| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.984771573604061, | |
| "eval_steps": 50, | |
| "global_step": 441, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0338409475465313, | |
| "grad_norm": 18.879603600162138, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7268, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0676818950930626, | |
| "grad_norm": 12.562690441872693, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5925, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10152284263959391, | |
| "grad_norm": 8.05925907057448, | |
| "learning_rate": 9.996679701338661e-07, | |
| "loss": 1.269, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1353637901861252, | |
| "grad_norm": 4.106846211376464, | |
| "learning_rate": 9.986723215107924e-07, | |
| "loss": 1.1436, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1692047377326565, | |
| "grad_norm": 4.065415826389198, | |
| "learning_rate": 9.97014376471095e-07, | |
| "loss": 1.0624, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.20304568527918782, | |
| "grad_norm": 3.5507350425506274, | |
| "learning_rate": 9.946963369638524e-07, | |
| "loss": 1.0302, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23688663282571912, | |
| "grad_norm": 3.620116416000427, | |
| "learning_rate": 9.917212816224536e-07, | |
| "loss": 1.0156, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2707275803722504, | |
| "grad_norm": 3.392556379414998, | |
| "learning_rate": 9.880931616758056e-07, | |
| "loss": 0.9731, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.30456852791878175, | |
| "grad_norm": 3.6283098411860224, | |
| "learning_rate": 9.838167957006293e-07, | |
| "loss": 0.9453, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "grad_norm": 3.643003094042682, | |
| "learning_rate": 9.788978632218138e-07, | |
| "loss": 0.9333, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "eval_loss": 0.9387685656547546, | |
| "eval_runtime": 73.7594, | |
| "eval_samples_per_second": 56.942, | |
| "eval_steps_per_second": 0.895, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.37225042301184436, | |
| "grad_norm": 3.5187728461211143, | |
| "learning_rate": 9.73342897169329e-07, | |
| "loss": 0.933, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.40609137055837563, | |
| "grad_norm": 3.221032807973684, | |
| "learning_rate": 9.671592752017137e-07, | |
| "loss": 0.9244, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.43993231810490696, | |
| "grad_norm": 3.6068076034822982, | |
| "learning_rate": 9.603552099076648e-07, | |
| "loss": 0.9217, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.47377326565143824, | |
| "grad_norm": 3.522728399930581, | |
| "learning_rate": 9.52939737898737e-07, | |
| "loss": 0.9251, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5076142131979695, | |
| "grad_norm": 3.4156164712414805, | |
| "learning_rate": 9.449227078076443e-07, | |
| "loss": 0.8951, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5414551607445008, | |
| "grad_norm": 3.5241377443178594, | |
| "learning_rate": 9.363147672080985e-07, | |
| "loss": 0.893, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5752961082910322, | |
| "grad_norm": 3.480593037844269, | |
| "learning_rate": 9.271273484735592e-07, | |
| "loss": 0.8702, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.6091370558375635, | |
| "grad_norm": 3.3606090648530906, | |
| "learning_rate": 9.173726535936766e-07, | |
| "loss": 0.8931, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6429780033840947, | |
| "grad_norm": 3.3802398568272545, | |
| "learning_rate": 9.070636379685915e-07, | |
| "loss": 0.8689, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "grad_norm": 3.7340957238116013, | |
| "learning_rate": 8.962139932026156e-07, | |
| "loss": 0.8736, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "eval_loss": 0.8820675015449524, | |
| "eval_runtime": 73.2211, | |
| "eval_samples_per_second": 57.361, | |
| "eval_steps_per_second": 0.901, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7106598984771574, | |
| "grad_norm": 3.491575137149294, | |
| "learning_rate": 8.848381289201459e-07, | |
| "loss": 0.8658, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.7445008460236887, | |
| "grad_norm": 3.3971585698046023, | |
| "learning_rate": 8.72951153627962e-07, | |
| "loss": 0.85, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7783417935702199, | |
| "grad_norm": 3.4606145381962574, | |
| "learning_rate": 8.605688546493238e-07, | |
| "loss": 0.8831, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.8121827411167513, | |
| "grad_norm": 3.312370481580055, | |
| "learning_rate": 8.477076771565202e-07, | |
| "loss": 0.8542, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8460236886632826, | |
| "grad_norm": 3.479568204762634, | |
| "learning_rate": 8.343847023297169e-07, | |
| "loss": 0.8538, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.8798646362098139, | |
| "grad_norm": 3.470397617651747, | |
| "learning_rate": 8.206176246711065e-07, | |
| "loss": 0.8568, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9137055837563451, | |
| "grad_norm": 3.4164690427045974, | |
| "learning_rate": 8.064247285044972e-07, | |
| "loss": 0.8566, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.9475465313028765, | |
| "grad_norm": 3.5885722316358994, | |
| "learning_rate": 7.918248636915459e-07, | |
| "loss": 0.8671, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9813874788494078, | |
| "grad_norm": 3.573742166264298, | |
| "learning_rate": 7.768374205968906e-07, | |
| "loss": 0.8607, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "grad_norm": 3.7440804811259376, | |
| "learning_rate": 7.614823043354285e-07, | |
| "loss": 0.8459, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "eval_loss": 0.8576342463493347, | |
| "eval_runtime": 73.2015, | |
| "eval_samples_per_second": 57.376, | |
| "eval_steps_per_second": 0.902, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0490693739424704, | |
| "grad_norm": 3.836769449038388, | |
| "learning_rate": 7.457799083359471e-07, | |
| "loss": 0.7928, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.0829103214890017, | |
| "grad_norm": 3.776692818561464, | |
| "learning_rate": 7.297510872562131e-07, | |
| "loss": 0.7801, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.116751269035533, | |
| "grad_norm": 3.5436254123817075, | |
| "learning_rate": 7.134171292854955e-07, | |
| "loss": 0.7915, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.1505922165820643, | |
| "grad_norm": 3.637774293664913, | |
| "learning_rate": 6.967997278713089e-07, | |
| "loss": 0.7831, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1844331641285957, | |
| "grad_norm": 3.6828430568432817, | |
| "learning_rate": 6.79920952907921e-07, | |
| "loss": 0.7911, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.218274111675127, | |
| "grad_norm": 3.3501836574132673, | |
| "learning_rate": 6.628032214248982e-07, | |
| "loss": 0.7816, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.252115059221658, | |
| "grad_norm": 3.678603436999248, | |
| "learning_rate": 6.454692678146119e-07, | |
| "loss": 0.7742, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.2859560067681894, | |
| "grad_norm": 3.469023528162151, | |
| "learning_rate": 6.279421136382494e-07, | |
| "loss": 0.7932, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3197969543147208, | |
| "grad_norm": 3.5311154474090007, | |
| "learning_rate": 6.102450370504299e-07, | |
| "loss": 0.7852, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "grad_norm": 3.509395500338986, | |
| "learning_rate": 5.924015418830354e-07, | |
| "loss": 0.7682, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "eval_loss": 0.8485246300697327, | |
| "eval_runtime": 73.3335, | |
| "eval_samples_per_second": 57.273, | |
| "eval_steps_per_second": 0.9, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3874788494077834, | |
| "grad_norm": 3.5828255600676226, | |
| "learning_rate": 5.74435326429313e-07, | |
| "loss": 0.7784, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.4213197969543148, | |
| "grad_norm": 3.8406423248810526, | |
| "learning_rate": 5.563702519697108e-07, | |
| "loss": 0.7789, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.455160744500846, | |
| "grad_norm": 3.5396413573689482, | |
| "learning_rate": 5.382303110812466e-07, | |
| "loss": 0.7843, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.4890016920473772, | |
| "grad_norm": 3.9366608589473184, | |
| "learning_rate": 5.200395957725005e-07, | |
| "loss": 0.7734, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.5228426395939088, | |
| "grad_norm": 3.6167320799092644, | |
| "learning_rate": 5.018222654865471e-07, | |
| "loss": 0.777, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.5566835871404399, | |
| "grad_norm": 3.508615499961612, | |
| "learning_rate": 4.836025150143318e-07, | |
| "loss": 0.7777, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5905245346869712, | |
| "grad_norm": 3.8323856939840133, | |
| "learning_rate": 4.654045423610952e-07, | |
| "loss": 0.7744, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.6243654822335025, | |
| "grad_norm": 3.48551950556471, | |
| "learning_rate": 4.4725251660853357e-07, | |
| "loss": 0.7812, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6582064297800339, | |
| "grad_norm": 3.2586294819951815, | |
| "learning_rate": 4.2917054581536926e-07, | |
| "loss": 0.768, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "grad_norm": 3.587138715732753, | |
| "learning_rate": 4.1118264499897003e-07, | |
| "loss": 0.7797, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "eval_loss": 0.8384872078895569, | |
| "eval_runtime": 73.1511, | |
| "eval_samples_per_second": 57.415, | |
| "eval_steps_per_second": 0.902, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7258883248730963, | |
| "grad_norm": 3.497575413714801, | |
| "learning_rate": 3.9331270424053616e-07, | |
| "loss": 0.7617, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.7597292724196278, | |
| "grad_norm": 3.410968439354172, | |
| "learning_rate": 3.755844569562191e-07, | |
| "loss": 0.7687, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.793570219966159, | |
| "grad_norm": 3.609149780281463, | |
| "learning_rate": 3.580214483763093e-07, | |
| "loss": 0.7602, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.8274111675126905, | |
| "grad_norm": 3.493512578978129, | |
| "learning_rate": 3.406470042743574e-07, | |
| "loss": 0.7717, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.8612521150592216, | |
| "grad_norm": 3.6002811799241714, | |
| "learning_rate": 3.23484199987761e-07, | |
| "loss": 0.7496, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.895093062605753, | |
| "grad_norm": 3.6855175892280303, | |
| "learning_rate": 3.065558297709588e-07, | |
| "loss": 0.7695, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.9289340101522843, | |
| "grad_norm": 3.6642552773973254, | |
| "learning_rate": 2.898843765219388e-07, | |
| "loss": 0.7628, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.9627749576988156, | |
| "grad_norm": 3.7846627756438913, | |
| "learning_rate": 2.7349198192226295e-07, | |
| "loss": 0.7732, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.996615905245347, | |
| "grad_norm": 3.4513200002016022, | |
| "learning_rate": 2.574004170302696e-07, | |
| "loss": 0.7684, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "grad_norm": 3.8311631709862968, | |
| "learning_rate": 2.4163105336650643e-07, | |
| "loss": 0.7246, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "eval_loss": 0.8336632251739502, | |
| "eval_runtime": 73.1743, | |
| "eval_samples_per_second": 57.397, | |
| "eval_steps_per_second": 0.902, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0642978003384096, | |
| "grad_norm": 3.9190275346300107, | |
| "learning_rate": 2.2620483452979887e-07, | |
| "loss": 0.7305, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.0981387478849407, | |
| "grad_norm": 3.6908653316747446, | |
| "learning_rate": 2.1114224838164806e-07, | |
| "loss": 0.7351, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1319796954314723, | |
| "grad_norm": 3.802531438077045, | |
| "learning_rate": 1.964632998359036e-07, | |
| "loss": 0.7367, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.1658206429780034, | |
| "grad_norm": 3.775491751588902, | |
| "learning_rate": 1.8218748428984782e-07, | |
| "loss": 0.7233, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.199661590524535, | |
| "grad_norm": 3.5862593065525, | |
| "learning_rate": 1.6833376173198005e-07, | |
| "loss": 0.7233, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.233502538071066, | |
| "grad_norm": 3.5427938243892, | |
| "learning_rate": 1.5492053156088498e-07, | |
| "loss": 0.7098, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.267343485617597, | |
| "grad_norm": 3.61152784910624, | |
| "learning_rate": 1.4196560814863567e-07, | |
| "loss": 0.7129, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 2.3011844331641287, | |
| "grad_norm": 3.807265016749999, | |
| "learning_rate": 1.294861971811773e-07, | |
| "loss": 0.7074, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.33502538071066, | |
| "grad_norm": 3.6719299115770783, | |
| "learning_rate": 1.1749887280712161e-07, | |
| "loss": 0.7121, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "grad_norm": 3.5588978514727305, | |
| "learning_rate": 1.0601955562529774e-07, | |
| "loss": 0.7082, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "eval_loss": 0.8359954357147217, | |
| "eval_runtime": 73.1022, | |
| "eval_samples_per_second": 57.454, | |
| "eval_steps_per_second": 0.903, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.4027072758037225, | |
| "grad_norm": 3.72273582563589, | |
| "learning_rate": 9.506349154029425e-08, | |
| "loss": 0.7117, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.436548223350254, | |
| "grad_norm": 3.777312137275298, | |
| "learning_rate": 8.46452315140772e-08, | |
| "loss": 0.7238, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.470389170896785, | |
| "grad_norm": 3.9355287879984404, | |
| "learning_rate": 7.477861224057403e-08, | |
| "loss": 0.7385, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.504230118443316, | |
| "grad_norm": 3.612978118989858, | |
| "learning_rate": 6.547673776889095e-08, | |
| "loss": 0.7239, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.5380710659898478, | |
| "grad_norm": 3.73334551587451, | |
| "learning_rate": 5.6751962099570396e-08, | |
| "loss": 0.6999, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.571912013536379, | |
| "grad_norm": 3.7980379085271974, | |
| "learning_rate": 4.861587277700274e-08, | |
| "loss": 0.7253, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.6057529610829104, | |
| "grad_norm": 3.8953973449204105, | |
| "learning_rate": 4.107927549978235e-08, | |
| "loss": 0.7267, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.6395939086294415, | |
| "grad_norm": 4.0344872537944525, | |
| "learning_rate": 3.4152179769449396e-08, | |
| "loss": 0.7179, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.673434856175973, | |
| "grad_norm": 3.8358198567516935, | |
| "learning_rate": 2.784378559667622e-08, | |
| "loss": 0.7043, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "grad_norm": 3.5246037765259532, | |
| "learning_rate": 2.2162471282553553e-08, | |
| "loss": 0.7007, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "eval_loss": 0.8346139788627625, | |
| "eval_runtime": 73.1876, | |
| "eval_samples_per_second": 57.387, | |
| "eval_steps_per_second": 0.902, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.7411167512690353, | |
| "grad_norm": 3.808011429657169, | |
| "learning_rate": 1.7115782291206082e-08, | |
| "loss": 0.7232, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.774957698815567, | |
| "grad_norm": 3.8868492324851003, | |
| "learning_rate": 1.2710421228514733e-08, | |
| "loss": 0.7213, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.808798646362098, | |
| "grad_norm": 3.8016855221337953, | |
| "learning_rate": 8.952238940255153e-09, | |
| "loss": 0.7114, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.8426395939086295, | |
| "grad_norm": 3.6712328006089763, | |
| "learning_rate": 5.846226741475557e-09, | |
| "loss": 0.7148, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.8764805414551606, | |
| "grad_norm": 3.7375800014295417, | |
| "learning_rate": 3.3965097874343872e-09, | |
| "loss": 0.731, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.910321489001692, | |
| "grad_norm": 3.6510115083695562, | |
| "learning_rate": 1.6063415949008618e-09, | |
| "loss": 0.7146, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.9441624365482233, | |
| "grad_norm": 4.038300563245875, | |
| "learning_rate": 4.780997210962478e-10, | |
| "loss": 0.7113, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.9780033840947544, | |
| "grad_norm": 3.6454952565628376, | |
| "learning_rate": 1.328260601385356e-11, | |
| "loss": 0.701, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.984771573604061, | |
| "step": 441, | |
| "total_flos": 2600101931384832.0, | |
| "train_loss": 0.8247822573125497, | |
| "train_runtime": 6397.793, | |
| "train_samples_per_second": 17.725, | |
| "train_steps_per_second": 0.069 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 441, | |
| "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": 2600101931384832.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |