| { | |
| "best_metric": 1.4004004001617432, | |
| "best_model_checkpoint": "finetuned_student_model/checkpoint-900", | |
| "epoch": 1.9969604863221884, | |
| "eval_steps": 100, | |
| "global_step": 904, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.022105554020447636, | |
| "grad_norm": 9.875, | |
| "learning_rate": 0.0002973392461197339, | |
| "loss": 3.2514, | |
| "mean_token_accuracy": 0.4204087435267866, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04421110804089527, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 0.0002940133037694013, | |
| "loss": 2.3718, | |
| "mean_token_accuracy": 0.5071762848645449, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06631666206134292, | |
| "grad_norm": 3.65625, | |
| "learning_rate": 0.0002906873614190687, | |
| "loss": 2.1087, | |
| "mean_token_accuracy": 0.5472177878022194, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08842221608179054, | |
| "grad_norm": 2.9375, | |
| "learning_rate": 0.0002873614190687361, | |
| "loss": 2.0888, | |
| "mean_token_accuracy": 0.5551092017441988, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11052777010223819, | |
| "grad_norm": 2.9375, | |
| "learning_rate": 0.0002840354767184035, | |
| "loss": 2.1327, | |
| "mean_token_accuracy": 0.5446394145488739, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13263332412268583, | |
| "grad_norm": 2.78125, | |
| "learning_rate": 0.0002807095343680709, | |
| "loss": 2.0511, | |
| "mean_token_accuracy": 0.5573025777935982, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15473887814313347, | |
| "grad_norm": 2.921875, | |
| "learning_rate": 0.0002773835920177383, | |
| "loss": 2.0515, | |
| "mean_token_accuracy": 0.5619197305291891, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1768444321635811, | |
| "grad_norm": 2.609375, | |
| "learning_rate": 0.0002740576496674058, | |
| "loss": 1.9854, | |
| "mean_token_accuracy": 0.5682128138840199, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19894998618402873, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 0.00027073170731707315, | |
| "loss": 1.9689, | |
| "mean_token_accuracy": 0.5708833433687687, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.22105554020447638, | |
| "grad_norm": 2.78125, | |
| "learning_rate": 0.0002674057649667406, | |
| "loss": 1.9959, | |
| "mean_token_accuracy": 0.5728900354355574, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22105554020447638, | |
| "eval_loss": 1.9453095197677612, | |
| "eval_mean_token_accuracy": 0.5785076600132566, | |
| "eval_runtime": 23.0365, | |
| "eval_samples_per_second": 34.293, | |
| "eval_steps_per_second": 4.298, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24316109422492402, | |
| "grad_norm": 2.5, | |
| "learning_rate": 0.00026407982261640795, | |
| "loss": 1.967, | |
| "mean_token_accuracy": 0.5708974912762642, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.26526664824537166, | |
| "grad_norm": 2.6875, | |
| "learning_rate": 0.0002607538802660754, | |
| "loss": 1.9176, | |
| "mean_token_accuracy": 0.5783881578594446, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2873722022658193, | |
| "grad_norm": 2.3125, | |
| "learning_rate": 0.00025742793791574275, | |
| "loss": 1.8481, | |
| "mean_token_accuracy": 0.5897452015429735, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.30947775628626695, | |
| "grad_norm": 2.484375, | |
| "learning_rate": 0.0002541019955654102, | |
| "loss": 1.8706, | |
| "mean_token_accuracy": 0.5862758502364158, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.33158331030671456, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 0.0002507760532150776, | |
| "loss": 1.8667, | |
| "mean_token_accuracy": 0.5883878566324711, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3536888643271622, | |
| "grad_norm": 2.421875, | |
| "learning_rate": 0.000247450110864745, | |
| "loss": 1.8101, | |
| "mean_token_accuracy": 0.5959903877228498, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.37579441834760985, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 0.00024412416851441238, | |
| "loss": 1.8353, | |
| "mean_token_accuracy": 0.5912164930254221, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.39789997236805746, | |
| "grad_norm": 2.25, | |
| "learning_rate": 0.00024079822616407978, | |
| "loss": 1.7821, | |
| "mean_token_accuracy": 0.6001696892082691, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.42000552638850513, | |
| "grad_norm": 2.21875, | |
| "learning_rate": 0.0002374722838137472, | |
| "loss": 1.8351, | |
| "mean_token_accuracy": 0.5922294020652771, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.44211108040895275, | |
| "grad_norm": 2.296875, | |
| "learning_rate": 0.0002341463414634146, | |
| "loss": 1.7342, | |
| "mean_token_accuracy": 0.6008941765874625, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.44211108040895275, | |
| "eval_loss": 1.7812447547912598, | |
| "eval_mean_token_accuracy": 0.6040364732645979, | |
| "eval_runtime": 21.5006, | |
| "eval_samples_per_second": 36.743, | |
| "eval_steps_per_second": 4.605, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.46421663442940037, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 0.000230820399113082, | |
| "loss": 1.7778, | |
| "mean_token_accuracy": 0.6000470589846373, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.48632218844984804, | |
| "grad_norm": 1.890625, | |
| "learning_rate": 0.0002274944567627494, | |
| "loss": 1.7515, | |
| "mean_token_accuracy": 0.6070282235741615, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5084277424702957, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 0.0002241685144124168, | |
| "loss": 1.7348, | |
| "mean_token_accuracy": 0.6033936321735383, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5305332964907433, | |
| "grad_norm": 1.8515625, | |
| "learning_rate": 0.0002208425720620842, | |
| "loss": 1.7198, | |
| "mean_token_accuracy": 0.6069234035909176, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5526388505111909, | |
| "grad_norm": 1.984375, | |
| "learning_rate": 0.00021751662971175166, | |
| "loss": 1.6866, | |
| "mean_token_accuracy": 0.6128835912793875, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5747444045316386, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 0.00021419068736141907, | |
| "loss": 1.6751, | |
| "mean_token_accuracy": 0.6139085631817579, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5968499585520862, | |
| "grad_norm": 2.203125, | |
| "learning_rate": 0.00021086474501108647, | |
| "loss": 1.7237, | |
| "mean_token_accuracy": 0.6062099590897561, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6189555125725339, | |
| "grad_norm": 2.09375, | |
| "learning_rate": 0.00020753880266075387, | |
| "loss": 1.6894, | |
| "mean_token_accuracy": 0.6122250266373157, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6410610665929815, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 0.00020421286031042127, | |
| "loss": 1.6325, | |
| "mean_token_accuracy": 0.6204121351242066, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6631666206134291, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 0.0002008869179600887, | |
| "loss": 1.6668, | |
| "mean_token_accuracy": 0.615133136883378, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6631666206134291, | |
| "eval_loss": 1.6475698947906494, | |
| "eval_mean_token_accuracy": 0.6219914049813242, | |
| "eval_runtime": 21.4767, | |
| "eval_samples_per_second": 36.784, | |
| "eval_steps_per_second": 4.61, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6852721746338768, | |
| "grad_norm": 1.8359375, | |
| "learning_rate": 0.0001975609756097561, | |
| "loss": 1.614, | |
| "mean_token_accuracy": 0.6179928559809923, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.7073777286543244, | |
| "grad_norm": 10.6875, | |
| "learning_rate": 0.0001942350332594235, | |
| "loss": 1.7166, | |
| "mean_token_accuracy": 0.6136065050959587, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.729483282674772, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 0.0001909090909090909, | |
| "loss": 1.657, | |
| "mean_token_accuracy": 0.6190799050033092, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7515888366952197, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 0.0001875831485587583, | |
| "loss": 1.6038, | |
| "mean_token_accuracy": 0.6287192285060883, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "grad_norm": 1.71875, | |
| "learning_rate": 0.0001842572062084257, | |
| "loss": 1.5914, | |
| "mean_token_accuracy": 0.6284744247794152, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7957999447361149, | |
| "grad_norm": 2.0, | |
| "learning_rate": 0.00018093126385809312, | |
| "loss": 1.6075, | |
| "mean_token_accuracy": 0.6283752031624317, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8179054987565626, | |
| "grad_norm": 2.0, | |
| "learning_rate": 0.00017760532150776052, | |
| "loss": 1.6082, | |
| "mean_token_accuracy": 0.6261149801313877, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8400110527770103, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 0.00017427937915742792, | |
| "loss": 1.5879, | |
| "mean_token_accuracy": 0.6261836618185044, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8621166067974578, | |
| "grad_norm": 2.515625, | |
| "learning_rate": 0.00017095343680709532, | |
| "loss": 1.5785, | |
| "mean_token_accuracy": 0.631008780002594, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8842221608179055, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 0.00016762749445676272, | |
| "loss": 1.4931, | |
| "mean_token_accuracy": 0.6441986732184887, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8842221608179055, | |
| "eval_loss": 1.5408236980438232, | |
| "eval_mean_token_accuracy": 0.6406004489070237, | |
| "eval_runtime": 21.4648, | |
| "eval_samples_per_second": 36.804, | |
| "eval_steps_per_second": 4.612, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9063277148383532, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 0.00016430155210643015, | |
| "loss": 1.5441, | |
| "mean_token_accuracy": 0.6400343291461468, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.9284332688588007, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 0.00016097560975609755, | |
| "loss": 1.5299, | |
| "mean_token_accuracy": 0.6389802560210228, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9505388228792484, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 0.00015764966740576495, | |
| "loss": 1.524, | |
| "mean_token_accuracy": 0.6369420018047094, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9726443768996961, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 0.00015432372505543235, | |
| "loss": 1.5423, | |
| "mean_token_accuracy": 0.6327314972877502, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.9947499309201436, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 0.00015099778270509975, | |
| "loss": 1.5033, | |
| "mean_token_accuracy": 0.6431983485817909, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.0154738878143132, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 0.00014767184035476718, | |
| "loss": 1.2191, | |
| "mean_token_accuracy": 0.6965674503644308, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.037579441834761, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 0.00014434589800443458, | |
| "loss": 1.0348, | |
| "mean_token_accuracy": 0.7245359934866429, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.0596849958552086, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 0.00014101995565410198, | |
| "loss": 1.0563, | |
| "mean_token_accuracy": 0.7208410792052746, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.0817905498756564, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 0.00013769401330376938, | |
| "loss": 1.0467, | |
| "mean_token_accuracy": 0.7243687815964222, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.103896103896104, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 0.00013436807095343678, | |
| "loss": 1.048, | |
| "mean_token_accuracy": 0.718181136995554, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.103896103896104, | |
| "eval_loss": 1.5222158432006836, | |
| "eval_mean_token_accuracy": 0.6474175338793282, | |
| "eval_runtime": 21.4789, | |
| "eval_samples_per_second": 36.78, | |
| "eval_steps_per_second": 4.609, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.1260016579165515, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 0.0001310421286031042, | |
| "loss": 1.0292, | |
| "mean_token_accuracy": 0.724801865965128, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.148107211936999, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 0.0001277161862527716, | |
| "loss": 1.0431, | |
| "mean_token_accuracy": 0.7216434337198734, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.1702127659574468, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 0.000124390243902439, | |
| "loss": 1.0261, | |
| "mean_token_accuracy": 0.7233444675803185, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.1923183199778944, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 0.00012106430155210642, | |
| "loss": 1.0001, | |
| "mean_token_accuracy": 0.7310704313218593, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.2144238739983422, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 0.00011773835920177382, | |
| "loss": 0.9996, | |
| "mean_token_accuracy": 0.7315604917705059, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.2365294280187897, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 0.00011441241685144122, | |
| "loss": 0.9732, | |
| "mean_token_accuracy": 0.7363451808691025, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.2586349820392373, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 0.00011108647450110864, | |
| "loss": 1.0185, | |
| "mean_token_accuracy": 0.7294765569269657, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.280740536059685, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 0.00010776053215077604, | |
| "loss": 0.9948, | |
| "mean_token_accuracy": 0.7329611636698246, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.3028460900801326, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 0.00010443458980044345, | |
| "loss": 0.9808, | |
| "mean_token_accuracy": 0.7348393484950065, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.3249516441005802, | |
| "grad_norm": 1.6875, | |
| "learning_rate": 0.00010110864745011085, | |
| "loss": 0.9809, | |
| "mean_token_accuracy": 0.7370587438344955, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3249516441005802, | |
| "eval_loss": 1.4869917631149292, | |
| "eval_mean_token_accuracy": 0.6575891068487456, | |
| "eval_runtime": 21.4699, | |
| "eval_samples_per_second": 36.796, | |
| "eval_steps_per_second": 4.611, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.347057198121028, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 9.778270509977825e-05, | |
| "loss": 0.9891, | |
| "mean_token_accuracy": 0.7354318417608738, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.3691627521414755, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 9.445676274944568e-05, | |
| "loss": 0.9653, | |
| "mean_token_accuracy": 0.7393273778259755, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.391268306161923, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 9.113082039911308e-05, | |
| "loss": 0.9552, | |
| "mean_token_accuracy": 0.7432896822690964, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.4133738601823709, | |
| "grad_norm": 1.6015625, | |
| "learning_rate": 8.780487804878047e-05, | |
| "loss": 0.9713, | |
| "mean_token_accuracy": 0.7419402815401555, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.4354794142028184, | |
| "grad_norm": 1.640625, | |
| "learning_rate": 8.44789356984479e-05, | |
| "loss": 0.9548, | |
| "mean_token_accuracy": 0.7421382986009121, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.4575849682232662, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 8.11529933481153e-05, | |
| "loss": 0.9824, | |
| "mean_token_accuracy": 0.737388264387846, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.4796905222437138, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 7.78270509977827e-05, | |
| "loss": 0.937, | |
| "mean_token_accuracy": 0.7477126508951187, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.5017960762641613, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 7.450110864745011e-05, | |
| "loss": 0.9433, | |
| "mean_token_accuracy": 0.744731155782938, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.523901630284609, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 7.117516629711751e-05, | |
| "loss": 0.9259, | |
| "mean_token_accuracy": 0.745546705648303, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.5460071843050567, | |
| "grad_norm": 1.625, | |
| "learning_rate": 6.784922394678491e-05, | |
| "loss": 0.9625, | |
| "mean_token_accuracy": 0.7406142510473728, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5460071843050567, | |
| "eval_loss": 1.429708480834961, | |
| "eval_mean_token_accuracy": 0.6674101424939705, | |
| "eval_runtime": 21.4679, | |
| "eval_samples_per_second": 36.799, | |
| "eval_steps_per_second": 4.612, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5681127383255042, | |
| "grad_norm": 1.75, | |
| "learning_rate": 6.452328159645232e-05, | |
| "loss": 0.9285, | |
| "mean_token_accuracy": 0.7476192332804203, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.590218292345952, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 6.119733924611973e-05, | |
| "loss": 0.9173, | |
| "mean_token_accuracy": 0.75327173396945, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.6123238463663996, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 5.787139689578713e-05, | |
| "loss": 0.9392, | |
| "mean_token_accuracy": 0.7470508739352226, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.6344294003868471, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 5.454545454545454e-05, | |
| "loss": 0.9148, | |
| "mean_token_accuracy": 0.7503846064209938, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.6565349544072947, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 5.121951219512195e-05, | |
| "loss": 0.9286, | |
| "mean_token_accuracy": 0.749035281687975, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.6786405084277425, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 4.7893569844789354e-05, | |
| "loss": 0.9312, | |
| "mean_token_accuracy": 0.748369749635458, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.7007460624481903, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 4.456762749445676e-05, | |
| "loss": 0.9585, | |
| "mean_token_accuracy": 0.742963894456625, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.7228516164686378, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 4.124168514412417e-05, | |
| "loss": 1.0002, | |
| "mean_token_accuracy": 0.7539634991437196, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.7449571704890854, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 3.791574279379157e-05, | |
| "loss": 0.93, | |
| "mean_token_accuracy": 0.7481286890804768, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.767062724509533, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 3.4589800443458975e-05, | |
| "loss": 0.9119, | |
| "mean_token_accuracy": 0.7515073113143445, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.767062724509533, | |
| "eval_loss": 1.4062376022338867, | |
| "eval_mean_token_accuracy": 0.6735276628022242, | |
| "eval_runtime": 21.4799, | |
| "eval_samples_per_second": 36.779, | |
| "eval_steps_per_second": 4.609, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7891682785299805, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 3.126385809312638e-05, | |
| "loss": 0.9018, | |
| "mean_token_accuracy": 0.7531212449073792, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.8112738325504283, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 2.793791574279379e-05, | |
| "loss": 0.9048, | |
| "mean_token_accuracy": 0.7560021050274373, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.833379386570876, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 2.4611973392461197e-05, | |
| "loss": 0.9028, | |
| "mean_token_accuracy": 0.7558302395045757, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.8554849405913236, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 2.12860310421286e-05, | |
| "loss": 0.9158, | |
| "mean_token_accuracy": 0.7508154392242432, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.8775904946117712, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 1.7960088691796008e-05, | |
| "loss": 0.8864, | |
| "mean_token_accuracy": 0.7576524101197719, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.8996960486322187, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 1.4634146341463413e-05, | |
| "loss": 0.9228, | |
| "mean_token_accuracy": 0.7508851245045662, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.9218016026526665, | |
| "grad_norm": 1.6015625, | |
| "learning_rate": 1.130820399113082e-05, | |
| "loss": 0.8975, | |
| "mean_token_accuracy": 0.7541297495365142, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.943907156673114, | |
| "grad_norm": 1.75, | |
| "learning_rate": 7.982261640798226e-06, | |
| "loss": 0.8971, | |
| "mean_token_accuracy": 0.7586074694991112, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.9660127106935619, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 4.656319290465632e-06, | |
| "loss": 0.9251, | |
| "mean_token_accuracy": 0.7490280956029892, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.9881182647140094, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 1.3303769401330375e-06, | |
| "loss": 0.9005, | |
| "mean_token_accuracy": 0.7555492661893368, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9881182647140094, | |
| "eval_loss": 1.4004004001617432, | |
| "eval_mean_token_accuracy": 0.6744482210188201, | |
| "eval_runtime": 21.4879, | |
| "eval_samples_per_second": 36.765, | |
| "eval_steps_per_second": 4.607, | |
| "step": 900 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 904, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7777625251882496e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |