| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9956108266276518, | |
| "eval_steps": 500, | |
| "global_step": 1023, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.029261155815654718, | |
| "grad_norm": 1.313577356108148, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7981, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.058522311631309436, | |
| "grad_norm": 1.3269701626701922, | |
| "learning_rate": 5e-06, | |
| "loss": 0.74, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08778346744696415, | |
| "grad_norm": 1.295468305567646, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7078, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11704462326261887, | |
| "grad_norm": 1.0952569971898316, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6971, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14630577907827358, | |
| "grad_norm": 1.1522366437551468, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6876, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1755669348939283, | |
| "grad_norm": 0.9337304723699321, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6773, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.20482809070958302, | |
| "grad_norm": 0.663457890098491, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6705, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.23408924652523774, | |
| "grad_norm": 0.45676131712073714, | |
| "learning_rate": 5e-06, | |
| "loss": 0.665, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26335040234089246, | |
| "grad_norm": 0.4288236647896689, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6621, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.29261155815654716, | |
| "grad_norm": 0.3968045535268243, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6545, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3218727139722019, | |
| "grad_norm": 0.4005298417378183, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6632, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3511338697878566, | |
| "grad_norm": 0.37524287200965606, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6682, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.38039502560351135, | |
| "grad_norm": 0.3838388476815443, | |
| "learning_rate": 5e-06, | |
| "loss": 0.652, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.40965618141916604, | |
| "grad_norm": 0.36960334592767113, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6527, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4389173372348208, | |
| "grad_norm": 0.3476368986908311, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6481, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4681784930504755, | |
| "grad_norm": 0.3852462502438018, | |
| "learning_rate": 5e-06, | |
| "loss": 0.65, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.49743964886613024, | |
| "grad_norm": 0.3669278975935387, | |
| "learning_rate": 5e-06, | |
| "loss": 0.655, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5267008046817849, | |
| "grad_norm": 0.3333018116429125, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6547, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5559619604974396, | |
| "grad_norm": 0.35791354795329805, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6432, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5852231163130943, | |
| "grad_norm": 0.34384449798393535, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6444, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6144842721287491, | |
| "grad_norm": 0.3389276664582108, | |
| "learning_rate": 5e-06, | |
| "loss": 0.651, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6437454279444038, | |
| "grad_norm": 0.3537068086163664, | |
| "learning_rate": 5e-06, | |
| "loss": 0.639, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6730065837600585, | |
| "grad_norm": 0.36001482083394, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6389, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7022677395757132, | |
| "grad_norm": 0.3260238185984492, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6458, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.731528895391368, | |
| "grad_norm": 0.3385112431681223, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6441, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7607900512070227, | |
| "grad_norm": 0.33741935574023685, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6467, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.7900512070226774, | |
| "grad_norm": 0.37090927004388813, | |
| "learning_rate": 5e-06, | |
| "loss": 0.649, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8193123628383321, | |
| "grad_norm": 0.34506670056383226, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6478, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8485735186539868, | |
| "grad_norm": 0.33968288467400815, | |
| "learning_rate": 5e-06, | |
| "loss": 0.64, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.8778346744696416, | |
| "grad_norm": 0.3468850912982488, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6319, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9070958302852963, | |
| "grad_norm": 0.35133173438186993, | |
| "learning_rate": 5e-06, | |
| "loss": 0.645, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.936356986100951, | |
| "grad_norm": 0.3350013657985102, | |
| "learning_rate": 5e-06, | |
| "loss": 0.637, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9656181419166057, | |
| "grad_norm": 0.3761018877235313, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6402, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9948792977322605, | |
| "grad_norm": 0.3507305461467966, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6362, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.9978054133138259, | |
| "eval_loss": 0.640738844871521, | |
| "eval_runtime": 341.9935, | |
| "eval_samples_per_second": 26.927, | |
| "eval_steps_per_second": 0.421, | |
| "step": 341 | |
| }, | |
| { | |
| "epoch": 1.025237746891002, | |
| "grad_norm": 0.37016017200964146, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6517, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.054498902706657, | |
| "grad_norm": 0.35451517967031254, | |
| "learning_rate": 5e-06, | |
| "loss": 0.606, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.0837600585223117, | |
| "grad_norm": 0.4093464489183681, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6093, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1130212143379663, | |
| "grad_norm": 0.3328609668211424, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6067, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.142282370153621, | |
| "grad_norm": 0.3929172278235903, | |
| "learning_rate": 5e-06, | |
| "loss": 0.612, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.1715435259692757, | |
| "grad_norm": 0.36956263451829535, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6135, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2008046817849305, | |
| "grad_norm": 0.35625492540137693, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6131, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2300658376005853, | |
| "grad_norm": 0.3805534450160043, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6013, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2593269934162399, | |
| "grad_norm": 0.35722847933137475, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6114, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.2885881492318947, | |
| "grad_norm": 0.34455850114898495, | |
| "learning_rate": 5e-06, | |
| "loss": 0.604, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3178493050475493, | |
| "grad_norm": 0.36847098245047766, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6078, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.347110460863204, | |
| "grad_norm": 0.373128115016406, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6055, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.3763716166788589, | |
| "grad_norm": 0.32106596594422865, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6028, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4056327724945135, | |
| "grad_norm": 0.36636054815544133, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6037, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.4348939283101683, | |
| "grad_norm": 0.3456331896000125, | |
| "learning_rate": 5e-06, | |
| "loss": 0.61, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.464155084125823, | |
| "grad_norm": 0.34045661745043854, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5984, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.4934162399414777, | |
| "grad_norm": 0.3438707843118531, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6063, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5226773957571325, | |
| "grad_norm": 0.3807401104516723, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6087, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.5519385515727873, | |
| "grad_norm": 0.3476316988283133, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6073, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.5811997073884418, | |
| "grad_norm": 0.37180861974943186, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6139, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.6104608632040964, | |
| "grad_norm": 0.3584155358816698, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6049, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6397220190197512, | |
| "grad_norm": 0.350475088364943, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6066, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.668983174835406, | |
| "grad_norm": 0.3217698359289029, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5972, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.6982443306510606, | |
| "grad_norm": 0.34896778513585464, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6066, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.7275054864667154, | |
| "grad_norm": 0.4043798899680667, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6068, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.7567666422823702, | |
| "grad_norm": 0.33940611207252186, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6037, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.7860277980980248, | |
| "grad_norm": 0.3427753606881015, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6061, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.8152889539136796, | |
| "grad_norm": 0.3213976735602389, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6066, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.8445501097293344, | |
| "grad_norm": 0.3527145844560201, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6056, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.873811265544989, | |
| "grad_norm": 0.32813676744746023, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6005, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.9030724213606436, | |
| "grad_norm": 0.3638376151394638, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6062, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.9323335771762986, | |
| "grad_norm": 0.34443917263076446, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6067, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.9615947329919532, | |
| "grad_norm": 0.3322566789880142, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5955, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.9908558888076078, | |
| "grad_norm": 0.35385249127498575, | |
| "learning_rate": 5e-06, | |
| "loss": 0.61, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.9967081199707388, | |
| "eval_loss": 0.633293867111206, | |
| "eval_runtime": 344.1509, | |
| "eval_samples_per_second": 26.759, | |
| "eval_steps_per_second": 0.418, | |
| "step": 682 | |
| }, | |
| { | |
| "epoch": 2.0212143379663496, | |
| "grad_norm": 0.39491322770292114, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6203, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.050475493782004, | |
| "grad_norm": 0.4352323735803224, | |
| "learning_rate": 5e-06, | |
| "loss": 0.568, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.0797366495976592, | |
| "grad_norm": 0.35548597343547844, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5701, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.108997805413314, | |
| "grad_norm": 0.36917361599584547, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5711, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.1382589612289684, | |
| "grad_norm": 0.35132763692404273, | |
| "learning_rate": 5e-06, | |
| "loss": 0.573, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.1675201170446234, | |
| "grad_norm": 0.35336419936754826, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5782, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.196781272860278, | |
| "grad_norm": 0.3142941595586366, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5769, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.2260424286759326, | |
| "grad_norm": 0.36678679238628564, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5744, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.255303584491587, | |
| "grad_norm": 0.3376459499493842, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5732, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.284564740307242, | |
| "grad_norm": 0.3329859631258577, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5722, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.313825896122897, | |
| "grad_norm": 0.3306554971721492, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5705, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.3430870519385514, | |
| "grad_norm": 0.35902042471365553, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5782, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.3723482077542064, | |
| "grad_norm": 0.3312419229224188, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5735, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.401609363569861, | |
| "grad_norm": 0.3327307706986044, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5798, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.4308705193855156, | |
| "grad_norm": 0.31672285917768106, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5721, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.4601316752011706, | |
| "grad_norm": 0.3517058777467751, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5725, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.489392831016825, | |
| "grad_norm": 0.349159379504596, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5747, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.5186539868324798, | |
| "grad_norm": 0.35617064549859445, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5677, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.547915142648135, | |
| "grad_norm": 0.39354050339723484, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5702, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.5771762984637894, | |
| "grad_norm": 0.34753777557024923, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5744, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.606437454279444, | |
| "grad_norm": 0.3680124888704586, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5737, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.6356986100950985, | |
| "grad_norm": 0.3413167162283287, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5783, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.6649597659107536, | |
| "grad_norm": 0.35574416904205275, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5784, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.694220921726408, | |
| "grad_norm": 0.358995127200404, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5728, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.723482077542063, | |
| "grad_norm": 0.35213516458107996, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5764, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.7527432333577178, | |
| "grad_norm": 0.3453618453932275, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5754, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.7820043891733723, | |
| "grad_norm": 0.33383973563266417, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5741, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.811265544989027, | |
| "grad_norm": 0.3717714129794388, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5779, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.840526700804682, | |
| "grad_norm": 0.3415472887200101, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5736, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.8697878566203365, | |
| "grad_norm": 0.37787318934941566, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5735, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.899049012435991, | |
| "grad_norm": 0.3302379886701318, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5665, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.928310168251646, | |
| "grad_norm": 0.3247773766219703, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5743, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.9575713240673007, | |
| "grad_norm": 0.355132160394971, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5661, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.9868324798829553, | |
| "grad_norm": 0.33688535053100965, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5777, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.9956108266276518, | |
| "eval_loss": 0.6330391764640808, | |
| "eval_runtime": 344.7903, | |
| "eval_samples_per_second": 26.709, | |
| "eval_steps_per_second": 0.418, | |
| "step": 1023 | |
| }, | |
| { | |
| "epoch": 2.9956108266276518, | |
| "step": 1023, | |
| "total_flos": 2144987064041472.0, | |
| "train_loss": 0.6145497837840403, | |
| "train_runtime": 54980.8916, | |
| "train_samples_per_second": 9.546, | |
| "train_steps_per_second": 0.019 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1023, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2144987064041472.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |