.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.726341629028321,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.4635,
|
|
"mean_token_accuracy": 0.1742193356156349,
|
|
"num_tokens": 5386602.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.805834245681763,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.5376,
|
|
"mean_token_accuracy": 0.16890167444944382,
|
|
"num_tokens": 5397749.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.771027374267578,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.5064,
|
|
"mean_token_accuracy": 0.16528881788253785,
|
|
"num_tokens": 5407666.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.756225061416626,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.5793,
|
|
"mean_token_accuracy": 0.16117600351572037,
|
|
"num_tokens": 5418668.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.756460332870484,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.4456,
|
|
"mean_token_accuracy": 0.1783042386174202,
|
|
"num_tokens": 5429098.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.706565999984742,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.3989,
|
|
"mean_token_accuracy": 0.17706974297761918,
|
|
"num_tokens": 5438997.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.765311002731323,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.4473,
|
|
"mean_token_accuracy": 0.17235431224107742,
|
|
"num_tokens": 5449315.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.753643417358399,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.16773738861083984,
|
|
"num_tokens": 5460523.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.830755090713501,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.5191,
|
|
"mean_token_accuracy": 0.17407918721437454,
|
|
"num_tokens": 5470012.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.681747961044311,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.4507,
|
|
"mean_token_accuracy": 0.17485994547605516,
|
|
"num_tokens": 5480364.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.875375699996948,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.5932,
|
|
"mean_token_accuracy": 0.1619901478290558,
|
|
"num_tokens": 5492138.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.729516458511353,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.3698,
|
|
"mean_token_accuracy": 0.1795819506049156,
|
|
"num_tokens": 5501935.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.675403738021851,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.5615,
|
|
"mean_token_accuracy": 0.16278308480978013,
|
|
"num_tokens": 5513993.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.840928411483764,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.4896,
|
|
"mean_token_accuracy": 0.172481207549572,
|
|
"num_tokens": 5524822.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.776667928695678,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.1702266275882721,
|
|
"num_tokens": 5535583.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.763628625869751,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.4528,
|
|
"mean_token_accuracy": 0.17762083411216736,
|
|
"num_tokens": 5546199.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.731304073333741,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.4217,
|
|
"mean_token_accuracy": 0.17636795341968536,
|
|
"num_tokens": 5556409.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.746342039108276,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.4949,
|
|
"mean_token_accuracy": 0.17396488785743713,
|
|
"num_tokens": 5565854.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.7632523536682125,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.16098303645849227,
|
|
"num_tokens": 5576194.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.767217111587525,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.435,
|
|
"mean_token_accuracy": 0.17658399194478988,
|
|
"num_tokens": 5586679.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.706618356704712,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.4678,
|
|
"mean_token_accuracy": 0.16140853315591813,
|
|
"num_tokens": 5597268.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.781574392318726,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.4453,
|
|
"mean_token_accuracy": 0.1683684453368187,
|
|
"num_tokens": 5608046.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.765046501159668,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.3535,
|
|
"mean_token_accuracy": 0.17913431525230408,
|
|
"num_tokens": 5617541.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.7153278350830075,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.42,
|
|
"mean_token_accuracy": 0.16958775669336318,
|
|
"num_tokens": 5628428.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.795324802398682,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.611,
|
|
"mean_token_accuracy": 0.1603081412613392,
|
|
"num_tokens": 5639557.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.821906805038452,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.5793,
|
|
"mean_token_accuracy": 0.17082345634698867,
|
|
"num_tokens": 5650586.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.813580751419067,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.17099237740039824,
|
|
"num_tokens": 5661525.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.7512657165527346,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.5151,
|
|
"mean_token_accuracy": 0.17055589258670806,
|
|
"num_tokens": 5671272.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.757832193374634,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.4371,
|
|
"mean_token_accuracy": 0.17337453812360765,
|
|
"num_tokens": 5680384.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.744565677642822,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.4569,
|
|
"mean_token_accuracy": 0.17362446933984757,
|
|
"num_tokens": 5690983.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.731819486618042,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.4706,
|
|
"mean_token_accuracy": 0.16703911125659943,
|
|
"num_tokens": 5702144.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.734309244155884,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.5364,
|
|
"mean_token_accuracy": 0.16374242752790452,
|
|
"num_tokens": 5713283.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.783908367156982,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.4904,
|
|
"mean_token_accuracy": 0.16583993136882783,
|
|
"num_tokens": 5724533.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.721942949295044,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.1701280266046524,
|
|
"num_tokens": 5735186.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.756659650802613,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.5114,
|
|
"mean_token_accuracy": 0.16702331006526946,
|
|
"num_tokens": 5746537.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.775514125823975,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.4821,
|
|
"mean_token_accuracy": 0.1666402906179428,
|
|
"num_tokens": 5756666.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.770417213439941,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.489,
|
|
"mean_token_accuracy": 0.16813197433948518,
|
|
"num_tokens": 5766870.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.729695224761963,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.4768,
|
|
"mean_token_accuracy": 0.1742140159010887,
|
|
"num_tokens": 5776775.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.826140737533569,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.4894,
|
|
"mean_token_accuracy": 0.1699837863445282,
|
|
"num_tokens": 5786993.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.704866170883179,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.4869,
|
|
"mean_token_accuracy": 0.17456516176462172,
|
|
"num_tokens": 5796230.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.8377176284790036,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.4793,
|
|
"mean_token_accuracy": 0.16887702643871308,
|
|
"num_tokens": 5807086.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.723038625717163,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.5184,
|
|
"mean_token_accuracy": 0.16765621900558472,
|
|
"num_tokens": 5817897.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.768254089355469,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.4054,
|
|
"mean_token_accuracy": 0.17909743785858154,
|
|
"num_tokens": 5827794.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.641049146652222,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.4935,
|
|
"mean_token_accuracy": 0.16935296952724457,
|
|
"num_tokens": 5837603.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.736937093734741,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.4024,
|
|
"mean_token_accuracy": 0.17678112238645555,
|
|
"num_tokens": 5847013.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.739947891235351,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.3962,
|
|
"mean_token_accuracy": 0.1710570216178894,
|
|
"num_tokens": 5856783.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.7007232189178465,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.4047,
|
|
"mean_token_accuracy": 0.17469320744276046,
|
|
"num_tokens": 5866241.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.66242561340332,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.4609,
|
|
"mean_token_accuracy": 0.17164679020643234,
|
|
"num_tokens": 5876135.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.7505950927734375,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.5089,
|
|
"mean_token_accuracy": 0.17529316395521163,
|
|
"num_tokens": 5886841.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.752353000640869,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.4435,
|
|
"mean_token_accuracy": 0.17276009321212768,
|
|
"num_tokens": 5896884.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.6858922958374025,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.4951,
|
|
"mean_token_accuracy": 0.17261727750301362,
|
|
"num_tokens": 5906194.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.732201480865479,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.4011,
|
|
"mean_token_accuracy": 0.17907560914754866,
|
|
"num_tokens": 5917217.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.697934770584107,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.3475,
|
|
"mean_token_accuracy": 0.17773282378911973,
|
|
"num_tokens": 5926307.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.704933404922485,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.4538,
|
|
"mean_token_accuracy": 0.17124793231487273,
|
|
"num_tokens": 5936718.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.638150882720947,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.3988,
|
|
"mean_token_accuracy": 0.1745395377278328,
|
|
"num_tokens": 5947870.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.737524318695068,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.5731,
|
|
"mean_token_accuracy": 0.16838356107473373,
|
|
"num_tokens": 5959090.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.822002315521241,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.5495,
|
|
"mean_token_accuracy": 0.16430439352989196,
|
|
"num_tokens": 5970435.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.6967254161834715,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.3781,
|
|
"mean_token_accuracy": 0.18003890067338943,
|
|
"num_tokens": 5979732.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.699376964569092,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.5337,
|
|
"mean_token_accuracy": 0.17351010888814927,
|
|
"num_tokens": 5989965.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.8251628398895265,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.5009,
|
|
"mean_token_accuracy": 0.17705196887254715,
|
|
"num_tokens": 6000752.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.717715692520142,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.4695,
|
|
"mean_token_accuracy": 0.16725421994924544,
|
|
"num_tokens": 6011318.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.7150286674499515,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.4717,
|
|
"mean_token_accuracy": 0.17503757178783416,
|
|
"num_tokens": 6022024.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.652255630493164,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.4324,
|
|
"mean_token_accuracy": 0.17356153279542924,
|
|
"num_tokens": 6032118.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.7666980743408205,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.4456,
|
|
"mean_token_accuracy": 0.17727430313825607,
|
|
"num_tokens": 6042607.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.664867258071899,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.457,
|
|
"mean_token_accuracy": 0.1739836573600769,
|
|
"num_tokens": 6053625.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.716910886764526,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.4276,
|
|
"mean_token_accuracy": 0.17799544036388398,
|
|
"num_tokens": 6063823.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.716484594345093,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.4083,
|
|
"mean_token_accuracy": 0.17423638105392455,
|
|
"num_tokens": 6073914.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.692704772949218,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.3949,
|
|
"mean_token_accuracy": 0.17471350133419036,
|
|
"num_tokens": 6085121.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.614338064193726,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.3177,
|
|
"mean_token_accuracy": 0.1846148192882538,
|
|
"num_tokens": 6095324.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.68001651763916,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.4584,
|
|
"mean_token_accuracy": 0.18262093365192414,
|
|
"num_tokens": 6104704.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.763574266433716,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.5411,
|
|
"mean_token_accuracy": 0.16811863780021669,
|
|
"num_tokens": 6116569.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.7841808795928955,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.5251,
|
|
"mean_token_accuracy": 0.17157238423824311,
|
|
"num_tokens": 6126166.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.709067296981812,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.3965,
|
|
"mean_token_accuracy": 0.17820557951927185,
|
|
"num_tokens": 6136169.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.717296457290649,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.4616,
|
|
"mean_token_accuracy": 0.177069228887558,
|
|
"num_tokens": 6146249.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.830132627487183,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.5397,
|
|
"mean_token_accuracy": 0.1752156972885132,
|
|
"num_tokens": 6156892.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.726925420761108,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.4199,
|
|
"mean_token_accuracy": 0.18360553979873656,
|
|
"num_tokens": 6166915.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.675100183486938,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.3677,
|
|
"mean_token_accuracy": 0.17813336104154587,
|
|
"num_tokens": 6176837.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.628390121459961,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.4544,
|
|
"mean_token_accuracy": 0.17339005023241044,
|
|
"num_tokens": 6188038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.720661163330078,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.4817,
|
|
"mean_token_accuracy": 0.17173578292131425,
|
|
"num_tokens": 6199835.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.7931312084197994,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.3881,
|
|
"mean_token_accuracy": 0.1742375299334526,
|
|
"num_tokens": 6209122.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.51928047801173,
|
|
"eval_loss": 5.816741943359375,
|
|
"eval_mean_token_accuracy": 0.16149290649350298,
|
|
"eval_num_tokens": 6209122.0,
|
|
"eval_runtime": 2.0336,
|
|
"eval_samples_per_second": 1689.087,
|
|
"eval_steps_per_second": 211.443,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.630403614044189,
|
|
"epoch": 2.729668332576102,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004568624876891796,
|
|
"loss": 5.4582,
|
|
"mean_token_accuracy": 0.1750628024339676,
|
|
"num_tokens": 6219317.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.744443941116333,
|
|
"epoch": 2.7342117219445705,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045665416380864587,
|
|
"loss": 5.4982,
|
|
"mean_token_accuracy": 0.1748853400349617,
|
|
"num_tokens": 6229561.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.750857496261597,
|
|
"epoch": 2.7387551113130395,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004564453917145206,
|
|
"loss": 5.47,
|
|
"mean_token_accuracy": 0.17052508145570755,
|
|
"num_tokens": 6240150.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.644581031799317,
|
|
"epoch": 2.7432985006815085,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004562361719219283,
|
|
"loss": 5.3541,
|
|
"mean_token_accuracy": 0.17477230727672577,
|
|
"num_tokens": 6250641.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.684276437759399,
|
|
"epoch": 2.747841890049977,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045602650494709786,
|
|
"loss": 5.4413,
|
|
"mean_token_accuracy": 0.17121798396110535,
|
|
"num_tokens": 6261239.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.697177171707153,
|
|
"epoch": 2.752385279418446,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045581639130736173,
|
|
"loss": 5.4512,
|
|
"mean_token_accuracy": 0.17007984220981598,
|
|
"num_tokens": 6270919.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.7353699684143065,
|
|
"epoch": 2.756928668786915,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004556058315211546,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.1678515136241913,
|
|
"num_tokens": 6283007.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.600628280639649,
|
|
"epoch": 2.761472058155384,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045539482610801156,
|
|
"loss": 5.3734,
|
|
"mean_token_accuracy": 0.18075916171073914,
|
|
"num_tokens": 6293596.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.701785564422607,
|
|
"epoch": 2.766015447523853,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004551833755885677,
|
|
"loss": 5.4211,
|
|
"mean_token_accuracy": 0.1742391288280487,
|
|
"num_tokens": 6304568.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.713755416870117,
|
|
"epoch": 2.7705588368923215,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004549714804845561,
|
|
"loss": 5.3921,
|
|
"mean_token_accuracy": 0.17213882356882096,
|
|
"num_tokens": 6315275.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.740856266021728,
|
|
"epoch": 2.7751022262607905,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004547591413188068,
|
|
"loss": 5.5018,
|
|
"mean_token_accuracy": 0.17504518181085588,
|
|
"num_tokens": 6324986.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.67020320892334,
|
|
"epoch": 2.7796456156292595,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004545463586152456,
|
|
"loss": 5.5076,
|
|
"mean_token_accuracy": 0.16599090695381163,
|
|
"num_tokens": 6336318.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.666765832901001,
|
|
"epoch": 2.784189004997728,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004543331328988927,
|
|
"loss": 5.3917,
|
|
"mean_token_accuracy": 0.1782316729426384,
|
|
"num_tokens": 6345986.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.7263648986816404,
|
|
"epoch": 2.788732394366197,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004541194646958614,
|
|
"loss": 5.4765,
|
|
"mean_token_accuracy": 0.17485484927892686,
|
|
"num_tokens": 6357025.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.699453401565552,
|
|
"epoch": 2.793275783734666,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045390535453335666,
|
|
"loss": 5.4684,
|
|
"mean_token_accuracy": 0.17214902639389038,
|
|
"num_tokens": 6367885.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.580998706817627,
|
|
"epoch": 2.797819173103135,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045369080293967414,
|
|
"loss": 5.3865,
|
|
"mean_token_accuracy": 0.1732836037874222,
|
|
"num_tokens": 6377729.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.644871044158935,
|
|
"epoch": 2.802362562471604,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045347581044419846,
|
|
"loss": 5.3845,
|
|
"mean_token_accuracy": 0.17982030361890794,
|
|
"num_tokens": 6387368.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.755515241622925,
|
|
"epoch": 2.806905951840073,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045326037757740216,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.16899367719888686,
|
|
"num_tokens": 6397771.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.728139019012451,
|
|
"epoch": 2.8114493412085415,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004530445048708447,
|
|
"loss": 5.3671,
|
|
"mean_token_accuracy": 0.18123927861452102,
|
|
"num_tokens": 6408238.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.689693593978882,
|
|
"epoch": 2.8159927305770105,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045282819285717024,
|
|
"loss": 5.5077,
|
|
"mean_token_accuracy": 0.17035059332847596,
|
|
"num_tokens": 6419196.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.6974506855010985,
|
|
"epoch": 2.8205361199454795,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00045261144207010736,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.17225333750247956,
|
|
"num_tokens": 6430342.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.664730215072632,
|
|
"epoch": 2.825079509313948,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045239425304446684,
|
|
"loss": 5.3624,
|
|
"mean_token_accuracy": 0.18250575959682463,
|
|
"num_tokens": 6440218.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.656542062759399,
|
|
"epoch": 2.829622898682417,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004521766263161413,
|
|
"loss": 5.3688,
|
|
"mean_token_accuracy": 0.1843707248568535,
|
|
"num_tokens": 6450128.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.653999519348145,
|
|
"epoch": 2.834166288050886,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000451958562422103,
|
|
"loss": 5.3778,
|
|
"mean_token_accuracy": 0.18543238192796707,
|
|
"num_tokens": 6459916.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.707833099365234,
|
|
"epoch": 2.838709677419355,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045174006190040267,
|
|
"loss": 5.4168,
|
|
"mean_token_accuracy": 0.17574355900287628,
|
|
"num_tokens": 6470250.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.690685749053955,
|
|
"epoch": 2.843253066787824,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000451521125290169,
|
|
"loss": 5.5111,
|
|
"mean_token_accuracy": 0.1676422730088234,
|
|
"num_tokens": 6481623.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.72694411277771,
|
|
"epoch": 2.8477964561562925,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000451301753131606,
|
|
"loss": 5.4067,
|
|
"mean_token_accuracy": 0.17806993573904037,
|
|
"num_tokens": 6492143.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.749693536758423,
|
|
"epoch": 2.8523398455247615,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045108194596599286,
|
|
"loss": 5.4264,
|
|
"mean_token_accuracy": 0.17895904928445816,
|
|
"num_tokens": 6501643.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.61626091003418,
|
|
"epoch": 2.8568832348932305,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004508617043356818,
|
|
"loss": 5.3662,
|
|
"mean_token_accuracy": 0.18171129375696182,
|
|
"num_tokens": 6511442.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.646936321258545,
|
|
"epoch": 2.861426624261699,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004506410287840972,
|
|
"loss": 5.2834,
|
|
"mean_token_accuracy": 0.18167806565761566,
|
|
"num_tokens": 6521303.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.628382539749145,
|
|
"epoch": 2.865970013630168,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000450419919855734,
|
|
"loss": 5.4212,
|
|
"mean_token_accuracy": 0.17254047691822053,
|
|
"num_tokens": 6531143.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.744618225097656,
|
|
"epoch": 2.870513402998637,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004501983780961566,
|
|
"loss": 5.4823,
|
|
"mean_token_accuracy": 0.17173656523227693,
|
|
"num_tokens": 6540899.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.8308022022247314,
|
|
"epoch": 2.875056792367106,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004499764040519972,
|
|
"loss": 5.531,
|
|
"mean_token_accuracy": 0.1689829409122467,
|
|
"num_tokens": 6552131.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.6553051471710205,
|
|
"epoch": 2.879600181735575,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044975399827095467,
|
|
"loss": 5.4437,
|
|
"mean_token_accuracy": 0.16968075484037398,
|
|
"num_tokens": 6563275.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.698030757904053,
|
|
"epoch": 2.8841435711040435,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044953116130179313,
|
|
"loss": 5.3142,
|
|
"mean_token_accuracy": 0.18318966925144195,
|
|
"num_tokens": 6573109.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.569756889343262,
|
|
"epoch": 2.8886869604725125,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004493078936943408,
|
|
"loss": 5.4244,
|
|
"mean_token_accuracy": 0.16793778985738755,
|
|
"num_tokens": 6584978.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.652692461013794,
|
|
"epoch": 2.8932303498409815,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004490841959994882,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.1768659383058548,
|
|
"num_tokens": 6595608.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.64333143234253,
|
|
"epoch": 2.89777373920945,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044886006876918716,
|
|
"loss": 5.3221,
|
|
"mean_token_accuracy": 0.1744397297501564,
|
|
"num_tokens": 6605983.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.658440685272216,
|
|
"epoch": 2.902317128577919,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004486355125564493,
|
|
"loss": 5.4439,
|
|
"mean_token_accuracy": 0.168673013150692,
|
|
"num_tokens": 6616682.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.8313366889953615,
|
|
"epoch": 2.906860517946388,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004484105279153449,
|
|
"loss": 5.466,
|
|
"mean_token_accuracy": 0.17580256462097169,
|
|
"num_tokens": 6628715.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.690737056732178,
|
|
"epoch": 2.911403907314857,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000448185115401001,
|
|
"loss": 5.4531,
|
|
"mean_token_accuracy": 0.171098530292511,
|
|
"num_tokens": 6639219.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.6329793453216555,
|
|
"epoch": 2.915947296683326,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004479592755696008,
|
|
"loss": 5.3503,
|
|
"mean_token_accuracy": 0.1818231984972954,
|
|
"num_tokens": 6649516.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.746676778793335,
|
|
"epoch": 2.9204906860517945,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044773300897838137,
|
|
"loss": 5.4603,
|
|
"mean_token_accuracy": 0.17531079798936844,
|
|
"num_tokens": 6659470.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.630096864700318,
|
|
"epoch": 2.9250340754202635,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004475063161856332,
|
|
"loss": 5.4167,
|
|
"mean_token_accuracy": 0.17711722254753112,
|
|
"num_tokens": 6669637.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.599526596069336,
|
|
"epoch": 2.9295774647887325,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004472791977506983,
|
|
"loss": 5.3779,
|
|
"mean_token_accuracy": 0.1744503155350685,
|
|
"num_tokens": 6680450.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.707645654678345,
|
|
"epoch": 2.934120854157201,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044705165423396855,
|
|
"loss": 5.4683,
|
|
"mean_token_accuracy": 0.17224285751581192,
|
|
"num_tokens": 6691526.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.701753950119018,
|
|
"epoch": 2.93866424352567,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044682368619688523,
|
|
"loss": 5.403,
|
|
"mean_token_accuracy": 0.17708637714385986,
|
|
"num_tokens": 6702543.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.6747335433959964,
|
|
"epoch": 2.943207632894139,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004465952942019367,
|
|
"loss": 5.4879,
|
|
"mean_token_accuracy": 0.1681745857000351,
|
|
"num_tokens": 6712374.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.70989351272583,
|
|
"epoch": 2.947751022262608,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044636647881265745,
|
|
"loss": 5.3799,
|
|
"mean_token_accuracy": 0.17257385700941086,
|
|
"num_tokens": 6722091.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.631851482391357,
|
|
"epoch": 2.952294411631077,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004461372405936269,
|
|
"loss": 5.3637,
|
|
"mean_token_accuracy": 0.18677607774734498,
|
|
"num_tokens": 6732024.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.725279092788696,
|
|
"epoch": 2.9568378009995455,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004459075801104673,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.17148397564888002,
|
|
"num_tokens": 6741480.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.711555862426758,
|
|
"epoch": 2.9613811903680145,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044567749792984325,
|
|
"loss": 5.3616,
|
|
"mean_token_accuracy": 0.17898045629262924,
|
|
"num_tokens": 6752504.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.685667896270752,
|
|
"epoch": 2.9659245797364835,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044544699461945966,
|
|
"loss": 5.4192,
|
|
"mean_token_accuracy": 0.1693477913737297,
|
|
"num_tokens": 6762945.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.635536813735962,
|
|
"epoch": 2.970467969104952,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044521607074806037,
|
|
"loss": 5.3931,
|
|
"mean_token_accuracy": 0.17766736894845964,
|
|
"num_tokens": 6772862.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.729037952423096,
|
|
"epoch": 2.975011358473421,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004449847268854272,
|
|
"loss": 5.4192,
|
|
"mean_token_accuracy": 0.1795768156647682,
|
|
"num_tokens": 6782912.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.732499122619629,
|
|
"epoch": 2.97955474784189,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004447529636023782,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.17771411538124085,
|
|
"num_tokens": 6793249.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.687345933914185,
|
|
"epoch": 2.984098137210359,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00044452078147076614,
|
|
"loss": 5.4553,
|
|
"mean_token_accuracy": 0.1770368903875351,
|
|
"num_tokens": 6803721.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.74183497428894,
|
|
"epoch": 2.988641526578828,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004442881810634774,
|
|
"loss": 5.4182,
|
|
"mean_token_accuracy": 0.1799023449420929,
|
|
"num_tokens": 6813637.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.662044906616211,
|
|
"epoch": 2.9931849159472965,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004440551629544303,
|
|
"loss": 5.4269,
|
|
"mean_token_accuracy": 0.17125516533851623,
|
|
"num_tokens": 6824171.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.615179538726807,
|
|
"epoch": 2.9977283053157655,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044382172771857383,
|
|
"loss": 5.2838,
|
|
"mean_token_accuracy": 0.18038785606622695,
|
|
"num_tokens": 6833668.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.607766734229194,
|
|
"epoch": 3.0018173557473875,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004435878759318864,
|
|
"loss": 5.2214,
|
|
"mean_token_accuracy": 0.18619143135017818,
|
|
"num_tokens": 6842297.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.708400058746338,
|
|
"epoch": 3.0063607451158565,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004433536081713738,
|
|
"loss": 5.1965,
|
|
"mean_token_accuracy": 0.18496198803186417,
|
|
"num_tokens": 6853124.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.683142566680909,
|
|
"epoch": 3.0109041344843255,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004431189250150686,
|
|
"loss": 5.0761,
|
|
"mean_token_accuracy": 0.19732627719640733,
|
|
"num_tokens": 6861990.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.643543529510498,
|
|
"epoch": 3.015447523852794,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044288382704202816,
|
|
"loss": 5.0725,
|
|
"mean_token_accuracy": 0.1970564529299736,
|
|
"num_tokens": 6872024.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.6079028129577635,
|
|
"epoch": 3.019990913221263,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044264831483233335,
|
|
"loss": 5.1409,
|
|
"mean_token_accuracy": 0.19019072204828263,
|
|
"num_tokens": 6881205.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.531577253341675,
|
|
"epoch": 3.024534302589732,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044241238896708716,
|
|
"loss": 5.0945,
|
|
"mean_token_accuracy": 0.19641977101564406,
|
|
"num_tokens": 6891741.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.542751836776733,
|
|
"epoch": 3.029077691958201,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004421760500284131,
|
|
"loss": 5.1675,
|
|
"mean_token_accuracy": 0.18492870330810546,
|
|
"num_tokens": 6902122.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.643613529205322,
|
|
"epoch": 3.0336210813266695,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004419392985994541,
|
|
"loss": 5.1294,
|
|
"mean_token_accuracy": 0.18925008177757263,
|
|
"num_tokens": 6912129.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.559886693954468,
|
|
"epoch": 3.0381644706951385,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044170213526437076,
|
|
"loss": 5.0153,
|
|
"mean_token_accuracy": 0.19870110154151915,
|
|
"num_tokens": 6921409.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.573070955276489,
|
|
"epoch": 3.0427078600636075,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044146456060833994,
|
|
"loss": 5.1324,
|
|
"mean_token_accuracy": 0.19372012466192245,
|
|
"num_tokens": 6932346.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.650528621673584,
|
|
"epoch": 3.0472512494320765,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044122657521755373,
|
|
"loss": 5.144,
|
|
"mean_token_accuracy": 0.1928184673190117,
|
|
"num_tokens": 6942139.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.623360443115234,
|
|
"epoch": 3.051794638800545,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044098817967921716,
|
|
"loss": 5.1476,
|
|
"mean_token_accuracy": 0.19087578505277633,
|
|
"num_tokens": 6952344.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.651640224456787,
|
|
"epoch": 3.056338028169014,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044074937458154763,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.19229988306760787,
|
|
"num_tokens": 6962430.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.654034042358399,
|
|
"epoch": 3.060881417537483,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00044051016051377303,
|
|
"loss": 5.1799,
|
|
"mean_token_accuracy": 0.18911581188440324,
|
|
"num_tokens": 6973213.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.538857269287109,
|
|
"epoch": 3.065424806905952,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004402705380661302,
|
|
"loss": 5.13,
|
|
"mean_token_accuracy": 0.18752226531505584,
|
|
"num_tokens": 6983896.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.611310291290283,
|
|
"epoch": 3.0699681962744205,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044003050782986377,
|
|
"loss": 5.1092,
|
|
"mean_token_accuracy": 0.19607766568660737,
|
|
"num_tokens": 6993549.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.579682397842407,
|
|
"epoch": 3.0745115856428895,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004397900703972245,
|
|
"loss": 5.1446,
|
|
"mean_token_accuracy": 0.19014745205640793,
|
|
"num_tokens": 7003893.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.648293733596802,
|
|
"epoch": 3.0790549750113585,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004395492263614677,
|
|
"loss": 5.2424,
|
|
"mean_token_accuracy": 0.18216302543878554,
|
|
"num_tokens": 7014826.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.678756046295166,
|
|
"epoch": 3.0835983643798275,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043930797631685234,
|
|
"loss": 5.1681,
|
|
"mean_token_accuracy": 0.19137286394834518,
|
|
"num_tokens": 7024758.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.5866437435150145,
|
|
"epoch": 3.088141753748296,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043906632085863876,
|
|
"loss": 5.136,
|
|
"mean_token_accuracy": 0.19628849029541015,
|
|
"num_tokens": 7034063.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.568763542175293,
|
|
"epoch": 3.092685143116765,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043882426058308775,
|
|
"loss": 5.1336,
|
|
"mean_token_accuracy": 0.18538877815008165,
|
|
"num_tokens": 7045137.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.520676231384277,
|
|
"epoch": 3.097228532485234,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00043858179608745915,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.19561208188533782,
|
|
"num_tokens": 7056381.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.698847913742066,
|
|
"epoch": 3.101771921853703,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004383389279700098,
|
|
"loss": 5.2015,
|
|
"mean_token_accuracy": 0.18250014185905455,
|
|
"num_tokens": 7066462.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.671674394607544,
|
|
"epoch": 3.1063153112221715,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043809565682999274,
|
|
"loss": 5.2231,
|
|
"mean_token_accuracy": 0.18543293178081513,
|
|
"num_tokens": 7078438.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.60929307937622,
|
|
"epoch": 3.1108587005906405,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00043785198326765535,
|
|
"loss": 5.1001,
|
|
"mean_token_accuracy": 0.192493799328804,
|
|
"num_tokens": 7088981.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.565306806564331,
|
|
"epoch": 3.1154020899591095,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043760790788423785,
|
|
"loss": 5.1574,
|
|
"mean_token_accuracy": 0.19066230803728104,
|
|
"num_tokens": 7099407.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.564446830749512,
|
|
"epoch": 3.1199454793275785,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043736343128197207,
|
|
"loss": 5.0399,
|
|
"mean_token_accuracy": 0.2005670338869095,
|
|
"num_tokens": 7109454.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.5073576927185055,
|
|
"epoch": 3.124488868696047,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004371185540640796,
|
|
"loss": 5.1398,
|
|
"mean_token_accuracy": 0.18948666602373124,
|
|
"num_tokens": 7119639.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.62153434753418,
|
|
"epoch": 3.129032258064516,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043687327683477077,
|
|
"loss": 5.1077,
|
|
"mean_token_accuracy": 0.19161372035741805,
|
|
"num_tokens": 7129872.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.553510665893555,
|
|
"epoch": 3.133575647432985,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004366276001992427,
|
|
"loss": 5.0572,
|
|
"mean_token_accuracy": 0.1962740823626518,
|
|
"num_tokens": 7140456.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.492085123062134,
|
|
"epoch": 3.138119036801454,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000436381524763678,
|
|
"loss": 5.0866,
|
|
"mean_token_accuracy": 0.19258544892072677,
|
|
"num_tokens": 7149787.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.56104006767273,
|
|
"epoch": 3.142662426169923,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004361350511352435,
|
|
"loss": 5.0703,
|
|
"mean_token_accuracy": 0.19688417315483092,
|
|
"num_tokens": 7160482.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.643281745910644,
|
|
"epoch": 3.1472058155383915,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004358881799220882,
|
|
"loss": 5.1104,
|
|
"mean_token_accuracy": 0.1937023714184761,
|
|
"num_tokens": 7170818.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.535250854492188,
|
|
"epoch": 3.1517492049068605,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043564091173334236,
|
|
"loss": 5.1345,
|
|
"mean_token_accuracy": 0.1935590460896492,
|
|
"num_tokens": 7180770.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.527993488311767,
|
|
"epoch": 3.1562925942753295,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043539324717911573,
|
|
"loss": 5.0564,
|
|
"mean_token_accuracy": 0.1996495321393013,
|
|
"num_tokens": 7190751.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.549295473098755,
|
|
"epoch": 3.1608359836437985,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043514518687049587,
|
|
"loss": 5.0693,
|
|
"mean_token_accuracy": 0.20097184926271439,
|
|
"num_tokens": 7200093.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.624794006347656,
|
|
"epoch": 3.165379373012267,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000434896731419547,
|
|
"loss": 5.1693,
|
|
"mean_token_accuracy": 0.19156304150819778,
|
|
"num_tokens": 7210526.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.512590074539185,
|
|
"epoch": 3.169922762380736,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004346478814393082,
|
|
"loss": 5.1222,
|
|
"mean_token_accuracy": 0.19338342547416687,
|
|
"num_tokens": 7222127.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.549171590805054,
|
|
"epoch": 3.174466151749205,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004343986375437921,
|
|
"loss": 5.0886,
|
|
"mean_token_accuracy": 0.19761230945587158,
|
|
"num_tokens": 7233418.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.546758079528809,
|
|
"epoch": 3.179009541117674,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043414900034798314,
|
|
"loss": 5.1424,
|
|
"mean_token_accuracy": 0.19149672836065293,
|
|
"num_tokens": 7243743.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.179009541117674,
|
|
"eval_entropy": 5.3101175984670945,
|
|
"eval_loss": 5.796391487121582,
|
|
"eval_mean_token_accuracy": 0.16366152772030165,
|
|
"eval_num_tokens": 7243743.0,
|
|
"eval_runtime": 2.0265,
|
|
"eval_samples_per_second": 1695.046,
|
|
"eval_steps_per_second": 212.189,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.5297328472137455,
|
|
"epoch": 3.1835529304861425,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004338989704678364,
|
|
"loss": 5.1766,
|
|
"mean_token_accuracy": 0.18635060936212539,
|
|
"num_tokens": 7254906.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.581528568267823,
|
|
"epoch": 3.1880963198546115,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043364854852027566,
|
|
"loss": 5.1295,
|
|
"mean_token_accuracy": 0.1920384258031845,
|
|
"num_tokens": 7264193.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.64374623298645,
|
|
"epoch": 3.1926397092230805,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004333977351231924,
|
|
"loss": 5.1828,
|
|
"mean_token_accuracy": 0.1904108554124832,
|
|
"num_tokens": 7273723.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.548360109329224,
|
|
"epoch": 3.1971830985915495,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043314653089544336,
|
|
"loss": 5.1979,
|
|
"mean_token_accuracy": 0.18939443975687026,
|
|
"num_tokens": 7284142.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.621256923675537,
|
|
"epoch": 3.201726487960018,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004328949364568504,
|
|
"loss": 5.0954,
|
|
"mean_token_accuracy": 0.19565156549215318,
|
|
"num_tokens": 7294228.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.573927927017212,
|
|
"epoch": 3.206269877328487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004326429524281975,
|
|
"loss": 5.1195,
|
|
"mean_token_accuracy": 0.191754150390625,
|
|
"num_tokens": 7304555.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.5959819793701175,
|
|
"epoch": 3.210813266696956,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004323905794312303,
|
|
"loss": 5.2104,
|
|
"mean_token_accuracy": 0.1781022861599922,
|
|
"num_tokens": 7315955.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.5397686004638675,
|
|
"epoch": 3.215356656065425,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043213781808865414,
|
|
"loss": 5.1518,
|
|
"mean_token_accuracy": 0.1881447434425354,
|
|
"num_tokens": 7326757.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.499069261550903,
|
|
"epoch": 3.2199000454338935,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004318846690241324,
|
|
"loss": 5.13,
|
|
"mean_token_accuracy": 0.19944306910037995,
|
|
"num_tokens": 7335969.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.553613758087158,
|
|
"epoch": 3.2244434348023625,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043163113286228533,
|
|
"loss": 5.1173,
|
|
"mean_token_accuracy": 0.19021477550268173,
|
|
"num_tokens": 7345824.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.624339914321899,
|
|
"epoch": 3.2289868241708315,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004313772102286882,
|
|
"loss": 5.21,
|
|
"mean_token_accuracy": 0.18719030320644378,
|
|
"num_tokens": 7356292.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.562908792495728,
|
|
"epoch": 3.2335302135393005,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004311229017498698,
|
|
"loss": 5.1838,
|
|
"mean_token_accuracy": 0.1888285219669342,
|
|
"num_tokens": 7366809.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.5770941257476805,
|
|
"epoch": 3.238073602907769,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000430868208053311,
|
|
"loss": 5.1184,
|
|
"mean_token_accuracy": 0.19796868562698364,
|
|
"num_tokens": 7376076.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.539325284957886,
|
|
"epoch": 3.242616992276238,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043061312976744326,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.18888795524835586,
|
|
"num_tokens": 7386338.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.5191186428070065,
|
|
"epoch": 3.247160381644707,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004303576675216469,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.1901090145111084,
|
|
"num_tokens": 7395792.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.534206247329712,
|
|
"epoch": 3.251703771013176,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043010182194624957,
|
|
"loss": 5.1388,
|
|
"mean_token_accuracy": 0.1897549256682396,
|
|
"num_tokens": 7405765.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.518572235107422,
|
|
"epoch": 3.2562471603816445,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004298455936725248,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.18658362030982972,
|
|
"num_tokens": 7415213.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.646478748321533,
|
|
"epoch": 3.2607905497501135,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004295889833326903,
|
|
"loss": 5.1625,
|
|
"mean_token_accuracy": 0.18602795898914337,
|
|
"num_tokens": 7425722.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.628579378128052,
|
|
"epoch": 3.2653339391185825,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00042933199155990675,
|
|
"loss": 5.2156,
|
|
"mean_token_accuracy": 0.19165341556072235,
|
|
"num_tokens": 7436240.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.516749715805053,
|
|
"epoch": 3.2698773284870515,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004290746189882757,
|
|
"loss": 5.1655,
|
|
"mean_token_accuracy": 0.18851610124111176,
|
|
"num_tokens": 7447878.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.58643102645874,
|
|
"epoch": 3.27442071785552,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042881686625283844,
|
|
"loss": 5.16,
|
|
"mean_token_accuracy": 0.19029032588005065,
|
|
"num_tokens": 7458635.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.596550846099854,
|
|
"epoch": 3.278964107223989,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004285587339895742,
|
|
"loss": 5.1862,
|
|
"mean_token_accuracy": 0.1820325121283531,
|
|
"num_tokens": 7468740.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.663819265365601,
|
|
"epoch": 3.283507496592458,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042830022283539863,
|
|
"loss": 5.2418,
|
|
"mean_token_accuracy": 0.18801323175430298,
|
|
"num_tokens": 7480977.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.554131126403808,
|
|
"epoch": 3.288050885960927,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004280413334281623,
|
|
"loss": 5.2345,
|
|
"mean_token_accuracy": 0.18782038539648055,
|
|
"num_tokens": 7491758.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.631022787094116,
|
|
"epoch": 3.292594275329396,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042778206640664915,
|
|
"loss": 5.1859,
|
|
"mean_token_accuracy": 0.1864585369825363,
|
|
"num_tokens": 7500486.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.494282627105713,
|
|
"epoch": 3.2971376646978645,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004275224224105748,
|
|
"loss": 5.1563,
|
|
"mean_token_accuracy": 0.19796231836080552,
|
|
"num_tokens": 7510371.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.585437297821045,
|
|
"epoch": 3.3016810540663335,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042726240208058484,
|
|
"loss": 5.2018,
|
|
"mean_token_accuracy": 0.19522295743227006,
|
|
"num_tokens": 7520697.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.531649971008301,
|
|
"epoch": 3.3062244434348025,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004270020060582538,
|
|
"loss": 5.1262,
|
|
"mean_token_accuracy": 0.18982232958078385,
|
|
"num_tokens": 7530757.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.5757252216339115,
|
|
"epoch": 3.310767832803271,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00042674123498608285,
|
|
"loss": 5.1564,
|
|
"mean_token_accuracy": 0.18989616334438325,
|
|
"num_tokens": 7541810.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.610417318344116,
|
|
"epoch": 3.31531122217174,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042648008950749877,
|
|
"loss": 5.344,
|
|
"mean_token_accuracy": 0.17718006819486617,
|
|
"num_tokens": 7552754.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.612136316299439,
|
|
"epoch": 3.319854611540209,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000426218570266852,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.1953836351633072,
|
|
"num_tokens": 7562906.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.476711845397949,
|
|
"epoch": 3.324398000908678,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042595667790941536,
|
|
"loss": 5.1415,
|
|
"mean_token_accuracy": 0.19171063899993895,
|
|
"num_tokens": 7574149.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.549911260604858,
|
|
"epoch": 3.328941390277147,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042569441308138204,
|
|
"loss": 5.1318,
|
|
"mean_token_accuracy": 0.1959947481751442,
|
|
"num_tokens": 7584610.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.531237411499023,
|
|
"epoch": 3.3334847796456155,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004254317764298645,
|
|
"loss": 5.1007,
|
|
"mean_token_accuracy": 0.1887877479195595,
|
|
"num_tokens": 7594954.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.601019716262817,
|
|
"epoch": 3.3380281690140845,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004251687686028927,
|
|
"loss": 5.2096,
|
|
"mean_token_accuracy": 0.18012717813253404,
|
|
"num_tokens": 7606069.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.489976835250855,
|
|
"epoch": 3.3425715583825535,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042490539024941216,
|
|
"loss": 5.0582,
|
|
"mean_token_accuracy": 0.19151556342840195,
|
|
"num_tokens": 7615671.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.5457679271698,
|
|
"epoch": 3.347114947751022,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042464164201928286,
|
|
"loss": 5.2538,
|
|
"mean_token_accuracy": 0.18164459019899368,
|
|
"num_tokens": 7626740.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.580724477767944,
|
|
"epoch": 3.351658337119491,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042437752456327734,
|
|
"loss": 5.1501,
|
|
"mean_token_accuracy": 0.1935126468539238,
|
|
"num_tokens": 7636459.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.619305562973023,
|
|
"epoch": 3.35620172648796,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042411303853307907,
|
|
"loss": 5.1354,
|
|
"mean_token_accuracy": 0.1886074110865593,
|
|
"num_tokens": 7646974.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.4681860446929935,
|
|
"epoch": 3.360745115856429,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004238481845812811,
|
|
"loss": 5.0903,
|
|
"mean_token_accuracy": 0.19271185696125032,
|
|
"num_tokens": 7657372.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.501412963867187,
|
|
"epoch": 3.365288505224898,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042358296336138436,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.189212603867054,
|
|
"num_tokens": 7668972.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.584800386428833,
|
|
"epoch": 3.3698318945933665,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004233173755277956,
|
|
"loss": 5.1421,
|
|
"mean_token_accuracy": 0.20665653944015502,
|
|
"num_tokens": 7679742.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.559856414794922,
|
|
"epoch": 3.3743752839618355,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042305142173582657,
|
|
"loss": 5.1867,
|
|
"mean_token_accuracy": 0.19379900693893432,
|
|
"num_tokens": 7690759.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.545025634765625,
|
|
"epoch": 3.3789186733303045,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042278510264169165,
|
|
"loss": 5.1391,
|
|
"mean_token_accuracy": 0.19267158657312394,
|
|
"num_tokens": 7700482.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.5073751449584964,
|
|
"epoch": 3.3834620626987735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042251841890250696,
|
|
"loss": 5.1209,
|
|
"mean_token_accuracy": 0.19036568254232406,
|
|
"num_tokens": 7710414.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.601136684417725,
|
|
"epoch": 3.388005452067242,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000422251371176288,
|
|
"loss": 5.156,
|
|
"mean_token_accuracy": 0.1919710785150528,
|
|
"num_tokens": 7721214.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.5292555809021,
|
|
"epoch": 3.392548841435711,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00042198396012194854,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.19231873750686646,
|
|
"num_tokens": 7732551.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.571150875091552,
|
|
"epoch": 3.39709223080418,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042171618639929865,
|
|
"loss": 5.2091,
|
|
"mean_token_accuracy": 0.191018545627594,
|
|
"num_tokens": 7742266.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.495872163772583,
|
|
"epoch": 3.401635620172649,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042144805066904345,
|
|
"loss": 5.0588,
|
|
"mean_token_accuracy": 0.18994052112102508,
|
|
"num_tokens": 7753152.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.516603899002075,
|
|
"epoch": 3.4061790095411175,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00042117955359278125,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.18809825032949448,
|
|
"num_tokens": 7764420.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.576537799835205,
|
|
"epoch": 3.4107223989095865,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00042091069583300177,
|
|
"loss": 5.1298,
|
|
"mean_token_accuracy": 0.19266747385263444,
|
|
"num_tokens": 7776044.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.5177184581756595,
|
|
"epoch": 3.4152657882780555,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004206414780530849,
|
|
"loss": 5.1603,
|
|
"mean_token_accuracy": 0.19181600511074065,
|
|
"num_tokens": 7786737.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.506136226654053,
|
|
"epoch": 3.4198091776465245,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004203719009172987,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.19453130811452865,
|
|
"num_tokens": 7796974.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.5970865249633786,
|
|
"epoch": 3.424352567014993,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000420101965090798,
|
|
"loss": 5.2717,
|
|
"mean_token_accuracy": 0.18344012647867203,
|
|
"num_tokens": 7808221.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.500562858581543,
|
|
"epoch": 3.428895956383462,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004198316712396226,
|
|
"loss": 5.0898,
|
|
"mean_token_accuracy": 0.19185245484113694,
|
|
"num_tokens": 7818422.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.540801000595093,
|
|
"epoch": 3.433439345751931,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00041956102003069573,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.19109012335538864,
|
|
"num_tokens": 7828313.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.5394628047943115,
|
|
"epoch": 3.4379827351204,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004192900121318223,
|
|
"loss": 5.1337,
|
|
"mean_token_accuracy": 0.19941307753324508,
|
|
"num_tokens": 7839493.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.543460845947266,
|
|
"epoch": 3.4425261244888685,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004190186482116874,
|
|
"loss": 5.134,
|
|
"mean_token_accuracy": 0.18883001506328584,
|
|
"num_tokens": 7849796.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.5130970001220705,
|
|
"epoch": 3.4470695138573375,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004187469289398545,
|
|
"loss": 5.1957,
|
|
"mean_token_accuracy": 0.18865694403648375,
|
|
"num_tokens": 7860865.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.49743332862854,
|
|
"epoch": 3.4516129032258065,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041847485498676385,
|
|
"loss": 5.1403,
|
|
"mean_token_accuracy": 0.19718256145715712,
|
|
"num_tokens": 7871144.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.522391748428345,
|
|
"epoch": 3.4561562925942755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041820242702373094,
|
|
"loss": 5.1888,
|
|
"mean_token_accuracy": 0.1942734718322754,
|
|
"num_tokens": 7880916.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.484518671035767,
|
|
"epoch": 3.460699681962744,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004179296457229448,
|
|
"loss": 5.1211,
|
|
"mean_token_accuracy": 0.1911142200231552,
|
|
"num_tokens": 7891813.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.585445737838745,
|
|
"epoch": 3.465243071331213,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004176565117574659,
|
|
"loss": 5.173,
|
|
"mean_token_accuracy": 0.1877262234687805,
|
|
"num_tokens": 7902697.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.537561893463135,
|
|
"epoch": 3.469786460699682,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004173830258012255,
|
|
"loss": 5.1464,
|
|
"mean_token_accuracy": 0.1930139735341072,
|
|
"num_tokens": 7912777.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.553113269805908,
|
|
"epoch": 3.474329850068151,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004171091885290226,
|
|
"loss": 5.1873,
|
|
"mean_token_accuracy": 0.19049507528543472,
|
|
"num_tokens": 7923255.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.537462568283081,
|
|
"epoch": 3.4788732394366195,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004168350006165238,
|
|
"loss": 5.0658,
|
|
"mean_token_accuracy": 0.19168077856302262,
|
|
"num_tokens": 7934268.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.432240676879883,
|
|
"epoch": 3.4834166288050885,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041656046274026034,
|
|
"loss": 5.0676,
|
|
"mean_token_accuracy": 0.19443341344594955,
|
|
"num_tokens": 7944526.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.554276704788208,
|
|
"epoch": 3.4879600181735575,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004162855755776272,
|
|
"loss": 5.1781,
|
|
"mean_token_accuracy": 0.18912502825260163,
|
|
"num_tokens": 7954512.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.529981517791748,
|
|
"epoch": 3.4925034075420265,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041601033980688106,
|
|
"loss": 5.196,
|
|
"mean_token_accuracy": 0.18088309913873674,
|
|
"num_tokens": 7965703.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.5635253429412845,
|
|
"epoch": 3.497046796910495,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004157347561071389,
|
|
"loss": 5.1868,
|
|
"mean_token_accuracy": 0.18955022394657134,
|
|
"num_tokens": 7977822.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.510582113265992,
|
|
"epoch": 3.501590186278964,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041545882515837604,
|
|
"loss": 5.0973,
|
|
"mean_token_accuracy": 0.18894753307104112,
|
|
"num_tokens": 7988318.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.550647926330567,
|
|
"epoch": 3.506133575647433,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041518254764142477,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.19622776061296462,
|
|
"num_tokens": 7998076.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.529185485839844,
|
|
"epoch": 3.510676965015902,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041490592423797226,
|
|
"loss": 5.1715,
|
|
"mean_token_accuracy": 0.19466058611869813,
|
|
"num_tokens": 8007913.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.489879989624024,
|
|
"epoch": 3.515220354384371,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004146289556305594,
|
|
"loss": 5.1338,
|
|
"mean_token_accuracy": 0.18910700529813768,
|
|
"num_tokens": 8018340.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.61398458480835,
|
|
"epoch": 3.5197637437528395,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004143516425025786,
|
|
"loss": 5.2044,
|
|
"mean_token_accuracy": 0.18639347851276397,
|
|
"num_tokens": 8028312.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.59273042678833,
|
|
"epoch": 3.5243071331213085,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004140739855382724,
|
|
"loss": 5.1965,
|
|
"mean_token_accuracy": 0.1921403169631958,
|
|
"num_tokens": 8038580.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.574572420120239,
|
|
"epoch": 3.5288505224897775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041379598542273197,
|
|
"loss": 5.1486,
|
|
"mean_token_accuracy": 0.18995873779058456,
|
|
"num_tokens": 8048365.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.552021026611328,
|
|
"epoch": 3.533393911858246,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00041351764284189464,
|
|
"loss": 5.2077,
|
|
"mean_token_accuracy": 0.1823919713497162,
|
|
"num_tokens": 8059638.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.579764318466187,
|
|
"epoch": 3.537937301226715,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004132389584825433,
|
|
"loss": 5.2502,
|
|
"mean_token_accuracy": 0.18523784279823302,
|
|
"num_tokens": 8070993.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.5283417224884035,
|
|
"epoch": 3.542480690595184,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041295993303230383,
|
|
"loss": 5.0907,
|
|
"mean_token_accuracy": 0.1978906959295273,
|
|
"num_tokens": 8081203.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.454664659500122,
|
|
"epoch": 3.547024079963653,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004126805671796438,
|
|
"loss": 5.0981,
|
|
"mean_token_accuracy": 0.1887076824903488,
|
|
"num_tokens": 8091564.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.5413196086883545,
|
|
"epoch": 3.551567469332122,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004124008616138707,
|
|
"loss": 5.087,
|
|
"mean_token_accuracy": 0.19312319606542588,
|
|
"num_tokens": 8100675.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.520606803894043,
|
|
"epoch": 3.5561108587005905,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041212081702513015,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.1945698454976082,
|
|
"num_tokens": 8110605.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.5288787364959715,
|
|
"epoch": 3.5606542480690595,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004118404341044044,
|
|
"loss": 5.1986,
|
|
"mean_token_accuracy": 0.18797775804996492,
|
|
"num_tokens": 8120910.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.55887804031372,
|
|
"epoch": 3.5651976374375285,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041155971354351034,
|
|
"loss": 5.1492,
|
|
"mean_token_accuracy": 0.19329776912927626,
|
|
"num_tokens": 8131827.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.540145587921143,
|
|
"epoch": 3.569741026805997,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041127865603509815,
|
|
"loss": 5.1235,
|
|
"mean_token_accuracy": 0.1942894384264946,
|
|
"num_tokens": 8141785.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.574447584152222,
|
|
"epoch": 3.574284416174466,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004109972622726493,
|
|
"loss": 5.1515,
|
|
"mean_token_accuracy": 0.1875311553478241,
|
|
"num_tokens": 8151732.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.563373231887818,
|
|
"epoch": 3.578827805542935,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00041071553295047485,
|
|
"loss": 5.2329,
|
|
"mean_token_accuracy": 0.18431088477373123,
|
|
"num_tokens": 8162334.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.535441780090332,
|
|
"epoch": 3.583371194911404,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004104334687637139,
|
|
"loss": 5.1146,
|
|
"mean_token_accuracy": 0.19262330383062362,
|
|
"num_tokens": 8172810.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.4998149394989015,
|
|
"epoch": 3.587914584279873,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00041015107040833187,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.1860748738050461,
|
|
"num_tokens": 8183240.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.54986252784729,
|
|
"epoch": 3.5924579736483415,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004098683385811185,
|
|
"loss": 5.103,
|
|
"mean_token_accuracy": 0.19000099450349808,
|
|
"num_tokens": 8193824.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.496000814437866,
|
|
"epoch": 3.5970013630168105,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040958527397968666,
|
|
"loss": 5.0843,
|
|
"mean_token_accuracy": 0.19927169978618622,
|
|
"num_tokens": 8203792.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.54337158203125,
|
|
"epoch": 3.6015447523852795,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004093018773024699,
|
|
"loss": 5.2784,
|
|
"mean_token_accuracy": 0.18759161233901978,
|
|
"num_tokens": 8214436.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.576527786254883,
|
|
"epoch": 3.606088141753748,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004090181492487215,
|
|
"loss": 5.1298,
|
|
"mean_token_accuracy": 0.1908451110124588,
|
|
"num_tokens": 8224518.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.563905239105225,
|
|
"epoch": 3.610631531122217,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040873409051851223,
|
|
"loss": 5.2793,
|
|
"mean_token_accuracy": 0.18164435774087906,
|
|
"num_tokens": 8235234.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.530511856079102,
|
|
"epoch": 3.615174920490686,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004084497018127286,
|
|
"loss": 5.1545,
|
|
"mean_token_accuracy": 0.18730711191892624,
|
|
"num_tokens": 8246612.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.554295444488526,
|
|
"epoch": 3.619718309859155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004081649838330717,
|
|
"loss": 5.1303,
|
|
"mean_token_accuracy": 0.19477790594100952,
|
|
"num_tokens": 8256720.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.533069324493408,
|
|
"epoch": 3.624261699227624,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00040787993728205464,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.1920352280139923,
|
|
"num_tokens": 8267666.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.5814110279083256,
|
|
"epoch": 3.6288050885960925,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040759456286300154,
|
|
"loss": 5.1561,
|
|
"mean_token_accuracy": 0.1906946375966072,
|
|
"num_tokens": 8277882.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.49630331993103,
|
|
"epoch": 3.6333484779645615,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040730886128004544,
|
|
"loss": 5.0717,
|
|
"mean_token_accuracy": 0.19409287124872207,
|
|
"num_tokens": 8286968.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.6333484779645615,
|
|
"eval_entropy": 5.366523528653522,
|
|
"eval_loss": 5.737530708312988,
|
|
"eval_mean_token_accuracy": 0.167983505424372,
|
|
"eval_num_tokens": 8286968.0,
|
|
"eval_runtime": 2.0103,
|
|
"eval_samples_per_second": 1708.737,
|
|
"eval_steps_per_second": 213.903,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"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": 1789765102725120.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|