| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9941972920696323, | |
| "eval_steps": 50, | |
| "global_step": 774, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.019342359767891684, | |
| "grad_norm": 17.769769703655175, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7186, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03868471953578337, | |
| "grad_norm": 13.081004196040046, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5925, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.058027079303675046, | |
| "grad_norm": 7.089375759826574, | |
| "learning_rate": 9.998943236640677e-07, | |
| "loss": 1.2917, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07736943907156674, | |
| "grad_norm": 4.226052382894484, | |
| "learning_rate": 9.995773393262229e-07, | |
| "loss": 1.152, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09671179883945841, | |
| "grad_norm": 4.275706578437442, | |
| "learning_rate": 9.99049180977439e-07, | |
| "loss": 1.0662, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.11605415860735009, | |
| "grad_norm": 3.8773098526703986, | |
| "learning_rate": 9.983100718730718e-07, | |
| "loss": 1.0205, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13539651837524178, | |
| "grad_norm": 3.5740777007667783, | |
| "learning_rate": 9.973603244384906e-07, | |
| "loss": 0.9948, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.15473887814313347, | |
| "grad_norm": 3.6003113336465318, | |
| "learning_rate": 9.9620034013701e-07, | |
| "loss": 0.9421, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.17408123791102514, | |
| "grad_norm": 3.583333545710018, | |
| "learning_rate": 9.948306093001933e-07, | |
| "loss": 0.9544, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "grad_norm": 3.7908364237614416, | |
| "learning_rate": 9.932517109205849e-07, | |
| "loss": 0.9541, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "eval_loss": 0.9273383021354675, | |
| "eval_runtime": 127.7915, | |
| "eval_samples_per_second": 57.516, | |
| "eval_steps_per_second": 0.9, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 3.613000346904536, | |
| "learning_rate": 9.914643124069666e-07, | |
| "loss": 0.906, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.23210831721470018, | |
| "grad_norm": 3.6167098335205994, | |
| "learning_rate": 9.89469169302242e-07, | |
| "loss": 0.9124, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2514506769825919, | |
| "grad_norm": 3.6118346250444553, | |
| "learning_rate": 9.872671249640626e-07, | |
| "loss": 0.9371, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.27079303675048355, | |
| "grad_norm": 3.5717388877832117, | |
| "learning_rate": 9.848591102083375e-07, | |
| "loss": 0.9202, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2901353965183752, | |
| "grad_norm": 3.713276627714593, | |
| "learning_rate": 9.822461429157716e-07, | |
| "loss": 0.9395, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.30947775628626695, | |
| "grad_norm": 3.8351121315501007, | |
| "learning_rate": 9.794293276016023e-07, | |
| "loss": 0.9039, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3288201160541586, | |
| "grad_norm": 3.3458211116161043, | |
| "learning_rate": 9.764098549487155e-07, | |
| "loss": 0.8935, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.3481624758220503, | |
| "grad_norm": 3.6453614828666105, | |
| "learning_rate": 9.731890013043367e-07, | |
| "loss": 0.8922, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.36750483558994196, | |
| "grad_norm": 3.67687317911342, | |
| "learning_rate": 9.697681281405128e-07, | |
| "loss": 0.8675, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "grad_norm": 3.4055043252450132, | |
| "learning_rate": 9.6614868147861e-07, | |
| "loss": 0.8621, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "eval_loss": 0.8719474077224731, | |
| "eval_runtime": 127.7997, | |
| "eval_samples_per_second": 57.512, | |
| "eval_steps_per_second": 0.9, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.40618955512572535, | |
| "grad_norm": 3.4010105290171797, | |
| "learning_rate": 9.623321912780744e-07, | |
| "loss": 0.8956, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.425531914893617, | |
| "grad_norm": 3.726255996119051, | |
| "learning_rate": 9.583202707897073e-07, | |
| "loss": 0.8645, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4448742746615087, | |
| "grad_norm": 3.428308495841723, | |
| "learning_rate": 9.54114615873738e-07, | |
| "loss": 0.8526, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.46421663442940037, | |
| "grad_norm": 3.4777454179113914, | |
| "learning_rate": 9.497170042829736e-07, | |
| "loss": 0.873, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4835589941972921, | |
| "grad_norm": 3.306069687386359, | |
| "learning_rate": 9.451292949113329e-07, | |
| "loss": 0.8646, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5029013539651838, | |
| "grad_norm": 3.6787006187120315, | |
| "learning_rate": 9.403534270080829e-07, | |
| "loss": 0.8668, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5222437137330754, | |
| "grad_norm": 3.8412614001222387, | |
| "learning_rate": 9.353914193581072e-07, | |
| "loss": 0.8522, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.5415860735009671, | |
| "grad_norm": 3.63279379492409, | |
| "learning_rate": 9.302453694285548e-07, | |
| "loss": 0.8552, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5609284332688588, | |
| "grad_norm": 3.557168960609258, | |
| "learning_rate": 9.249174524822305e-07, | |
| "loss": 0.8569, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "grad_norm": 3.6248960952352673, | |
| "learning_rate": 9.19409920658098e-07, | |
| "loss": 0.8657, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "eval_loss": 0.8459084033966064, | |
| "eval_runtime": 127.5971, | |
| "eval_samples_per_second": 57.603, | |
| "eval_steps_per_second": 0.901, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5996131528046421, | |
| "grad_norm": 3.4411079524968815, | |
| "learning_rate": 9.137251020192907e-07, | |
| "loss": 0.8513, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.6189555125725339, | |
| "grad_norm": 3.399830662079635, | |
| "learning_rate": 9.078653995690246e-07, | |
| "loss": 0.8375, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6382978723404256, | |
| "grad_norm": 3.60600371924574, | |
| "learning_rate": 9.018332902348388e-07, | |
| "loss": 0.8505, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.6576402321083172, | |
| "grad_norm": 3.331304480330792, | |
| "learning_rate": 8.956313238215823e-07, | |
| "loss": 0.8472, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.6769825918762089, | |
| "grad_norm": 3.6483016146418876, | |
| "learning_rate": 8.892621219336e-07, | |
| "loss": 0.8489, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.6963249516441006, | |
| "grad_norm": 3.421730346610911, | |
| "learning_rate": 8.827283768665648e-07, | |
| "loss": 0.8463, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7156673114119922, | |
| "grad_norm": 3.7149068640377694, | |
| "learning_rate": 8.760328504694317e-07, | |
| "loss": 0.8303, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.7350096711798839, | |
| "grad_norm": 3.5148473544431953, | |
| "learning_rate": 8.691783729769873e-07, | |
| "loss": 0.8285, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.7543520309477756, | |
| "grad_norm": 3.8143875519937227, | |
| "learning_rate": 8.621678418134963e-07, | |
| "loss": 0.8267, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "grad_norm": 3.665364587167835, | |
| "learning_rate": 8.550042203679439e-07, | |
| "loss": 0.8232, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "eval_loss": 0.8286838531494141, | |
| "eval_runtime": 127.6464, | |
| "eval_samples_per_second": 57.581, | |
| "eval_steps_per_second": 0.901, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.793036750483559, | |
| "grad_norm": 3.6010898270602216, | |
| "learning_rate": 8.476905367413957e-07, | |
| "loss": 0.8321, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.8123791102514507, | |
| "grad_norm": 3.4203770888001173, | |
| "learning_rate": 8.402298824670029e-07, | |
| "loss": 0.8261, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8317214700193424, | |
| "grad_norm": 3.5555049932560707, | |
| "learning_rate": 8.326254112031949e-07, | |
| "loss": 0.83, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.851063829787234, | |
| "grad_norm": 3.5446967824907465, | |
| "learning_rate": 8.248803374006113e-07, | |
| "loss": 0.8263, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.8704061895551257, | |
| "grad_norm": 3.582387564660827, | |
| "learning_rate": 8.169979349433358e-07, | |
| "loss": 0.8233, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.8897485493230174, | |
| "grad_norm": 3.477179043854488, | |
| "learning_rate": 8.089815357650089e-07, | |
| "loss": 0.8224, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 3.5228566797833176, | |
| "learning_rate": 8.008345284404003e-07, | |
| "loss": 0.823, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.9284332688588007, | |
| "grad_norm": 3.5831989534603235, | |
| "learning_rate": 7.925603567530418e-07, | |
| "loss": 0.8363, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9477756286266924, | |
| "grad_norm": 3.5406680784685003, | |
| "learning_rate": 7.841625182395206e-07, | |
| "loss": 0.8396, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "grad_norm": 3.586661139550379, | |
| "learning_rate": 7.756445627110522e-07, | |
| "loss": 0.8179, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "eval_loss": 0.8162312507629395, | |
| "eval_runtime": 127.7269, | |
| "eval_samples_per_second": 57.545, | |
| "eval_steps_per_second": 0.9, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9864603481624759, | |
| "grad_norm": 3.583872815575711, | |
| "learning_rate": 7.670100907529557e-07, | |
| "loss": 0.8267, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.0058027079303675, | |
| "grad_norm": 3.4543546378461354, | |
| "learning_rate": 7.582627522026685e-07, | |
| "loss": 0.7992, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.0251450676982592, | |
| "grad_norm": 3.4033484910260197, | |
| "learning_rate": 7.49406244606939e-07, | |
| "loss": 0.7297, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.0444874274661509, | |
| "grad_norm": 3.4147333856503272, | |
| "learning_rate": 7.404443116588547e-07, | |
| "loss": 0.7652, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 3.4496152121107486, | |
| "learning_rate": 7.31380741615363e-07, | |
| "loss": 0.7412, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.0831721470019342, | |
| "grad_norm": 3.327696184514914, | |
| "learning_rate": 7.222193656959546e-07, | |
| "loss": 0.7637, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.1025145067698259, | |
| "grad_norm": 3.7061830831988227, | |
| "learning_rate": 7.129640564631863e-07, | |
| "loss": 0.7634, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.1218568665377175, | |
| "grad_norm": 4.192820471653476, | |
| "learning_rate": 7.036187261857288e-07, | |
| "loss": 0.7502, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1411992263056092, | |
| "grad_norm": 3.6982147551712887, | |
| "learning_rate": 6.941873251846293e-07, | |
| "loss": 0.7737, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "grad_norm": 3.6871631539510807, | |
| "learning_rate": 6.846738401634898e-07, | |
| "loss": 0.7558, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "eval_loss": 0.8122316598892212, | |
| "eval_runtime": 127.708, | |
| "eval_samples_per_second": 57.553, | |
| "eval_steps_per_second": 0.9, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1798839458413926, | |
| "grad_norm": 3.6376198365580934, | |
| "learning_rate": 6.750822925232663e-07, | |
| "loss": 0.7701, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.1992263056092844, | |
| "grad_norm": 3.5708682961272316, | |
| "learning_rate": 6.654167366624008e-07, | |
| "loss": 0.7641, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.218568665377176, | |
| "grad_norm": 3.71858745567317, | |
| "learning_rate": 6.556812582630059e-07, | |
| "loss": 0.7481, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.2379110251450678, | |
| "grad_norm": 3.3987189299432217, | |
| "learning_rate": 6.458799725638248e-07, | |
| "loss": 0.7494, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.2572533849129595, | |
| "grad_norm": 3.582889984477799, | |
| "learning_rate": 6.36017022620698e-07, | |
| "loss": 0.7291, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.2765957446808511, | |
| "grad_norm": 3.7598968173069776, | |
| "learning_rate": 6.260965775552713e-07, | |
| "loss": 0.7643, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.2959381044487428, | |
| "grad_norm": 3.489163050088922, | |
| "learning_rate": 6.161228307926858e-07, | |
| "loss": 0.7374, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.3152804642166345, | |
| "grad_norm": 3.75863899391321, | |
| "learning_rate": 6.060999982889954e-07, | |
| "loss": 0.7568, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.3346228239845261, | |
| "grad_norm": 3.3559127612246233, | |
| "learning_rate": 5.960323167490588e-07, | |
| "loss": 0.7453, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "grad_norm": 3.490356836359392, | |
| "learning_rate": 5.859240418356614e-07, | |
| "loss": 0.7602, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "eval_loss": 0.8059322834014893, | |
| "eval_runtime": 127.7998, | |
| "eval_samples_per_second": 57.512, | |
| "eval_steps_per_second": 0.9, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3733075435203095, | |
| "grad_norm": 3.6757853156395606, | |
| "learning_rate": 5.757794463706253e-07, | |
| "loss": 0.7523, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.3926499032882012, | |
| "grad_norm": 3.5063904211356385, | |
| "learning_rate": 5.656028185286637e-07, | |
| "loss": 0.7496, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.4119922630560928, | |
| "grad_norm": 3.4960857334989806, | |
| "learning_rate": 5.553984600247463e-07, | |
| "loss": 0.7647, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.4313346228239845, | |
| "grad_norm": 3.6440901222517152, | |
| "learning_rate": 5.451706842957421e-07, | |
| "loss": 0.718, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.4506769825918762, | |
| "grad_norm": 3.7711569042281576, | |
| "learning_rate": 5.349238146771061e-07, | |
| "loss": 0.7583, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.4700193423597678, | |
| "grad_norm": 3.4413207786627957, | |
| "learning_rate": 5.246621825753827e-07, | |
| "loss": 0.7435, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.4893617021276595, | |
| "grad_norm": 3.6324458377021265, | |
| "learning_rate": 5.143901256372967e-07, | |
| "loss": 0.7397, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.5087040618955512, | |
| "grad_norm": 3.5785578278016485, | |
| "learning_rate": 5.041119859162068e-07, | |
| "loss": 0.7295, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.528046421663443, | |
| "grad_norm": 3.5486429985289116, | |
| "learning_rate": 4.938321080366968e-07, | |
| "loss": 0.7484, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "grad_norm": 3.563717839228555, | |
| "learning_rate": 4.835548373580792e-07, | |
| "loss": 0.771, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "eval_loss": 0.7998485565185547, | |
| "eval_runtime": 127.8334, | |
| "eval_samples_per_second": 57.497, | |
| "eval_steps_per_second": 0.9, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5667311411992264, | |
| "grad_norm": 3.531231527366412, | |
| "learning_rate": 4.73284518137589e-07, | |
| "loss": 0.7472, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.5860735009671179, | |
| "grad_norm": 3.4703134540594354, | |
| "learning_rate": 4.630254916940423e-07, | |
| "loss": 0.7349, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.6054158607350097, | |
| "grad_norm": 3.571312117827743, | |
| "learning_rate": 4.5278209457273825e-07, | |
| "loss": 0.7539, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.6247582205029012, | |
| "grad_norm": 3.6699695284667304, | |
| "learning_rate": 4.425586567123779e-07, | |
| "loss": 0.7561, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.644100580270793, | |
| "grad_norm": 3.444249068608954, | |
| "learning_rate": 4.3235949961477627e-07, | |
| "loss": 0.7456, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.6634429400386848, | |
| "grad_norm": 3.42086843430337, | |
| "learning_rate": 4.2218893451814e-07, | |
| "loss": 0.7458, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.6827852998065764, | |
| "grad_norm": 3.3865347869200857, | |
| "learning_rate": 4.120512605746842e-07, | |
| "loss": 0.7394, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.702127659574468, | |
| "grad_norm": 3.677592972618459, | |
| "learning_rate": 4.019507630333577e-07, | |
| "loss": 0.7365, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.7214700193423598, | |
| "grad_norm": 3.6136504230202435, | |
| "learning_rate": 3.9189171142844553e-07, | |
| "loss": 0.7559, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "grad_norm": 3.572140075959248, | |
| "learning_rate": 3.8187835777481375e-07, | |
| "loss": 0.753, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "eval_loss": 0.7954617738723755, | |
| "eval_runtime": 127.7956, | |
| "eval_samples_per_second": 57.514, | |
| "eval_steps_per_second": 0.9, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.760154738878143, | |
| "grad_norm": 3.4557270180643984, | |
| "learning_rate": 3.7191493477056086e-07, | |
| "loss": 0.7416, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 1.7794970986460348, | |
| "grad_norm": 3.5454869396159294, | |
| "learning_rate": 3.620056540078323e-07, | |
| "loss": 0.7465, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.7988394584139265, | |
| "grad_norm": 3.5193296679089197, | |
| "learning_rate": 3.5215470419255897e-07, | |
| "loss": 0.7381, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 3.6300180309499055, | |
| "learning_rate": 3.423662493738687e-07, | |
| "loss": 0.7393, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.8375241779497098, | |
| "grad_norm": 3.6128465808218673, | |
| "learning_rate": 3.3264442718392014e-07, | |
| "loss": 0.7361, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.8568665377176017, | |
| "grad_norm": 3.8823278499726266, | |
| "learning_rate": 3.229933470889038e-07, | |
| "loss": 0.743, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.8762088974854931, | |
| "grad_norm": 3.4195861313789444, | |
| "learning_rate": 3.134170886519486e-07, | |
| "loss": 0.722, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 1.895551257253385, | |
| "grad_norm": 3.6240796400929005, | |
| "learning_rate": 3.039196998086687e-07, | |
| "loss": 0.7292, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.9148936170212765, | |
| "grad_norm": 3.602124125688223, | |
| "learning_rate": 2.9450519515607963e-07, | |
| "loss": 0.7402, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "grad_norm": 3.8410929027045757, | |
| "learning_rate": 2.8517755425560663e-07, | |
| "loss": 0.7479, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "eval_loss": 0.7919633984565735, | |
| "eval_runtime": 127.71, | |
| "eval_samples_per_second": 57.552, | |
| "eval_steps_per_second": 0.9, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9535783365570598, | |
| "grad_norm": 3.553013732530746, | |
| "learning_rate": 2.7594071995090283e-07, | |
| "loss": 0.7414, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 1.9729206963249517, | |
| "grad_norm": 3.726389314193614, | |
| "learning_rate": 2.667985967011878e-07, | |
| "loss": 0.7442, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.9922630560928434, | |
| "grad_norm": 3.6891205891519196, | |
| "learning_rate": 2.577550489308123e-07, | |
| "loss": 0.7445, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.011605415860735, | |
| "grad_norm": 4.183763879873082, | |
| "learning_rate": 2.488138993957452e-07, | |
| "loss": 0.7238, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.0309477756286265, | |
| "grad_norm": 3.5335122665404533, | |
| "learning_rate": 2.3997892756767394e-07, | |
| "loss": 0.7148, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.0502901353965184, | |
| "grad_norm": 3.877744660356318, | |
| "learning_rate": 2.3125386803640183e-07, | |
| "loss": 0.6955, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.0696324951644103, | |
| "grad_norm": 3.7193372520055847, | |
| "learning_rate": 2.226424089312174e-07, | |
| "loss": 0.7016, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.0889748549323017, | |
| "grad_norm": 3.6986475178104063, | |
| "learning_rate": 2.1414819036190157e-07, | |
| "loss": 0.7054, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.1083172147001936, | |
| "grad_norm": 3.7985188083265258, | |
| "learning_rate": 2.057748028800344e-07, | |
| "loss": 0.6801, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 3.812215835510966, | |
| "learning_rate": 1.9752578596124952e-07, | |
| "loss": 0.6735, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_loss": 0.7972328066825867, | |
| "eval_runtime": 127.877, | |
| "eval_samples_per_second": 57.477, | |
| "eval_steps_per_second": 0.899, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.147001934235977, | |
| "grad_norm": 3.556112328434251, | |
| "learning_rate": 1.8940462650907912e-07, | |
| "loss": 0.6887, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 2.1663442940038684, | |
| "grad_norm": 4.031706169414501, | |
| "learning_rate": 1.8141475738102086e-07, | |
| "loss": 0.6829, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.1856866537717603, | |
| "grad_norm": 3.9038438081899636, | |
| "learning_rate": 1.735595559374508e-07, | |
| "loss": 0.701, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 2.2050290135396517, | |
| "grad_norm": 3.802389298862028, | |
| "learning_rate": 1.6584234261399532e-07, | |
| "loss": 0.6962, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.2243713733075436, | |
| "grad_norm": 3.95265234731058, | |
| "learning_rate": 1.5826637951796474e-07, | |
| "loss": 0.704, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.243713733075435, | |
| "grad_norm": 3.920278049666238, | |
| "learning_rate": 1.5083486904944387e-07, | |
| "loss": 0.6866, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.263056092843327, | |
| "grad_norm": 3.917007172675932, | |
| "learning_rate": 1.4355095254761974e-07, | |
| "loss": 0.6921, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.2823984526112184, | |
| "grad_norm": 3.7991457510336915, | |
| "learning_rate": 1.3641770896292082e-07, | |
| "loss": 0.6813, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.3017408123791103, | |
| "grad_norm": 3.687358486996167, | |
| "learning_rate": 1.2943815355552851e-07, | |
| "loss": 0.6928, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "grad_norm": 3.6543009032276026, | |
| "learning_rate": 1.226152366208104e-07, | |
| "loss": 0.7116, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "eval_loss": 0.7951865196228027, | |
| "eval_runtime": 127.85, | |
| "eval_samples_per_second": 57.489, | |
| "eval_steps_per_second": 0.899, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.3404255319148937, | |
| "grad_norm": 3.8730917233152384, | |
| "learning_rate": 1.1595184224221466e-07, | |
| "loss": 0.692, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.359767891682785, | |
| "grad_norm": 4.03531309851571, | |
| "learning_rate": 1.0945078707215221e-07, | |
| "loss": 0.6865, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.379110251450677, | |
| "grad_norm": 3.872705003238953, | |
| "learning_rate": 1.0311481914138371e-07, | |
| "loss": 0.6985, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.398452611218569, | |
| "grad_norm": 3.720403831965154, | |
| "learning_rate": 9.6946616697411e-08, | |
| "loss": 0.6968, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.4177949709864603, | |
| "grad_norm": 3.8402045858860214, | |
| "learning_rate": 9.094878707236841e-08, | |
| "loss": 0.7002, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.437137330754352, | |
| "grad_norm": 3.870172963176105, | |
| "learning_rate": 8.512386558088919e-08, | |
| "loss": 0.7018, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.4564796905222437, | |
| "grad_norm": 3.844232920183782, | |
| "learning_rate": 7.947431444841452e-08, | |
| "loss": 0.6953, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.4758220502901356, | |
| "grad_norm": 3.9369438805219357, | |
| "learning_rate": 7.400252177039784e-08, | |
| "loss": 0.7023, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.495164410058027, | |
| "grad_norm": 3.7400884688082217, | |
| "learning_rate": 6.871080050284394e-08, | |
| "loss": 0.6822, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "grad_norm": 3.6809188260562498, | |
| "learning_rate": 6.360138748461013e-08, | |
| "loss": 0.6822, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "eval_loss": 0.794310986995697, | |
| "eval_runtime": 128.0518, | |
| "eval_samples_per_second": 57.399, | |
| "eval_steps_per_second": 0.898, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.5338491295938104, | |
| "grad_norm": 4.16744242512269, | |
| "learning_rate": 5.867644249188247e-08, | |
| "loss": 0.6991, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 2.5531914893617023, | |
| "grad_norm": 3.851977029205454, | |
| "learning_rate": 5.3938047325226944e-08, | |
| "loss": 0.6986, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.5725338491295937, | |
| "grad_norm": 4.101916973231146, | |
| "learning_rate": 4.9388204929601326e-08, | |
| "loss": 0.6854, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.5918762088974856, | |
| "grad_norm": 3.8486651679301023, | |
| "learning_rate": 4.5028838547699346e-08, | |
| "loss": 0.6924, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.611218568665377, | |
| "grad_norm": 3.717552659671349, | |
| "learning_rate": 4.0861790906985884e-08, | |
| "loss": 0.6888, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.630560928433269, | |
| "grad_norm": 3.9259495383255314, | |
| "learning_rate": 3.6888823440766214e-08, | |
| "loss": 0.6914, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.6499032882011604, | |
| "grad_norm": 3.627999141521453, | |
| "learning_rate": 3.311161554361874e-08, | |
| "loss": 0.6773, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 2.6692456479690523, | |
| "grad_norm": 3.9122944066082934, | |
| "learning_rate": 2.9531763861505964e-08, | |
| "loss": 0.6908, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.6885880077369437, | |
| "grad_norm": 3.7957015499021547, | |
| "learning_rate": 2.6150781616863794e-08, | |
| "loss": 0.6638, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 2.7079303675048356, | |
| "grad_norm": 3.9116633525167033, | |
| "learning_rate": 2.2970097968953994e-08, | |
| "loss": 0.6861, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.7079303675048356, | |
| "eval_loss": 0.7937687635421753, | |
| "eval_runtime": 127.6414, | |
| "eval_samples_per_second": 57.583, | |
| "eval_steps_per_second": 0.901, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 3.7729712553297987, | |
| "learning_rate": 1.9991057409751267e-08, | |
| "loss": 0.6911, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 2.746615087040619, | |
| "grad_norm": 3.834540719290449, | |
| "learning_rate": 1.7214919195619125e-08, | |
| "loss": 0.6909, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.7659574468085104, | |
| "grad_norm": 3.7151903198724, | |
| "learning_rate": 1.4642856815015758e-08, | |
| "loss": 0.6999, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 2.7852998065764023, | |
| "grad_norm": 3.973205161583112, | |
| "learning_rate": 1.2275957492453692e-08, | |
| "loss": 0.6933, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.804642166344294, | |
| "grad_norm": 3.8696685699188507, | |
| "learning_rate": 1.0115221728924706e-08, | |
| "loss": 0.6775, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.8239845261121856, | |
| "grad_norm": 4.076911148311998, | |
| "learning_rate": 8.161562878982398e-09, | |
| "loss": 0.6748, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.843326885880077, | |
| "grad_norm": 3.6416916556998817, | |
| "learning_rate": 6.415806764662524e-09, | |
| "loss": 0.6759, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 2.862669245647969, | |
| "grad_norm": 3.9225986402250754, | |
| "learning_rate": 4.8786913264033945e-09, | |
| "loss": 0.698, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.882011605415861, | |
| "grad_norm": 4.048565845552162, | |
| "learning_rate": 3.5508663111147306e-09, | |
| "loss": 0.6856, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 2.9013539651837523, | |
| "grad_norm": 4.082712170990346, | |
| "learning_rate": 2.432892997526026e-09, | |
| "loss": 0.6989, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.9013539651837523, | |
| "eval_loss": 0.7935436367988586, | |
| "eval_runtime": 127.7461, | |
| "eval_samples_per_second": 57.536, | |
| "eval_steps_per_second": 0.9, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.920696324951644, | |
| "grad_norm": 3.8537422532400827, | |
| "learning_rate": 1.5252439589311107e-09, | |
| "loss": 0.6929, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 2.9400386847195357, | |
| "grad_norm": 3.841072555465488, | |
| "learning_rate": 8.283028634287203e-10, | |
| "loss": 0.6963, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.9593810444874276, | |
| "grad_norm": 3.741338861355809, | |
| "learning_rate": 3.4236431174428094e-10, | |
| "loss": 0.683, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 2.978723404255319, | |
| "grad_norm": 3.962628716770498, | |
| "learning_rate": 6.763371270035457e-11, | |
| "loss": 0.684, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.9941972920696323, | |
| "step": 774, | |
| "total_flos": 4563620855808000.0, | |
| "train_loss": 0.7856120132968716, | |
| "train_runtime": 12058.8177, | |
| "train_samples_per_second": 16.457, | |
| "train_steps_per_second": 0.064 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 774, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4563620855808000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |