| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 3694, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0027070925825663237, | |
| "grad_norm": 5.65625, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.7343, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005414185165132647, | |
| "grad_norm": 8.9375, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.864, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008121277747698972, | |
| "grad_norm": 9.0625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.7618, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.010828370330265295, | |
| "grad_norm": 7.375, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.749, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01353546291283162, | |
| "grad_norm": 5.4375, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7434, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.016242555495397944, | |
| "grad_norm": 5.34375, | |
| "learning_rate": 6e-06, | |
| "loss": 0.7079, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.018949648077964266, | |
| "grad_norm": 4.25, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.7368, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.02165674066053059, | |
| "grad_norm": 3.828125, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.682, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.024363833243096916, | |
| "grad_norm": 2.875, | |
| "learning_rate": 9e-06, | |
| "loss": 0.6489, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.02707092582566324, | |
| "grad_norm": 2.453125, | |
| "learning_rate": 1e-05, | |
| "loss": 0.589, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02707092582566324, | |
| "eval_loss": 0.6064066290855408, | |
| "eval_runtime": 155.4734, | |
| "eval_samples_per_second": 3.293, | |
| "eval_steps_per_second": 0.103, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02977801840822956, | |
| "grad_norm": 2.1875, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.5723, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.03248511099079589, | |
| "grad_norm": 3.640625, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.5525, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.03519220357336221, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.5503, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.03789929615592853, | |
| "grad_norm": 2.3125, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.4926, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.040606388738494856, | |
| "grad_norm": 2.578125, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.4831, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04331348132106118, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.4413, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.0460205739036275, | |
| "grad_norm": 2.671875, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.4562, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.04872766648619383, | |
| "grad_norm": 2.203125, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.4348, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.051434759068760154, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.4386, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.05414185165132648, | |
| "grad_norm": 2.09375, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4009, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05414185165132648, | |
| "eval_loss": 0.4071368873119354, | |
| "eval_runtime": 104.5667, | |
| "eval_samples_per_second": 4.896, | |
| "eval_steps_per_second": 0.153, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.0568489442338928, | |
| "grad_norm": 2.25, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.4026, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.05955603681645912, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.3774, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.062263129399025445, | |
| "grad_norm": 1.8828125, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.3734, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.06497022198159177, | |
| "grad_norm": 1.5703125, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.331, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.0676773145641581, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3381, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.07038440714672442, | |
| "grad_norm": 1.8515625, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.3392, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.07309149972929074, | |
| "grad_norm": 1.625, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.3165, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.07579859231185707, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.3162, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.07850568489442339, | |
| "grad_norm": 1.953125, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.2844, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.08121277747698971, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3093, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08121277747698971, | |
| "eval_loss": 0.2839646637439728, | |
| "eval_runtime": 103.4698, | |
| "eval_samples_per_second": 4.948, | |
| "eval_steps_per_second": 0.155, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08391987005955603, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.2753, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.08662696264212236, | |
| "grad_norm": 1.4609375, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.2625, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.08933405522468868, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.2525, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.092041147807255, | |
| "grad_norm": 1.9453125, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.2897, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.09474824038982133, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.243, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.09745533297238766, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.2442, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.10016242555495398, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.2139, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.10286951813752031, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.2312, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.10557661072008663, | |
| "grad_norm": 2.703125, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.2202, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.10828370330265295, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2115, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.10828370330265295, | |
| "eval_loss": 0.23121848702430725, | |
| "eval_runtime": 134.1955, | |
| "eval_samples_per_second": 3.815, | |
| "eval_steps_per_second": 0.119, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.11099079588521928, | |
| "grad_norm": 1.828125, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.1976, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.1136978884677856, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2182, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.11640498105035192, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.1767, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.11911207363291824, | |
| "grad_norm": 3.078125, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.2058, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.12181916621548457, | |
| "grad_norm": 2.25, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.168, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.12452625879805089, | |
| "grad_norm": 2.5625, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.1728, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.12723335138061723, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.1927, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.12994044396318355, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.1433, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.13264753654574987, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.1728, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.1353546291283162, | |
| "grad_norm": 2.4375, | |
| "learning_rate": 5e-05, | |
| "loss": 0.176, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1353546291283162, | |
| "eval_loss": 0.19780531525611877, | |
| "eval_runtime": 102.6349, | |
| "eval_samples_per_second": 4.989, | |
| "eval_steps_per_second": 0.156, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13806172171088252, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 4.984345648090169e-05, | |
| "loss": 0.1639, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.14076881429344884, | |
| "grad_norm": 1.5078125, | |
| "learning_rate": 4.9686912961803384e-05, | |
| "loss": 0.1598, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.14347590687601516, | |
| "grad_norm": 5.625, | |
| "learning_rate": 4.9530369442705075e-05, | |
| "loss": 0.17, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.1461829994585815, | |
| "grad_norm": 2.390625, | |
| "learning_rate": 4.9373825923606765e-05, | |
| "loss": 0.1601, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.1488900920411478, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 4.9217282404508456e-05, | |
| "loss": 0.2067, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.15159718462371413, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 4.906073888541015e-05, | |
| "loss": 0.1396, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.15430427720628045, | |
| "grad_norm": 2.3125, | |
| "learning_rate": 4.890419536631184e-05, | |
| "loss": 0.1379, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.15701136978884678, | |
| "grad_norm": 1.984375, | |
| "learning_rate": 4.874765184721353e-05, | |
| "loss": 0.1722, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.1597184623714131, | |
| "grad_norm": 2.796875, | |
| "learning_rate": 4.859110832811522e-05, | |
| "loss": 0.1608, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.16242555495397942, | |
| "grad_norm": 2.4375, | |
| "learning_rate": 4.843456480901691e-05, | |
| "loss": 0.1576, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.16242555495397942, | |
| "eval_loss": 0.1739654242992401, | |
| "eval_runtime": 102.4755, | |
| "eval_samples_per_second": 4.996, | |
| "eval_steps_per_second": 0.156, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.16513264753654575, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 4.82780212899186e-05, | |
| "loss": 0.1292, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.16783974011911207, | |
| "grad_norm": 2.296875, | |
| "learning_rate": 4.812147777082029e-05, | |
| "loss": 0.1421, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.1705468327016784, | |
| "grad_norm": 3.53125, | |
| "learning_rate": 4.796493425172198e-05, | |
| "loss": 0.1473, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.17325392528424471, | |
| "grad_norm": 1.6015625, | |
| "learning_rate": 4.780839073262367e-05, | |
| "loss": 0.1313, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.17596101786681104, | |
| "grad_norm": 2.75, | |
| "learning_rate": 4.765184721352536e-05, | |
| "loss": 0.0987, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.17866811044937736, | |
| "grad_norm": 2.453125, | |
| "learning_rate": 4.7495303694427054e-05, | |
| "loss": 0.1301, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.18137520303194368, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 4.7338760175328744e-05, | |
| "loss": 0.118, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.18408229561451, | |
| "grad_norm": 2.703125, | |
| "learning_rate": 4.7182216656230435e-05, | |
| "loss": 0.1384, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.18678938819707633, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 4.7025673137132126e-05, | |
| "loss": 0.0855, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.18949648077964265, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 4.6869129618033816e-05, | |
| "loss": 0.148, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.18949648077964265, | |
| "eval_loss": 0.1687919795513153, | |
| "eval_runtime": 102.502, | |
| "eval_samples_per_second": 4.995, | |
| "eval_steps_per_second": 0.156, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.19220357336220897, | |
| "grad_norm": 0.81640625, | |
| "learning_rate": 4.671258609893551e-05, | |
| "loss": 0.134, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.19491066594477532, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 4.65560425798372e-05, | |
| "loss": 0.1222, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.19761775852734165, | |
| "grad_norm": 2.5625, | |
| "learning_rate": 4.639949906073889e-05, | |
| "loss": 0.1511, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.20032485110990797, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 4.624295554164057e-05, | |
| "loss": 0.1661, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.2030319436924743, | |
| "grad_norm": 1.9140625, | |
| "learning_rate": 4.608641202254227e-05, | |
| "loss": 0.1381, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.20573903627504062, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 4.5929868503443954e-05, | |
| "loss": 0.1499, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.20844612885760694, | |
| "grad_norm": 2.265625, | |
| "learning_rate": 4.577332498434565e-05, | |
| "loss": 0.1692, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.21115322144017326, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 4.561678146524734e-05, | |
| "loss": 0.1226, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.21386031402273958, | |
| "grad_norm": 2.484375, | |
| "learning_rate": 4.546023794614903e-05, | |
| "loss": 0.1594, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.2165674066053059, | |
| "grad_norm": 0.76171875, | |
| "learning_rate": 4.5303694427050724e-05, | |
| "loss": 0.0962, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.2165674066053059, | |
| "eval_loss": 0.15033257007598877, | |
| "eval_runtime": 102.8567, | |
| "eval_samples_per_second": 4.978, | |
| "eval_steps_per_second": 0.156, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.21927449918787223, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 4.5147150907952414e-05, | |
| "loss": 0.1581, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.22198159177043855, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 4.4990607388854105e-05, | |
| "loss": 0.155, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.22468868435300487, | |
| "grad_norm": 0.921875, | |
| "learning_rate": 4.4834063869755796e-05, | |
| "loss": 0.1103, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.2273957769355712, | |
| "grad_norm": 0.90234375, | |
| "learning_rate": 4.4677520350657486e-05, | |
| "loss": 0.1323, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.23010286951813752, | |
| "grad_norm": 0.953125, | |
| "learning_rate": 4.452097683155918e-05, | |
| "loss": 0.1073, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.23280996210070384, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 4.436443331246087e-05, | |
| "loss": 0.1333, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.23551705468327017, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 4.420788979336256e-05, | |
| "loss": 0.1429, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.2382241472658365, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 4.405134627426425e-05, | |
| "loss": 0.0988, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.2409312398484028, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 4.389480275516594e-05, | |
| "loss": 0.1184, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.24363833243096913, | |
| "grad_norm": 1.0234375, | |
| "learning_rate": 4.373825923606763e-05, | |
| "loss": 0.118, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.24363833243096913, | |
| "eval_loss": 0.14425741136074066, | |
| "eval_runtime": 115.407, | |
| "eval_samples_per_second": 4.436, | |
| "eval_steps_per_second": 0.139, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.24634542501353546, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 4.358171571696932e-05, | |
| "loss": 0.1032, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.24905251759610178, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 4.342517219787101e-05, | |
| "loss": 0.1242, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.2517596101786681, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 4.3268628678772696e-05, | |
| "loss": 0.131, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.25446670276123445, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 4.3112085159674393e-05, | |
| "loss": 0.1058, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.25717379534380075, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 4.295554164057608e-05, | |
| "loss": 0.155, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.2598808879263671, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 4.2798998121477775e-05, | |
| "loss": 0.1103, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.2625879805089334, | |
| "grad_norm": 4.59375, | |
| "learning_rate": 4.264245460237946e-05, | |
| "loss": 0.1288, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.26529507309149974, | |
| "grad_norm": 0.90234375, | |
| "learning_rate": 4.2485911083281156e-05, | |
| "loss": 0.0967, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.26800216567406604, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 4.232936756418284e-05, | |
| "loss": 0.1523, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.2707092582566324, | |
| "grad_norm": 2.90625, | |
| "learning_rate": 4.217282404508454e-05, | |
| "loss": 0.1364, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2707092582566324, | |
| "eval_loss": 0.14593610167503357, | |
| "eval_runtime": 104.6862, | |
| "eval_samples_per_second": 4.891, | |
| "eval_steps_per_second": 0.153, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2734163508391987, | |
| "grad_norm": 2.125, | |
| "learning_rate": 4.201628052598622e-05, | |
| "loss": 0.1228, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.27612344342176504, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 4.185973700688792e-05, | |
| "loss": 0.1114, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.27883053600433133, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 4.170319348778961e-05, | |
| "loss": 0.1016, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.2815376285868977, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 4.15466499686913e-05, | |
| "loss": 0.1023, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.284244721169464, | |
| "grad_norm": 1.5859375, | |
| "learning_rate": 4.139010644959299e-05, | |
| "loss": 0.1097, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.2869518137520303, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 4.123356293049468e-05, | |
| "loss": 0.1522, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.2896589063345966, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 4.107701941139637e-05, | |
| "loss": 0.1218, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.292365998917163, | |
| "grad_norm": 1.9765625, | |
| "learning_rate": 4.092047589229806e-05, | |
| "loss": 0.1491, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.29507309149972927, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 4.0763932373199754e-05, | |
| "loss": 0.1103, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.2977801840822956, | |
| "grad_norm": 1.96875, | |
| "learning_rate": 4.0607388854101445e-05, | |
| "loss": 0.1419, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.2977801840822956, | |
| "eval_loss": 0.14324142038822174, | |
| "eval_runtime": 105.9997, | |
| "eval_samples_per_second": 4.83, | |
| "eval_steps_per_second": 0.151, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.3004872766648619, | |
| "grad_norm": 0.8984375, | |
| "learning_rate": 4.0450845335003135e-05, | |
| "loss": 0.0784, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.30319436924742826, | |
| "grad_norm": 0.9296875, | |
| "learning_rate": 4.029430181590482e-05, | |
| "loss": 0.0869, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.30590146182999456, | |
| "grad_norm": 0.6015625, | |
| "learning_rate": 4.013775829680652e-05, | |
| "loss": 0.1135, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.3086085544125609, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 3.99812147777082e-05, | |
| "loss": 0.1296, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.31131564699512726, | |
| "grad_norm": 0.8203125, | |
| "learning_rate": 3.98246712586099e-05, | |
| "loss": 0.122, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.31402273957769355, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 3.966812773951158e-05, | |
| "loss": 0.1102, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.3167298321602599, | |
| "grad_norm": 0.98046875, | |
| "learning_rate": 3.951158422041328e-05, | |
| "loss": 0.1284, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.3194369247428262, | |
| "grad_norm": 2.390625, | |
| "learning_rate": 3.9355040701314964e-05, | |
| "loss": 0.139, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.32214401732539255, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 3.919849718221666e-05, | |
| "loss": 0.1405, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.32485110990795885, | |
| "grad_norm": 0.75390625, | |
| "learning_rate": 3.9041953663118345e-05, | |
| "loss": 0.1359, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.32485110990795885, | |
| "eval_loss": 0.13801878690719604, | |
| "eval_runtime": 105.4772, | |
| "eval_samples_per_second": 4.854, | |
| "eval_steps_per_second": 0.152, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.3275582024905252, | |
| "grad_norm": 0.9609375, | |
| "learning_rate": 3.888541014402004e-05, | |
| "loss": 0.0934, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.3302652950730915, | |
| "grad_norm": 1.9921875, | |
| "learning_rate": 3.8728866624921726e-05, | |
| "loss": 0.1272, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.33297238765565784, | |
| "grad_norm": 2.734375, | |
| "learning_rate": 3.8572323105823424e-05, | |
| "loss": 0.1099, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.33567948023822414, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 3.841577958672511e-05, | |
| "loss": 0.1019, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.3383865728207905, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 3.8259236067626805e-05, | |
| "loss": 0.1503, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.3410936654033568, | |
| "grad_norm": 0.490234375, | |
| "learning_rate": 3.810269254852849e-05, | |
| "loss": 0.0857, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.34380075798592313, | |
| "grad_norm": 0.69140625, | |
| "learning_rate": 3.794614902943019e-05, | |
| "loss": 0.0949, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.34650785056848943, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 3.778960551033187e-05, | |
| "loss": 0.1242, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.3492149431510558, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 3.763306199123356e-05, | |
| "loss": 0.1041, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.3519220357336221, | |
| "grad_norm": 1.5234375, | |
| "learning_rate": 3.747651847213526e-05, | |
| "loss": 0.0727, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.3519220357336221, | |
| "eval_loss": 0.13924407958984375, | |
| "eval_runtime": 103.9612, | |
| "eval_samples_per_second": 4.925, | |
| "eval_steps_per_second": 0.154, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.3546291283161884, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 3.731997495303694e-05, | |
| "loss": 0.0898, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.3573362208987547, | |
| "grad_norm": 2.609375, | |
| "learning_rate": 3.716343143393864e-05, | |
| "loss": 0.1697, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.36004331348132107, | |
| "grad_norm": 1.1015625, | |
| "learning_rate": 3.7006887914840324e-05, | |
| "loss": 0.1111, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.36275040606388737, | |
| "grad_norm": 0.44921875, | |
| "learning_rate": 3.685034439574202e-05, | |
| "loss": 0.1194, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.3654574986464537, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 3.6693800876643706e-05, | |
| "loss": 0.1019, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.36816459122902, | |
| "grad_norm": 1.25, | |
| "learning_rate": 3.65372573575454e-05, | |
| "loss": 0.0984, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.37087168381158636, | |
| "grad_norm": 0.86328125, | |
| "learning_rate": 3.638071383844709e-05, | |
| "loss": 0.1298, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.37357877639415266, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 3.6224170319348784e-05, | |
| "loss": 0.1242, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.376285868976719, | |
| "grad_norm": 0.72265625, | |
| "learning_rate": 3.606762680025047e-05, | |
| "loss": 0.094, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.3789929615592853, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 3.5911083281152166e-05, | |
| "loss": 0.1286, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.3789929615592853, | |
| "eval_loss": 0.13358080387115479, | |
| "eval_runtime": 120.0689, | |
| "eval_samples_per_second": 4.264, | |
| "eval_steps_per_second": 0.133, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.38170005414185165, | |
| "grad_norm": 2.703125, | |
| "learning_rate": 3.575453976205385e-05, | |
| "loss": 0.1594, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 0.38440714672441795, | |
| "grad_norm": 2.0, | |
| "learning_rate": 3.559799624295555e-05, | |
| "loss": 0.0771, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.3871142393069843, | |
| "grad_norm": 2.328125, | |
| "learning_rate": 3.544145272385723e-05, | |
| "loss": 0.1485, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 0.38982133188955065, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 3.528490920475893e-05, | |
| "loss": 0.0938, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 0.39252842447211694, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 3.512836568566061e-05, | |
| "loss": 0.1372, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.3952355170546833, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 3.497182216656231e-05, | |
| "loss": 0.1033, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 0.3979426096372496, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 3.4815278647463994e-05, | |
| "loss": 0.1345, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 0.40064970221981594, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 3.4658735128365685e-05, | |
| "loss": 0.103, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 0.40335679480238223, | |
| "grad_norm": 1.125, | |
| "learning_rate": 3.4502191609267375e-05, | |
| "loss": 0.083, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 0.4060638873849486, | |
| "grad_norm": 0.9296875, | |
| "learning_rate": 3.4345648090169066e-05, | |
| "loss": 0.1024, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4060638873849486, | |
| "eval_loss": 0.13767841458320618, | |
| "eval_runtime": 104.1987, | |
| "eval_samples_per_second": 4.914, | |
| "eval_steps_per_second": 0.154, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4087709799675149, | |
| "grad_norm": 1.71875, | |
| "learning_rate": 3.418910457107076e-05, | |
| "loss": 0.1319, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 0.41147807255008123, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 3.403256105197245e-05, | |
| "loss": 0.1152, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 0.4141851651326475, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 3.387601753287414e-05, | |
| "loss": 0.1072, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 0.4168922577152139, | |
| "grad_norm": 2.171875, | |
| "learning_rate": 3.371947401377583e-05, | |
| "loss": 0.1408, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 0.41959935029778017, | |
| "grad_norm": 1.53125, | |
| "learning_rate": 3.3562930494677526e-05, | |
| "loss": 0.1125, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.4223064428803465, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 3.340638697557921e-05, | |
| "loss": 0.1144, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.4250135354629128, | |
| "grad_norm": 1.6015625, | |
| "learning_rate": 3.324984345648091e-05, | |
| "loss": 0.1092, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 0.42772062804547917, | |
| "grad_norm": 0.408203125, | |
| "learning_rate": 3.309329993738259e-05, | |
| "loss": 0.1271, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 0.43042772062804546, | |
| "grad_norm": 0.72265625, | |
| "learning_rate": 3.293675641828429e-05, | |
| "loss": 0.0882, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 0.4331348132106118, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 3.278021289918597e-05, | |
| "loss": 0.1141, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.4331348132106118, | |
| "eval_loss": 0.13831546902656555, | |
| "eval_runtime": 104.916, | |
| "eval_samples_per_second": 4.88, | |
| "eval_steps_per_second": 0.153, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.4358419057931781, | |
| "grad_norm": 0.462890625, | |
| "learning_rate": 3.262366938008767e-05, | |
| "loss": 0.0874, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 0.43854899837574446, | |
| "grad_norm": 1.921875, | |
| "learning_rate": 3.2467125860989355e-05, | |
| "loss": 0.1288, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 0.44125609095831075, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 3.231058234189105e-05, | |
| "loss": 0.1171, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 0.4439631835408771, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 3.2154038822792736e-05, | |
| "loss": 0.0845, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 0.4466702761234434, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 3.1997495303694433e-05, | |
| "loss": 0.1052, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.44937736870600975, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 3.184095178459612e-05, | |
| "loss": 0.1187, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 0.45208446128857604, | |
| "grad_norm": 2.8125, | |
| "learning_rate": 3.168440826549781e-05, | |
| "loss": 0.1051, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 0.4547915538711424, | |
| "grad_norm": 1.9765625, | |
| "learning_rate": 3.15278647463995e-05, | |
| "loss": 0.1415, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 0.4574986464537087, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 3.137132122730119e-05, | |
| "loss": 0.1313, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 0.46020573903627504, | |
| "grad_norm": 0.8359375, | |
| "learning_rate": 3.121477770820288e-05, | |
| "loss": 0.1129, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.46020573903627504, | |
| "eval_loss": 0.13771241903305054, | |
| "eval_runtime": 104.8503, | |
| "eval_samples_per_second": 4.883, | |
| "eval_steps_per_second": 0.153, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.4629128316188414, | |
| "grad_norm": 1.8359375, | |
| "learning_rate": 3.105823418910457e-05, | |
| "loss": 0.1618, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 0.4656199242014077, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 3.090169067000626e-05, | |
| "loss": 0.1025, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 0.46832701678397404, | |
| "grad_norm": 4.0625, | |
| "learning_rate": 3.074514715090795e-05, | |
| "loss": 0.1253, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 0.47103410936654033, | |
| "grad_norm": 2.25, | |
| "learning_rate": 3.058860363180964e-05, | |
| "loss": 0.1098, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 0.4737412019491067, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 3.0432060112711337e-05, | |
| "loss": 0.0868, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.476448294531673, | |
| "grad_norm": 1.9375, | |
| "learning_rate": 3.0275516593613024e-05, | |
| "loss": 0.1427, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 0.47915538711423933, | |
| "grad_norm": 2.546875, | |
| "learning_rate": 3.011897307451472e-05, | |
| "loss": 0.1272, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 0.4818624796968056, | |
| "grad_norm": 2.703125, | |
| "learning_rate": 2.9962429555416406e-05, | |
| "loss": 0.16, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 0.484569572279372, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 2.9805886036318097e-05, | |
| "loss": 0.1079, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 0.48727666486193827, | |
| "grad_norm": 0.75, | |
| "learning_rate": 2.9649342517219787e-05, | |
| "loss": 0.0907, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.48727666486193827, | |
| "eval_loss": 0.14291706681251526, | |
| "eval_runtime": 105.4966, | |
| "eval_samples_per_second": 4.853, | |
| "eval_steps_per_second": 0.152, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.4899837574445046, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 2.9492798998121478e-05, | |
| "loss": 0.1137, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 0.4926908500270709, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 2.9336255479023172e-05, | |
| "loss": 0.114, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 0.49539794260963727, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 2.917971195992486e-05, | |
| "loss": 0.1263, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 0.49810503519220356, | |
| "grad_norm": 2.171875, | |
| "learning_rate": 2.9023168440826553e-05, | |
| "loss": 0.1095, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 0.5008121277747699, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 2.886662492172824e-05, | |
| "loss": 0.1609, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.5035192203573362, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 2.8710081402629935e-05, | |
| "loss": 0.1079, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 0.5062263129399025, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 2.8553537883531622e-05, | |
| "loss": 0.1114, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 0.5089334055224689, | |
| "grad_norm": 1.9140625, | |
| "learning_rate": 2.8396994364433316e-05, | |
| "loss": 0.1163, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 0.5116404981050352, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 2.8240450845335004e-05, | |
| "loss": 0.105, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 0.5143475906876015, | |
| "grad_norm": 0.76171875, | |
| "learning_rate": 2.8083907326236698e-05, | |
| "loss": 0.0931, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.5143475906876015, | |
| "eval_loss": 0.1378411054611206, | |
| "eval_runtime": 122.4527, | |
| "eval_samples_per_second": 4.181, | |
| "eval_steps_per_second": 0.131, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.5170546832701678, | |
| "grad_norm": 2.125, | |
| "learning_rate": 2.7927363807138385e-05, | |
| "loss": 0.1137, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 0.5197617758527342, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 2.777082028804008e-05, | |
| "loss": 0.1127, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 0.5224688684353005, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 2.7614276768941766e-05, | |
| "loss": 0.1082, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 0.5251759610178668, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 2.745773324984346e-05, | |
| "loss": 0.1147, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 0.5278830536004331, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 2.7301189730745148e-05, | |
| "loss": 0.1027, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.5305901461829995, | |
| "grad_norm": 0.86328125, | |
| "learning_rate": 2.7144646211646842e-05, | |
| "loss": 0.108, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 0.5332972387655658, | |
| "grad_norm": 0.59375, | |
| "learning_rate": 2.698810269254853e-05, | |
| "loss": 0.0917, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 0.5360043313481321, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 2.683155917345022e-05, | |
| "loss": 0.078, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 0.5387114239306985, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 2.667501565435191e-05, | |
| "loss": 0.0845, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 0.5414185165132648, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 2.65184721352536e-05, | |
| "loss": 0.0888, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5414185165132648, | |
| "eval_loss": 0.14294707775115967, | |
| "eval_runtime": 103.5797, | |
| "eval_samples_per_second": 4.943, | |
| "eval_steps_per_second": 0.154, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5441256090958311, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 2.636192861615529e-05, | |
| "loss": 0.099, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 0.5468327016783974, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 2.6205385097056983e-05, | |
| "loss": 0.1048, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 0.5495397942609638, | |
| "grad_norm": 1.984375, | |
| "learning_rate": 2.604884157795867e-05, | |
| "loss": 0.1117, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 0.5522468868435301, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 2.5892298058860364e-05, | |
| "loss": 0.1291, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 0.5549539794260964, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 2.573575453976205e-05, | |
| "loss": 0.1529, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.5576610720086627, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 2.5579211020663746e-05, | |
| "loss": 0.0559, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 0.5603681645912291, | |
| "grad_norm": 1.40625, | |
| "learning_rate": 2.5422667501565433e-05, | |
| "loss": 0.0992, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 0.5630752571737954, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 2.5266123982467127e-05, | |
| "loss": 0.0876, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 0.5657823497563617, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 2.510958046336882e-05, | |
| "loss": 0.1442, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 0.568489442338928, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 2.495303694427051e-05, | |
| "loss": 0.073, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.568489442338928, | |
| "eval_loss": 0.1371915191411972, | |
| "eval_runtime": 102.7955, | |
| "eval_samples_per_second": 4.981, | |
| "eval_steps_per_second": 0.156, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.5711965349214944, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 2.47964934251722e-05, | |
| "loss": 0.0818, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 0.5739036275040607, | |
| "grad_norm": 1.0, | |
| "learning_rate": 2.463994990607389e-05, | |
| "loss": 0.0785, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 0.576610720086627, | |
| "grad_norm": 2.09375, | |
| "learning_rate": 2.448340638697558e-05, | |
| "loss": 0.0966, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 0.5793178126691932, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 2.432686286787727e-05, | |
| "loss": 0.0805, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 0.5820249052517596, | |
| "grad_norm": 0.9375, | |
| "learning_rate": 2.4170319348778962e-05, | |
| "loss": 0.1032, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.584731997834326, | |
| "grad_norm": 1.9765625, | |
| "learning_rate": 2.4013775829680653e-05, | |
| "loss": 0.1314, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 0.5874390904168922, | |
| "grad_norm": 1.5078125, | |
| "learning_rate": 2.3857232310582343e-05, | |
| "loss": 0.1026, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 0.5901461829994585, | |
| "grad_norm": 0.58984375, | |
| "learning_rate": 2.3700688791484034e-05, | |
| "loss": 0.0858, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 0.5928532755820249, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 2.3544145272385725e-05, | |
| "loss": 0.069, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 0.5955603681645912, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 2.3387601753287412e-05, | |
| "loss": 0.1238, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.5955603681645912, | |
| "eval_loss": 0.1365373730659485, | |
| "eval_runtime": 103.5823, | |
| "eval_samples_per_second": 4.943, | |
| "eval_steps_per_second": 0.154, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.5982674607471575, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 2.3231058234189106e-05, | |
| "loss": 0.1196, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 0.6009745533297238, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 2.3074514715090797e-05, | |
| "loss": 0.111, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 0.6036816459122902, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 2.2917971195992488e-05, | |
| "loss": 0.0875, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 0.6063887384948565, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 2.2761427676894178e-05, | |
| "loss": 0.1029, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 0.6090958310774228, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 2.260488415779587e-05, | |
| "loss": 0.0933, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.6118029236599891, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 2.244834063869756e-05, | |
| "loss": 0.1009, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 0.6145100162425555, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 2.229179711959925e-05, | |
| "loss": 0.135, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 0.6172171088251218, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 2.213525360050094e-05, | |
| "loss": 0.0773, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 0.6199242014076881, | |
| "grad_norm": 0.9375, | |
| "learning_rate": 2.1978710081402632e-05, | |
| "loss": 0.0732, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 0.6226312939902545, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 2.1822166562304323e-05, | |
| "loss": 0.0851, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.6226312939902545, | |
| "eval_loss": 0.1353287696838379, | |
| "eval_runtime": 102.3815, | |
| "eval_samples_per_second": 5.001, | |
| "eval_steps_per_second": 0.156, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.6253383865728208, | |
| "grad_norm": 0.71484375, | |
| "learning_rate": 2.1665623043206013e-05, | |
| "loss": 0.0633, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 0.6280454791553871, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 2.1509079524107704e-05, | |
| "loss": 0.0825, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 0.6307525717379534, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 2.1352536005009395e-05, | |
| "loss": 0.1016, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 0.6334596643205198, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 2.1195992485911085e-05, | |
| "loss": 0.0942, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 0.6361667569030861, | |
| "grad_norm": 1.75, | |
| "learning_rate": 2.1039448966812776e-05, | |
| "loss": 0.086, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.6388738494856524, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 2.0882905447714467e-05, | |
| "loss": 0.1095, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 0.6415809420682187, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 2.0726361928616157e-05, | |
| "loss": 0.0711, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 0.6442880346507851, | |
| "grad_norm": 1.3515625, | |
| "learning_rate": 2.0569818409517845e-05, | |
| "loss": 0.0677, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 0.6469951272333514, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 2.0413274890419535e-05, | |
| "loss": 0.0892, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 0.6497022198159177, | |
| "grad_norm": 1.640625, | |
| "learning_rate": 2.0256731371321226e-05, | |
| "loss": 0.0955, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.6497022198159177, | |
| "eval_loss": 0.13519829511642456, | |
| "eval_runtime": 121.253, | |
| "eval_samples_per_second": 4.223, | |
| "eval_steps_per_second": 0.132, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.652409312398484, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 2.0100187852222917e-05, | |
| "loss": 0.1012, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 0.6551164049810504, | |
| "grad_norm": 0.984375, | |
| "learning_rate": 1.9943644333124608e-05, | |
| "loss": 0.1008, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 0.6578234975636167, | |
| "grad_norm": 0.9453125, | |
| "learning_rate": 1.9787100814026298e-05, | |
| "loss": 0.1026, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 0.660530590146183, | |
| "grad_norm": 0.95703125, | |
| "learning_rate": 1.963055729492799e-05, | |
| "loss": 0.1021, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 0.6632376827287493, | |
| "grad_norm": 1.4375, | |
| "learning_rate": 1.947401377582968e-05, | |
| "loss": 0.1098, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.6659447753113157, | |
| "grad_norm": 0.375, | |
| "learning_rate": 1.931747025673137e-05, | |
| "loss": 0.0828, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 0.668651867893882, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 1.9160926737633064e-05, | |
| "loss": 0.0612, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 0.6713589604764483, | |
| "grad_norm": 3.21875, | |
| "learning_rate": 1.9004383218534755e-05, | |
| "loss": 0.1367, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 0.6740660530590146, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 1.8847839699436446e-05, | |
| "loss": 0.0647, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 0.676773145641581, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 1.8691296180338137e-05, | |
| "loss": 0.0973, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.676773145641581, | |
| "eval_loss": 0.13495835661888123, | |
| "eval_runtime": 105.5301, | |
| "eval_samples_per_second": 4.852, | |
| "eval_steps_per_second": 0.152, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6794802382241473, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 1.8534752661239827e-05, | |
| "loss": 0.0775, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 0.6821873308067136, | |
| "grad_norm": 0.96484375, | |
| "learning_rate": 1.8378209142141518e-05, | |
| "loss": 0.0836, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 0.6848944233892799, | |
| "grad_norm": 2.5625, | |
| "learning_rate": 1.822166562304321e-05, | |
| "loss": 0.1118, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 0.6876015159718463, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 1.80651221039449e-05, | |
| "loss": 0.0889, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 0.6903086085544126, | |
| "grad_norm": 0.71484375, | |
| "learning_rate": 1.790857858484659e-05, | |
| "loss": 0.0982, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.6930157011369789, | |
| "grad_norm": 2.125, | |
| "learning_rate": 1.775203506574828e-05, | |
| "loss": 0.1131, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 0.6957227937195453, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 1.7595491546649968e-05, | |
| "loss": 0.1142, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 0.6984298863021116, | |
| "grad_norm": 0.3671875, | |
| "learning_rate": 1.743894802755166e-05, | |
| "loss": 0.0911, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 0.7011369788846779, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 1.728240450845335e-05, | |
| "loss": 0.0774, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 0.7038440714672441, | |
| "grad_norm": 0.9453125, | |
| "learning_rate": 1.712586098935504e-05, | |
| "loss": 0.0933, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.7038440714672441, | |
| "eval_loss": 0.13393962383270264, | |
| "eval_runtime": 103.5576, | |
| "eval_samples_per_second": 4.944, | |
| "eval_steps_per_second": 0.155, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.7065511640498106, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 1.696931747025673e-05, | |
| "loss": 0.1073, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 0.7092582566323768, | |
| "grad_norm": 2.203125, | |
| "learning_rate": 1.681277395115842e-05, | |
| "loss": 0.097, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 0.7119653492149431, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 1.6656230432060112e-05, | |
| "loss": 0.0772, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 0.7146724417975094, | |
| "grad_norm": 2.953125, | |
| "learning_rate": 1.6499686912961803e-05, | |
| "loss": 0.0935, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 0.7173795343800758, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 1.6343143393863494e-05, | |
| "loss": 0.0924, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.7200866269626421, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 1.6186599874765184e-05, | |
| "loss": 0.1057, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 0.7227937195452084, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 1.6030056355666875e-05, | |
| "loss": 0.0865, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 0.7255008121277747, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 1.5873512836568566e-05, | |
| "loss": 0.0957, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 0.7282079047103411, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 1.5716969317470257e-05, | |
| "loss": 0.107, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 0.7309149972929074, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 1.5560425798371947e-05, | |
| "loss": 0.0976, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.7309149972929074, | |
| "eval_loss": 0.13289867341518402, | |
| "eval_runtime": 104.5232, | |
| "eval_samples_per_second": 4.898, | |
| "eval_steps_per_second": 0.153, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.7336220898754737, | |
| "grad_norm": 0.34765625, | |
| "learning_rate": 1.5403882279273638e-05, | |
| "loss": 0.0918, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 0.73632918245804, | |
| "grad_norm": 1.6875, | |
| "learning_rate": 1.5247338760175329e-05, | |
| "loss": 0.1113, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 0.7390362750406064, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 1.5090795241077021e-05, | |
| "loss": 0.1221, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 0.7417433676231727, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 1.4934251721978712e-05, | |
| "loss": 0.1236, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 0.744450460205739, | |
| "grad_norm": 0.95703125, | |
| "learning_rate": 1.4777708202880403e-05, | |
| "loss": 0.119, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.7471575527883053, | |
| "grad_norm": 3.125, | |
| "learning_rate": 1.4621164683782093e-05, | |
| "loss": 0.1141, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 0.7498646453708717, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 1.4464621164683784e-05, | |
| "loss": 0.106, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 0.752571737953438, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 1.4308077645585475e-05, | |
| "loss": 0.0963, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 0.7552788305360043, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 1.4151534126487165e-05, | |
| "loss": 0.0946, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 0.7579859231185706, | |
| "grad_norm": 1.71875, | |
| "learning_rate": 1.3994990607388856e-05, | |
| "loss": 0.0985, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.7579859231185706, | |
| "eval_loss": 0.13054493069648743, | |
| "eval_runtime": 102.656, | |
| "eval_samples_per_second": 4.988, | |
| "eval_steps_per_second": 0.156, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.760693015701137, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 1.3838447088290547e-05, | |
| "loss": 0.1329, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 0.7634001082837033, | |
| "grad_norm": 2.21875, | |
| "learning_rate": 1.3681903569192236e-05, | |
| "loss": 0.1072, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 0.7661072008662696, | |
| "grad_norm": 1.390625, | |
| "learning_rate": 1.3525360050093926e-05, | |
| "loss": 0.0953, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 0.7688142934488359, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 1.3368816530995617e-05, | |
| "loss": 0.0912, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 0.7715213860314023, | |
| "grad_norm": 0.984375, | |
| "learning_rate": 1.3212273011897308e-05, | |
| "loss": 0.1132, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.7742284786139686, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 1.3055729492798999e-05, | |
| "loss": 0.1059, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 0.7769355711965349, | |
| "grad_norm": 0.99609375, | |
| "learning_rate": 1.289918597370069e-05, | |
| "loss": 0.1182, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 0.7796426637791013, | |
| "grad_norm": 0.921875, | |
| "learning_rate": 1.274264245460238e-05, | |
| "loss": 0.0647, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 0.7823497563616676, | |
| "grad_norm": 1.609375, | |
| "learning_rate": 1.258609893550407e-05, | |
| "loss": 0.1049, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 0.7850568489442339, | |
| "grad_norm": 2.53125, | |
| "learning_rate": 1.2429555416405761e-05, | |
| "loss": 0.1223, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.7850568489442339, | |
| "eval_loss": 0.13052764534950256, | |
| "eval_runtime": 114.8255, | |
| "eval_samples_per_second": 4.459, | |
| "eval_steps_per_second": 0.139, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.7877639415268002, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 1.2273011897307452e-05, | |
| "loss": 0.0768, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 0.7904710341093666, | |
| "grad_norm": 1.0, | |
| "learning_rate": 1.2116468378209143e-05, | |
| "loss": 0.0902, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 0.7931781266919329, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 1.1959924859110834e-05, | |
| "loss": 0.1059, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 0.7958852192744992, | |
| "grad_norm": 0.66796875, | |
| "learning_rate": 1.1803381340012524e-05, | |
| "loss": 0.0919, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 0.7985923118570655, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 1.1646837820914215e-05, | |
| "loss": 0.0779, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.8012994044396319, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 1.1490294301815906e-05, | |
| "loss": 0.1019, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 0.8040064970221982, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 1.1333750782717596e-05, | |
| "loss": 0.0909, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 0.8067135896047645, | |
| "grad_norm": 0.435546875, | |
| "learning_rate": 1.1177207263619287e-05, | |
| "loss": 0.0973, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 0.8094206821873308, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 1.1020663744520978e-05, | |
| "loss": 0.0875, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 0.8121277747698972, | |
| "grad_norm": 0.71484375, | |
| "learning_rate": 1.0864120225422668e-05, | |
| "loss": 0.1053, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.8121277747698972, | |
| "eval_loss": 0.13060268759727478, | |
| "eval_runtime": 108.9657, | |
| "eval_samples_per_second": 4.699, | |
| "eval_steps_per_second": 0.147, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.8148348673524635, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 1.070757670632436e-05, | |
| "loss": 0.0808, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 0.8175419599350298, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 1.0551033187226048e-05, | |
| "loss": 0.0766, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 0.8202490525175961, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 1.0394489668127739e-05, | |
| "loss": 0.073, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 0.8229561451001625, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 1.023794614902943e-05, | |
| "loss": 0.108, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 0.8256632376827288, | |
| "grad_norm": 0.69140625, | |
| "learning_rate": 1.008140262993112e-05, | |
| "loss": 0.0815, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.828370330265295, | |
| "grad_norm": 0.80859375, | |
| "learning_rate": 9.924859110832813e-06, | |
| "loss": 0.0965, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 0.8310774228478613, | |
| "grad_norm": 1.953125, | |
| "learning_rate": 9.768315591734503e-06, | |
| "loss": 0.1145, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 0.8337845154304278, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 9.611772072636194e-06, | |
| "loss": 0.1319, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 0.836491608012994, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 9.455228553537885e-06, | |
| "loss": 0.0871, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 0.8391987005955603, | |
| "grad_norm": 1.515625, | |
| "learning_rate": 9.298685034439576e-06, | |
| "loss": 0.1207, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.8391987005955603, | |
| "eval_loss": 0.1296384036540985, | |
| "eval_runtime": 103.7944, | |
| "eval_samples_per_second": 4.933, | |
| "eval_steps_per_second": 0.154, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.8419057931781266, | |
| "grad_norm": 2.203125, | |
| "learning_rate": 9.142141515341266e-06, | |
| "loss": 0.1177, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 0.844612885760693, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 8.985597996242955e-06, | |
| "loss": 0.1111, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 0.8473199783432593, | |
| "grad_norm": 0.384765625, | |
| "learning_rate": 8.829054477144646e-06, | |
| "loss": 0.0774, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 0.8500270709258256, | |
| "grad_norm": 1.3984375, | |
| "learning_rate": 8.672510958046337e-06, | |
| "loss": 0.0864, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 0.852734163508392, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 8.515967438948027e-06, | |
| "loss": 0.0914, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.8554412560909583, | |
| "grad_norm": 0.453125, | |
| "learning_rate": 8.359423919849718e-06, | |
| "loss": 0.0947, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 0.8581483486735246, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 8.202880400751409e-06, | |
| "loss": 0.0812, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 0.8608554412560909, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 8.0463368816531e-06, | |
| "loss": 0.0924, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 0.8635625338386573, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 7.889793362554792e-06, | |
| "loss": 0.1047, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 0.8662696264212236, | |
| "grad_norm": 2.0, | |
| "learning_rate": 7.733249843456483e-06, | |
| "loss": 0.1071, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.8662696264212236, | |
| "eval_loss": 0.128614142537117, | |
| "eval_runtime": 105.0877, | |
| "eval_samples_per_second": 4.872, | |
| "eval_steps_per_second": 0.152, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.8689767190037899, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 7.576706324358172e-06, | |
| "loss": 0.0834, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 0.8716838115863562, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 7.420162805259863e-06, | |
| "loss": 0.1098, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 0.8743909041689226, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 7.263619286161554e-06, | |
| "loss": 0.1152, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 0.8770979967514889, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 7.107075767063244e-06, | |
| "loss": 0.1093, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 0.8798050893340552, | |
| "grad_norm": 0.93359375, | |
| "learning_rate": 6.950532247964934e-06, | |
| "loss": 0.1116, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.8825121819166215, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 6.793988728866625e-06, | |
| "loss": 0.0824, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 0.8852192744991879, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 6.637445209768316e-06, | |
| "loss": 0.1241, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 0.8879263670817542, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 6.4809016906700065e-06, | |
| "loss": 0.0967, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 0.8906334596643205, | |
| "grad_norm": 1.125, | |
| "learning_rate": 6.324358171571697e-06, | |
| "loss": 0.0893, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 0.8933405522468868, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 6.167814652473388e-06, | |
| "loss": 0.1081, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.8933405522468868, | |
| "eval_loss": 0.12801626324653625, | |
| "eval_runtime": 103.0737, | |
| "eval_samples_per_second": 4.967, | |
| "eval_steps_per_second": 0.155, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.8960476448294532, | |
| "grad_norm": 1.3984375, | |
| "learning_rate": 6.011271133375079e-06, | |
| "loss": 0.1199, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 0.8987547374120195, | |
| "grad_norm": 1.640625, | |
| "learning_rate": 5.854727614276769e-06, | |
| "loss": 0.0997, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 0.9014618299945858, | |
| "grad_norm": 0.921875, | |
| "learning_rate": 5.69818409517846e-06, | |
| "loss": 0.1118, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 0.9041689225771521, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 5.54164057608015e-06, | |
| "loss": 0.083, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 0.9068760151597185, | |
| "grad_norm": 1.609375, | |
| "learning_rate": 5.3850970569818414e-06, | |
| "loss": 0.0765, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.9095831077422848, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 5.228553537883532e-06, | |
| "loss": 0.0925, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 0.9122902003248511, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 5.072010018785223e-06, | |
| "loss": 0.0941, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 0.9149972929074174, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 4.9154664996869136e-06, | |
| "loss": 0.1193, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 0.9177043854899838, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 4.758922980588603e-06, | |
| "loss": 0.1124, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 0.9204114780725501, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 4.602379461490294e-06, | |
| "loss": 0.0924, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.9204114780725501, | |
| "eval_loss": 0.12878485023975372, | |
| "eval_runtime": 123.0744, | |
| "eval_samples_per_second": 4.16, | |
| "eval_steps_per_second": 0.13, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.9231185706551164, | |
| "grad_norm": 0.86328125, | |
| "learning_rate": 4.445835942391985e-06, | |
| "loss": 0.0938, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 0.9258256632376828, | |
| "grad_norm": 2.328125, | |
| "learning_rate": 4.289292423293676e-06, | |
| "loss": 0.0839, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 0.9285327558202491, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 4.132748904195367e-06, | |
| "loss": 0.1368, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 0.9312398484028154, | |
| "grad_norm": 2.515625, | |
| "learning_rate": 3.976205385097057e-06, | |
| "loss": 0.1154, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 0.9339469409853817, | |
| "grad_norm": 0.205078125, | |
| "learning_rate": 3.819661865998748e-06, | |
| "loss": 0.095, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.9366540335679481, | |
| "grad_norm": 0.87890625, | |
| "learning_rate": 3.6631183469004384e-06, | |
| "loss": 0.0872, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 0.9393611261505144, | |
| "grad_norm": 2.515625, | |
| "learning_rate": 3.506574827802129e-06, | |
| "loss": 0.092, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 0.9420682187330807, | |
| "grad_norm": 1.125, | |
| "learning_rate": 3.35003130870382e-06, | |
| "loss": 0.0944, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 0.944775311315647, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 3.193487789605511e-06, | |
| "loss": 0.1366, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 0.9474824038982134, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 3.036944270507201e-06, | |
| "loss": 0.0948, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.9474824038982134, | |
| "eval_loss": 0.1291881799697876, | |
| "eval_runtime": 104.6704, | |
| "eval_samples_per_second": 4.892, | |
| "eval_steps_per_second": 0.153, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.9501894964807797, | |
| "grad_norm": 1.375, | |
| "learning_rate": 2.880400751408892e-06, | |
| "loss": 0.1181, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 0.952896589063346, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 2.7238572323105826e-06, | |
| "loss": 0.0819, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 0.9556036816459123, | |
| "grad_norm": 0.9765625, | |
| "learning_rate": 2.5673137132122733e-06, | |
| "loss": 0.1134, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 0.9583107742284787, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 2.410770194113964e-06, | |
| "loss": 0.1183, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 0.961017866811045, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 2.2542266750156543e-06, | |
| "loss": 0.1021, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.9637249593936112, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 2.0976831559173454e-06, | |
| "loss": 0.0871, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 0.9664320519761775, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 1.9411396368190357e-06, | |
| "loss": 0.1036, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 0.969139144558744, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 1.7845961177207264e-06, | |
| "loss": 0.0874, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 0.9718462371413102, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 1.6280525986224169e-06, | |
| "loss": 0.085, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 0.9745533297238765, | |
| "grad_norm": 0.90234375, | |
| "learning_rate": 1.4715090795241078e-06, | |
| "loss": 0.1162, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.9745533297238765, | |
| "eval_loss": 0.12912487983703613, | |
| "eval_runtime": 106.4924, | |
| "eval_samples_per_second": 4.808, | |
| "eval_steps_per_second": 0.15, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.9772604223064428, | |
| "grad_norm": 0.74609375, | |
| "learning_rate": 1.3149655604257985e-06, | |
| "loss": 0.1066, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 0.9799675148890092, | |
| "grad_norm": 1.4609375, | |
| "learning_rate": 1.1584220413274892e-06, | |
| "loss": 0.1533, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 0.9826746074715755, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 1.0018785222291797e-06, | |
| "loss": 0.0717, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 0.9853817000541418, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 8.453350031308704e-07, | |
| "loss": 0.0667, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 0.9880887926367081, | |
| "grad_norm": 0.427734375, | |
| "learning_rate": 6.887914840325611e-07, | |
| "loss": 0.1175, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.9907958852192745, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 5.322479649342517e-07, | |
| "loss": 0.0871, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 0.9935029778018408, | |
| "grad_norm": 1.8515625, | |
| "learning_rate": 3.757044458359424e-07, | |
| "loss": 0.122, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 0.9962100703844071, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 2.1916092673763307e-07, | |
| "loss": 0.0922, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 0.9989171629669734, | |
| "grad_norm": 3.140625, | |
| "learning_rate": 6.261740763932373e-08, | |
| "loss": 0.1331, | |
| "step": 3690 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 3694, | |
| "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": 1.0990508415397056e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |