| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.999438727782975, | |
| "eval_steps": 500, | |
| "global_step": 1002, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.029934518241347054, | |
| "grad_norm": 1.21135244631254, | |
| "learning_rate": 8e-06, | |
| "loss": 0.7943, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05986903648269411, | |
| "grad_norm": 0.9257888985815798, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6917, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08980355472404115, | |
| "grad_norm": 0.7841012581900926, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6676, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11973807296538821, | |
| "grad_norm": 0.8186897658494727, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6642, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14967259120673526, | |
| "grad_norm": 0.9640808091551717, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6563, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1796071094480823, | |
| "grad_norm": 0.6573771107768287, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6486, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.20954162768942938, | |
| "grad_norm": 0.7451852720331203, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6459, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.23947614593077643, | |
| "grad_norm": 0.6760171899376464, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6469, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2694106641721235, | |
| "grad_norm": 0.9185224784748138, | |
| "learning_rate": 8e-06, | |
| "loss": 0.638, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2993451824134705, | |
| "grad_norm": 0.8310714141547997, | |
| "learning_rate": 8e-06, | |
| "loss": 0.64, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3292797006548176, | |
| "grad_norm": 0.6968599344480478, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6357, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3592142188961646, | |
| "grad_norm": 0.7783107345226523, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6344, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3891487371375117, | |
| "grad_norm": 0.7320920547129813, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6347, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.41908325537885877, | |
| "grad_norm": 0.8310187430809345, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6284, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4490177736202058, | |
| "grad_norm": 0.742852301838313, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6305, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.47895229186155286, | |
| "grad_norm": 0.7072242715008016, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6294, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5088868101028999, | |
| "grad_norm": 0.698214271861064, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6239, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.538821328344247, | |
| "grad_norm": 0.6662946690363835, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6253, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.568755846585594, | |
| "grad_norm": 0.6538562990065714, | |
| "learning_rate": 8e-06, | |
| "loss": 0.623, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.598690364826941, | |
| "grad_norm": 0.7549312483543436, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6308, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6286248830682881, | |
| "grad_norm": 0.806924453966631, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6247, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6585594013096352, | |
| "grad_norm": 0.8525309679457572, | |
| "learning_rate": 8e-06, | |
| "loss": 0.623, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6884939195509823, | |
| "grad_norm": 0.7201883248635861, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6257, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7184284377923292, | |
| "grad_norm": 0.6818329042197464, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6247, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7483629560336763, | |
| "grad_norm": 0.6151687407500573, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6212, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7782974742750234, | |
| "grad_norm": 0.6622047387016661, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6198, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8082319925163705, | |
| "grad_norm": 0.661479958293304, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6272, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8381665107577175, | |
| "grad_norm": 0.7220669462375645, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6174, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8681010289990645, | |
| "grad_norm": 0.6245633494447357, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6142, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.8980355472404116, | |
| "grad_norm": 0.7369753450517131, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6155, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9279700654817586, | |
| "grad_norm": 0.684240331799871, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6142, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9579045837231057, | |
| "grad_norm": 0.6592793440033525, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6165, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9878391019644528, | |
| "grad_norm": 0.7596640728440148, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6094, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9998129092609915, | |
| "eval_loss": 0.6207359433174133, | |
| "eval_runtime": 516.5507, | |
| "eval_samples_per_second": 17.425, | |
| "eval_steps_per_second": 0.546, | |
| "step": 334 | |
| }, | |
| { | |
| "epoch": 1.0177736202057999, | |
| "grad_norm": 1.0914053476567482, | |
| "learning_rate": 8e-06, | |
| "loss": 0.627, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.047708138447147, | |
| "grad_norm": 0.9628992294558122, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5324, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.077642656688494, | |
| "grad_norm": 0.7102957206524368, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5276, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1075771749298409, | |
| "grad_norm": 0.6396993471164553, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5282, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.137511693171188, | |
| "grad_norm": 0.6680934494961647, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5287, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.167446211412535, | |
| "grad_norm": 0.6517910836945853, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5311, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.197380729653882, | |
| "grad_norm": 0.6720240409719913, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5318, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2273152478952292, | |
| "grad_norm": 0.6803368260593882, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5347, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2572497661365762, | |
| "grad_norm": 0.6884652022904354, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5383, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2871842843779233, | |
| "grad_norm": 0.7329284898776822, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5338, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3171188026192704, | |
| "grad_norm": 0.6392516489495297, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5359, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3470533208606175, | |
| "grad_norm": 0.6262376414200107, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5306, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.3769878391019645, | |
| "grad_norm": 0.6361024950573079, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5411, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4069223573433116, | |
| "grad_norm": 0.6335891265421159, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5407, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4368568755846587, | |
| "grad_norm": 0.6848282816282614, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5428, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.4667913938260055, | |
| "grad_norm": 0.6098388511595073, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5424, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.4967259120673526, | |
| "grad_norm": 0.6168997234574297, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5386, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5266604303086997, | |
| "grad_norm": 0.6272985821202596, | |
| "learning_rate": 8e-06, | |
| "loss": 0.542, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5565949485500468, | |
| "grad_norm": 0.6696253220648498, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5402, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.5865294667913938, | |
| "grad_norm": 0.6447015112747386, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5408, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.616463985032741, | |
| "grad_norm": 0.6296023955791409, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5449, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.646398503274088, | |
| "grad_norm": 0.6086610727578513, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5442, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6763330215154348, | |
| "grad_norm": 0.6145121966059982, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5352, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.706267539756782, | |
| "grad_norm": 0.6924376605157173, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5371, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.736202057998129, | |
| "grad_norm": 0.6574403524622597, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5387, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.766136576239476, | |
| "grad_norm": 0.6340547035740907, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5417, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.7960710944808231, | |
| "grad_norm": 0.6875606923110796, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5422, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8260056127221702, | |
| "grad_norm": 0.601235347108058, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5404, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.8559401309635173, | |
| "grad_norm": 0.6002851968034844, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5388, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.8858746492048644, | |
| "grad_norm": 0.6476875772663511, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5402, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9158091674462114, | |
| "grad_norm": 0.6584929467438523, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5419, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.9457436856875585, | |
| "grad_norm": 0.635864107312744, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5405, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.9756782039289056, | |
| "grad_norm": 0.6971331062500022, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5452, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.999625818521983, | |
| "eval_loss": 0.6230265498161316, | |
| "eval_runtime": 517.4275, | |
| "eval_samples_per_second": 17.396, | |
| "eval_steps_per_second": 0.545, | |
| "step": 668 | |
| }, | |
| { | |
| "epoch": 2.0056127221702527, | |
| "grad_norm": 1.1175717427173644, | |
| "learning_rate": 8e-06, | |
| "loss": 0.5812, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.0355472404115997, | |
| "grad_norm": 0.9014193389938453, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4457, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.065481758652947, | |
| "grad_norm": 0.7610276344475498, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4401, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.095416276894294, | |
| "grad_norm": 0.7005120545670707, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4442, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.125350795135641, | |
| "grad_norm": 0.694947791996958, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4449, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.155285313376988, | |
| "grad_norm": 0.8511316426645517, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4488, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.185219831618335, | |
| "grad_norm": 0.8835380141763632, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4493, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.2151543498596817, | |
| "grad_norm": 0.711769327296804, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4504, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.245088868101029, | |
| "grad_norm": 0.8058645279970468, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4527, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.275023386342376, | |
| "grad_norm": 0.6752264716430066, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4497, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.304957904583723, | |
| "grad_norm": 0.7456444513922865, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4543, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.33489242282507, | |
| "grad_norm": 0.6829933479220452, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4517, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.364826941066417, | |
| "grad_norm": 0.6401921390853881, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4542, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.394761459307764, | |
| "grad_norm": 0.7127587062318401, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4565, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4246959775491113, | |
| "grad_norm": 0.7066267427455665, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4632, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.4546304957904583, | |
| "grad_norm": 0.7226062602383946, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4611, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.4845650140318054, | |
| "grad_norm": 0.7793500688968047, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4613, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5144995322731525, | |
| "grad_norm": 0.6820846969489297, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4591, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.5444340505144996, | |
| "grad_norm": 0.6470236236262259, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4629, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.5743685687558466, | |
| "grad_norm": 0.706376237919755, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4607, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.6043030869971937, | |
| "grad_norm": 0.6665109583977343, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4596, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.634237605238541, | |
| "grad_norm": 0.6583544195294855, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4571, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.664172123479888, | |
| "grad_norm": 0.6766815167737668, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4634, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.694106641721235, | |
| "grad_norm": 0.6485676514451294, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4661, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.724041159962582, | |
| "grad_norm": 0.6787774852448176, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4605, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.753975678203929, | |
| "grad_norm": 0.6424941259370515, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4619, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.7839101964452757, | |
| "grad_norm": 0.6314188313732944, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4676, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.8138447146866232, | |
| "grad_norm": 0.654263916123419, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4651, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.84377923292797, | |
| "grad_norm": 0.7342445113024143, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4657, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.8737137511693174, | |
| "grad_norm": 0.749469773156686, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4674, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.903648269410664, | |
| "grad_norm": 0.6461650603511763, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4638, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.933582787652011, | |
| "grad_norm": 0.6858936516605968, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4688, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.963517305893358, | |
| "grad_norm": 0.6772228156335088, | |
| "learning_rate": 8e-06, | |
| "loss": 0.4707, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.9934518241347052, | |
| "grad_norm": 0.677502927868203, | |
| "learning_rate": 8e-06, | |
| "loss": 0.472, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.999438727782975, | |
| "eval_loss": 0.6610371470451355, | |
| "eval_runtime": 518.6132, | |
| "eval_samples_per_second": 17.356, | |
| "eval_steps_per_second": 0.544, | |
| "step": 1002 | |
| }, | |
| { | |
| "epoch": 2.999438727782975, | |
| "step": 1002, | |
| "total_flos": 3818092983484416.0, | |
| "train_loss": 0.545673328840328, | |
| "train_runtime": 90899.0902, | |
| "train_samples_per_second": 5.644, | |
| "train_steps_per_second": 0.011 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1002, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3818092983484416.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |