| { | |
| "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": 18.11811168155941, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7276, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0338409475465313, | |
| "grad_norm": 12.465284537803942, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5961, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.050761421319796954, | |
| "grad_norm": 8.029594619011075, | |
| "learning_rate": 9.99919433964529e-07, | |
| "loss": 1.2868, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0676818950930626, | |
| "grad_norm": 4.322443518606483, | |
| "learning_rate": 9.996777618216605e-07, | |
| "loss": 1.1317, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08460236886632826, | |
| "grad_norm": 3.908382290362389, | |
| "learning_rate": 9.992750614536604e-07, | |
| "loss": 1.067, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.10152284263959391, | |
| "grad_norm": 3.3966731740840395, | |
| "learning_rate": 9.98711462636417e-07, | |
| "loss": 1.0087, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11844331641285956, | |
| "grad_norm": 3.404122104363119, | |
| "learning_rate": 9.979871469976195e-07, | |
| "loss": 0.9993, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.1353637901861252, | |
| "grad_norm": 3.5130122180648637, | |
| "learning_rate": 9.971023479582256e-07, | |
| "loss": 0.9808, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15228426395939088, | |
| "grad_norm": 3.8017404235610677, | |
| "learning_rate": 9.960573506572389e-07, | |
| "loss": 0.9589, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1692047377326565, | |
| "grad_norm": 3.7077817350880475, | |
| "learning_rate": 9.948524918598173e-07, | |
| "loss": 0.9418, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1692047377326565, | |
| "eval_loss": 0.9316250085830688, | |
| "eval_runtime": 146.1453, | |
| "eval_samples_per_second": 57.477, | |
| "eval_steps_per_second": 0.903, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18612521150592218, | |
| "grad_norm": 3.8413604284063037, | |
| "learning_rate": 9.934881598487478e-07, | |
| "loss": 0.9442, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.20304568527918782, | |
| "grad_norm": 3.4583776939446076, | |
| "learning_rate": 9.919647942993147e-07, | |
| "loss": 0.928, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21996615905245348, | |
| "grad_norm": 4.21857142186165, | |
| "learning_rate": 9.9028288613761e-07, | |
| "loss": 0.9234, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.23688663282571912, | |
| "grad_norm": 3.6660260059909637, | |
| "learning_rate": 9.884429773823236e-07, | |
| "loss": 0.9051, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.25380710659898476, | |
| "grad_norm": 3.3907413557630837, | |
| "learning_rate": 9.864456609700723e-07, | |
| "loss": 0.8896, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.2707275803722504, | |
| "grad_norm": 3.4712295626998713, | |
| "learning_rate": 9.842915805643156e-07, | |
| "loss": 0.8869, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2876480541455161, | |
| "grad_norm": 3.51256446643522, | |
| "learning_rate": 9.819814303479267e-07, | |
| "loss": 0.877, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.30456852791878175, | |
| "grad_norm": 3.3964554088877685, | |
| "learning_rate": 9.795159547994828e-07, | |
| "loss": 0.8925, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.32148900169204736, | |
| "grad_norm": 3.372268647607599, | |
| "learning_rate": 9.76895948453346e-07, | |
| "loss": 0.893, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "grad_norm": 3.5879097796760644, | |
| "learning_rate": 9.74122255643613e-07, | |
| "loss": 0.8739, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "eval_loss": 0.8766918778419495, | |
| "eval_runtime": 146.0181, | |
| "eval_samples_per_second": 57.527, | |
| "eval_steps_per_second": 0.904, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3553299492385787, | |
| "grad_norm": 3.6525162260417656, | |
| "learning_rate": 9.711957702320174e-07, | |
| "loss": 0.8943, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.37225042301184436, | |
| "grad_norm": 3.825829336255867, | |
| "learning_rate": 9.681174353198686e-07, | |
| "loss": 0.869, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.38917089678510997, | |
| "grad_norm": 3.5161921261333418, | |
| "learning_rate": 9.648882429441256e-07, | |
| "loss": 0.8732, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.40609137055837563, | |
| "grad_norm": 3.754342765222721, | |
| "learning_rate": 9.615092337576987e-07, | |
| "loss": 0.8695, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4230118443316413, | |
| "grad_norm": 3.8927983233298966, | |
| "learning_rate": 9.579814966940833e-07, | |
| "loss": 0.8676, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.43993231810490696, | |
| "grad_norm": 3.313844438659833, | |
| "learning_rate": 9.543061686164372e-07, | |
| "loss": 0.8448, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.45685279187817257, | |
| "grad_norm": 3.5030033367963354, | |
| "learning_rate": 9.504844339512094e-07, | |
| "loss": 0.848, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.47377326565143824, | |
| "grad_norm": 3.482428868379046, | |
| "learning_rate": 9.465175243064428e-07, | |
| "loss": 0.853, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4906937394247039, | |
| "grad_norm": 3.4957931844026326, | |
| "learning_rate": 9.424067180748691e-07, | |
| "loss": 0.8768, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.5076142131979695, | |
| "grad_norm": 3.6539662236846318, | |
| "learning_rate": 9.381533400219317e-07, | |
| "loss": 0.8408, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5076142131979695, | |
| "eval_loss": 0.8505008220672607, | |
| "eval_runtime": 145.9826, | |
| "eval_samples_per_second": 57.541, | |
| "eval_steps_per_second": 0.904, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5245346869712352, | |
| "grad_norm": 3.5394178909999634, | |
| "learning_rate": 9.337587608588588e-07, | |
| "loss": 0.8527, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.5414551607445008, | |
| "grad_norm": 3.3947598872414386, | |
| "learning_rate": 9.29224396800933e-07, | |
| "loss": 0.8408, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5583756345177665, | |
| "grad_norm": 3.4873778908245434, | |
| "learning_rate": 9.245517091110968e-07, | |
| "loss": 0.8367, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.5752961082910322, | |
| "grad_norm": 3.50355137921012, | |
| "learning_rate": 9.197422036290386e-07, | |
| "loss": 0.8654, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5922165820642978, | |
| "grad_norm": 3.665320201247955, | |
| "learning_rate": 9.147974302859156e-07, | |
| "loss": 0.8467, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.6091370558375635, | |
| "grad_norm": 3.5032254780990266, | |
| "learning_rate": 9.097189826048659e-07, | |
| "loss": 0.8456, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.626057529610829, | |
| "grad_norm": 3.3742001228223693, | |
| "learning_rate": 9.045084971874737e-07, | |
| "loss": 0.8186, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.6429780033840947, | |
| "grad_norm": 3.5363272285750442, | |
| "learning_rate": 8.991676531863507e-07, | |
| "loss": 0.8354, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6598984771573604, | |
| "grad_norm": 3.6685050093786975, | |
| "learning_rate": 8.93698171764006e-07, | |
| "loss": 0.8415, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "grad_norm": 3.580018631107888, | |
| "learning_rate": 8.881018155381765e-07, | |
| "loss": 0.8415, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "eval_loss": 0.8333195447921753, | |
| "eval_runtime": 145.929, | |
| "eval_samples_per_second": 57.562, | |
| "eval_steps_per_second": 0.905, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6937394247038917, | |
| "grad_norm": 3.4977190055813225, | |
| "learning_rate": 8.823803880137992e-07, | |
| "loss": 0.8425, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.7106598984771574, | |
| "grad_norm": 3.424911861726825, | |
| "learning_rate": 8.765357330018055e-07, | |
| "loss": 0.8197, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.727580372250423, | |
| "grad_norm": 3.5180014609137804, | |
| "learning_rate": 8.705697340249274e-07, | |
| "loss": 0.84, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.7445008460236887, | |
| "grad_norm": 3.269933787195742, | |
| "learning_rate": 8.644843137107057e-07, | |
| "loss": 0.8192, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7614213197969543, | |
| "grad_norm": 3.490737285532297, | |
| "learning_rate": 8.58281433171896e-07, | |
| "loss": 0.8392, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.7783417935702199, | |
| "grad_norm": 3.6006380297077745, | |
| "learning_rate": 8.519630913744724e-07, | |
| "loss": 0.8285, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7952622673434856, | |
| "grad_norm": 3.5073056320425744, | |
| "learning_rate": 8.455313244934324e-07, | |
| "loss": 0.8157, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.8121827411167513, | |
| "grad_norm": 3.384395618192681, | |
| "learning_rate": 8.389882052566105e-07, | |
| "loss": 0.8205, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8291032148900169, | |
| "grad_norm": 3.4511971640264494, | |
| "learning_rate": 8.323358422767128e-07, | |
| "loss": 0.8236, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.8460236886632826, | |
| "grad_norm": 3.454441667633359, | |
| "learning_rate": 8.255763793717867e-07, | |
| "loss": 0.8121, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8460236886632826, | |
| "eval_loss": 0.8217905163764954, | |
| "eval_runtime": 146.2621, | |
| "eval_samples_per_second": 57.431, | |
| "eval_steps_per_second": 0.902, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8629441624365483, | |
| "grad_norm": 3.4938458445163985, | |
| "learning_rate": 8.187119948743449e-07, | |
| "loss": 0.8432, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.8798646362098139, | |
| "grad_norm": 3.6058018876142834, | |
| "learning_rate": 8.117449009293668e-07, | |
| "loss": 0.8015, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8967851099830795, | |
| "grad_norm": 3.5735964122028037, | |
| "learning_rate": 8.046773427814041e-07, | |
| "loss": 0.8385, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.9137055837563451, | |
| "grad_norm": 3.6543886570209496, | |
| "learning_rate": 7.975115980510185e-07, | |
| "loss": 0.8056, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.9306260575296108, | |
| "grad_norm": 3.3189069792277515, | |
| "learning_rate": 7.902499760007867e-07, | |
| "loss": 0.8269, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.9475465313028765, | |
| "grad_norm": 3.3073330589639034, | |
| "learning_rate": 7.828948167911073e-07, | |
| "loss": 0.8222, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.9644670050761421, | |
| "grad_norm": 3.4922958115443046, | |
| "learning_rate": 7.754484907260512e-07, | |
| "loss": 0.8149, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.9813874788494078, | |
| "grad_norm": 3.327799524845609, | |
| "learning_rate": 7.679133974894982e-07, | |
| "loss": 0.846, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9983079526226735, | |
| "grad_norm": 3.3720725389154973, | |
| "learning_rate": 7.602919653718043e-07, | |
| "loss": 0.8155, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "grad_norm": 3.440889104712857, | |
| "learning_rate": 7.525866504872506e-07, | |
| "loss": 0.7579, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "eval_loss": 0.8133670687675476, | |
| "eval_runtime": 146.0887, | |
| "eval_samples_per_second": 57.499, | |
| "eval_steps_per_second": 0.904, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0321489001692048, | |
| "grad_norm": 3.523402139992769, | |
| "learning_rate": 7.447999359825262e-07, | |
| "loss": 0.757, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.0490693739424704, | |
| "grad_norm": 3.736820649423743, | |
| "learning_rate": 7.369343312364993e-07, | |
| "loss": 0.7555, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.0659898477157361, | |
| "grad_norm": 3.63625396256149, | |
| "learning_rate": 7.289923710515338e-07, | |
| "loss": 0.7562, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.0829103214890017, | |
| "grad_norm": 3.444640371019277, | |
| "learning_rate": 7.209766148366134e-07, | |
| "loss": 0.7431, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0998307952622675, | |
| "grad_norm": 3.33725792795571, | |
| "learning_rate": 7.128896457825363e-07, | |
| "loss": 0.7419, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.116751269035533, | |
| "grad_norm": 3.463759633912316, | |
| "learning_rate": 7.047340700294453e-07, | |
| "loss": 0.7359, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.1336717428087986, | |
| "grad_norm": 3.6911828143235175, | |
| "learning_rate": 6.965125158269618e-07, | |
| "loss": 0.7483, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.1505922165820643, | |
| "grad_norm": 3.425162825369402, | |
| "learning_rate": 6.882276326871959e-07, | |
| "loss": 0.7523, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.16751269035533, | |
| "grad_norm": 3.534755736267555, | |
| "learning_rate": 6.798820905309035e-07, | |
| "loss": 0.7477, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.1844331641285957, | |
| "grad_norm": 3.355585464713691, | |
| "learning_rate": 6.714785788270657e-07, | |
| "loss": 0.7469, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1844331641285957, | |
| "eval_loss": 0.8087317943572998, | |
| "eval_runtime": 146.194, | |
| "eval_samples_per_second": 57.458, | |
| "eval_steps_per_second": 0.903, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.2013536379018612, | |
| "grad_norm": 3.7479898725924947, | |
| "learning_rate": 6.630198057261709e-07, | |
| "loss": 0.7506, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.218274111675127, | |
| "grad_norm": 3.57646668569766, | |
| "learning_rate": 6.545084971874736e-07, | |
| "loss": 0.7418, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.2351945854483926, | |
| "grad_norm": 3.4779353040269227, | |
| "learning_rate": 6.459473961005168e-07, | |
| "loss": 0.7385, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.252115059221658, | |
| "grad_norm": 3.5090655067073677, | |
| "learning_rate": 6.373392614011951e-07, | |
| "loss": 0.7361, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.2690355329949239, | |
| "grad_norm": 3.5457643005290005, | |
| "learning_rate": 6.286868671826511e-07, | |
| "loss": 0.7667, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.2859560067681894, | |
| "grad_norm": 3.700629853039733, | |
| "learning_rate": 6.199930018012829e-07, | |
| "loss": 0.751, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.3028764805414552, | |
| "grad_norm": 3.7733854906847215, | |
| "learning_rate": 6.112604669781572e-07, | |
| "loss": 0.7435, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.3197969543147208, | |
| "grad_norm": 3.609967851597449, | |
| "learning_rate": 6.024920768961152e-07, | |
| "loss": 0.7506, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.3367174280879865, | |
| "grad_norm": 3.5839373687018057, | |
| "learning_rate": 5.936906572928624e-07, | |
| "loss": 0.7525, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "grad_norm": 3.7104702064535697, | |
| "learning_rate": 5.848590445503344e-07, | |
| "loss": 0.7411, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "eval_loss": 0.803146243095398, | |
| "eval_runtime": 146.173, | |
| "eval_samples_per_second": 57.466, | |
| "eval_steps_per_second": 0.903, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3705583756345177, | |
| "grad_norm": 3.586978273247657, | |
| "learning_rate": 5.760000847806337e-07, | |
| "loss": 0.7262, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.3874788494077834, | |
| "grad_norm": 3.5534067032813033, | |
| "learning_rate": 5.671166329088277e-07, | |
| "loss": 0.7478, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.404399323181049, | |
| "grad_norm": 3.5702928476693505, | |
| "learning_rate": 5.582115517529114e-07, | |
| "loss": 0.7482, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.4213197969543148, | |
| "grad_norm": 3.497025840273878, | |
| "learning_rate": 5.492877111012218e-07, | |
| "loss": 0.7305, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.4382402707275803, | |
| "grad_norm": 3.8312529281102865, | |
| "learning_rate": 5.403479867876087e-07, | |
| "loss": 0.7417, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.455160744500846, | |
| "grad_norm": 3.5923517702731833, | |
| "learning_rate": 5.313952597646567e-07, | |
| "loss": 0.7552, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.4720812182741116, | |
| "grad_norm": 3.441823126296519, | |
| "learning_rate": 5.224324151752575e-07, | |
| "loss": 0.7504, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.4890016920473772, | |
| "grad_norm": 3.5174370036638534, | |
| "learning_rate": 5.134623414228315e-07, | |
| "loss": 0.7511, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.505922165820643, | |
| "grad_norm": 4.018568644922475, | |
| "learning_rate": 5.044879292404989e-07, | |
| "loss": 0.7493, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 1.5228426395939088, | |
| "grad_norm": 3.5869525261938184, | |
| "learning_rate": 4.95512070759501e-07, | |
| "loss": 0.7378, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5228426395939088, | |
| "eval_loss": 0.798000693321228, | |
| "eval_runtime": 146.3627, | |
| "eval_samples_per_second": 57.392, | |
| "eval_steps_per_second": 0.902, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5397631133671743, | |
| "grad_norm": 3.749721313687578, | |
| "learning_rate": 4.865376585771687e-07, | |
| "loss": 0.762, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 1.5566835871404399, | |
| "grad_norm": 3.680890544965231, | |
| "learning_rate": 4.775675848247427e-07, | |
| "loss": 0.7417, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.5736040609137056, | |
| "grad_norm": 3.5281531142193874, | |
| "learning_rate": 4.686047402353433e-07, | |
| "loss": 0.736, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.5905245346869712, | |
| "grad_norm": 3.555081148710941, | |
| "learning_rate": 4.596520132123914e-07, | |
| "loss": 0.7351, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.6074450084602367, | |
| "grad_norm": 3.4493804884671944, | |
| "learning_rate": 4.507122888987782e-07, | |
| "loss": 0.7514, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.6243654822335025, | |
| "grad_norm": 3.7270963408742452, | |
| "learning_rate": 4.417884482470886e-07, | |
| "loss": 0.7374, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.6412859560067683, | |
| "grad_norm": 3.845425412960905, | |
| "learning_rate": 4.328833670911724e-07, | |
| "loss": 0.7547, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 1.6582064297800339, | |
| "grad_norm": 3.4772313441838505, | |
| "learning_rate": 4.239999152193664e-07, | |
| "loss": 0.7345, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.6751269035532994, | |
| "grad_norm": 3.8728376091114436, | |
| "learning_rate": 4.1514095544966557e-07, | |
| "loss": 0.731, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "grad_norm": 3.5535644482480633, | |
| "learning_rate": 4.0630934270713755e-07, | |
| "loss": 0.7423, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "eval_loss": 0.7937961220741272, | |
| "eval_runtime": 146.133, | |
| "eval_samples_per_second": 57.482, | |
| "eval_steps_per_second": 0.903, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.708967851099831, | |
| "grad_norm": 3.60894521319071, | |
| "learning_rate": 3.9750792310388483e-07, | |
| "loss": 0.7264, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 1.7258883248730963, | |
| "grad_norm": 3.5752693076457227, | |
| "learning_rate": 3.8873953302184283e-07, | |
| "loss": 0.7129, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.742808798646362, | |
| "grad_norm": 3.5319598859647208, | |
| "learning_rate": 3.80006998198717e-07, | |
| "loss": 0.7379, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 1.7597292724196278, | |
| "grad_norm": 3.7064342382307136, | |
| "learning_rate": 3.713131328173489e-07, | |
| "loss": 0.7291, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.7766497461928934, | |
| "grad_norm": 3.6278369087716413, | |
| "learning_rate": 3.62660738598805e-07, | |
| "loss": 0.7248, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.793570219966159, | |
| "grad_norm": 3.888359889647679, | |
| "learning_rate": 3.5405260389948333e-07, | |
| "loss": 0.7516, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.8104906937394247, | |
| "grad_norm": 3.4352787394554727, | |
| "learning_rate": 3.454915028125263e-07, | |
| "loss": 0.7398, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 1.8274111675126905, | |
| "grad_norm": 3.5354090229387714, | |
| "learning_rate": 3.369801942738291e-07, | |
| "loss": 0.7415, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.844331641285956, | |
| "grad_norm": 3.6324342637431912, | |
| "learning_rate": 3.285214211729343e-07, | |
| "loss": 0.7343, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 1.8612521150592216, | |
| "grad_norm": 4.0734439950626555, | |
| "learning_rate": 3.2011790946909666e-07, | |
| "loss": 0.7363, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8612521150592216, | |
| "eval_loss": 0.7899667620658875, | |
| "eval_runtime": 146.2033, | |
| "eval_samples_per_second": 57.454, | |
| "eval_steps_per_second": 0.903, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8781725888324874, | |
| "grad_norm": 3.488509809559841, | |
| "learning_rate": 3.11772367312804e-07, | |
| "loss": 0.722, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 1.895093062605753, | |
| "grad_norm": 3.568702481260837, | |
| "learning_rate": 3.034874841730382e-07, | |
| "loss": 0.7268, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.9120135363790185, | |
| "grad_norm": 3.5437718957391584, | |
| "learning_rate": 2.9526592997055483e-07, | |
| "loss": 0.7397, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 1.9289340101522843, | |
| "grad_norm": 3.467818973564197, | |
| "learning_rate": 2.8711035421746363e-07, | |
| "loss": 0.7438, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.94585448392555, | |
| "grad_norm": 3.7876015240002396, | |
| "learning_rate": 2.7902338516338674e-07, | |
| "loss": 0.7221, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.9627749576988156, | |
| "grad_norm": 3.8139203476929824, | |
| "learning_rate": 2.7100762894846627e-07, | |
| "loss": 0.7257, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.9796954314720812, | |
| "grad_norm": 3.6272530624543853, | |
| "learning_rate": 2.6306566876350067e-07, | |
| "loss": 0.7254, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 1.996615905245347, | |
| "grad_norm": 3.676365257262644, | |
| "learning_rate": 2.5520006401747395e-07, | |
| "loss": 0.7208, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.0135363790186127, | |
| "grad_norm": 3.8206014953262084, | |
| "learning_rate": 2.474133495127494e-07, | |
| "loss": 0.7074, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "grad_norm": 3.876566280546555, | |
| "learning_rate": 2.3970803462819583e-07, | |
| "loss": 0.6778, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "eval_loss": 0.7937635183334351, | |
| "eval_runtime": 146.134, | |
| "eval_samples_per_second": 57.481, | |
| "eval_steps_per_second": 0.903, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.047377326565144, | |
| "grad_norm": 3.9168575166004937, | |
| "learning_rate": 2.3208660251050156e-07, | |
| "loss": 0.6807, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.0642978003384096, | |
| "grad_norm": 3.894680280362581, | |
| "learning_rate": 2.2455150927394878e-07, | |
| "loss": 0.6935, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.081218274111675, | |
| "grad_norm": 3.9693943038444104, | |
| "learning_rate": 2.1710518320889276e-07, | |
| "loss": 0.691, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.0981387478849407, | |
| "grad_norm": 3.794663051980087, | |
| "learning_rate": 2.097500239992132e-07, | |
| "loss": 0.6957, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.1150592216582065, | |
| "grad_norm": 3.9622114678017306, | |
| "learning_rate": 2.0248840194898155e-07, | |
| "loss": 0.6752, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.1319796954314723, | |
| "grad_norm": 3.7595934491117573, | |
| "learning_rate": 1.9532265721859597e-07, | |
| "loss": 0.6771, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.1489001692047376, | |
| "grad_norm": 3.9388724645377162, | |
| "learning_rate": 1.8825509907063326e-07, | |
| "loss": 0.6858, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.1658206429780034, | |
| "grad_norm": 3.7788201106321972, | |
| "learning_rate": 1.812880051256551e-07, | |
| "loss": 0.6982, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.182741116751269, | |
| "grad_norm": 3.947050032362458, | |
| "learning_rate": 1.744236206282132e-07, | |
| "loss": 0.6895, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.199661590524535, | |
| "grad_norm": 3.600017791091893, | |
| "learning_rate": 1.6766415772328728e-07, | |
| "loss": 0.6716, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.199661590524535, | |
| "eval_loss": 0.7936422824859619, | |
| "eval_runtime": 146.1456, | |
| "eval_samples_per_second": 57.477, | |
| "eval_steps_per_second": 0.903, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.2165820642978002, | |
| "grad_norm": 4.081340915582073, | |
| "learning_rate": 1.6101179474338966e-07, | |
| "loss": 0.6804, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 2.233502538071066, | |
| "grad_norm": 3.7657350706904342, | |
| "learning_rate": 1.5446867550656767e-07, | |
| "loss": 0.6899, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.250423011844332, | |
| "grad_norm": 3.9022387275901536, | |
| "learning_rate": 1.4803690862552753e-07, | |
| "loss": 0.6798, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.267343485617597, | |
| "grad_norm": 3.682924093787561, | |
| "learning_rate": 1.4171856682810384e-07, | |
| "loss": 0.6814, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.284263959390863, | |
| "grad_norm": 3.6877950643476938, | |
| "learning_rate": 1.3551568628929432e-07, | |
| "loss": 0.6739, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.3011844331641287, | |
| "grad_norm": 3.935618521241239, | |
| "learning_rate": 1.2943026597507267e-07, | |
| "loss": 0.6929, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.3181049069373945, | |
| "grad_norm": 3.5924620792590534, | |
| "learning_rate": 1.2346426699819456e-07, | |
| "loss": 0.6951, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 2.33502538071066, | |
| "grad_norm": 3.8665100002377204, | |
| "learning_rate": 1.176196119862008e-07, | |
| "loss": 0.6861, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.3519458544839256, | |
| "grad_norm": 3.7323294473109843, | |
| "learning_rate": 1.1189818446182358e-07, | |
| "loss": 0.6934, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "grad_norm": 3.689893104326803, | |
| "learning_rate": 1.0630182823599399e-07, | |
| "loss": 0.6809, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "eval_loss": 0.7926481366157532, | |
| "eval_runtime": 146.1872, | |
| "eval_samples_per_second": 57.461, | |
| "eval_steps_per_second": 0.903, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.3857868020304567, | |
| "grad_norm": 3.724488260369369, | |
| "learning_rate": 1.0083234681364932e-07, | |
| "loss": 0.6874, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 2.4027072758037225, | |
| "grad_norm": 4.0223025153408285, | |
| "learning_rate": 9.549150281252632e-08, | |
| "loss": 0.6796, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.4196277495769882, | |
| "grad_norm": 3.8561396496122247, | |
| "learning_rate": 9.028101739513405e-08, | |
| "loss": 0.6843, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 2.436548223350254, | |
| "grad_norm": 3.801934759112675, | |
| "learning_rate": 8.520256971408452e-08, | |
| "loss": 0.685, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.4534686971235193, | |
| "grad_norm": 3.83257434946296, | |
| "learning_rate": 8.025779637096137e-08, | |
| "loss": 0.6828, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.470389170896785, | |
| "grad_norm": 3.6401708717024146, | |
| "learning_rate": 7.544829088890325e-08, | |
| "loss": 0.6717, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.487309644670051, | |
| "grad_norm": 3.9122520606711917, | |
| "learning_rate": 7.077560319906694e-08, | |
| "loss": 0.6897, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 2.504230118443316, | |
| "grad_norm": 3.9524113141623154, | |
| "learning_rate": 6.624123914114122e-08, | |
| "loss": 0.6848, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.521150592216582, | |
| "grad_norm": 3.670900471362486, | |
| "learning_rate": 6.184665997806831e-08, | |
| "loss": 0.6847, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 2.5380710659898478, | |
| "grad_norm": 3.9559042523399115, | |
| "learning_rate": 5.759328192513074e-08, | |
| "loss": 0.6973, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.5380710659898478, | |
| "eval_loss": 0.7919026017189026, | |
| "eval_runtime": 146.2281, | |
| "eval_samples_per_second": 57.444, | |
| "eval_steps_per_second": 0.903, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.5549915397631136, | |
| "grad_norm": 3.8423336960320538, | |
| "learning_rate": 5.348247569355735e-08, | |
| "loss": 0.6949, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 2.571912013536379, | |
| "grad_norm": 3.7975655402725943, | |
| "learning_rate": 4.951556604879048e-08, | |
| "loss": 0.6866, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.5888324873096447, | |
| "grad_norm": 3.568623425566606, | |
| "learning_rate": 4.569383138356275e-08, | |
| "loss": 0.683, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 2.6057529610829104, | |
| "grad_norm": 3.776851694466855, | |
| "learning_rate": 4.201850330591677e-08, | |
| "loss": 0.6749, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.6226734348561758, | |
| "grad_norm": 3.7650644536804894, | |
| "learning_rate": 3.8490766242301353e-08, | |
| "loss": 0.6824, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.6395939086294415, | |
| "grad_norm": 4.1315575147110515, | |
| "learning_rate": 3.5111757055874326e-08, | |
| "loss": 0.6915, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.6565143824027073, | |
| "grad_norm": 3.945733483403889, | |
| "learning_rate": 3.188256468013139e-08, | |
| "loss": 0.6948, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 2.673434856175973, | |
| "grad_norm": 4.0702928933504445, | |
| "learning_rate": 2.8804229767982636e-08, | |
| "loss": 0.7135, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.6903553299492384, | |
| "grad_norm": 4.049036919724828, | |
| "learning_rate": 2.587774435638679e-08, | |
| "loss": 0.7004, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "grad_norm": 3.686063905885555, | |
| "learning_rate": 2.3104051546654013e-08, | |
| "loss": 0.6804, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "eval_loss": 0.7912485599517822, | |
| "eval_runtime": 146.0875, | |
| "eval_samples_per_second": 57.5, | |
| "eval_steps_per_second": 0.904, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.72419627749577, | |
| "grad_norm": 3.7450509400766707, | |
| "learning_rate": 2.048404520051722e-08, | |
| "loss": 0.6934, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 2.7411167512690353, | |
| "grad_norm": 3.7829378959141673, | |
| "learning_rate": 1.8018569652073378e-08, | |
| "loss": 0.66, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.758037225042301, | |
| "grad_norm": 3.790421803380348, | |
| "learning_rate": 1.570841943568446e-08, | |
| "loss": 0.6893, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 2.774957698815567, | |
| "grad_norm": 3.724677293686406, | |
| "learning_rate": 1.3554339029927531e-08, | |
| "loss": 0.6814, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.7918781725888326, | |
| "grad_norm": 3.9688985767320704, | |
| "learning_rate": 1.1557022617676216e-08, | |
| "loss": 0.673, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.808798646362098, | |
| "grad_norm": 4.22397815551477, | |
| "learning_rate": 9.717113862389992e-09, | |
| "loss": 0.689, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.8257191201353637, | |
| "grad_norm": 3.740661431646974, | |
| "learning_rate": 8.035205700685165e-09, | |
| "loss": 0.6853, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 2.8426395939086295, | |
| "grad_norm": 4.039653126712931, | |
| "learning_rate": 6.511840151252168e-09, | |
| "loss": 0.6823, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.859560067681895, | |
| "grad_norm": 3.665159603413248, | |
| "learning_rate": 5.147508140182555e-09, | |
| "loss": 0.6717, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 2.8764805414551606, | |
| "grad_norm": 3.7704278001935525, | |
| "learning_rate": 3.9426493427611175e-09, | |
| "loss": 0.6684, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.8764805414551606, | |
| "eval_loss": 0.791114091873169, | |
| "eval_runtime": 146.2505, | |
| "eval_samples_per_second": 57.436, | |
| "eval_steps_per_second": 0.903, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.8934010152284264, | |
| "grad_norm": 3.6709944458168975, | |
| "learning_rate": 2.897652041774279e-09, | |
| "loss": 0.6849, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 2.910321489001692, | |
| "grad_norm": 3.8441912749301244, | |
| "learning_rate": 2.0128530023804656e-09, | |
| "loss": 0.6833, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.927241962774958, | |
| "grad_norm": 3.8325829088626713, | |
| "learning_rate": 1.2885373635829754e-09, | |
| "loss": 0.691, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 2.9441624365482233, | |
| "grad_norm": 3.6989879423314207, | |
| "learning_rate": 7.249385463395374e-10, | |
| "loss": 0.6847, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.961082910321489, | |
| "grad_norm": 3.697758295388761, | |
| "learning_rate": 3.22238178339318e-10, | |
| "loss": 0.6721, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.9780033840947544, | |
| "grad_norm": 4.082760711386711, | |
| "learning_rate": 8.056603547090812e-11, | |
| "loss": 0.6798, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.99492385786802, | |
| "grad_norm": 4.0230745119972795, | |
| "learning_rate": 0.0, | |
| "loss": 0.7023, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 2.99492385786802, | |
| "step": 885, | |
| "total_flos": 5218127163949056.0, | |
| "train_loss": 0.7775604002893308, | |
| "train_runtime": 14098.6672, | |
| "train_samples_per_second": 16.086, | |
| "train_steps_per_second": 0.063 | |
| } | |
| ], | |
| "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 | |
| } | |