| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.992874109263658, | |
| "eval_steps": 500, | |
| "global_step": 945, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03167062549485352, | |
| "grad_norm": 1.8530284877370395, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8166, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06334125098970704, | |
| "grad_norm": 1.5337420189302975, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7464, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09501187648456057, | |
| "grad_norm": 1.4029937472724519, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7243, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12668250197941408, | |
| "grad_norm": 0.9360700334222073, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7146, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1583531274742676, | |
| "grad_norm": 0.994137693760135, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6895, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19002375296912113, | |
| "grad_norm": 0.9347536063077458, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6778, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.22169437846397466, | |
| "grad_norm": 1.4913361116122625, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6594, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.25336500395882816, | |
| "grad_norm": 0.7126258978453617, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6637, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2850356294536817, | |
| "grad_norm": 0.9899069063293856, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6456, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3167062549485352, | |
| "grad_norm": 0.8045731477224973, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6537, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.34837688044338877, | |
| "grad_norm": 0.7578866653732541, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6481, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.38004750593824227, | |
| "grad_norm": 0.5676407147308244, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6419, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4117181314330958, | |
| "grad_norm": 0.8078622210774858, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6479, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4433887569279493, | |
| "grad_norm": 0.7647299971015771, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6439, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4750593824228028, | |
| "grad_norm": 0.5575084199732572, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6388, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5067300079176563, | |
| "grad_norm": 0.5128045187828547, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6347, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5384006334125099, | |
| "grad_norm": 0.8861637952775342, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6354, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5700712589073634, | |
| "grad_norm": 0.7708516251715861, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6365, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.601741884402217, | |
| "grad_norm": 0.6980158820500171, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6339, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6334125098970704, | |
| "grad_norm": 0.5994527923990384, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6384, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.665083135391924, | |
| "grad_norm": 1.2967423762938355, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6369, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6967537608867775, | |
| "grad_norm": 0.47535091611506025, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6321, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.728424386381631, | |
| "grad_norm": 0.6353037224955166, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6241, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7600950118764845, | |
| "grad_norm": 0.6601833755256151, | |
| "learning_rate": 5e-06, | |
| "loss": 0.623, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7917656373713381, | |
| "grad_norm": 0.6122026802056224, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6263, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8234362628661916, | |
| "grad_norm": 0.5075094381024849, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6301, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8551068883610451, | |
| "grad_norm": 0.5422031773802576, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6292, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8867775138558986, | |
| "grad_norm": 0.42516469360864606, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6177, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.9184481393507522, | |
| "grad_norm": 0.7465232508313683, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6197, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9501187648456056, | |
| "grad_norm": 0.5164060465761252, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6286, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9817893903404592, | |
| "grad_norm": 0.6125601748565044, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6301, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.997624703087886, | |
| "eval_loss": 0.6208310723304749, | |
| "eval_runtime": 170.2387, | |
| "eval_samples_per_second": 49.965, | |
| "eval_steps_per_second": 0.394, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.0134600158353126, | |
| "grad_norm": 0.8577950377026092, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6042, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0451306413301662, | |
| "grad_norm": 0.563239613051996, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5769, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.0768012668250198, | |
| "grad_norm": 0.5011250455738395, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5753, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.1084718923198733, | |
| "grad_norm": 0.507779510609032, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5755, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1401425178147269, | |
| "grad_norm": 0.5864390033721468, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5731, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1718131433095804, | |
| "grad_norm": 0.5651111499269338, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5687, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.203483768804434, | |
| "grad_norm": 0.6150558670403593, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5729, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.2351543942992875, | |
| "grad_norm": 0.5808136283436993, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5718, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2668250197941409, | |
| "grad_norm": 0.5830305205374973, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5761, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2984956452889944, | |
| "grad_norm": 0.5467014072656838, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5799, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.330166270783848, | |
| "grad_norm": 0.5214522227982674, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5734, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.3618368962787015, | |
| "grad_norm": 0.7855825418949094, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5712, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.393507521773555, | |
| "grad_norm": 0.5656871920545978, | |
| "learning_rate": 5e-06, | |
| "loss": 0.576, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.4251781472684084, | |
| "grad_norm": 0.5944245365102998, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5708, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.4568487727632622, | |
| "grad_norm": 0.5333261966632318, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5671, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4885193982581155, | |
| "grad_norm": 0.491104016024968, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5743, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.520190023752969, | |
| "grad_norm": 0.5014163060947014, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5728, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.5518606492478226, | |
| "grad_norm": 0.4892383437610088, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5802, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5835312747426762, | |
| "grad_norm": 0.4989580296588211, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5751, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6152019002375297, | |
| "grad_norm": 0.616735446474154, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5716, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.646872525732383, | |
| "grad_norm": 0.5387044899734632, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5751, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.6785431512272369, | |
| "grad_norm": 0.5526914425760312, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5684, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.7102137767220902, | |
| "grad_norm": 0.6545410316398991, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5702, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.7418844022169437, | |
| "grad_norm": 0.5254133456615973, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5674, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.7735550277117973, | |
| "grad_norm": 0.521939360967283, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5724, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.8052256532066508, | |
| "grad_norm": 0.463070184825861, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5818, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.8368962787015044, | |
| "grad_norm": 0.6381561992890791, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5688, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.8685669041963577, | |
| "grad_norm": 0.6496360141809046, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5646, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.9002375296912115, | |
| "grad_norm": 0.5324989809239741, | |
| "learning_rate": 5e-06, | |
| "loss": 0.566, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.9319081551860648, | |
| "grad_norm": 0.5111765382529705, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5708, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.9635787806809184, | |
| "grad_norm": 0.5610135178229989, | |
| "learning_rate": 5e-06, | |
| "loss": 0.574, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.995249406175772, | |
| "grad_norm": 0.6915334000048331, | |
| "learning_rate": 5e-06, | |
| "loss": 0.58, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9984164687252575, | |
| "eval_loss": 0.6117470860481262, | |
| "eval_runtime": 171.4849, | |
| "eval_samples_per_second": 49.602, | |
| "eval_steps_per_second": 0.391, | |
| "step": 631 | |
| }, | |
| { | |
| "epoch": 2.0269200316706253, | |
| "grad_norm": 0.616676539774963, | |
| "learning_rate": 5e-06, | |
| "loss": 0.528, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.058590657165479, | |
| "grad_norm": 0.6073136045120692, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5195, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0902612826603324, | |
| "grad_norm": 0.5470687590939235, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5258, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.121931908155186, | |
| "grad_norm": 0.5476384369380188, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5207, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.1536025336500395, | |
| "grad_norm": 0.5807654736471419, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5229, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.1852731591448933, | |
| "grad_norm": 0.5658725818639244, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5155, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.2169437846397466, | |
| "grad_norm": 0.5270348810540799, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5257, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.2486144101346, | |
| "grad_norm": 0.6710965861689625, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5148, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.2802850356294537, | |
| "grad_norm": 0.5096235564504402, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5181, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.311955661124307, | |
| "grad_norm": 0.5550769313636378, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5171, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.343626286619161, | |
| "grad_norm": 0.7122071616039661, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5181, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.375296912114014, | |
| "grad_norm": 0.5158288008674774, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5255, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.406967537608868, | |
| "grad_norm": 0.5744842064968558, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5232, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.4386381631037213, | |
| "grad_norm": 0.48569927712881583, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5231, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.470308788598575, | |
| "grad_norm": 0.6323853806061133, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5199, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.5019794140934284, | |
| "grad_norm": 0.5506449580962853, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5192, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.5336500395882817, | |
| "grad_norm": 0.4790039532718334, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5179, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.5653206650831355, | |
| "grad_norm": 0.5390096335172841, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5269, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.596991290577989, | |
| "grad_norm": 0.5217848609501268, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5229, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.6286619160728426, | |
| "grad_norm": 0.5196866879306132, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5217, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.660332541567696, | |
| "grad_norm": 0.5333118708551611, | |
| "learning_rate": 5e-06, | |
| "loss": 0.525, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.6920031670625493, | |
| "grad_norm": 0.5483398687319767, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5224, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.723673792557403, | |
| "grad_norm": 0.5117263088559509, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5283, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.7553444180522564, | |
| "grad_norm": 0.5377436931762913, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5232, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.78701504354711, | |
| "grad_norm": 0.5662585202942261, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5274, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.8186856690419635, | |
| "grad_norm": 0.48954404185612554, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5273, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.850356294536817, | |
| "grad_norm": 0.5387435059221906, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5181, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.8820269200316706, | |
| "grad_norm": 0.5630237900783817, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5228, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.9136975455265244, | |
| "grad_norm": 0.4995116097709295, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5247, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.9453681710213777, | |
| "grad_norm": 0.49890480333211984, | |
| "learning_rate": 5e-06, | |
| "loss": 0.521, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.977038796516231, | |
| "grad_norm": 0.5217828385490265, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5261, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.992874109263658, | |
| "eval_loss": 0.6157404184341431, | |
| "eval_runtime": 171.3558, | |
| "eval_samples_per_second": 49.639, | |
| "eval_steps_per_second": 0.391, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 2.992874109263658, | |
| "step": 945, | |
| "total_flos": 1582491437629440.0, | |
| "train_loss": 0.5832562999119835, | |
| "train_runtime": 28511.6322, | |
| "train_samples_per_second": 17.004, | |
| "train_steps_per_second": 0.033 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 945, | |
| "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": 1582491437629440.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |