| { | |
| "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": 18.29429795265829, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7218, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03868471953578337, | |
| "grad_norm": 12.374896347790715, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5873, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.058027079303675046, | |
| "grad_norm": 7.9779911036982005, | |
| "learning_rate": 9.998943236640677e-07, | |
| "loss": 1.2946, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07736943907156674, | |
| "grad_norm": 4.26597805437282, | |
| "learning_rate": 9.995773393262229e-07, | |
| "loss": 1.1398, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09671179883945841, | |
| "grad_norm": 3.7901464214490552, | |
| "learning_rate": 9.99049180977439e-07, | |
| "loss": 1.0768, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.11605415860735009, | |
| "grad_norm": 3.719211053411099, | |
| "learning_rate": 9.983100718730718e-07, | |
| "loss": 1.0361, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13539651837524178, | |
| "grad_norm": 3.467191764227361, | |
| "learning_rate": 9.973603244384906e-07, | |
| "loss": 0.9909, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.15473887814313347, | |
| "grad_norm": 3.3499969223078385, | |
| "learning_rate": 9.9620034013701e-07, | |
| "loss": 0.9577, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.17408123791102514, | |
| "grad_norm": 3.53015393903578, | |
| "learning_rate": 9.948306093001933e-07, | |
| "loss": 0.9516, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "grad_norm": 3.39919184233192, | |
| "learning_rate": 9.932517109205849e-07, | |
| "loss": 0.9353, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "eval_loss": 0.9258183836936951, | |
| "eval_runtime": 127.8238, | |
| "eval_samples_per_second": 57.501, | |
| "eval_steps_per_second": 0.9, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 3.2060428601111997, | |
| "learning_rate": 9.914643124069666e-07, | |
| "loss": 0.9063, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.23210831721470018, | |
| "grad_norm": 3.666917330552863, | |
| "learning_rate": 9.89469169302242e-07, | |
| "loss": 0.9336, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2514506769825919, | |
| "grad_norm": 3.9838409320760357, | |
| "learning_rate": 9.872671249640626e-07, | |
| "loss": 0.9231, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.27079303675048355, | |
| "grad_norm": 3.518667919651293, | |
| "learning_rate": 9.848591102083375e-07, | |
| "loss": 0.9306, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2901353965183752, | |
| "grad_norm": 3.2598329874793666, | |
| "learning_rate": 9.822461429157716e-07, | |
| "loss": 0.9022, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.30947775628626695, | |
| "grad_norm": 3.496771288813608, | |
| "learning_rate": 9.794293276016023e-07, | |
| "loss": 0.8818, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3288201160541586, | |
| "grad_norm": 3.5451094564430847, | |
| "learning_rate": 9.764098549487155e-07, | |
| "loss": 0.8897, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.3481624758220503, | |
| "grad_norm": 3.2267999853614593, | |
| "learning_rate": 9.731890013043367e-07, | |
| "loss": 0.8789, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.36750483558994196, | |
| "grad_norm": 3.5002585177526213, | |
| "learning_rate": 9.697681281405128e-07, | |
| "loss": 0.8748, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "grad_norm": 3.365687264198066, | |
| "learning_rate": 9.6614868147861e-07, | |
| "loss": 0.8526, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "eval_loss": 0.8687371015548706, | |
| "eval_runtime": 127.6254, | |
| "eval_samples_per_second": 57.59, | |
| "eval_steps_per_second": 0.901, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.40618955512572535, | |
| "grad_norm": 3.379303752602587, | |
| "learning_rate": 9.623321912780744e-07, | |
| "loss": 0.865, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.425531914893617, | |
| "grad_norm": 3.544869473720659, | |
| "learning_rate": 9.583202707897073e-07, | |
| "loss": 0.8807, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4448742746615087, | |
| "grad_norm": 3.7015145875218574, | |
| "learning_rate": 9.54114615873738e-07, | |
| "loss": 0.8707, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.46421663442940037, | |
| "grad_norm": 3.6640282382892218, | |
| "learning_rate": 9.497170042829736e-07, | |
| "loss": 0.8644, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4835589941972921, | |
| "grad_norm": 3.5307683407864134, | |
| "learning_rate": 9.451292949113329e-07, | |
| "loss": 0.8794, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5029013539651838, | |
| "grad_norm": 3.402033918475512, | |
| "learning_rate": 9.403534270080829e-07, | |
| "loss": 0.872, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5222437137330754, | |
| "grad_norm": 3.310826176140146, | |
| "learning_rate": 9.353914193581072e-07, | |
| "loss": 0.8677, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.5415860735009671, | |
| "grad_norm": 3.2406150003457452, | |
| "learning_rate": 9.302453694285548e-07, | |
| "loss": 0.8474, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5609284332688588, | |
| "grad_norm": 3.4974393211031907, | |
| "learning_rate": 9.249174524822305e-07, | |
| "loss": 0.8465, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "grad_norm": 3.4284093385969925, | |
| "learning_rate": 9.19409920658098e-07, | |
| "loss": 0.8383, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "eval_loss": 0.8431525826454163, | |
| "eval_runtime": 127.7056, | |
| "eval_samples_per_second": 57.554, | |
| "eval_steps_per_second": 0.901, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5996131528046421, | |
| "grad_norm": 3.6968279151532384, | |
| "learning_rate": 9.137251020192907e-07, | |
| "loss": 0.8423, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.6189555125725339, | |
| "grad_norm": 3.50833637505395, | |
| "learning_rate": 9.078653995690246e-07, | |
| "loss": 0.8387, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6382978723404256, | |
| "grad_norm": 3.3457806997461987, | |
| "learning_rate": 9.018332902348388e-07, | |
| "loss": 0.8633, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.6576402321083172, | |
| "grad_norm": 3.804547305635231, | |
| "learning_rate": 8.956313238215823e-07, | |
| "loss": 0.8376, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.6769825918762089, | |
| "grad_norm": 3.640636148585517, | |
| "learning_rate": 8.892621219336e-07, | |
| "loss": 0.8438, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.6963249516441006, | |
| "grad_norm": 3.4756125888954723, | |
| "learning_rate": 8.827283768665648e-07, | |
| "loss": 0.8607, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7156673114119922, | |
| "grad_norm": 3.3875203995409215, | |
| "learning_rate": 8.760328504694317e-07, | |
| "loss": 0.8301, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.7350096711798839, | |
| "grad_norm": 3.372082258308208, | |
| "learning_rate": 8.691783729769873e-07, | |
| "loss": 0.8263, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.7543520309477756, | |
| "grad_norm": 3.3557984138015637, | |
| "learning_rate": 8.621678418134963e-07, | |
| "loss": 0.8187, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "grad_norm": 3.335888461384703, | |
| "learning_rate": 8.550042203679439e-07, | |
| "loss": 0.8264, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "eval_loss": 0.8269398808479309, | |
| "eval_runtime": 127.5697, | |
| "eval_samples_per_second": 57.616, | |
| "eval_steps_per_second": 0.901, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.793036750483559, | |
| "grad_norm": 3.423751371323696, | |
| "learning_rate": 8.476905367413957e-07, | |
| "loss": 0.8231, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.8123791102514507, | |
| "grad_norm": 3.512146342996446, | |
| "learning_rate": 8.402298824670029e-07, | |
| "loss": 0.8383, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8317214700193424, | |
| "grad_norm": 3.679961917362473, | |
| "learning_rate": 8.326254112031949e-07, | |
| "loss": 0.8352, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.851063829787234, | |
| "grad_norm": 3.5282426701829825, | |
| "learning_rate": 8.248803374006113e-07, | |
| "loss": 0.8121, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.8704061895551257, | |
| "grad_norm": 3.6512031815177486, | |
| "learning_rate": 8.169979349433358e-07, | |
| "loss": 0.8376, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.8897485493230174, | |
| "grad_norm": 3.235414274362894, | |
| "learning_rate": 8.089815357650089e-07, | |
| "loss": 0.8303, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 3.253247851140818, | |
| "learning_rate": 8.008345284404003e-07, | |
| "loss": 0.8302, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.9284332688588007, | |
| "grad_norm": 3.398661312301685, | |
| "learning_rate": 7.925603567530418e-07, | |
| "loss": 0.8388, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9477756286266924, | |
| "grad_norm": 3.4458534561793144, | |
| "learning_rate": 7.841625182395206e-07, | |
| "loss": 0.8354, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "grad_norm": 3.52003203310209, | |
| "learning_rate": 7.756445627110522e-07, | |
| "loss": 0.825, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "eval_loss": 0.8141016960144043, | |
| "eval_runtime": 127.5727, | |
| "eval_samples_per_second": 57.614, | |
| "eval_steps_per_second": 0.901, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9864603481624759, | |
| "grad_norm": 3.5618840305239043, | |
| "learning_rate": 7.670100907529557e-07, | |
| "loss": 0.8222, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.0058027079303675, | |
| "grad_norm": 3.8599831213139533, | |
| "learning_rate": 7.582627522026685e-07, | |
| "loss": 0.8146, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.0251450676982592, | |
| "grad_norm": 3.541052361485808, | |
| "learning_rate": 7.49406244606939e-07, | |
| "loss": 0.7562, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.0444874274661509, | |
| "grad_norm": 3.6633110448636996, | |
| "learning_rate": 7.404443116588547e-07, | |
| "loss": 0.7751, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 3.488605973668302, | |
| "learning_rate": 7.31380741615363e-07, | |
| "loss": 0.7661, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.0831721470019342, | |
| "grad_norm": 3.6599674759653698, | |
| "learning_rate": 7.222193656959546e-07, | |
| "loss": 0.7692, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.1025145067698259, | |
| "grad_norm": 3.7997115771096177, | |
| "learning_rate": 7.129640564631863e-07, | |
| "loss": 0.7556, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.1218568665377175, | |
| "grad_norm": 3.6182500776322213, | |
| "learning_rate": 7.036187261857288e-07, | |
| "loss": 0.7641, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1411992263056092, | |
| "grad_norm": 3.4091769100972584, | |
| "learning_rate": 6.941873251846293e-07, | |
| "loss": 0.7636, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "grad_norm": 3.5604176730475974, | |
| "learning_rate": 6.846738401634898e-07, | |
| "loss": 0.7583, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "eval_loss": 0.8098444938659668, | |
| "eval_runtime": 127.8067, | |
| "eval_samples_per_second": 57.509, | |
| "eval_steps_per_second": 0.9, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1798839458413926, | |
| "grad_norm": 3.549344381739553, | |
| "learning_rate": 6.750822925232663e-07, | |
| "loss": 0.7756, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.1992263056092844, | |
| "grad_norm": 3.925155393010541, | |
| "learning_rate": 6.654167366624008e-07, | |
| "loss": 0.7463, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.218568665377176, | |
| "grad_norm": 3.4515420712536202, | |
| "learning_rate": 6.556812582630059e-07, | |
| "loss": 0.7564, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.2379110251450678, | |
| "grad_norm": 3.516145369254403, | |
| "learning_rate": 6.458799725638248e-07, | |
| "loss": 0.7619, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.2572533849129595, | |
| "grad_norm": 3.588101859460772, | |
| "learning_rate": 6.36017022620698e-07, | |
| "loss": 0.7579, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.2765957446808511, | |
| "grad_norm": 3.499006102072433, | |
| "learning_rate": 6.260965775552713e-07, | |
| "loss": 0.7517, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.2959381044487428, | |
| "grad_norm": 3.8452645432049355, | |
| "learning_rate": 6.161228307926858e-07, | |
| "loss": 0.7615, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.3152804642166345, | |
| "grad_norm": 3.530157481700969, | |
| "learning_rate": 6.060999982889954e-07, | |
| "loss": 0.7349, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.3346228239845261, | |
| "grad_norm": 3.511064558056812, | |
| "learning_rate": 5.960323167490588e-07, | |
| "loss": 0.7453, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "grad_norm": 3.681202337191524, | |
| "learning_rate": 5.859240418356614e-07, | |
| "loss": 0.7459, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "eval_loss": 0.8035895228385925, | |
| "eval_runtime": 127.6066, | |
| "eval_samples_per_second": 57.599, | |
| "eval_steps_per_second": 0.901, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3733075435203095, | |
| "grad_norm": 3.7050410147150594, | |
| "learning_rate": 5.757794463706253e-07, | |
| "loss": 0.7603, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.3926499032882012, | |
| "grad_norm": 3.6286211129571107, | |
| "learning_rate": 5.656028185286637e-07, | |
| "loss": 0.7581, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.4119922630560928, | |
| "grad_norm": 3.705421837944294, | |
| "learning_rate": 5.553984600247463e-07, | |
| "loss": 0.7422, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.4313346228239845, | |
| "grad_norm": 3.7953601373705026, | |
| "learning_rate": 5.451706842957421e-07, | |
| "loss": 0.7502, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.4506769825918762, | |
| "grad_norm": 3.8014187386742777, | |
| "learning_rate": 5.349238146771061e-07, | |
| "loss": 0.7483, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.4700193423597678, | |
| "grad_norm": 3.504282700491077, | |
| "learning_rate": 5.246621825753827e-07, | |
| "loss": 0.7396, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.4893617021276595, | |
| "grad_norm": 3.5095582192022743, | |
| "learning_rate": 5.143901256372967e-07, | |
| "loss": 0.7377, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.5087040618955512, | |
| "grad_norm": 3.4475672003161257, | |
| "learning_rate": 5.041119859162068e-07, | |
| "loss": 0.7393, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.528046421663443, | |
| "grad_norm": 3.571290931797336, | |
| "learning_rate": 4.938321080366968e-07, | |
| "loss": 0.7338, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "grad_norm": 3.6205663721700914, | |
| "learning_rate": 4.835548373580792e-07, | |
| "loss": 0.7534, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "eval_loss": 0.7977527379989624, | |
| "eval_runtime": 127.6, | |
| "eval_samples_per_second": 57.602, | |
| "eval_steps_per_second": 0.901, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5667311411992264, | |
| "grad_norm": 3.5535922972652334, | |
| "learning_rate": 4.73284518137589e-07, | |
| "loss": 0.7492, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.5860735009671179, | |
| "grad_norm": 3.5273195524585454, | |
| "learning_rate": 4.630254916940423e-07, | |
| "loss": 0.7479, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.6054158607350097, | |
| "grad_norm": 3.5266217127055888, | |
| "learning_rate": 4.5278209457273825e-07, | |
| "loss": 0.732, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.6247582205029012, | |
| "grad_norm": 3.9206092134824697, | |
| "learning_rate": 4.425586567123779e-07, | |
| "loss": 0.7312, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.644100580270793, | |
| "grad_norm": 3.5562647759590305, | |
| "learning_rate": 4.3235949961477627e-07, | |
| "loss": 0.7482, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.6634429400386848, | |
| "grad_norm": 3.4662308137356646, | |
| "learning_rate": 4.2218893451814e-07, | |
| "loss": 0.7393, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.6827852998065764, | |
| "grad_norm": 3.7128827415530004, | |
| "learning_rate": 4.120512605746842e-07, | |
| "loss": 0.7432, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.702127659574468, | |
| "grad_norm": 3.5592695862370074, | |
| "learning_rate": 4.019507630333577e-07, | |
| "loss": 0.7404, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.7214700193423598, | |
| "grad_norm": 3.710326956161109, | |
| "learning_rate": 3.9189171142844553e-07, | |
| "loss": 0.7559, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "grad_norm": 3.523728094197245, | |
| "learning_rate": 3.8187835777481375e-07, | |
| "loss": 0.7429, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "eval_loss": 0.7935256361961365, | |
| "eval_runtime": 127.7706, | |
| "eval_samples_per_second": 57.525, | |
| "eval_steps_per_second": 0.9, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.760154738878143, | |
| "grad_norm": 3.524155101820067, | |
| "learning_rate": 3.7191493477056086e-07, | |
| "loss": 0.721, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 1.7794970986460348, | |
| "grad_norm": 3.307755294406217, | |
| "learning_rate": 3.620056540078323e-07, | |
| "loss": 0.7279, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.7988394584139265, | |
| "grad_norm": 3.450615128725478, | |
| "learning_rate": 3.5215470419255897e-07, | |
| "loss": 0.7329, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 3.4244105953540442, | |
| "learning_rate": 3.423662493738687e-07, | |
| "loss": 0.7502, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.8375241779497098, | |
| "grad_norm": 3.6595088524698576, | |
| "learning_rate": 3.3264442718392014e-07, | |
| "loss": 0.741, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.8568665377176017, | |
| "grad_norm": 3.898153864979138, | |
| "learning_rate": 3.229933470889038e-07, | |
| "loss": 0.7573, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.8762088974854931, | |
| "grad_norm": 3.446217110493551, | |
| "learning_rate": 3.134170886519486e-07, | |
| "loss": 0.7414, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 1.895551257253385, | |
| "grad_norm": 3.640466931045702, | |
| "learning_rate": 3.039196998086687e-07, | |
| "loss": 0.7201, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.9148936170212765, | |
| "grad_norm": 3.8276180248573355, | |
| "learning_rate": 2.9450519515607963e-07, | |
| "loss": 0.7475, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "grad_norm": 3.6601723064769893, | |
| "learning_rate": 2.8517755425560663e-07, | |
| "loss": 0.7409, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "eval_loss": 0.7896685004234314, | |
| "eval_runtime": 127.5357, | |
| "eval_samples_per_second": 57.631, | |
| "eval_steps_per_second": 0.902, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9535783365570598, | |
| "grad_norm": 3.5497802979146456, | |
| "learning_rate": 2.7594071995090283e-07, | |
| "loss": 0.7451, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 1.9729206963249517, | |
| "grad_norm": 3.402165439915187, | |
| "learning_rate": 2.667985967011878e-07, | |
| "loss": 0.7266, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.9922630560928434, | |
| "grad_norm": 3.7514406502780964, | |
| "learning_rate": 2.577550489308123e-07, | |
| "loss": 0.7307, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.011605415860735, | |
| "grad_norm": 4.09122639389021, | |
| "learning_rate": 2.488138993957452e-07, | |
| "loss": 0.7063, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.0309477756286265, | |
| "grad_norm": 4.141130274306359, | |
| "learning_rate": 2.3997892756767394e-07, | |
| "loss": 0.6951, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.0502901353965184, | |
| "grad_norm": 3.67252888168734, | |
| "learning_rate": 2.3125386803640183e-07, | |
| "loss": 0.7084, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.0696324951644103, | |
| "grad_norm": 3.5582226093164433, | |
| "learning_rate": 2.226424089312174e-07, | |
| "loss": 0.7051, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.0889748549323017, | |
| "grad_norm": 3.696880117962318, | |
| "learning_rate": 2.1414819036190157e-07, | |
| "loss": 0.6955, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.1083172147001936, | |
| "grad_norm": 3.6604807985161303, | |
| "learning_rate": 2.057748028800344e-07, | |
| "loss": 0.6965, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 3.751745079490738, | |
| "learning_rate": 1.9752578596124952e-07, | |
| "loss": 0.6927, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_loss": 0.7948961853981018, | |
| "eval_runtime": 127.4872, | |
| "eval_samples_per_second": 57.653, | |
| "eval_steps_per_second": 0.902, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.147001934235977, | |
| "grad_norm": 3.9167862522448957, | |
| "learning_rate": 1.8940462650907912e-07, | |
| "loss": 0.6882, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 2.1663442940038684, | |
| "grad_norm": 3.560294152990905, | |
| "learning_rate": 1.8141475738102086e-07, | |
| "loss": 0.6858, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.1856866537717603, | |
| "grad_norm": 3.642012709130625, | |
| "learning_rate": 1.735595559374508e-07, | |
| "loss": 0.6763, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 2.2050290135396517, | |
| "grad_norm": 3.8138504169072984, | |
| "learning_rate": 1.6584234261399532e-07, | |
| "loss": 0.703, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.2243713733075436, | |
| "grad_norm": 4.103706228250993, | |
| "learning_rate": 1.5826637951796474e-07, | |
| "loss": 0.6956, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.243713733075435, | |
| "grad_norm": 3.5059744293726336, | |
| "learning_rate": 1.5083486904944387e-07, | |
| "loss": 0.6892, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.263056092843327, | |
| "grad_norm": 4.028201423509813, | |
| "learning_rate": 1.4355095254761974e-07, | |
| "loss": 0.6863, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.2823984526112184, | |
| "grad_norm": 3.8119585236671893, | |
| "learning_rate": 1.3641770896292082e-07, | |
| "loss": 0.6946, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.3017408123791103, | |
| "grad_norm": 3.9781395096904575, | |
| "learning_rate": 1.2943815355552851e-07, | |
| "loss": 0.6938, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "grad_norm": 3.8219447395207364, | |
| "learning_rate": 1.226152366208104e-07, | |
| "loss": 0.6889, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "eval_loss": 0.7934096455574036, | |
| "eval_runtime": 127.5736, | |
| "eval_samples_per_second": 57.614, | |
| "eval_steps_per_second": 0.901, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.3404255319148937, | |
| "grad_norm": 3.784762679951525, | |
| "learning_rate": 1.1595184224221466e-07, | |
| "loss": 0.6897, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.359767891682785, | |
| "grad_norm": 4.023132354100149, | |
| "learning_rate": 1.0945078707215221e-07, | |
| "loss": 0.6873, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.379110251450677, | |
| "grad_norm": 3.7520760586749953, | |
| "learning_rate": 1.0311481914138371e-07, | |
| "loss": 0.7062, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.398452611218569, | |
| "grad_norm": 3.777209444957414, | |
| "learning_rate": 9.6946616697411e-08, | |
| "loss": 0.6909, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.4177949709864603, | |
| "grad_norm": 3.7303835054301593, | |
| "learning_rate": 9.094878707236841e-08, | |
| "loss": 0.6934, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.437137330754352, | |
| "grad_norm": 3.5630150859580425, | |
| "learning_rate": 8.512386558088919e-08, | |
| "loss": 0.7, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.4564796905222437, | |
| "grad_norm": 3.813667924982544, | |
| "learning_rate": 7.947431444841452e-08, | |
| "loss": 0.6933, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.4758220502901356, | |
| "grad_norm": 3.819792577178714, | |
| "learning_rate": 7.400252177039784e-08, | |
| "loss": 0.6949, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.495164410058027, | |
| "grad_norm": 3.9105258158809204, | |
| "learning_rate": 6.871080050284394e-08, | |
| "loss": 0.6971, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "grad_norm": 3.864980953874583, | |
| "learning_rate": 6.360138748461013e-08, | |
| "loss": 0.7015, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "eval_loss": 0.7927743792533875, | |
| "eval_runtime": 127.5567, | |
| "eval_samples_per_second": 57.621, | |
| "eval_steps_per_second": 0.902, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.5338491295938104, | |
| "grad_norm": 3.8434289557296815, | |
| "learning_rate": 5.867644249188247e-08, | |
| "loss": 0.7014, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 2.5531914893617023, | |
| "grad_norm": 3.908654718811004, | |
| "learning_rate": 5.3938047325226944e-08, | |
| "loss": 0.6839, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.5725338491295937, | |
| "grad_norm": 3.778555696115051, | |
| "learning_rate": 4.9388204929601326e-08, | |
| "loss": 0.6854, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.5918762088974856, | |
| "grad_norm": 3.6881527806646406, | |
| "learning_rate": 4.5028838547699346e-08, | |
| "loss": 0.7061, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.611218568665377, | |
| "grad_norm": 3.8412216704216475, | |
| "learning_rate": 4.0861790906985884e-08, | |
| "loss": 0.676, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.630560928433269, | |
| "grad_norm": 3.769050419414534, | |
| "learning_rate": 3.6888823440766214e-08, | |
| "loss": 0.7027, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.6499032882011604, | |
| "grad_norm": 3.84774623904604, | |
| "learning_rate": 3.311161554361874e-08, | |
| "loss": 0.6912, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 2.6692456479690523, | |
| "grad_norm": 3.9446313135020596, | |
| "learning_rate": 2.9531763861505964e-08, | |
| "loss": 0.6756, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.6885880077369437, | |
| "grad_norm": 3.5761684361674004, | |
| "learning_rate": 2.6150781616863794e-08, | |
| "loss": 0.6914, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 2.7079303675048356, | |
| "grad_norm": 4.1393792533061715, | |
| "learning_rate": 2.2970097968953994e-08, | |
| "loss": 0.6837, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.7079303675048356, | |
| "eval_loss": 0.7920587658882141, | |
| "eval_runtime": 127.7157, | |
| "eval_samples_per_second": 57.55, | |
| "eval_steps_per_second": 0.9, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 3.9729587208028274, | |
| "learning_rate": 1.9991057409751267e-08, | |
| "loss": 0.6905, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 2.746615087040619, | |
| "grad_norm": 3.7929506007765017, | |
| "learning_rate": 1.7214919195619125e-08, | |
| "loss": 0.6752, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.7659574468085104, | |
| "grad_norm": 3.615765661856518, | |
| "learning_rate": 1.4642856815015758e-08, | |
| "loss": 0.6991, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 2.7852998065764023, | |
| "grad_norm": 3.884082014060051, | |
| "learning_rate": 1.2275957492453692e-08, | |
| "loss": 0.6846, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.804642166344294, | |
| "grad_norm": 3.590900524493669, | |
| "learning_rate": 1.0115221728924706e-08, | |
| "loss": 0.6801, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.8239845261121856, | |
| "grad_norm": 3.8191804987004025, | |
| "learning_rate": 8.161562878982398e-09, | |
| "loss": 0.7049, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.843326885880077, | |
| "grad_norm": 4.18976948653072, | |
| "learning_rate": 6.415806764662524e-09, | |
| "loss": 0.6871, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 2.862669245647969, | |
| "grad_norm": 3.7105498242235786, | |
| "learning_rate": 4.8786913264033945e-09, | |
| "loss": 0.6774, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.882011605415861, | |
| "grad_norm": 3.655342612843487, | |
| "learning_rate": 3.5508663111147306e-09, | |
| "loss": 0.6913, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 2.9013539651837523, | |
| "grad_norm": 3.64997500663328, | |
| "learning_rate": 2.432892997526026e-09, | |
| "loss": 0.687, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.9013539651837523, | |
| "eval_loss": 0.7918885350227356, | |
| "eval_runtime": 127.697, | |
| "eval_samples_per_second": 57.558, | |
| "eval_steps_per_second": 0.901, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.920696324951644, | |
| "grad_norm": 3.720532393294838, | |
| "learning_rate": 1.5252439589311107e-09, | |
| "loss": 0.6823, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 2.9400386847195357, | |
| "grad_norm": 3.67654814171945, | |
| "learning_rate": 8.283028634287203e-10, | |
| "loss": 0.6823, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.9593810444874276, | |
| "grad_norm": 3.756704932951535, | |
| "learning_rate": 3.4236431174428094e-10, | |
| "loss": 0.6766, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 2.978723404255319, | |
| "grad_norm": 3.716812044061899, | |
| "learning_rate": 6.763371270035457e-11, | |
| "loss": 0.6781, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.9941972920696323, | |
| "step": 774, | |
| "total_flos": 4563620855808000.0, | |
| "train_loss": 0.7852395317043137, | |
| "train_runtime": 12070.3191, | |
| "train_samples_per_second": 16.441, | |
| "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 | |
| } | |