| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.99492385786802, | |
| "eval_steps": 50, | |
| "global_step": 885, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01692047377326565, | |
| "grad_norm": 17.95940849112974, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7425, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0338409475465313, | |
| "grad_norm": 12.23967055037602, | |
| "learning_rate": 1e-06, | |
| "loss": 1.599, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.050761421319796954, | |
| "grad_norm": 7.600510364072396, | |
| "learning_rate": 9.99919433964529e-07, | |
| "loss": 1.2976, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0676818950930626, | |
| "grad_norm": 4.134278668785207, | |
| "learning_rate": 9.996777618216605e-07, | |
| "loss": 1.1572, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08460236886632826, | |
| "grad_norm": 3.846352071719652, | |
| "learning_rate": 9.992750614536604e-07, | |
| "loss": 1.0495, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.10152284263959391, | |
| "grad_norm": 3.4799081252721886, | |
| "learning_rate": 9.98711462636417e-07, | |
| "loss": 1.0222, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11844331641285956, | |
| "grad_norm": 3.6435583580883644, | |
| "learning_rate": 9.979871469976195e-07, | |
| "loss": 0.982, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.1353637901861252, | |
| "grad_norm": 3.474443331449369, | |
| "learning_rate": 9.971023479582256e-07, | |
| "loss": 0.9659, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15228426395939088, | |
| "grad_norm": 3.5291718369580094, | |
| "learning_rate": 9.960573506572389e-07, | |
| "loss": 0.9517, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1692047377326565, | |
| "grad_norm": 3.6607844369820763, | |
| "learning_rate": 9.948524918598173e-07, | |
| "loss": 0.9744, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1692047377326565, | |
| "eval_loss": 0.9363481402397156, | |
| "eval_runtime": 147.9244, | |
| "eval_samples_per_second": 56.786, | |
| "eval_steps_per_second": 0.892, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18612521150592218, | |
| "grad_norm": 3.5029465776623305, | |
| "learning_rate": 9.934881598487478e-07, | |
| "loss": 0.9291, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.20304568527918782, | |
| "grad_norm": 3.420583685374325, | |
| "learning_rate": 9.919647942993147e-07, | |
| "loss": 0.9373, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21996615905245348, | |
| "grad_norm": 3.639006411388483, | |
| "learning_rate": 9.9028288613761e-07, | |
| "loss": 0.9371, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.23688663282571912, | |
| "grad_norm": 3.739200188919654, | |
| "learning_rate": 9.884429773823236e-07, | |
| "loss": 0.9168, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.25380710659898476, | |
| "grad_norm": 3.6961021681290647, | |
| "learning_rate": 9.864456609700723e-07, | |
| "loss": 0.9036, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.2707275803722504, | |
| "grad_norm": 3.469126484018746, | |
| "learning_rate": 9.842915805643156e-07, | |
| "loss": 0.8789, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2876480541455161, | |
| "grad_norm": 3.4540644342177953, | |
| "learning_rate": 9.819814303479267e-07, | |
| "loss": 0.8843, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.30456852791878175, | |
| "grad_norm": 3.491589457047051, | |
| "learning_rate": 9.795159547994828e-07, | |
| "loss": 0.878, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.32148900169204736, | |
| "grad_norm": 3.3672158895857582, | |
| "learning_rate": 9.76895948453346e-07, | |
| "loss": 0.8817, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "grad_norm": 3.4204437907535032, | |
| "learning_rate": 9.74122255643613e-07, | |
| "loss": 0.8749, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "eval_loss": 0.8794726729393005, | |
| "eval_runtime": 146.5584, | |
| "eval_samples_per_second": 57.315, | |
| "eval_steps_per_second": 0.901, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3553299492385787, | |
| "grad_norm": 3.499218396193934, | |
| "learning_rate": 9.711957702320174e-07, | |
| "loss": 0.8689, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.37225042301184436, | |
| "grad_norm": 3.575982847323226, | |
| "learning_rate": 9.681174353198686e-07, | |
| "loss": 0.8552, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.38917089678510997, | |
| "grad_norm": 3.5820322461640624, | |
| "learning_rate": 9.648882429441256e-07, | |
| "loss": 0.8723, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.40609137055837563, | |
| "grad_norm": 3.428649928071515, | |
| "learning_rate": 9.615092337576987e-07, | |
| "loss": 0.8737, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4230118443316413, | |
| "grad_norm": 3.439041413094408, | |
| "learning_rate": 9.579814966940833e-07, | |
| "loss": 0.8574, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.43993231810490696, | |
| "grad_norm": 3.611321063385662, | |
| "learning_rate": 9.543061686164372e-07, | |
| "loss": 0.8774, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.45685279187817257, | |
| "grad_norm": 3.3003493132482657, | |
| "learning_rate": 9.504844339512094e-07, | |
| "loss": 0.8594, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.47377326565143824, | |
| "grad_norm": 3.4741051861556684, | |
| "learning_rate": 9.465175243064428e-07, | |
| "loss": 0.8674, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4906937394247039, | |
| "grad_norm": 3.2142657401436416, | |
| "learning_rate": 9.424067180748691e-07, | |
| "loss": 0.8648, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.5076142131979695, | |
| "grad_norm": 3.5722793490057336, | |
| "learning_rate": 9.381533400219317e-07, | |
| "loss": 0.8423, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5076142131979695, | |
| "eval_loss": 0.8532436490058899, | |
| "eval_runtime": 146.5502, | |
| "eval_samples_per_second": 57.318, | |
| "eval_steps_per_second": 0.901, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5245346869712352, | |
| "grad_norm": 3.387238895689527, | |
| "learning_rate": 9.337587608588588e-07, | |
| "loss": 0.8344, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.5414551607445008, | |
| "grad_norm": 3.3257959351398165, | |
| "learning_rate": 9.29224396800933e-07, | |
| "loss": 0.8296, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5583756345177665, | |
| "grad_norm": 3.4206543705067425, | |
| "learning_rate": 9.245517091110968e-07, | |
| "loss": 0.8281, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.5752961082910322, | |
| "grad_norm": 3.2917606038623672, | |
| "learning_rate": 9.197422036290386e-07, | |
| "loss": 0.8388, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5922165820642978, | |
| "grad_norm": 3.6356342856497874, | |
| "learning_rate": 9.147974302859156e-07, | |
| "loss": 0.8479, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.6091370558375635, | |
| "grad_norm": 3.531537582517598, | |
| "learning_rate": 9.097189826048659e-07, | |
| "loss": 0.8465, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.626057529610829, | |
| "grad_norm": 3.534890106733192, | |
| "learning_rate": 9.045084971874737e-07, | |
| "loss": 0.8338, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.6429780033840947, | |
| "grad_norm": 3.5162595243845476, | |
| "learning_rate": 8.991676531863507e-07, | |
| "loss": 0.8454, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6598984771573604, | |
| "grad_norm": 3.6712481126005607, | |
| "learning_rate": 8.93698171764006e-07, | |
| "loss": 0.8239, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "grad_norm": 3.3780652656023573, | |
| "learning_rate": 8.881018155381765e-07, | |
| "loss": 0.8269, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "eval_loss": 0.8365465998649597, | |
| "eval_runtime": 146.6071, | |
| "eval_samples_per_second": 57.296, | |
| "eval_steps_per_second": 0.9, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6937394247038917, | |
| "grad_norm": 3.3425859477068123, | |
| "learning_rate": 8.823803880137992e-07, | |
| "loss": 0.8382, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.7106598984771574, | |
| "grad_norm": 3.2738301474432956, | |
| "learning_rate": 8.765357330018055e-07, | |
| "loss": 0.8456, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.727580372250423, | |
| "grad_norm": 3.563151004829781, | |
| "learning_rate": 8.705697340249274e-07, | |
| "loss": 0.8266, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.7445008460236887, | |
| "grad_norm": 3.1683419089159126, | |
| "learning_rate": 8.644843137107057e-07, | |
| "loss": 0.8403, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7614213197969543, | |
| "grad_norm": 3.609357791743884, | |
| "learning_rate": 8.58281433171896e-07, | |
| "loss": 0.8244, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.7783417935702199, | |
| "grad_norm": 3.421312500448169, | |
| "learning_rate": 8.519630913744724e-07, | |
| "loss": 0.8288, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7952622673434856, | |
| "grad_norm": 3.577171842922122, | |
| "learning_rate": 8.455313244934324e-07, | |
| "loss": 0.8167, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.8121827411167513, | |
| "grad_norm": 3.325456007325782, | |
| "learning_rate": 8.389882052566105e-07, | |
| "loss": 0.8118, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8291032148900169, | |
| "grad_norm": 3.5627054194832914, | |
| "learning_rate": 8.323358422767128e-07, | |
| "loss": 0.8378, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.8460236886632826, | |
| "grad_norm": 3.57666960479274, | |
| "learning_rate": 8.255763793717867e-07, | |
| "loss": 0.8223, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8460236886632826, | |
| "eval_loss": 0.8226217031478882, | |
| "eval_runtime": 146.6337, | |
| "eval_samples_per_second": 57.286, | |
| "eval_steps_per_second": 0.9, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8629441624365483, | |
| "grad_norm": 3.4619536067260244, | |
| "learning_rate": 8.187119948743449e-07, | |
| "loss": 0.8215, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.8798646362098139, | |
| "grad_norm": 3.3755438705432805, | |
| "learning_rate": 8.117449009293668e-07, | |
| "loss": 0.7959, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8967851099830795, | |
| "grad_norm": 3.3504490747841595, | |
| "learning_rate": 8.046773427814041e-07, | |
| "loss": 0.8198, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.9137055837563451, | |
| "grad_norm": 3.652293584185451, | |
| "learning_rate": 7.975115980510185e-07, | |
| "loss": 0.8198, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.9306260575296108, | |
| "grad_norm": 3.497277962331386, | |
| "learning_rate": 7.902499760007867e-07, | |
| "loss": 0.8181, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.9475465313028765, | |
| "grad_norm": 3.5043433150139336, | |
| "learning_rate": 7.828948167911073e-07, | |
| "loss": 0.8151, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.9644670050761421, | |
| "grad_norm": 3.5662591692739034, | |
| "learning_rate": 7.754484907260512e-07, | |
| "loss": 0.8192, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.9813874788494078, | |
| "grad_norm": 3.741586297285277, | |
| "learning_rate": 7.679133974894982e-07, | |
| "loss": 0.7975, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9983079526226735, | |
| "grad_norm": 3.2550447105418057, | |
| "learning_rate": 7.602919653718043e-07, | |
| "loss": 0.7885, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "grad_norm": 3.3766404706055613, | |
| "learning_rate": 7.525866504872506e-07, | |
| "loss": 0.7651, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "eval_loss": 0.8148965239524841, | |
| "eval_runtime": 146.61, | |
| "eval_samples_per_second": 57.295, | |
| "eval_steps_per_second": 0.9, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0321489001692048, | |
| "grad_norm": 3.5183190070892314, | |
| "learning_rate": 7.447999359825262e-07, | |
| "loss": 0.7393, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.0490693739424704, | |
| "grad_norm": 3.7614891711473657, | |
| "learning_rate": 7.369343312364993e-07, | |
| "loss": 0.7621, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.0659898477157361, | |
| "grad_norm": 3.5318777711133125, | |
| "learning_rate": 7.289923710515338e-07, | |
| "loss": 0.7546, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.0829103214890017, | |
| "grad_norm": 3.5586679876971754, | |
| "learning_rate": 7.209766148366134e-07, | |
| "loss": 0.759, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0998307952622675, | |
| "grad_norm": 3.3557844891140305, | |
| "learning_rate": 7.128896457825363e-07, | |
| "loss": 0.7445, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.116751269035533, | |
| "grad_norm": 3.439711853714508, | |
| "learning_rate": 7.047340700294453e-07, | |
| "loss": 0.7406, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.1336717428087986, | |
| "grad_norm": 3.522824978614251, | |
| "learning_rate": 6.965125158269618e-07, | |
| "loss": 0.7368, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.1505922165820643, | |
| "grad_norm": 4.008601044386287, | |
| "learning_rate": 6.882276326871959e-07, | |
| "loss": 0.7578, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.16751269035533, | |
| "grad_norm": 3.6557873733426955, | |
| "learning_rate": 6.798820905309035e-07, | |
| "loss": 0.7332, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.1844331641285957, | |
| "grad_norm": 3.5152732593214515, | |
| "learning_rate": 6.714785788270657e-07, | |
| "loss": 0.7388, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1844331641285957, | |
| "eval_loss": 0.8107805252075195, | |
| "eval_runtime": 146.5199, | |
| "eval_samples_per_second": 57.33, | |
| "eval_steps_per_second": 0.901, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.2013536379018612, | |
| "grad_norm": 3.7338182802228093, | |
| "learning_rate": 6.630198057261709e-07, | |
| "loss": 0.7406, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.218274111675127, | |
| "grad_norm": 3.5135812697699724, | |
| "learning_rate": 6.545084971874736e-07, | |
| "loss": 0.7421, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.2351945854483926, | |
| "grad_norm": 3.508021675469905, | |
| "learning_rate": 6.459473961005168e-07, | |
| "loss": 0.7755, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.252115059221658, | |
| "grad_norm": 3.5287017860167196, | |
| "learning_rate": 6.373392614011951e-07, | |
| "loss": 0.7408, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.2690355329949239, | |
| "grad_norm": 3.6233235029794093, | |
| "learning_rate": 6.286868671826511e-07, | |
| "loss": 0.751, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.2859560067681894, | |
| "grad_norm": 3.5669498367227304, | |
| "learning_rate": 6.199930018012829e-07, | |
| "loss": 0.7276, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.3028764805414552, | |
| "grad_norm": 3.7287000280408176, | |
| "learning_rate": 6.112604669781572e-07, | |
| "loss": 0.7278, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.3197969543147208, | |
| "grad_norm": 3.824405237133237, | |
| "learning_rate": 6.024920768961152e-07, | |
| "loss": 0.743, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.3367174280879865, | |
| "grad_norm": 3.5197677626280965, | |
| "learning_rate": 5.936906572928624e-07, | |
| "loss": 0.7159, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "grad_norm": 3.759524343808812, | |
| "learning_rate": 5.848590445503344e-07, | |
| "loss": 0.7429, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "eval_loss": 0.805133044719696, | |
| "eval_runtime": 146.7971, | |
| "eval_samples_per_second": 57.222, | |
| "eval_steps_per_second": 0.899, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3705583756345177, | |
| "grad_norm": 3.797267695279564, | |
| "learning_rate": 5.760000847806337e-07, | |
| "loss": 0.7464, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.3874788494077834, | |
| "grad_norm": 3.439223330784389, | |
| "learning_rate": 5.671166329088277e-07, | |
| "loss": 0.725, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.404399323181049, | |
| "grad_norm": 3.6761682639396653, | |
| "learning_rate": 5.582115517529114e-07, | |
| "loss": 0.7311, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.4213197969543148, | |
| "grad_norm": 3.571768390407566, | |
| "learning_rate": 5.492877111012218e-07, | |
| "loss": 0.7393, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.4382402707275803, | |
| "grad_norm": 3.8046958424761623, | |
| "learning_rate": 5.403479867876087e-07, | |
| "loss": 0.758, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.455160744500846, | |
| "grad_norm": 3.552061598209118, | |
| "learning_rate": 5.313952597646567e-07, | |
| "loss": 0.741, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.4720812182741116, | |
| "grad_norm": 3.5137582048526546, | |
| "learning_rate": 5.224324151752575e-07, | |
| "loss": 0.736, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.4890016920473772, | |
| "grad_norm": 3.6806640730520046, | |
| "learning_rate": 5.134623414228315e-07, | |
| "loss": 0.7414, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.505922165820643, | |
| "grad_norm": 3.7306988391241203, | |
| "learning_rate": 5.044879292404989e-07, | |
| "loss": 0.7578, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 1.5228426395939088, | |
| "grad_norm": 3.5044826704791543, | |
| "learning_rate": 4.95512070759501e-07, | |
| "loss": 0.7481, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5228426395939088, | |
| "eval_loss": 0.8002220392227173, | |
| "eval_runtime": 146.6261, | |
| "eval_samples_per_second": 57.289, | |
| "eval_steps_per_second": 0.9, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5397631133671743, | |
| "grad_norm": 3.5593238876932416, | |
| "learning_rate": 4.865376585771687e-07, | |
| "loss": 0.741, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 1.5566835871404399, | |
| "grad_norm": 3.9021045537145174, | |
| "learning_rate": 4.775675848247427e-07, | |
| "loss": 0.7462, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.5736040609137056, | |
| "grad_norm": 3.603020142861588, | |
| "learning_rate": 4.686047402353433e-07, | |
| "loss": 0.7344, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.5905245346869712, | |
| "grad_norm": 3.5798855947417247, | |
| "learning_rate": 4.596520132123914e-07, | |
| "loss": 0.7246, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.6074450084602367, | |
| "grad_norm": 3.392440988553216, | |
| "learning_rate": 4.507122888987782e-07, | |
| "loss": 0.7304, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.6243654822335025, | |
| "grad_norm": 3.7346005543444307, | |
| "learning_rate": 4.417884482470886e-07, | |
| "loss": 0.7329, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.6412859560067683, | |
| "grad_norm": 3.929271128512869, | |
| "learning_rate": 4.328833670911724e-07, | |
| "loss": 0.7529, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 1.6582064297800339, | |
| "grad_norm": 3.5171536414776163, | |
| "learning_rate": 4.239999152193664e-07, | |
| "loss": 0.7531, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.6751269035532994, | |
| "grad_norm": 3.574806818948794, | |
| "learning_rate": 4.1514095544966557e-07, | |
| "loss": 0.7418, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "grad_norm": 3.5129274484405486, | |
| "learning_rate": 4.0630934270713755e-07, | |
| "loss": 0.7308, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "eval_loss": 0.795360267162323, | |
| "eval_runtime": 146.6381, | |
| "eval_samples_per_second": 57.284, | |
| "eval_steps_per_second": 0.9, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.708967851099831, | |
| "grad_norm": 3.483481666306759, | |
| "learning_rate": 3.9750792310388483e-07, | |
| "loss": 0.7311, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 1.7258883248730963, | |
| "grad_norm": 3.5163371793641387, | |
| "learning_rate": 3.8873953302184283e-07, | |
| "loss": 0.7268, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.742808798646362, | |
| "grad_norm": 3.746899781553285, | |
| "learning_rate": 3.80006998198717e-07, | |
| "loss": 0.7471, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 1.7597292724196278, | |
| "grad_norm": 3.6600329176410735, | |
| "learning_rate": 3.713131328173489e-07, | |
| "loss": 0.7426, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.7766497461928934, | |
| "grad_norm": 3.715103932641678, | |
| "learning_rate": 3.62660738598805e-07, | |
| "loss": 0.7452, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.793570219966159, | |
| "grad_norm": 3.6219953866879036, | |
| "learning_rate": 3.5405260389948333e-07, | |
| "loss": 0.7447, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.8104906937394247, | |
| "grad_norm": 3.5483251792927866, | |
| "learning_rate": 3.454915028125263e-07, | |
| "loss": 0.7219, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 1.8274111675126905, | |
| "grad_norm": 3.687667827428392, | |
| "learning_rate": 3.369801942738291e-07, | |
| "loss": 0.7297, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.844331641285956, | |
| "grad_norm": 3.538997656633232, | |
| "learning_rate": 3.285214211729343e-07, | |
| "loss": 0.7498, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 1.8612521150592216, | |
| "grad_norm": 3.763523903194986, | |
| "learning_rate": 3.2011790946909666e-07, | |
| "loss": 0.7306, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8612521150592216, | |
| "eval_loss": 0.7919500470161438, | |
| "eval_runtime": 146.647, | |
| "eval_samples_per_second": 57.28, | |
| "eval_steps_per_second": 0.9, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8781725888324874, | |
| "grad_norm": 4.154646770046085, | |
| "learning_rate": 3.11772367312804e-07, | |
| "loss": 0.7364, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 1.895093062605753, | |
| "grad_norm": 3.6964867615997874, | |
| "learning_rate": 3.034874841730382e-07, | |
| "loss": 0.7357, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.9120135363790185, | |
| "grad_norm": 3.6098682228440024, | |
| "learning_rate": 2.9526592997055483e-07, | |
| "loss": 0.7435, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 1.9289340101522843, | |
| "grad_norm": 3.5510859742021514, | |
| "learning_rate": 2.8711035421746363e-07, | |
| "loss": 0.7401, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.94585448392555, | |
| "grad_norm": 3.5339412485916477, | |
| "learning_rate": 2.7902338516338674e-07, | |
| "loss": 0.7196, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.9627749576988156, | |
| "grad_norm": 3.654769871220945, | |
| "learning_rate": 2.7100762894846627e-07, | |
| "loss": 0.7427, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.9796954314720812, | |
| "grad_norm": 3.5585570542699485, | |
| "learning_rate": 2.6306566876350067e-07, | |
| "loss": 0.7549, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 1.996615905245347, | |
| "grad_norm": 3.5792060121886804, | |
| "learning_rate": 2.5520006401747395e-07, | |
| "loss": 0.7306, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.0135363790186127, | |
| "grad_norm": 3.629067553958508, | |
| "learning_rate": 2.474133495127494e-07, | |
| "loss": 0.7062, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "grad_norm": 3.646688738000953, | |
| "learning_rate": 2.3970803462819583e-07, | |
| "loss": 0.7065, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "eval_loss": 0.7942918539047241, | |
| "eval_runtime": 304.0198, | |
| "eval_samples_per_second": 27.63, | |
| "eval_steps_per_second": 0.434, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.047377326565144, | |
| "grad_norm": 4.052211962047984, | |
| "learning_rate": 2.3208660251050156e-07, | |
| "loss": 0.675, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.0642978003384096, | |
| "grad_norm": 3.9688880359901444, | |
| "learning_rate": 2.2455150927394878e-07, | |
| "loss": 0.6934, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.081218274111675, | |
| "grad_norm": 3.8191830448574575, | |
| "learning_rate": 2.1710518320889276e-07, | |
| "loss": 0.695, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.0981387478849407, | |
| "grad_norm": 3.9122017489827488, | |
| "learning_rate": 2.097500239992132e-07, | |
| "loss": 0.6909, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.1150592216582065, | |
| "grad_norm": 4.035937936283847, | |
| "learning_rate": 2.0248840194898155e-07, | |
| "loss": 0.6869, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.1319796954314723, | |
| "grad_norm": 4.0592719997150875, | |
| "learning_rate": 1.9532265721859597e-07, | |
| "loss": 0.6758, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.1489001692047376, | |
| "grad_norm": 3.99048820587498, | |
| "learning_rate": 1.8825509907063326e-07, | |
| "loss": 0.6717, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.1658206429780034, | |
| "grad_norm": 3.6085759994280413, | |
| "learning_rate": 1.812880051256551e-07, | |
| "loss": 0.7084, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.182741116751269, | |
| "grad_norm": 3.8339000315450633, | |
| "learning_rate": 1.744236206282132e-07, | |
| "loss": 0.6795, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.199661590524535, | |
| "grad_norm": 3.8042578709482937, | |
| "learning_rate": 1.6766415772328728e-07, | |
| "loss": 0.695, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.199661590524535, | |
| "eval_loss": 0.7947296500205994, | |
| "eval_runtime": 146.6011, | |
| "eval_samples_per_second": 57.298, | |
| "eval_steps_per_second": 0.9, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.2165820642978002, | |
| "grad_norm": 3.6822683929855318, | |
| "learning_rate": 1.6101179474338966e-07, | |
| "loss": 0.6637, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 2.233502538071066, | |
| "grad_norm": 3.8036837748646746, | |
| "learning_rate": 1.5446867550656767e-07, | |
| "loss": 0.6846, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.250423011844332, | |
| "grad_norm": 3.7154882715692747, | |
| "learning_rate": 1.4803690862552753e-07, | |
| "loss": 0.6761, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.267343485617597, | |
| "grad_norm": 3.8188043343483167, | |
| "learning_rate": 1.4171856682810384e-07, | |
| "loss": 0.6834, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.284263959390863, | |
| "grad_norm": 3.7763683799555494, | |
| "learning_rate": 1.3551568628929432e-07, | |
| "loss": 0.682, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.3011844331641287, | |
| "grad_norm": 3.7184968099032742, | |
| "learning_rate": 1.2943026597507267e-07, | |
| "loss": 0.6758, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.3181049069373945, | |
| "grad_norm": 3.6752732128517978, | |
| "learning_rate": 1.2346426699819456e-07, | |
| "loss": 0.6778, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 2.33502538071066, | |
| "grad_norm": 3.914291257516614, | |
| "learning_rate": 1.176196119862008e-07, | |
| "loss": 0.6915, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.3519458544839256, | |
| "grad_norm": 3.6920364954305307, | |
| "learning_rate": 1.1189818446182358e-07, | |
| "loss": 0.6858, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "grad_norm": 3.848115462041246, | |
| "learning_rate": 1.0630182823599399e-07, | |
| "loss": 0.7013, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "eval_loss": 0.7938902378082275, | |
| "eval_runtime": 146.616, | |
| "eval_samples_per_second": 57.292, | |
| "eval_steps_per_second": 0.9, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.3857868020304567, | |
| "grad_norm": 3.7700574896430665, | |
| "learning_rate": 1.0083234681364932e-07, | |
| "loss": 0.6637, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 2.4027072758037225, | |
| "grad_norm": 3.662805050537354, | |
| "learning_rate": 9.549150281252632e-08, | |
| "loss": 0.6885, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.4196277495769882, | |
| "grad_norm": 3.7513042616130616, | |
| "learning_rate": 9.028101739513405e-08, | |
| "loss": 0.6949, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 2.436548223350254, | |
| "grad_norm": 3.5781011423869193, | |
| "learning_rate": 8.520256971408452e-08, | |
| "loss": 0.6796, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.4534686971235193, | |
| "grad_norm": 3.8550128332718465, | |
| "learning_rate": 8.025779637096137e-08, | |
| "loss": 0.6724, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.470389170896785, | |
| "grad_norm": 3.908604741906314, | |
| "learning_rate": 7.544829088890325e-08, | |
| "loss": 0.6789, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.487309644670051, | |
| "grad_norm": 3.815438363979128, | |
| "learning_rate": 7.077560319906694e-08, | |
| "loss": 0.6878, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 2.504230118443316, | |
| "grad_norm": 3.678241183457789, | |
| "learning_rate": 6.624123914114122e-08, | |
| "loss": 0.6758, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.521150592216582, | |
| "grad_norm": 3.766677918686516, | |
| "learning_rate": 6.184665997806831e-08, | |
| "loss": 0.6743, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 2.5380710659898478, | |
| "grad_norm": 3.7999340905324974, | |
| "learning_rate": 5.759328192513074e-08, | |
| "loss": 0.6743, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.5380710659898478, | |
| "eval_loss": 0.7932254672050476, | |
| "eval_runtime": 146.605, | |
| "eval_samples_per_second": 57.297, | |
| "eval_steps_per_second": 0.9, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.5549915397631136, | |
| "grad_norm": 3.865893271080869, | |
| "learning_rate": 5.348247569355735e-08, | |
| "loss": 0.6872, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 2.571912013536379, | |
| "grad_norm": 4.144060563570108, | |
| "learning_rate": 4.951556604879048e-08, | |
| "loss": 0.694, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.5888324873096447, | |
| "grad_norm": 3.780680669356305, | |
| "learning_rate": 4.569383138356275e-08, | |
| "loss": 0.6826, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 2.6057529610829104, | |
| "grad_norm": 4.114587718099576, | |
| "learning_rate": 4.201850330591677e-08, | |
| "loss": 0.68, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.6226734348561758, | |
| "grad_norm": 3.4623751046204285, | |
| "learning_rate": 3.8490766242301353e-08, | |
| "loss": 0.6658, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.6395939086294415, | |
| "grad_norm": 3.9196860968662137, | |
| "learning_rate": 3.5111757055874326e-08, | |
| "loss": 0.6898, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.6565143824027073, | |
| "grad_norm": 3.820036546151991, | |
| "learning_rate": 3.188256468013139e-08, | |
| "loss": 0.6732, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 2.673434856175973, | |
| "grad_norm": 4.231574698760902, | |
| "learning_rate": 2.8804229767982636e-08, | |
| "loss": 0.6728, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.6903553299492384, | |
| "grad_norm": 4.115986044491551, | |
| "learning_rate": 2.587774435638679e-08, | |
| "loss": 0.6755, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "grad_norm": 3.724220098163462, | |
| "learning_rate": 2.3104051546654013e-08, | |
| "loss": 0.6778, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "eval_loss": 0.7928686141967773, | |
| "eval_runtime": 146.5361, | |
| "eval_samples_per_second": 57.324, | |
| "eval_steps_per_second": 0.901, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.72419627749577, | |
| "grad_norm": 4.055902737307998, | |
| "learning_rate": 2.048404520051722e-08, | |
| "loss": 0.6846, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 2.7411167512690353, | |
| "grad_norm": 3.761429857208659, | |
| "learning_rate": 1.8018569652073378e-08, | |
| "loss": 0.6865, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.758037225042301, | |
| "grad_norm": 3.828128027906773, | |
| "learning_rate": 1.570841943568446e-08, | |
| "loss": 0.6712, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 2.774957698815567, | |
| "grad_norm": 3.765063878680342, | |
| "learning_rate": 1.3554339029927531e-08, | |
| "loss": 0.6816, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.7918781725888326, | |
| "grad_norm": 3.827232987418165, | |
| "learning_rate": 1.1557022617676216e-08, | |
| "loss": 0.6789, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.808798646362098, | |
| "grad_norm": 3.9934159183359106, | |
| "learning_rate": 9.717113862389992e-09, | |
| "loss": 0.6961, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.8257191201353637, | |
| "grad_norm": 3.905375697814281, | |
| "learning_rate": 8.035205700685165e-09, | |
| "loss": 0.6794, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 2.8426395939086295, | |
| "grad_norm": 3.8176339677184687, | |
| "learning_rate": 6.511840151252168e-09, | |
| "loss": 0.6858, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.859560067681895, | |
| "grad_norm": 3.669045035086845, | |
| "learning_rate": 5.147508140182555e-09, | |
| "loss": 0.6709, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 2.8764805414551606, | |
| "grad_norm": 3.7873007524702644, | |
| "learning_rate": 3.9426493427611175e-09, | |
| "loss": 0.6951, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.8764805414551606, | |
| "eval_loss": 0.7926760911941528, | |
| "eval_runtime": 146.7334, | |
| "eval_samples_per_second": 57.247, | |
| "eval_steps_per_second": 0.9, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.8934010152284264, | |
| "grad_norm": 3.6861972611151548, | |
| "learning_rate": 2.897652041774279e-09, | |
| "loss": 0.6878, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 2.910321489001692, | |
| "grad_norm": 3.6981706590030003, | |
| "learning_rate": 2.0128530023804656e-09, | |
| "loss": 0.6979, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.927241962774958, | |
| "grad_norm": 3.788952889279561, | |
| "learning_rate": 1.2885373635829754e-09, | |
| "loss": 0.6897, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 2.9441624365482233, | |
| "grad_norm": 3.818071711218172, | |
| "learning_rate": 7.249385463395374e-10, | |
| "loss": 0.6883, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.961082910321489, | |
| "grad_norm": 3.612189636258872, | |
| "learning_rate": 3.22238178339318e-10, | |
| "loss": 0.6879, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.9780033840947544, | |
| "grad_norm": 3.6807850492378975, | |
| "learning_rate": 8.056603547090812e-11, | |
| "loss": 0.6952, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.99492385786802, | |
| "grad_norm": 3.843102017586158, | |
| "learning_rate": 0.0, | |
| "loss": 0.6819, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 2.99492385786802, | |
| "step": 885, | |
| "total_flos": 5218127163949056.0, | |
| "train_loss": 0.776407975396194, | |
| "train_runtime": 14269.6597, | |
| "train_samples_per_second": 15.893, | |
| "train_steps_per_second": 0.062 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 885, | |
| "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": 5218127163949056.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |