| { | |
| "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": 18.04123876140778, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7306, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0676818950930626, | |
| "grad_norm": 12.44365503200701, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5847, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10152284263959391, | |
| "grad_norm": 7.167811765970724, | |
| "learning_rate": 9.996679701338661e-07, | |
| "loss": 1.2931, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1353637901861252, | |
| "grad_norm": 4.021885562012876, | |
| "learning_rate": 9.986723215107924e-07, | |
| "loss": 1.1281, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1692047377326565, | |
| "grad_norm": 3.7108202673486153, | |
| "learning_rate": 9.97014376471095e-07, | |
| "loss": 1.0586, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.20304568527918782, | |
| "grad_norm": 3.5766280565164448, | |
| "learning_rate": 9.946963369638524e-07, | |
| "loss": 1.0329, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23688663282571912, | |
| "grad_norm": 3.593877812578378, | |
| "learning_rate": 9.917212816224536e-07, | |
| "loss": 0.9888, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2707275803722504, | |
| "grad_norm": 3.505437201913201, | |
| "learning_rate": 9.880931616758056e-07, | |
| "loss": 0.9608, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.30456852791878175, | |
| "grad_norm": 3.625778496171611, | |
| "learning_rate": 9.838167957006293e-07, | |
| "loss": 0.959, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "grad_norm": 3.4571136622580982, | |
| "learning_rate": 9.788978632218138e-07, | |
| "loss": 0.9399, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "eval_loss": 0.9205839037895203, | |
| "eval_runtime": 73.53, | |
| "eval_samples_per_second": 57.12, | |
| "eval_steps_per_second": 0.898, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.37225042301184436, | |
| "grad_norm": 3.54912813814943, | |
| "learning_rate": 9.73342897169329e-07, | |
| "loss": 0.9271, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.40609137055837563, | |
| "grad_norm": 3.5131112236149744, | |
| "learning_rate": 9.671592752017137e-07, | |
| "loss": 0.932, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.43993231810490696, | |
| "grad_norm": 3.4825792800162825, | |
| "learning_rate": 9.603552099076648e-07, | |
| "loss": 0.9033, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.47377326565143824, | |
| "grad_norm": 3.9017822285715247, | |
| "learning_rate": 9.52939737898737e-07, | |
| "loss": 0.8903, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5076142131979695, | |
| "grad_norm": 3.5887670616037903, | |
| "learning_rate": 9.449227078076443e-07, | |
| "loss": 0.8957, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5414551607445008, | |
| "grad_norm": 3.3830625519729782, | |
| "learning_rate": 9.363147672080985e-07, | |
| "loss": 0.9122, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5752961082910322, | |
| "grad_norm": 3.455530835875489, | |
| "learning_rate": 9.271273484735592e-07, | |
| "loss": 0.903, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.6091370558375635, | |
| "grad_norm": 3.5443952071489804, | |
| "learning_rate": 9.173726535936766e-07, | |
| "loss": 0.888, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6429780033840947, | |
| "grad_norm": 3.5814893753397525, | |
| "learning_rate": 9.070636379685915e-07, | |
| "loss": 0.8774, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "grad_norm": 3.665506397682717, | |
| "learning_rate": 8.962139932026156e-07, | |
| "loss": 0.8917, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "eval_loss": 0.8663063049316406, | |
| "eval_runtime": 73.3117, | |
| "eval_samples_per_second": 57.29, | |
| "eval_steps_per_second": 0.9, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7106598984771574, | |
| "grad_norm": 3.602061147245115, | |
| "learning_rate": 8.848381289201459e-07, | |
| "loss": 0.8735, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.7445008460236887, | |
| "grad_norm": 3.4514188732332345, | |
| "learning_rate": 8.72951153627962e-07, | |
| "loss": 0.8888, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7783417935702199, | |
| "grad_norm": 3.6738383323517363, | |
| "learning_rate": 8.605688546493238e-07, | |
| "loss": 0.8855, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.8121827411167513, | |
| "grad_norm": 3.4523022569374935, | |
| "learning_rate": 8.477076771565202e-07, | |
| "loss": 0.8682, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8460236886632826, | |
| "grad_norm": 3.5605291777094137, | |
| "learning_rate": 8.343847023297169e-07, | |
| "loss": 0.8652, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.8798646362098139, | |
| "grad_norm": 3.4424891069882606, | |
| "learning_rate": 8.206176246711065e-07, | |
| "loss": 0.8647, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9137055837563451, | |
| "grad_norm": 3.4364000143518996, | |
| "learning_rate": 8.064247285044972e-07, | |
| "loss": 0.8706, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.9475465313028765, | |
| "grad_norm": 3.4611546295144153, | |
| "learning_rate": 7.918248636915459e-07, | |
| "loss": 0.8603, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9813874788494078, | |
| "grad_norm": 3.5575449321623824, | |
| "learning_rate": 7.768374205968906e-07, | |
| "loss": 0.8622, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "grad_norm": 3.4436002362743054, | |
| "learning_rate": 7.614823043354285e-07, | |
| "loss": 0.818, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "eval_loss": 0.8412903547286987, | |
| "eval_runtime": 73.3325, | |
| "eval_samples_per_second": 57.273, | |
| "eval_steps_per_second": 0.9, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0490693739424704, | |
| "grad_norm": 3.4712844894054182, | |
| "learning_rate": 7.457799083359471e-07, | |
| "loss": 0.7931, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.0829103214890017, | |
| "grad_norm": 3.581045212972793, | |
| "learning_rate": 7.297510872562131e-07, | |
| "loss": 0.7903, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.116751269035533, | |
| "grad_norm": 3.304018905030938, | |
| "learning_rate": 7.134171292854955e-07, | |
| "loss": 0.7805, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.1505922165820643, | |
| "grad_norm": 3.5760772107880694, | |
| "learning_rate": 6.967997278713089e-07, | |
| "loss": 0.8039, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1844331641285957, | |
| "grad_norm": 3.4996799034828943, | |
| "learning_rate": 6.79920952907921e-07, | |
| "loss": 0.7773, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.218274111675127, | |
| "grad_norm": 3.6909558079169376, | |
| "learning_rate": 6.628032214248982e-07, | |
| "loss": 0.7803, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.252115059221658, | |
| "grad_norm": 3.821600482215546, | |
| "learning_rate": 6.454692678146119e-07, | |
| "loss": 0.7881, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.2859560067681894, | |
| "grad_norm": 3.7317289544669983, | |
| "learning_rate": 6.279421136382494e-07, | |
| "loss": 0.7611, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3197969543147208, | |
| "grad_norm": 3.3653007194840487, | |
| "learning_rate": 6.102450370504299e-07, | |
| "loss": 0.7722, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "grad_norm": 3.448429862793799, | |
| "learning_rate": 5.924015418830354e-07, | |
| "loss": 0.77, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "eval_loss": 0.8311071395874023, | |
| "eval_runtime": 73.3623, | |
| "eval_samples_per_second": 57.25, | |
| "eval_steps_per_second": 0.9, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3874788494077834, | |
| "grad_norm": 3.5126750577909993, | |
| "learning_rate": 5.74435326429313e-07, | |
| "loss": 0.7882, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.4213197969543148, | |
| "grad_norm": 3.498420719768302, | |
| "learning_rate": 5.563702519697108e-07, | |
| "loss": 0.7799, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.455160744500846, | |
| "grad_norm": 3.596937767899288, | |
| "learning_rate": 5.382303110812466e-07, | |
| "loss": 0.7807, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.4890016920473772, | |
| "grad_norm": 3.4351691894382292, | |
| "learning_rate": 5.200395957725005e-07, | |
| "loss": 0.7766, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.5228426395939088, | |
| "grad_norm": 3.7180060178057177, | |
| "learning_rate": 5.018222654865471e-07, | |
| "loss": 0.7934, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.5566835871404399, | |
| "grad_norm": 3.5160283267974193, | |
| "learning_rate": 4.836025150143318e-07, | |
| "loss": 0.7866, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5905245346869712, | |
| "grad_norm": 3.4704636157746798, | |
| "learning_rate": 4.654045423610952e-07, | |
| "loss": 0.7702, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.6243654822335025, | |
| "grad_norm": 3.774960092850383, | |
| "learning_rate": 4.4725251660853357e-07, | |
| "loss": 0.7894, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6582064297800339, | |
| "grad_norm": 3.5691138942613314, | |
| "learning_rate": 4.2917054581536926e-07, | |
| "loss": 0.7753, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "grad_norm": 3.696717245730972, | |
| "learning_rate": 4.1118264499897003e-07, | |
| "loss": 0.7434, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "eval_loss": 0.8217452168464661, | |
| "eval_runtime": 73.3181, | |
| "eval_samples_per_second": 57.285, | |
| "eval_steps_per_second": 0.9, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7258883248730963, | |
| "grad_norm": 3.4881049254451524, | |
| "learning_rate": 3.9331270424053616e-07, | |
| "loss": 0.7688, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.7597292724196278, | |
| "grad_norm": 3.5846352074260244, | |
| "learning_rate": 3.755844569562191e-07, | |
| "loss": 0.7488, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.793570219966159, | |
| "grad_norm": 3.732110081747128, | |
| "learning_rate": 3.580214483763093e-07, | |
| "loss": 0.7662, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.8274111675126905, | |
| "grad_norm": 3.625640468934038, | |
| "learning_rate": 3.406470042743574e-07, | |
| "loss": 0.7786, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.8612521150592216, | |
| "grad_norm": 3.629745300594503, | |
| "learning_rate": 3.23484199987761e-07, | |
| "loss": 0.7756, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.895093062605753, | |
| "grad_norm": 3.711354708797325, | |
| "learning_rate": 3.065558297709588e-07, | |
| "loss": 0.7719, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.9289340101522843, | |
| "grad_norm": 3.4735681051737646, | |
| "learning_rate": 2.898843765219388e-07, | |
| "loss": 0.7658, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.9627749576988156, | |
| "grad_norm": 3.6997948049416194, | |
| "learning_rate": 2.7349198192226295e-07, | |
| "loss": 0.7638, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.996615905245347, | |
| "grad_norm": 3.552802450824442, | |
| "learning_rate": 2.574004170302696e-07, | |
| "loss": 0.7739, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "grad_norm": 3.5676179776962464, | |
| "learning_rate": 2.4163105336650643e-07, | |
| "loss": 0.7295, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "eval_loss": 0.817137598991394, | |
| "eval_runtime": 73.3807, | |
| "eval_samples_per_second": 57.236, | |
| "eval_steps_per_second": 0.899, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0642978003384096, | |
| "grad_norm": 3.584817764435053, | |
| "learning_rate": 2.2620483452979887e-07, | |
| "loss": 0.7347, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.0981387478849407, | |
| "grad_norm": 3.638087907826697, | |
| "learning_rate": 2.1114224838164806e-07, | |
| "loss": 0.7016, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1319796954314723, | |
| "grad_norm": 3.6585520356041585, | |
| "learning_rate": 1.964632998359036e-07, | |
| "loss": 0.7277, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.1658206429780034, | |
| "grad_norm": 3.6487421821744883, | |
| "learning_rate": 1.8218748428984782e-07, | |
| "loss": 0.738, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.199661590524535, | |
| "grad_norm": 3.7920215108101387, | |
| "learning_rate": 1.6833376173198005e-07, | |
| "loss": 0.7081, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.233502538071066, | |
| "grad_norm": 3.891656855680349, | |
| "learning_rate": 1.5492053156088498e-07, | |
| "loss": 0.7063, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.267343485617597, | |
| "grad_norm": 3.839396725585509, | |
| "learning_rate": 1.4196560814863567e-07, | |
| "loss": 0.7089, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 2.3011844331641287, | |
| "grad_norm": 3.9684244326616582, | |
| "learning_rate": 1.294861971811773e-07, | |
| "loss": 0.7281, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.33502538071066, | |
| "grad_norm": 4.044882143357874, | |
| "learning_rate": 1.1749887280712161e-07, | |
| "loss": 0.7029, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "grad_norm": 3.7761464031547507, | |
| "learning_rate": 1.0601955562529774e-07, | |
| "loss": 0.7105, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "eval_loss": 0.8195010423660278, | |
| "eval_runtime": 73.2987, | |
| "eval_samples_per_second": 57.3, | |
| "eval_steps_per_second": 0.9, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.4027072758037225, | |
| "grad_norm": 3.952702608923679, | |
| "learning_rate": 9.506349154029425e-08, | |
| "loss": 0.7319, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.436548223350254, | |
| "grad_norm": 3.7564660732893436, | |
| "learning_rate": 8.46452315140772e-08, | |
| "loss": 0.7166, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.470389170896785, | |
| "grad_norm": 3.6940657432889914, | |
| "learning_rate": 7.477861224057403e-08, | |
| "loss": 0.7135, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.504230118443316, | |
| "grad_norm": 3.971509239172655, | |
| "learning_rate": 6.547673776889095e-08, | |
| "loss": 0.723, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.5380710659898478, | |
| "grad_norm": 3.525966906531149, | |
| "learning_rate": 5.6751962099570396e-08, | |
| "loss": 0.7136, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.571912013536379, | |
| "grad_norm": 3.861645295828095, | |
| "learning_rate": 4.861587277700274e-08, | |
| "loss": 0.7065, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.6057529610829104, | |
| "grad_norm": 3.717743918270453, | |
| "learning_rate": 4.107927549978235e-08, | |
| "loss": 0.7175, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.6395939086294415, | |
| "grad_norm": 3.6557649436864743, | |
| "learning_rate": 3.4152179769449396e-08, | |
| "loss": 0.7127, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.673434856175973, | |
| "grad_norm": 3.710305859402267, | |
| "learning_rate": 2.784378559667622e-08, | |
| "loss": 0.7129, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "grad_norm": 3.586323419338086, | |
| "learning_rate": 2.2162471282553553e-08, | |
| "loss": 0.7202, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "eval_loss": 0.8184880018234253, | |
| "eval_runtime": 73.3278, | |
| "eval_samples_per_second": 57.277, | |
| "eval_steps_per_second": 0.9, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.7411167512690353, | |
| "grad_norm": 3.813268548866539, | |
| "learning_rate": 1.7115782291206082e-08, | |
| "loss": 0.7184, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.774957698815567, | |
| "grad_norm": 3.8521495700670796, | |
| "learning_rate": 1.2710421228514733e-08, | |
| "loss": 0.7031, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.808798646362098, | |
| "grad_norm": 3.829733158101215, | |
| "learning_rate": 8.952238940255153e-09, | |
| "loss": 0.7175, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.8426395939086295, | |
| "grad_norm": 3.788240055268367, | |
| "learning_rate": 5.846226741475557e-09, | |
| "loss": 0.7199, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.8764805414551606, | |
| "grad_norm": 3.8402837662797285, | |
| "learning_rate": 3.3965097874343872e-09, | |
| "loss": 0.735, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.910321489001692, | |
| "grad_norm": 3.7954951315305645, | |
| "learning_rate": 1.6063415949008618e-09, | |
| "loss": 0.7267, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.9441624365482233, | |
| "grad_norm": 3.610630533441302, | |
| "learning_rate": 4.780997210962478e-10, | |
| "loss": 0.7217, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.9780033840947544, | |
| "grad_norm": 3.6564021061260137, | |
| "learning_rate": 1.328260601385356e-11, | |
| "loss": 0.7201, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.984771573604061, | |
| "step": 441, | |
| "total_flos": 2600101931384832.0, | |
| "train_loss": 0.8260096437266083, | |
| "train_runtime": 6398.3881, | |
| "train_samples_per_second": 17.723, | |
| "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 | |
| } | |