| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 1847, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005414185165132647, | |
| "grad_norm": 4.75, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.8236, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.010828370330265295, | |
| "grad_norm": 4.59375, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.8484, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.016242555495397944, | |
| "grad_norm": 5.21875, | |
| "learning_rate": 6e-06, | |
| "loss": 0.8014, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02165674066053059, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.779, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02707092582566324, | |
| "grad_norm": 2.875, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7153, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03248511099079589, | |
| "grad_norm": 2.328125, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.6692, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03789929615592853, | |
| "grad_norm": 2.359375, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.672, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.04331348132106118, | |
| "grad_norm": 1.9765625, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.6071, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.04872766648619383, | |
| "grad_norm": 1.8046875, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.5757, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05414185165132648, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5318, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05414185165132648, | |
| "eval_loss": 0.4426738917827606, | |
| "eval_runtime": 58.7167, | |
| "eval_samples_per_second": 8.515, | |
| "eval_steps_per_second": 0.545, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05955603681645912, | |
| "grad_norm": 3.078125, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.5026, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.06497022198159177, | |
| "grad_norm": 3.234375, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.4925, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.07038440714672442, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.4557, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.07579859231185707, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.4558, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.08121277747698971, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 3e-05, | |
| "loss": 0.426, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.08662696264212236, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.3799, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.092041147807255, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.3866, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.09745533297238766, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.3668, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.10286951813752031, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.3396, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.10828370330265295, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3133, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.10828370330265295, | |
| "eval_loss": 0.23219549655914307, | |
| "eval_runtime": 60.3361, | |
| "eval_samples_per_second": 8.287, | |
| "eval_steps_per_second": 0.53, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1136978884677856, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2862, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.11911207363291824, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.2726, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.12452625879805089, | |
| "grad_norm": 0.9765625, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.2823, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.12994044396318355, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.2724, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1353546291283162, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2563, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.14076881429344884, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 4.9686912961803384e-05, | |
| "loss": 0.2243, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1461829994585815, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 4.9373825923606765e-05, | |
| "loss": 0.2297, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.15159718462371413, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 4.906073888541015e-05, | |
| "loss": 0.207, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.15701136978884678, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 4.874765184721353e-05, | |
| "loss": 0.2255, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.16242555495397942, | |
| "grad_norm": 1.1953125, | |
| "learning_rate": 4.843456480901691e-05, | |
| "loss": 0.1946, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.16242555495397942, | |
| "eval_loss": 0.1316749006509781, | |
| "eval_runtime": 58.9058, | |
| "eval_samples_per_second": 8.488, | |
| "eval_steps_per_second": 0.543, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.16783974011911207, | |
| "grad_norm": 1.8671875, | |
| "learning_rate": 4.812147777082029e-05, | |
| "loss": 0.2035, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.17325392528424471, | |
| "grad_norm": 1.40625, | |
| "learning_rate": 4.780839073262367e-05, | |
| "loss": 0.1852, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.17866811044937736, | |
| "grad_norm": 1.375, | |
| "learning_rate": 4.7495303694427054e-05, | |
| "loss": 0.2054, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.18408229561451, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 4.7182216656230435e-05, | |
| "loss": 0.1803, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.18949648077964265, | |
| "grad_norm": 1.6875, | |
| "learning_rate": 4.6869129618033816e-05, | |
| "loss": 0.1993, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.19491066594477532, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 4.65560425798372e-05, | |
| "loss": 0.1954, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.20032485110990797, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 4.624295554164057e-05, | |
| "loss": 0.1802, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.20573903627504062, | |
| "grad_norm": 0.87890625, | |
| "learning_rate": 4.5929868503443954e-05, | |
| "loss": 0.1941, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.21115322144017326, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 4.561678146524734e-05, | |
| "loss": 0.1706, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.2165674066053059, | |
| "grad_norm": 0.9765625, | |
| "learning_rate": 4.5303694427050724e-05, | |
| "loss": 0.1845, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2165674066053059, | |
| "eval_loss": 0.11701580137014389, | |
| "eval_runtime": 58.716, | |
| "eval_samples_per_second": 8.516, | |
| "eval_steps_per_second": 0.545, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.22198159177043855, | |
| "grad_norm": 1.5078125, | |
| "learning_rate": 4.4990607388854105e-05, | |
| "loss": 0.1386, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.2273957769355712, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 4.4677520350657486e-05, | |
| "loss": 0.1599, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.23280996210070384, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 4.436443331246087e-05, | |
| "loss": 0.1856, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.2382241472658365, | |
| "grad_norm": 0.8203125, | |
| "learning_rate": 4.405134627426425e-05, | |
| "loss": 0.1766, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.24363833243096913, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 4.373825923606763e-05, | |
| "loss": 0.1448, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.24905251759610178, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 4.342517219787101e-05, | |
| "loss": 0.1838, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.25446670276123445, | |
| "grad_norm": 0.87109375, | |
| "learning_rate": 4.3112085159674393e-05, | |
| "loss": 0.1698, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.2598808879263671, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 4.2798998121477775e-05, | |
| "loss": 0.1392, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.26529507309149974, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 4.2485911083281156e-05, | |
| "loss": 0.1769, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.2707092582566324, | |
| "grad_norm": 1.25, | |
| "learning_rate": 4.217282404508454e-05, | |
| "loss": 0.1506, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2707092582566324, | |
| "eval_loss": 0.11454860121011734, | |
| "eval_runtime": 58.3011, | |
| "eval_samples_per_second": 8.576, | |
| "eval_steps_per_second": 0.549, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.27612344342176504, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 4.185973700688792e-05, | |
| "loss": 0.1297, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2815376285868977, | |
| "grad_norm": 1.3515625, | |
| "learning_rate": 4.15466499686913e-05, | |
| "loss": 0.1459, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.2869518137520303, | |
| "grad_norm": 1.3515625, | |
| "learning_rate": 4.123356293049468e-05, | |
| "loss": 0.1565, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.292365998917163, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 4.092047589229806e-05, | |
| "loss": 0.1444, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.2977801840822956, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 4.0607388854101445e-05, | |
| "loss": 0.1581, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.30319436924742826, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 4.029430181590482e-05, | |
| "loss": 0.1781, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.3086085544125609, | |
| "grad_norm": 0.8359375, | |
| "learning_rate": 3.99812147777082e-05, | |
| "loss": 0.1333, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.31402273957769355, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 3.966812773951158e-05, | |
| "loss": 0.1254, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.3194369247428262, | |
| "grad_norm": 1.625, | |
| "learning_rate": 3.9355040701314964e-05, | |
| "loss": 0.2092, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.32485110990795885, | |
| "grad_norm": 1.5234375, | |
| "learning_rate": 3.9041953663118345e-05, | |
| "loss": 0.1269, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.32485110990795885, | |
| "eval_loss": 0.1084875538945198, | |
| "eval_runtime": 61.8367, | |
| "eval_samples_per_second": 8.086, | |
| "eval_steps_per_second": 0.517, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.3302652950730915, | |
| "grad_norm": 0.8203125, | |
| "learning_rate": 3.8728866624921726e-05, | |
| "loss": 0.1448, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.33567948023822414, | |
| "grad_norm": 0.90625, | |
| "learning_rate": 3.841577958672511e-05, | |
| "loss": 0.1529, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.3410936654033568, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 3.810269254852849e-05, | |
| "loss": 0.1434, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.34650785056848943, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 3.778960551033187e-05, | |
| "loss": 0.1679, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.3519220357336221, | |
| "grad_norm": 1.1953125, | |
| "learning_rate": 3.747651847213526e-05, | |
| "loss": 0.1832, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.3573362208987547, | |
| "grad_norm": 0.96875, | |
| "learning_rate": 3.716343143393864e-05, | |
| "loss": 0.1634, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.36275040606388737, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 3.685034439574202e-05, | |
| "loss": 0.1799, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.36816459122902, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 3.65372573575454e-05, | |
| "loss": 0.1308, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.37357877639415266, | |
| "grad_norm": 0.71875, | |
| "learning_rate": 3.6224170319348784e-05, | |
| "loss": 0.1282, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.3789929615592853, | |
| "grad_norm": 0.99609375, | |
| "learning_rate": 3.5911083281152166e-05, | |
| "loss": 0.159, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.3789929615592853, | |
| "eval_loss": 0.11033277213573456, | |
| "eval_runtime": 59.5923, | |
| "eval_samples_per_second": 8.39, | |
| "eval_steps_per_second": 0.537, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.38440714672441795, | |
| "grad_norm": 1.9453125, | |
| "learning_rate": 3.559799624295555e-05, | |
| "loss": 0.1718, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.38982133188955065, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 3.528490920475893e-05, | |
| "loss": 0.1584, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.3952355170546833, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 3.497182216656231e-05, | |
| "loss": 0.1556, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.40064970221981594, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 3.4658735128365685e-05, | |
| "loss": 0.1379, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.4060638873849486, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 3.4345648090169066e-05, | |
| "loss": 0.1542, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.41147807255008123, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 3.403256105197245e-05, | |
| "loss": 0.142, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.4168922577152139, | |
| "grad_norm": 0.92578125, | |
| "learning_rate": 3.371947401377583e-05, | |
| "loss": 0.1463, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.4223064428803465, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 3.340638697557921e-05, | |
| "loss": 0.1317, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.42772062804547917, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 3.309329993738259e-05, | |
| "loss": 0.1709, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.4331348132106118, | |
| "grad_norm": 0.9296875, | |
| "learning_rate": 3.278021289918597e-05, | |
| "loss": 0.1518, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.4331348132106118, | |
| "eval_loss": 0.10848626494407654, | |
| "eval_runtime": 58.5847, | |
| "eval_samples_per_second": 8.535, | |
| "eval_steps_per_second": 0.546, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.43854899837574446, | |
| "grad_norm": 0.921875, | |
| "learning_rate": 3.2467125860989355e-05, | |
| "loss": 0.1298, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.4439631835408771, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 3.2154038822792736e-05, | |
| "loss": 0.1291, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.44937736870600975, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 3.184095178459612e-05, | |
| "loss": 0.1715, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.4547915538711424, | |
| "grad_norm": 1.375, | |
| "learning_rate": 3.15278647463995e-05, | |
| "loss": 0.1607, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.46020573903627504, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 3.121477770820288e-05, | |
| "loss": 0.1261, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.4656199242014077, | |
| "grad_norm": 1.3984375, | |
| "learning_rate": 3.090169067000626e-05, | |
| "loss": 0.15, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.47103410936654033, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 3.058860363180964e-05, | |
| "loss": 0.1501, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.476448294531673, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 3.0275516593613024e-05, | |
| "loss": 0.1372, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.4818624796968056, | |
| "grad_norm": 1.0, | |
| "learning_rate": 2.9962429555416406e-05, | |
| "loss": 0.1725, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.48727666486193827, | |
| "grad_norm": 0.9609375, | |
| "learning_rate": 2.9649342517219787e-05, | |
| "loss": 0.1451, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.48727666486193827, | |
| "eval_loss": 0.1076415479183197, | |
| "eval_runtime": 58.8592, | |
| "eval_samples_per_second": 8.495, | |
| "eval_steps_per_second": 0.544, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.4926908500270709, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 2.9336255479023172e-05, | |
| "loss": 0.1423, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.49810503519220356, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 2.9023168440826553e-05, | |
| "loss": 0.1326, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.5035192203573362, | |
| "grad_norm": 0.9921875, | |
| "learning_rate": 2.8710081402629935e-05, | |
| "loss": 0.141, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.5089334055224689, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 2.8396994364433316e-05, | |
| "loss": 0.1311, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.5143475906876015, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 2.8083907326236698e-05, | |
| "loss": 0.1154, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.5197617758527342, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 2.777082028804008e-05, | |
| "loss": 0.1433, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.5251759610178668, | |
| "grad_norm": 1.546875, | |
| "learning_rate": 2.745773324984346e-05, | |
| "loss": 0.1601, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.5305901461829995, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 2.7144646211646842e-05, | |
| "loss": 0.1343, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.5360043313481321, | |
| "grad_norm": 0.953125, | |
| "learning_rate": 2.683155917345022e-05, | |
| "loss": 0.1467, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.5414185165132648, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 2.65184721352536e-05, | |
| "loss": 0.1562, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5414185165132648, | |
| "eval_loss": 0.10366573929786682, | |
| "eval_runtime": 60.3546, | |
| "eval_samples_per_second": 8.284, | |
| "eval_steps_per_second": 0.53, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5468327016783974, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 2.6205385097056983e-05, | |
| "loss": 0.1299, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.5522468868435301, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 2.5892298058860364e-05, | |
| "loss": 0.1386, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.5576610720086627, | |
| "grad_norm": 0.9921875, | |
| "learning_rate": 2.5579211020663746e-05, | |
| "loss": 0.1212, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.5630752571737954, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 2.5266123982467127e-05, | |
| "loss": 0.1251, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.568489442338928, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 2.495303694427051e-05, | |
| "loss": 0.1275, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.5739036275040607, | |
| "grad_norm": 0.66796875, | |
| "learning_rate": 2.463994990607389e-05, | |
| "loss": 0.1098, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.5793178126691932, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 2.432686286787727e-05, | |
| "loss": 0.1297, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.584731997834326, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 2.4013775829680653e-05, | |
| "loss": 0.1487, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.5901461829994585, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 2.3700688791484034e-05, | |
| "loss": 0.1286, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.5955603681645912, | |
| "grad_norm": 0.875, | |
| "learning_rate": 2.3387601753287412e-05, | |
| "loss": 0.1456, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.5955603681645912, | |
| "eval_loss": 0.09888758510351181, | |
| "eval_runtime": 62.4618, | |
| "eval_samples_per_second": 8.005, | |
| "eval_steps_per_second": 0.512, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.6009745533297238, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 2.3074514715090797e-05, | |
| "loss": 0.1473, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.6063887384948565, | |
| "grad_norm": 1.1953125, | |
| "learning_rate": 2.2761427676894178e-05, | |
| "loss": 0.155, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.6118029236599891, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 2.244834063869756e-05, | |
| "loss": 0.1251, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.6172171088251218, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 2.213525360050094e-05, | |
| "loss": 0.1495, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.6226312939902545, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 2.1822166562304323e-05, | |
| "loss": 0.142, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.6280454791553871, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 2.1509079524107704e-05, | |
| "loss": 0.1303, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.6334596643205198, | |
| "grad_norm": 0.9453125, | |
| "learning_rate": 2.1195992485911085e-05, | |
| "loss": 0.1263, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.6388738494856524, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 2.0882905447714467e-05, | |
| "loss": 0.1133, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.6442880346507851, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 2.0569818409517845e-05, | |
| "loss": 0.1411, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.6497022198159177, | |
| "grad_norm": 0.90234375, | |
| "learning_rate": 2.0256731371321226e-05, | |
| "loss": 0.1571, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.6497022198159177, | |
| "eval_loss": 0.09719575196504593, | |
| "eval_runtime": 61.5385, | |
| "eval_samples_per_second": 8.125, | |
| "eval_steps_per_second": 0.52, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.6551164049810504, | |
| "grad_norm": 0.6328125, | |
| "learning_rate": 1.9943644333124608e-05, | |
| "loss": 0.1521, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.660530590146183, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 1.963055729492799e-05, | |
| "loss": 0.1128, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.6659447753113157, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 1.931747025673137e-05, | |
| "loss": 0.134, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.6713589604764483, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 1.9004383218534755e-05, | |
| "loss": 0.1263, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.676773145641581, | |
| "grad_norm": 0.92578125, | |
| "learning_rate": 1.8691296180338137e-05, | |
| "loss": 0.1273, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.6821873308067136, | |
| "grad_norm": 0.75, | |
| "learning_rate": 1.8378209142141518e-05, | |
| "loss": 0.1373, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.6876015159718463, | |
| "grad_norm": 0.58984375, | |
| "learning_rate": 1.80651221039449e-05, | |
| "loss": 0.1242, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.6930157011369789, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 1.775203506574828e-05, | |
| "loss": 0.1424, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.6984298863021116, | |
| "grad_norm": 0.984375, | |
| "learning_rate": 1.743894802755166e-05, | |
| "loss": 0.1053, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.7038440714672441, | |
| "grad_norm": 0.97265625, | |
| "learning_rate": 1.712586098935504e-05, | |
| "loss": 0.1195, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.7038440714672441, | |
| "eval_loss": 0.09898315370082855, | |
| "eval_runtime": 59.819, | |
| "eval_samples_per_second": 8.359, | |
| "eval_steps_per_second": 0.535, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.7092582566323768, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 1.681277395115842e-05, | |
| "loss": 0.1339, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.7146724417975094, | |
| "grad_norm": 0.92578125, | |
| "learning_rate": 1.6499686912961803e-05, | |
| "loss": 0.155, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.7200866269626421, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 1.6186599874765184e-05, | |
| "loss": 0.1275, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.7255008121277747, | |
| "grad_norm": 1.25, | |
| "learning_rate": 1.5873512836568566e-05, | |
| "loss": 0.1454, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.7309149972929074, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 1.5560425798371947e-05, | |
| "loss": 0.1184, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.73632918245804, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 1.5247338760175329e-05, | |
| "loss": 0.1231, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.7417433676231727, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 1.4934251721978712e-05, | |
| "loss": 0.1245, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.7471575527883053, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 1.4621164683782093e-05, | |
| "loss": 0.1388, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.752571737953438, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 1.4308077645585475e-05, | |
| "loss": 0.1599, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.7579859231185706, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 1.3994990607388856e-05, | |
| "loss": 0.1481, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.7579859231185706, | |
| "eval_loss": 0.09811025857925415, | |
| "eval_runtime": 58.4243, | |
| "eval_samples_per_second": 8.558, | |
| "eval_steps_per_second": 0.548, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.7634001082837033, | |
| "grad_norm": 0.72265625, | |
| "learning_rate": 1.3681903569192236e-05, | |
| "loss": 0.1228, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 0.7688142934488359, | |
| "grad_norm": 0.8359375, | |
| "learning_rate": 1.3368816530995617e-05, | |
| "loss": 0.1356, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.7742284786139686, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 1.3055729492798999e-05, | |
| "loss": 0.1431, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 0.7796426637791013, | |
| "grad_norm": 0.9140625, | |
| "learning_rate": 1.274264245460238e-05, | |
| "loss": 0.1154, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 0.7850568489442339, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 1.2429555416405761e-05, | |
| "loss": 0.1106, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.7904710341093666, | |
| "grad_norm": 0.8046875, | |
| "learning_rate": 1.2116468378209143e-05, | |
| "loss": 0.1172, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 0.7958852192744992, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 1.1803381340012524e-05, | |
| "loss": 0.1424, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 0.8012994044396319, | |
| "grad_norm": 1.0, | |
| "learning_rate": 1.1490294301815906e-05, | |
| "loss": 0.1328, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 0.8067135896047645, | |
| "grad_norm": 2.171875, | |
| "learning_rate": 1.1177207263619287e-05, | |
| "loss": 0.1363, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 0.8121277747698972, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 1.0864120225422668e-05, | |
| "loss": 0.1255, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8121277747698972, | |
| "eval_loss": 0.0966864824295044, | |
| "eval_runtime": 61.9578, | |
| "eval_samples_per_second": 8.07, | |
| "eval_steps_per_second": 0.516, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8175419599350298, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 1.0551033187226048e-05, | |
| "loss": 0.1397, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 0.8229561451001625, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 1.023794614902943e-05, | |
| "loss": 0.1289, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 0.828370330265295, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 9.924859110832813e-06, | |
| "loss": 0.1231, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 0.8337845154304278, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 9.611772072636194e-06, | |
| "loss": 0.1477, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 0.8391987005955603, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 9.298685034439576e-06, | |
| "loss": 0.1228, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.844612885760693, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 8.985597996242955e-06, | |
| "loss": 0.1745, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.8500270709258256, | |
| "grad_norm": 0.859375, | |
| "learning_rate": 8.672510958046337e-06, | |
| "loss": 0.1565, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 0.8554412560909583, | |
| "grad_norm": 0.95703125, | |
| "learning_rate": 8.359423919849718e-06, | |
| "loss": 0.1503, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 0.8608554412560909, | |
| "grad_norm": 1.6875, | |
| "learning_rate": 8.0463368816531e-06, | |
| "loss": 0.1569, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 0.8662696264212236, | |
| "grad_norm": 0.96875, | |
| "learning_rate": 7.733249843456483e-06, | |
| "loss": 0.1408, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.8662696264212236, | |
| "eval_loss": 0.09672338515520096, | |
| "eval_runtime": 60.7387, | |
| "eval_samples_per_second": 8.232, | |
| "eval_steps_per_second": 0.527, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.8716838115863562, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 7.420162805259863e-06, | |
| "loss": 0.1313, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 0.8770979967514889, | |
| "grad_norm": 0.482421875, | |
| "learning_rate": 7.107075767063244e-06, | |
| "loss": 0.1506, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 0.8825121819166215, | |
| "grad_norm": 0.796875, | |
| "learning_rate": 6.793988728866625e-06, | |
| "loss": 0.1274, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 0.8879263670817542, | |
| "grad_norm": 1.515625, | |
| "learning_rate": 6.4809016906700065e-06, | |
| "loss": 0.1471, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 0.8933405522468868, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 6.167814652473388e-06, | |
| "loss": 0.142, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.8987547374120195, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 5.854727614276769e-06, | |
| "loss": 0.1084, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 0.9041689225771521, | |
| "grad_norm": 0.97265625, | |
| "learning_rate": 5.54164057608015e-06, | |
| "loss": 0.1044, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 0.9095831077422848, | |
| "grad_norm": 0.96484375, | |
| "learning_rate": 5.228553537883532e-06, | |
| "loss": 0.147, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 0.9149972929074174, | |
| "grad_norm": 0.4765625, | |
| "learning_rate": 4.9154664996869136e-06, | |
| "loss": 0.1254, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 0.9204114780725501, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 4.602379461490294e-06, | |
| "loss": 0.1208, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.9204114780725501, | |
| "eval_loss": 0.09627044945955276, | |
| "eval_runtime": 61.5631, | |
| "eval_samples_per_second": 8.122, | |
| "eval_steps_per_second": 0.52, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.9258256632376828, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 4.289292423293676e-06, | |
| "loss": 0.1468, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 0.9312398484028154, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 3.976205385097057e-06, | |
| "loss": 0.1231, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 0.9366540335679481, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 3.6631183469004384e-06, | |
| "loss": 0.1548, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 0.9420682187330807, | |
| "grad_norm": 1.125, | |
| "learning_rate": 3.35003130870382e-06, | |
| "loss": 0.1332, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 0.9474824038982134, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 3.036944270507201e-06, | |
| "loss": 0.1382, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.952896589063346, | |
| "grad_norm": 0.72265625, | |
| "learning_rate": 2.7238572323105826e-06, | |
| "loss": 0.1203, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 0.9583107742284787, | |
| "grad_norm": 0.96875, | |
| "learning_rate": 2.410770194113964e-06, | |
| "loss": 0.1206, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 0.9637249593936112, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 2.0976831559173454e-06, | |
| "loss": 0.1512, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 0.969139144558744, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 1.7845961177207264e-06, | |
| "loss": 0.1522, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 0.9745533297238765, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 1.4715090795241078e-06, | |
| "loss": 0.1228, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.9745533297238765, | |
| "eval_loss": 0.09588468819856644, | |
| "eval_runtime": 61.1963, | |
| "eval_samples_per_second": 8.17, | |
| "eval_steps_per_second": 0.523, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.9799675148890092, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 1.1584220413274892e-06, | |
| "loss": 0.1349, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 0.9853817000541418, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 8.453350031308704e-07, | |
| "loss": 0.1268, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 0.9907958852192745, | |
| "grad_norm": 1.4609375, | |
| "learning_rate": 5.322479649342517e-07, | |
| "loss": 0.1416, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 0.9962100703844071, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 2.1916092673763307e-07, | |
| "loss": 0.1308, | |
| "step": 1840 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1847, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 1.1050860758570842e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |