| { | |
| "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": 18.527687354581595, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7356, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03868471953578337, | |
| "grad_norm": 12.936955115293438, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5854, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.058027079303675046, | |
| "grad_norm": 7.321839613340823, | |
| "learning_rate": 9.998943236640677e-07, | |
| "loss": 1.2859, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07736943907156674, | |
| "grad_norm": 4.100947188595897, | |
| "learning_rate": 9.995773393262229e-07, | |
| "loss": 1.1468, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09671179883945841, | |
| "grad_norm": 3.5921467819144097, | |
| "learning_rate": 9.99049180977439e-07, | |
| "loss": 1.0832, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.11605415860735009, | |
| "grad_norm": 3.700008683460077, | |
| "learning_rate": 9.983100718730718e-07, | |
| "loss": 1.0181, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13539651837524178, | |
| "grad_norm": 3.413586927958152, | |
| "learning_rate": 9.973603244384906e-07, | |
| "loss": 0.9956, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.15473887814313347, | |
| "grad_norm": 3.668981471014926, | |
| "learning_rate": 9.9620034013701e-07, | |
| "loss": 0.9636, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.17408123791102514, | |
| "grad_norm": 3.55289668457361, | |
| "learning_rate": 9.948306093001933e-07, | |
| "loss": 0.9538, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "grad_norm": 3.3904476145405393, | |
| "learning_rate": 9.932517109205849e-07, | |
| "loss": 0.9216, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "eval_loss": 0.9377365708351135, | |
| "eval_runtime": 129.4079, | |
| "eval_samples_per_second": 56.797, | |
| "eval_steps_per_second": 0.889, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 3.4681388676869433, | |
| "learning_rate": 9.914643124069666e-07, | |
| "loss": 0.923, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.23210831721470018, | |
| "grad_norm": 3.29526552312586, | |
| "learning_rate": 9.89469169302242e-07, | |
| "loss": 0.9115, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2514506769825919, | |
| "grad_norm": 3.3040218603649225, | |
| "learning_rate": 9.872671249640626e-07, | |
| "loss": 0.9168, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.27079303675048355, | |
| "grad_norm": 3.315572544161604, | |
| "learning_rate": 9.848591102083375e-07, | |
| "loss": 0.8954, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2901353965183752, | |
| "grad_norm": 3.2544174349742825, | |
| "learning_rate": 9.822461429157716e-07, | |
| "loss": 0.9181, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.30947775628626695, | |
| "grad_norm": 3.4125559699409056, | |
| "learning_rate": 9.794293276016023e-07, | |
| "loss": 0.9121, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3288201160541586, | |
| "grad_norm": 3.6466175557331986, | |
| "learning_rate": 9.764098549487155e-07, | |
| "loss": 0.8976, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.3481624758220503, | |
| "grad_norm": 3.4442797818420234, | |
| "learning_rate": 9.731890013043367e-07, | |
| "loss": 0.8697, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.36750483558994196, | |
| "grad_norm": 3.4952948347778796, | |
| "learning_rate": 9.697681281405128e-07, | |
| "loss": 0.8797, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "grad_norm": 3.3555597902003944, | |
| "learning_rate": 9.6614868147861e-07, | |
| "loss": 0.872, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "eval_loss": 0.8798893690109253, | |
| "eval_runtime": 127.7553, | |
| "eval_samples_per_second": 57.532, | |
| "eval_steps_per_second": 0.9, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.40618955512572535, | |
| "grad_norm": 3.5367127004254324, | |
| "learning_rate": 9.623321912780744e-07, | |
| "loss": 0.8812, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.425531914893617, | |
| "grad_norm": 3.578326025613384, | |
| "learning_rate": 9.583202707897073e-07, | |
| "loss": 0.8779, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4448742746615087, | |
| "grad_norm": 3.296044526686552, | |
| "learning_rate": 9.54114615873738e-07, | |
| "loss": 0.8618, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.46421663442940037, | |
| "grad_norm": 3.4854773143444535, | |
| "learning_rate": 9.497170042829736e-07, | |
| "loss": 0.8771, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4835589941972921, | |
| "grad_norm": 3.49978332955871, | |
| "learning_rate": 9.451292949113329e-07, | |
| "loss": 0.8678, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5029013539651838, | |
| "grad_norm": 3.8225654739598895, | |
| "learning_rate": 9.403534270080829e-07, | |
| "loss": 0.8884, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5222437137330754, | |
| "grad_norm": 3.2475969413565635, | |
| "learning_rate": 9.353914193581072e-07, | |
| "loss": 0.8575, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.5415860735009671, | |
| "grad_norm": 3.6725820783332774, | |
| "learning_rate": 9.302453694285548e-07, | |
| "loss": 0.8553, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5609284332688588, | |
| "grad_norm": 3.2370725122807484, | |
| "learning_rate": 9.249174524822305e-07, | |
| "loss": 0.868, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "grad_norm": 3.669185365661775, | |
| "learning_rate": 9.19409920658098e-07, | |
| "loss": 0.8452, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "eval_loss": 0.854381799697876, | |
| "eval_runtime": 127.6808, | |
| "eval_samples_per_second": 57.565, | |
| "eval_steps_per_second": 0.901, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5996131528046421, | |
| "grad_norm": 3.538171559620098, | |
| "learning_rate": 9.137251020192907e-07, | |
| "loss": 0.8327, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.6189555125725339, | |
| "grad_norm": 3.45820833220196, | |
| "learning_rate": 9.078653995690246e-07, | |
| "loss": 0.8577, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6382978723404256, | |
| "grad_norm": 3.52275765783797, | |
| "learning_rate": 9.018332902348388e-07, | |
| "loss": 0.8288, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.6576402321083172, | |
| "grad_norm": 3.5694200958956097, | |
| "learning_rate": 8.956313238215823e-07, | |
| "loss": 0.8297, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.6769825918762089, | |
| "grad_norm": 3.623437037037063, | |
| "learning_rate": 8.892621219336e-07, | |
| "loss": 0.8376, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.6963249516441006, | |
| "grad_norm": 3.4631672962599755, | |
| "learning_rate": 8.827283768665648e-07, | |
| "loss": 0.8401, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7156673114119922, | |
| "grad_norm": 3.5857126637962726, | |
| "learning_rate": 8.760328504694317e-07, | |
| "loss": 0.8447, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.7350096711798839, | |
| "grad_norm": 3.4134278990568436, | |
| "learning_rate": 8.691783729769873e-07, | |
| "loss": 0.8223, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.7543520309477756, | |
| "grad_norm": 3.527262510580844, | |
| "learning_rate": 8.621678418134963e-07, | |
| "loss": 0.8527, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "grad_norm": 3.524465193791854, | |
| "learning_rate": 8.550042203679439e-07, | |
| "loss": 0.8444, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "eval_loss": 0.8374524116516113, | |
| "eval_runtime": 127.8349, | |
| "eval_samples_per_second": 57.496, | |
| "eval_steps_per_second": 0.9, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.793036750483559, | |
| "grad_norm": 3.5379630468540806, | |
| "learning_rate": 8.476905367413957e-07, | |
| "loss": 0.8291, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.8123791102514507, | |
| "grad_norm": 3.427913081867794, | |
| "learning_rate": 8.402298824670029e-07, | |
| "loss": 0.822, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8317214700193424, | |
| "grad_norm": 3.6293945120927464, | |
| "learning_rate": 8.326254112031949e-07, | |
| "loss": 0.8253, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.851063829787234, | |
| "grad_norm": 3.1613190541906797, | |
| "learning_rate": 8.248803374006113e-07, | |
| "loss": 0.8266, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.8704061895551257, | |
| "grad_norm": 3.516439765275824, | |
| "learning_rate": 8.169979349433358e-07, | |
| "loss": 0.8281, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.8897485493230174, | |
| "grad_norm": 3.358282538404413, | |
| "learning_rate": 8.089815357650089e-07, | |
| "loss": 0.8358, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 3.496336629563945, | |
| "learning_rate": 8.008345284404003e-07, | |
| "loss": 0.8189, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.9284332688588007, | |
| "grad_norm": 3.550035619548321, | |
| "learning_rate": 7.925603567530418e-07, | |
| "loss": 0.8191, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9477756286266924, | |
| "grad_norm": 3.6931325798521857, | |
| "learning_rate": 7.841625182395206e-07, | |
| "loss": 0.8241, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "grad_norm": 3.3871785206976863, | |
| "learning_rate": 7.756445627110522e-07, | |
| "loss": 0.8236, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "eval_loss": 0.8247936964035034, | |
| "eval_runtime": 127.7676, | |
| "eval_samples_per_second": 57.526, | |
| "eval_steps_per_second": 0.9, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9864603481624759, | |
| "grad_norm": 3.4188093303718965, | |
| "learning_rate": 7.670100907529557e-07, | |
| "loss": 0.8353, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.0058027079303675, | |
| "grad_norm": 3.636501521011555, | |
| "learning_rate": 7.582627522026685e-07, | |
| "loss": 0.8063, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.0251450676982592, | |
| "grad_norm": 3.621680508418713, | |
| "learning_rate": 7.49406244606939e-07, | |
| "loss": 0.7737, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.0444874274661509, | |
| "grad_norm": 3.581741353253455, | |
| "learning_rate": 7.404443116588547e-07, | |
| "loss": 0.7829, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 3.383739671366194, | |
| "learning_rate": 7.31380741615363e-07, | |
| "loss": 0.7543, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.0831721470019342, | |
| "grad_norm": 3.943776092351626, | |
| "learning_rate": 7.222193656959546e-07, | |
| "loss": 0.7892, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.1025145067698259, | |
| "grad_norm": 3.449730443330265, | |
| "learning_rate": 7.129640564631863e-07, | |
| "loss": 0.7594, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.1218568665377175, | |
| "grad_norm": 3.5920269631980477, | |
| "learning_rate": 7.036187261857288e-07, | |
| "loss": 0.759, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1411992263056092, | |
| "grad_norm": 3.590692121258084, | |
| "learning_rate": 6.941873251846293e-07, | |
| "loss": 0.7602, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "grad_norm": 3.7238466219935935, | |
| "learning_rate": 6.846738401634898e-07, | |
| "loss": 0.776, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "eval_loss": 0.8193865418434143, | |
| "eval_runtime": 127.7613, | |
| "eval_samples_per_second": 57.529, | |
| "eval_steps_per_second": 0.9, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1798839458413926, | |
| "grad_norm": 3.3793218130154643, | |
| "learning_rate": 6.750822925232663e-07, | |
| "loss": 0.7495, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.1992263056092844, | |
| "grad_norm": 3.56289301999189, | |
| "learning_rate": 6.654167366624008e-07, | |
| "loss": 0.7579, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.218568665377176, | |
| "grad_norm": 3.55989595423253, | |
| "learning_rate": 6.556812582630059e-07, | |
| "loss": 0.7715, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.2379110251450678, | |
| "grad_norm": 3.5614329702070995, | |
| "learning_rate": 6.458799725638248e-07, | |
| "loss": 0.7583, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.2572533849129595, | |
| "grad_norm": 3.3734819969541463, | |
| "learning_rate": 6.36017022620698e-07, | |
| "loss": 0.758, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.2765957446808511, | |
| "grad_norm": 3.3947723934851015, | |
| "learning_rate": 6.260965775552713e-07, | |
| "loss": 0.7671, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.2959381044487428, | |
| "grad_norm": 3.5148140089765434, | |
| "learning_rate": 6.161228307926858e-07, | |
| "loss": 0.7435, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.3152804642166345, | |
| "grad_norm": 3.79895410453979, | |
| "learning_rate": 6.060999982889954e-07, | |
| "loss": 0.7515, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.3346228239845261, | |
| "grad_norm": 3.2855087244873533, | |
| "learning_rate": 5.960323167490588e-07, | |
| "loss": 0.7294, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "grad_norm": 3.8271832901061886, | |
| "learning_rate": 5.859240418356614e-07, | |
| "loss": 0.7598, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "eval_loss": 0.8137449622154236, | |
| "eval_runtime": 127.7492, | |
| "eval_samples_per_second": 57.535, | |
| "eval_steps_per_second": 0.9, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3733075435203095, | |
| "grad_norm": 3.932445955468858, | |
| "learning_rate": 5.757794463706253e-07, | |
| "loss": 0.7626, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.3926499032882012, | |
| "grad_norm": 3.5885849985690035, | |
| "learning_rate": 5.656028185286637e-07, | |
| "loss": 0.7548, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.4119922630560928, | |
| "grad_norm": 3.654106726967214, | |
| "learning_rate": 5.553984600247463e-07, | |
| "loss": 0.7446, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.4313346228239845, | |
| "grad_norm": 3.512502382810576, | |
| "learning_rate": 5.451706842957421e-07, | |
| "loss": 0.7378, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.4506769825918762, | |
| "grad_norm": 3.7978244792994285, | |
| "learning_rate": 5.349238146771061e-07, | |
| "loss": 0.7436, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.4700193423597678, | |
| "grad_norm": 3.542579158199173, | |
| "learning_rate": 5.246621825753827e-07, | |
| "loss": 0.7549, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.4893617021276595, | |
| "grad_norm": 3.538084352067931, | |
| "learning_rate": 5.143901256372967e-07, | |
| "loss": 0.7495, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.5087040618955512, | |
| "grad_norm": 3.617738720729932, | |
| "learning_rate": 5.041119859162068e-07, | |
| "loss": 0.7396, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.528046421663443, | |
| "grad_norm": 3.673050568373933, | |
| "learning_rate": 4.938321080366968e-07, | |
| "loss": 0.7436, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "grad_norm": 3.6088143658238208, | |
| "learning_rate": 4.835548373580792e-07, | |
| "loss": 0.7539, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "eval_loss": 0.8074547648429871, | |
| "eval_runtime": 127.7746, | |
| "eval_samples_per_second": 57.523, | |
| "eval_steps_per_second": 0.9, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5667311411992264, | |
| "grad_norm": 3.605893887637476, | |
| "learning_rate": 4.73284518137589e-07, | |
| "loss": 0.7371, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.5860735009671179, | |
| "grad_norm": 3.5111892276140098, | |
| "learning_rate": 4.630254916940423e-07, | |
| "loss": 0.7502, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.6054158607350097, | |
| "grad_norm": 3.7061953647378085, | |
| "learning_rate": 4.5278209457273825e-07, | |
| "loss": 0.7455, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.6247582205029012, | |
| "grad_norm": 3.7197608595379426, | |
| "learning_rate": 4.425586567123779e-07, | |
| "loss": 0.7348, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.644100580270793, | |
| "grad_norm": 3.950557508016625, | |
| "learning_rate": 4.3235949961477627e-07, | |
| "loss": 0.7528, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.6634429400386848, | |
| "grad_norm": 3.4882683728185966, | |
| "learning_rate": 4.2218893451814e-07, | |
| "loss": 0.7233, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.6827852998065764, | |
| "grad_norm": 3.7564603432326535, | |
| "learning_rate": 4.120512605746842e-07, | |
| "loss": 0.762, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.702127659574468, | |
| "grad_norm": 3.735016928935352, | |
| "learning_rate": 4.019507630333577e-07, | |
| "loss": 0.7333, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.7214700193423598, | |
| "grad_norm": 3.666531076735895, | |
| "learning_rate": 3.9189171142844553e-07, | |
| "loss": 0.7343, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "grad_norm": 3.751741635827122, | |
| "learning_rate": 3.8187835777481375e-07, | |
| "loss": 0.7273, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "eval_loss": 0.8021160364151001, | |
| "eval_runtime": 127.7286, | |
| "eval_samples_per_second": 57.544, | |
| "eval_steps_per_second": 0.9, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.760154738878143, | |
| "grad_norm": 3.6597963453482656, | |
| "learning_rate": 3.7191493477056086e-07, | |
| "loss": 0.7416, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 1.7794970986460348, | |
| "grad_norm": 3.7522608563050537, | |
| "learning_rate": 3.620056540078323e-07, | |
| "loss": 0.7464, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.7988394584139265, | |
| "grad_norm": 3.6374420398476457, | |
| "learning_rate": 3.5215470419255897e-07, | |
| "loss": 0.7279, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 3.7809155789430045, | |
| "learning_rate": 3.423662493738687e-07, | |
| "loss": 0.7464, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.8375241779497098, | |
| "grad_norm": 3.5450074284083444, | |
| "learning_rate": 3.3264442718392014e-07, | |
| "loss": 0.7295, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.8568665377176017, | |
| "grad_norm": 3.676696426875523, | |
| "learning_rate": 3.229933470889038e-07, | |
| "loss": 0.7281, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.8762088974854931, | |
| "grad_norm": 3.589693765306483, | |
| "learning_rate": 3.134170886519486e-07, | |
| "loss": 0.7319, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 1.895551257253385, | |
| "grad_norm": 3.6200404091466933, | |
| "learning_rate": 3.039196998086687e-07, | |
| "loss": 0.7344, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.9148936170212765, | |
| "grad_norm": 3.583752869993935, | |
| "learning_rate": 2.9450519515607963e-07, | |
| "loss": 0.7269, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "grad_norm": 3.769352285404306, | |
| "learning_rate": 2.8517755425560663e-07, | |
| "loss": 0.7314, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "eval_loss": 0.7983231544494629, | |
| "eval_runtime": 127.8024, | |
| "eval_samples_per_second": 57.511, | |
| "eval_steps_per_second": 0.9, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9535783365570598, | |
| "grad_norm": 3.5961562887933747, | |
| "learning_rate": 2.7594071995090283e-07, | |
| "loss": 0.7424, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 1.9729206963249517, | |
| "grad_norm": 3.821907492658478, | |
| "learning_rate": 2.667985967011878e-07, | |
| "loss": 0.7332, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.9922630560928434, | |
| "grad_norm": 3.8816672682352626, | |
| "learning_rate": 2.577550489308123e-07, | |
| "loss": 0.7328, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.011605415860735, | |
| "grad_norm": 4.108838347130274, | |
| "learning_rate": 2.488138993957452e-07, | |
| "loss": 0.7194, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.0309477756286265, | |
| "grad_norm": 3.5548470436404433, | |
| "learning_rate": 2.3997892756767394e-07, | |
| "loss": 0.6922, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.0502901353965184, | |
| "grad_norm": 3.7279675160489827, | |
| "learning_rate": 2.3125386803640183e-07, | |
| "loss": 0.6936, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.0696324951644103, | |
| "grad_norm": 3.884838129631813, | |
| "learning_rate": 2.226424089312174e-07, | |
| "loss": 0.6973, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.0889748549323017, | |
| "grad_norm": 3.592031200352461, | |
| "learning_rate": 2.1414819036190157e-07, | |
| "loss": 0.6941, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.1083172147001936, | |
| "grad_norm": 3.807636152221673, | |
| "learning_rate": 2.057748028800344e-07, | |
| "loss": 0.7032, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 3.791470897609836, | |
| "learning_rate": 1.9752578596124952e-07, | |
| "loss": 0.7094, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_loss": 0.8028796911239624, | |
| "eval_runtime": 127.4685, | |
| "eval_samples_per_second": 57.661, | |
| "eval_steps_per_second": 0.902, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.147001934235977, | |
| "grad_norm": 3.801001026224913, | |
| "learning_rate": 1.8940462650907912e-07, | |
| "loss": 0.6816, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 2.1663442940038684, | |
| "grad_norm": 3.7897529261988403, | |
| "learning_rate": 1.8141475738102086e-07, | |
| "loss": 0.6795, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.1856866537717603, | |
| "grad_norm": 3.873193867336735, | |
| "learning_rate": 1.735595559374508e-07, | |
| "loss": 0.6968, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 2.2050290135396517, | |
| "grad_norm": 4.010411940489978, | |
| "learning_rate": 1.6584234261399532e-07, | |
| "loss": 0.6982, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.2243713733075436, | |
| "grad_norm": 3.689366962655481, | |
| "learning_rate": 1.5826637951796474e-07, | |
| "loss": 0.684, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.243713733075435, | |
| "grad_norm": 3.702408168586401, | |
| "learning_rate": 1.5083486904944387e-07, | |
| "loss": 0.678, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.263056092843327, | |
| "grad_norm": 3.7789654460337467, | |
| "learning_rate": 1.4355095254761974e-07, | |
| "loss": 0.7011, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.2823984526112184, | |
| "grad_norm": 3.757564264693129, | |
| "learning_rate": 1.3641770896292082e-07, | |
| "loss": 0.7105, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.3017408123791103, | |
| "grad_norm": 3.6208442620229238, | |
| "learning_rate": 1.2943815355552851e-07, | |
| "loss": 0.6939, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "grad_norm": 3.661781095521188, | |
| "learning_rate": 1.226152366208104e-07, | |
| "loss": 0.7073, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "eval_loss": 0.8018428683280945, | |
| "eval_runtime": 127.8893, | |
| "eval_samples_per_second": 57.472, | |
| "eval_steps_per_second": 0.899, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.3404255319148937, | |
| "grad_norm": 3.680478717422678, | |
| "learning_rate": 1.1595184224221466e-07, | |
| "loss": 0.6853, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.359767891682785, | |
| "grad_norm": 4.069897412011099, | |
| "learning_rate": 1.0945078707215221e-07, | |
| "loss": 0.6988, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.379110251450677, | |
| "grad_norm": 3.616210595264312, | |
| "learning_rate": 1.0311481914138371e-07, | |
| "loss": 0.6789, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.398452611218569, | |
| "grad_norm": 3.8468836483394195, | |
| "learning_rate": 9.6946616697411e-08, | |
| "loss": 0.6861, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.4177949709864603, | |
| "grad_norm": 3.719763619027049, | |
| "learning_rate": 9.094878707236841e-08, | |
| "loss": 0.701, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.437137330754352, | |
| "grad_norm": 3.817198136247863, | |
| "learning_rate": 8.512386558088919e-08, | |
| "loss": 0.7061, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.4564796905222437, | |
| "grad_norm": 3.9942569258409915, | |
| "learning_rate": 7.947431444841452e-08, | |
| "loss": 0.7032, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.4758220502901356, | |
| "grad_norm": 3.752182756744305, | |
| "learning_rate": 7.400252177039784e-08, | |
| "loss": 0.6791, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.495164410058027, | |
| "grad_norm": 3.7498129520839885, | |
| "learning_rate": 6.871080050284394e-08, | |
| "loss": 0.6837, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "grad_norm": 3.796492759569209, | |
| "learning_rate": 6.360138748461013e-08, | |
| "loss": 0.6944, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "eval_loss": 0.8010697364807129, | |
| "eval_runtime": 127.6709, | |
| "eval_samples_per_second": 57.57, | |
| "eval_steps_per_second": 0.901, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.5338491295938104, | |
| "grad_norm": 3.761294248355449, | |
| "learning_rate": 5.867644249188247e-08, | |
| "loss": 0.69, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 2.5531914893617023, | |
| "grad_norm": 3.972673839505174, | |
| "learning_rate": 5.3938047325226944e-08, | |
| "loss": 0.7015, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.5725338491295937, | |
| "grad_norm": 3.6874503743702918, | |
| "learning_rate": 4.9388204929601326e-08, | |
| "loss": 0.6942, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.5918762088974856, | |
| "grad_norm": 3.8115465970994156, | |
| "learning_rate": 4.5028838547699346e-08, | |
| "loss": 0.6851, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.611218568665377, | |
| "grad_norm": 3.7125576734640138, | |
| "learning_rate": 4.0861790906985884e-08, | |
| "loss": 0.6952, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.630560928433269, | |
| "grad_norm": 3.7316375539263666, | |
| "learning_rate": 3.6888823440766214e-08, | |
| "loss": 0.6766, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.6499032882011604, | |
| "grad_norm": 3.821854585997512, | |
| "learning_rate": 3.311161554361874e-08, | |
| "loss": 0.6766, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 2.6692456479690523, | |
| "grad_norm": 3.790693563798963, | |
| "learning_rate": 2.9531763861505964e-08, | |
| "loss": 0.6832, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.6885880077369437, | |
| "grad_norm": 3.528383170970932, | |
| "learning_rate": 2.6150781616863794e-08, | |
| "loss": 0.6797, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 2.7079303675048356, | |
| "grad_norm": 3.703907558068266, | |
| "learning_rate": 2.2970097968953994e-08, | |
| "loss": 0.6841, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.7079303675048356, | |
| "eval_loss": 0.8003172278404236, | |
| "eval_runtime": 127.7485, | |
| "eval_samples_per_second": 57.535, | |
| "eval_steps_per_second": 0.9, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 4.045927762778011, | |
| "learning_rate": 1.9991057409751267e-08, | |
| "loss": 0.7023, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 2.746615087040619, | |
| "grad_norm": 3.7217800342328053, | |
| "learning_rate": 1.7214919195619125e-08, | |
| "loss": 0.6809, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.7659574468085104, | |
| "grad_norm": 3.849855952936902, | |
| "learning_rate": 1.4642856815015758e-08, | |
| "loss": 0.6918, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 2.7852998065764023, | |
| "grad_norm": 3.4541052934042704, | |
| "learning_rate": 1.2275957492453692e-08, | |
| "loss": 0.6672, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.804642166344294, | |
| "grad_norm": 3.752950947257332, | |
| "learning_rate": 1.0115221728924706e-08, | |
| "loss": 0.6954, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.8239845261121856, | |
| "grad_norm": 3.8044625608243603, | |
| "learning_rate": 8.161562878982398e-09, | |
| "loss": 0.7087, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.843326885880077, | |
| "grad_norm": 3.877886572852321, | |
| "learning_rate": 6.415806764662524e-09, | |
| "loss": 0.694, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 2.862669245647969, | |
| "grad_norm": 3.905892889322174, | |
| "learning_rate": 4.8786913264033945e-09, | |
| "loss": 0.6904, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.882011605415861, | |
| "grad_norm": 3.6119800064006884, | |
| "learning_rate": 3.5508663111147306e-09, | |
| "loss": 0.6895, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 2.9013539651837523, | |
| "grad_norm": 3.5602581412399625, | |
| "learning_rate": 2.432892997526026e-09, | |
| "loss": 0.6832, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.9013539651837523, | |
| "eval_loss": 0.8001306056976318, | |
| "eval_runtime": 127.7405, | |
| "eval_samples_per_second": 57.539, | |
| "eval_steps_per_second": 0.9, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.920696324951644, | |
| "grad_norm": 4.020301585342686, | |
| "learning_rate": 1.5252439589311107e-09, | |
| "loss": 0.6995, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 2.9400386847195357, | |
| "grad_norm": 3.794246694730058, | |
| "learning_rate": 8.283028634287203e-10, | |
| "loss": 0.69, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.9593810444874276, | |
| "grad_norm": 3.7696382426574115, | |
| "learning_rate": 3.4236431174428094e-10, | |
| "loss": 0.6655, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 2.978723404255319, | |
| "grad_norm": 3.7732134199260745, | |
| "learning_rate": 6.763371270035457e-11, | |
| "loss": 0.678, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.9941972920696323, | |
| "step": 774, | |
| "total_flos": 4563620855808000.0, | |
| "train_loss": 0.7853651687464357, | |
| "train_runtime": 12073.1716, | |
| "train_samples_per_second": 16.437, | |
| "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 | |
| } | |