diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,42015 @@ +{ + "best_metric": 0.9675748705246566, + "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-59946", + "epoch": 2.999924934317528, + "eval_steps": 500, + "global_step": 59946, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0005004378831477543, + "grad_norm": 24.303993225097656, + "learning_rate": 8.340283569641367e-08, + "loss": 1.5187, + "step": 10 + }, + { + "epoch": 0.0010008757662955086, + "grad_norm": 29.885019302368164, + "learning_rate": 1.6680567139282735e-07, + "loss": 1.5257, + "step": 20 + }, + { + "epoch": 0.0015013136494432628, + "grad_norm": 47.42334747314453, + "learning_rate": 2.5020850708924106e-07, + "loss": 1.4778, + "step": 30 + }, + { + "epoch": 0.0020017515325910173, + "grad_norm": 38.371612548828125, + "learning_rate": 3.336113427856547e-07, + "loss": 1.4342, + "step": 40 + }, + { + "epoch": 0.0025021894157387712, + "grad_norm": 24.121122360229492, + "learning_rate": 4.170141784820684e-07, + "loss": 1.5607, + "step": 50 + }, + { + "epoch": 0.0030026272988865257, + "grad_norm": 35.837249755859375, + "learning_rate": 5.004170141784821e-07, + "loss": 1.4844, + "step": 60 + }, + { + "epoch": 0.00350306518203428, + "grad_norm": 27.676471710205078, + "learning_rate": 5.838198498748957e-07, + "loss": 1.4639, + "step": 70 + }, + { + "epoch": 0.0040035030651820345, + "grad_norm": 30.950366973876953, + "learning_rate": 6.672226855713094e-07, + "loss": 1.495, + "step": 80 + }, + { + "epoch": 0.0045039409483297885, + "grad_norm": 22.30168914794922, + "learning_rate": 7.506255212677231e-07, + "loss": 1.4673, + "step": 90 + }, + { + "epoch": 0.0050043788314775425, + "grad_norm": 16.999237060546875, + "learning_rate": 8.340283569641368e-07, + "loss": 1.5575, + "step": 100 + }, + { + "epoch": 0.005504816714625297, + "grad_norm": 39.305023193359375, + "learning_rate": 9.174311926605506e-07, + "loss": 1.4464, + "step": 110 + }, + { + "epoch": 0.006005254597773051, + "grad_norm": 46.71931838989258, + "learning_rate": 1.0008340283569642e-06, + "loss": 1.3992, + "step": 120 + }, + { + "epoch": 0.006505692480920805, + "grad_norm": 26.411865234375, + "learning_rate": 1.084236864053378e-06, + "loss": 1.377, + "step": 130 + }, + { + "epoch": 0.00700613036406856, + "grad_norm": 20.488250732421875, + "learning_rate": 1.1676396997497914e-06, + "loss": 1.4191, + "step": 140 + }, + { + "epoch": 0.007506568247216314, + "grad_norm": 16.906946182250977, + "learning_rate": 1.2510425354462053e-06, + "loss": 1.4821, + "step": 150 + }, + { + "epoch": 0.008007006130364069, + "grad_norm": 19.9748592376709, + "learning_rate": 1.3344453711426188e-06, + "loss": 1.3907, + "step": 160 + }, + { + "epoch": 0.008507444013511823, + "grad_norm": 33.63947296142578, + "learning_rate": 1.4178482068390327e-06, + "loss": 1.4826, + "step": 170 + }, + { + "epoch": 0.009007881896659577, + "grad_norm": 17.32013511657715, + "learning_rate": 1.5012510425354462e-06, + "loss": 1.374, + "step": 180 + }, + { + "epoch": 0.009508319779807331, + "grad_norm": 24.52235221862793, + "learning_rate": 1.58465387823186e-06, + "loss": 1.3353, + "step": 190 + }, + { + "epoch": 0.010008757662955085, + "grad_norm": 29.115766525268555, + "learning_rate": 1.6680567139282735e-06, + "loss": 1.3277, + "step": 200 + }, + { + "epoch": 0.01050919554610284, + "grad_norm": 35.404964447021484, + "learning_rate": 1.7514595496246872e-06, + "loss": 1.4092, + "step": 210 + }, + { + "epoch": 0.011009633429250595, + "grad_norm": 34.37901306152344, + "learning_rate": 1.8348623853211011e-06, + "loss": 1.3217, + "step": 220 + }, + { + "epoch": 0.011510071312398349, + "grad_norm": 50.12200164794922, + "learning_rate": 1.918265221017515e-06, + "loss": 1.3067, + "step": 230 + }, + { + "epoch": 0.012010509195546103, + "grad_norm": 25.301254272460938, + "learning_rate": 2.0016680567139285e-06, + "loss": 1.2497, + "step": 240 + }, + { + "epoch": 0.012510947078693857, + "grad_norm": 23.887237548828125, + "learning_rate": 2.0850708924103418e-06, + "loss": 1.3173, + "step": 250 + }, + { + "epoch": 0.01301138496184161, + "grad_norm": 69.04029846191406, + "learning_rate": 2.168473728106756e-06, + "loss": 1.2979, + "step": 260 + }, + { + "epoch": 0.013511822844989366, + "grad_norm": 15.67342758178711, + "learning_rate": 2.2518765638031695e-06, + "loss": 1.2539, + "step": 270 + }, + { + "epoch": 0.01401226072813712, + "grad_norm": 17.66701889038086, + "learning_rate": 2.335279399499583e-06, + "loss": 1.2638, + "step": 280 + }, + { + "epoch": 0.014512698611284874, + "grad_norm": 36.377662658691406, + "learning_rate": 2.418682235195997e-06, + "loss": 1.2183, + "step": 290 + }, + { + "epoch": 0.015013136494432628, + "grad_norm": 15.283369064331055, + "learning_rate": 2.5020850708924106e-06, + "loss": 1.212, + "step": 300 + }, + { + "epoch": 0.015513574377580382, + "grad_norm": 23.049930572509766, + "learning_rate": 2.5854879065888243e-06, + "loss": 1.293, + "step": 310 + }, + { + "epoch": 0.016014012260728138, + "grad_norm": 15.622237205505371, + "learning_rate": 2.6688907422852376e-06, + "loss": 1.2436, + "step": 320 + }, + { + "epoch": 0.01651445014387589, + "grad_norm": 73.09809112548828, + "learning_rate": 2.7522935779816517e-06, + "loss": 1.3138, + "step": 330 + }, + { + "epoch": 0.017014888027023646, + "grad_norm": 36.036922454833984, + "learning_rate": 2.8356964136780654e-06, + "loss": 1.2595, + "step": 340 + }, + { + "epoch": 0.0175153259101714, + "grad_norm": 25.95122528076172, + "learning_rate": 2.919099249374479e-06, + "loss": 1.0891, + "step": 350 + }, + { + "epoch": 0.018015763793319154, + "grad_norm": 31.78199577331543, + "learning_rate": 3.0025020850708923e-06, + "loss": 1.2088, + "step": 360 + }, + { + "epoch": 0.01851620167646691, + "grad_norm": 34.28477096557617, + "learning_rate": 3.0859049207673064e-06, + "loss": 1.2346, + "step": 370 + }, + { + "epoch": 0.019016639559614662, + "grad_norm": 36.921974182128906, + "learning_rate": 3.16930775646372e-06, + "loss": 1.1984, + "step": 380 + }, + { + "epoch": 0.019517077442762418, + "grad_norm": 32.475669860839844, + "learning_rate": 3.252710592160134e-06, + "loss": 1.2, + "step": 390 + }, + { + "epoch": 0.02001751532591017, + "grad_norm": 27.83840560913086, + "learning_rate": 3.336113427856547e-06, + "loss": 1.2048, + "step": 400 + }, + { + "epoch": 0.020517953209057926, + "grad_norm": 31.08649444580078, + "learning_rate": 3.419516263552961e-06, + "loss": 1.1298, + "step": 410 + }, + { + "epoch": 0.02101839109220568, + "grad_norm": 35.307559967041016, + "learning_rate": 3.5029190992493744e-06, + "loss": 1.0511, + "step": 420 + }, + { + "epoch": 0.021518828975353434, + "grad_norm": 16.019481658935547, + "learning_rate": 3.586321934945788e-06, + "loss": 1.1649, + "step": 430 + }, + { + "epoch": 0.02201926685850119, + "grad_norm": 26.41493034362793, + "learning_rate": 3.6697247706422022e-06, + "loss": 1.2166, + "step": 440 + }, + { + "epoch": 0.02251970474164894, + "grad_norm": 25.900371551513672, + "learning_rate": 3.7531276063386155e-06, + "loss": 1.1236, + "step": 450 + }, + { + "epoch": 0.023020142624796697, + "grad_norm": 26.47336196899414, + "learning_rate": 3.83653044203503e-06, + "loss": 1.0567, + "step": 460 + }, + { + "epoch": 0.023520580507944453, + "grad_norm": 19.386762619018555, + "learning_rate": 3.919933277731443e-06, + "loss": 0.9619, + "step": 470 + }, + { + "epoch": 0.024021018391092205, + "grad_norm": 48.911983489990234, + "learning_rate": 4.003336113427857e-06, + "loss": 1.0638, + "step": 480 + }, + { + "epoch": 0.02452145627423996, + "grad_norm": 21.044219970703125, + "learning_rate": 4.086738949124271e-06, + "loss": 1.0993, + "step": 490 + }, + { + "epoch": 0.025021894157387713, + "grad_norm": 40.69071578979492, + "learning_rate": 4.1701417848206835e-06, + "loss": 1.0736, + "step": 500 + }, + { + "epoch": 0.02552233204053547, + "grad_norm": 21.76827049255371, + "learning_rate": 4.253544620517098e-06, + "loss": 0.9627, + "step": 510 + }, + { + "epoch": 0.02602276992368322, + "grad_norm": 14.909859657287598, + "learning_rate": 4.336947456213512e-06, + "loss": 1.0344, + "step": 520 + }, + { + "epoch": 0.026523207806830977, + "grad_norm": 28.29896354675293, + "learning_rate": 4.4203502919099246e-06, + "loss": 0.9905, + "step": 530 + }, + { + "epoch": 0.027023645689978733, + "grad_norm": 21.939638137817383, + "learning_rate": 4.503753127606339e-06, + "loss": 1.0076, + "step": 540 + }, + { + "epoch": 0.027524083573126485, + "grad_norm": 30.79641342163086, + "learning_rate": 4.587155963302753e-06, + "loss": 0.9605, + "step": 550 + }, + { + "epoch": 0.02802452145627424, + "grad_norm": 38.55771255493164, + "learning_rate": 4.670558798999166e-06, + "loss": 0.9963, + "step": 560 + }, + { + "epoch": 0.028524959339421993, + "grad_norm": 26.077362060546875, + "learning_rate": 4.75396163469558e-06, + "loss": 0.9092, + "step": 570 + }, + { + "epoch": 0.02902539722256975, + "grad_norm": 33.923362731933594, + "learning_rate": 4.837364470391994e-06, + "loss": 0.9774, + "step": 580 + }, + { + "epoch": 0.029525835105717504, + "grad_norm": 23.031034469604492, + "learning_rate": 4.9207673060884075e-06, + "loss": 1.0908, + "step": 590 + }, + { + "epoch": 0.030026272988865257, + "grad_norm": 19.210575103759766, + "learning_rate": 5.004170141784821e-06, + "loss": 0.8988, + "step": 600 + }, + { + "epoch": 0.030526710872013012, + "grad_norm": 40.24635696411133, + "learning_rate": 5.087572977481234e-06, + "loss": 0.7881, + "step": 610 + }, + { + "epoch": 0.031027148755160765, + "grad_norm": 45.417110443115234, + "learning_rate": 5.170975813177649e-06, + "loss": 0.8571, + "step": 620 + }, + { + "epoch": 0.03152758663830852, + "grad_norm": 17.2496337890625, + "learning_rate": 5.254378648874062e-06, + "loss": 0.8629, + "step": 630 + }, + { + "epoch": 0.032028024521456276, + "grad_norm": 34.87730026245117, + "learning_rate": 5.337781484570475e-06, + "loss": 0.9388, + "step": 640 + }, + { + "epoch": 0.03252846240460403, + "grad_norm": 27.393306732177734, + "learning_rate": 5.42118432026689e-06, + "loss": 0.8621, + "step": 650 + }, + { + "epoch": 0.03302890028775178, + "grad_norm": 37.76310729980469, + "learning_rate": 5.504587155963303e-06, + "loss": 0.9663, + "step": 660 + }, + { + "epoch": 0.033529338170899536, + "grad_norm": 31.907419204711914, + "learning_rate": 5.587989991659716e-06, + "loss": 0.9002, + "step": 670 + }, + { + "epoch": 0.03402977605404729, + "grad_norm": 74.0341796875, + "learning_rate": 5.671392827356131e-06, + "loss": 0.8941, + "step": 680 + }, + { + "epoch": 0.03453021393719505, + "grad_norm": 57.13862228393555, + "learning_rate": 5.7547956630525436e-06, + "loss": 0.8073, + "step": 690 + }, + { + "epoch": 0.0350306518203428, + "grad_norm": 13.777112007141113, + "learning_rate": 5.838198498748958e-06, + "loss": 0.9078, + "step": 700 + }, + { + "epoch": 0.03553108970349055, + "grad_norm": 26.033964157104492, + "learning_rate": 5.921601334445372e-06, + "loss": 0.8891, + "step": 710 + }, + { + "epoch": 0.03603152758663831, + "grad_norm": 23.96412467956543, + "learning_rate": 6.005004170141785e-06, + "loss": 0.7603, + "step": 720 + }, + { + "epoch": 0.036531965469786064, + "grad_norm": 128.5401153564453, + "learning_rate": 6.088407005838199e-06, + "loss": 0.7518, + "step": 730 + }, + { + "epoch": 0.03703240335293382, + "grad_norm": 37.72187805175781, + "learning_rate": 6.171809841534613e-06, + "loss": 0.8742, + "step": 740 + }, + { + "epoch": 0.03753284123608157, + "grad_norm": 18.735960006713867, + "learning_rate": 6.255212677231026e-06, + "loss": 0.9233, + "step": 750 + }, + { + "epoch": 0.038033279119229324, + "grad_norm": 22.319923400878906, + "learning_rate": 6.33861551292744e-06, + "loss": 0.9806, + "step": 760 + }, + { + "epoch": 0.03853371700237708, + "grad_norm": 27.96265983581543, + "learning_rate": 6.422018348623854e-06, + "loss": 0.7985, + "step": 770 + }, + { + "epoch": 0.039034154885524835, + "grad_norm": 31.731229782104492, + "learning_rate": 6.505421184320268e-06, + "loss": 0.884, + "step": 780 + }, + { + "epoch": 0.03953459276867259, + "grad_norm": 15.225069046020508, + "learning_rate": 6.5888240200166804e-06, + "loss": 0.7686, + "step": 790 + }, + { + "epoch": 0.04003503065182034, + "grad_norm": 44.5091552734375, + "learning_rate": 6.672226855713094e-06, + "loss": 0.8282, + "step": 800 + }, + { + "epoch": 0.040535468534968096, + "grad_norm": 24.462493896484375, + "learning_rate": 6.755629691409509e-06, + "loss": 0.6836, + "step": 810 + }, + { + "epoch": 0.04103590641811585, + "grad_norm": 23.01093101501465, + "learning_rate": 6.839032527105922e-06, + "loss": 0.7766, + "step": 820 + }, + { + "epoch": 0.04153634430126361, + "grad_norm": 69.99420928955078, + "learning_rate": 6.922435362802336e-06, + "loss": 0.8201, + "step": 830 + }, + { + "epoch": 0.04203678218441136, + "grad_norm": 24.048906326293945, + "learning_rate": 7.005838198498749e-06, + "loss": 0.7838, + "step": 840 + }, + { + "epoch": 0.04253722006755911, + "grad_norm": 24.80336570739746, + "learning_rate": 7.0892410341951625e-06, + "loss": 0.71, + "step": 850 + }, + { + "epoch": 0.04303765795070687, + "grad_norm": 31.412073135375977, + "learning_rate": 7.172643869891576e-06, + "loss": 0.8782, + "step": 860 + }, + { + "epoch": 0.04353809583385462, + "grad_norm": 20.90207290649414, + "learning_rate": 7.256046705587991e-06, + "loss": 0.8251, + "step": 870 + }, + { + "epoch": 0.04403853371700238, + "grad_norm": 40.478153228759766, + "learning_rate": 7.3394495412844045e-06, + "loss": 0.7006, + "step": 880 + }, + { + "epoch": 0.044538971600150135, + "grad_norm": 24.657398223876953, + "learning_rate": 7.422852376980818e-06, + "loss": 0.7654, + "step": 890 + }, + { + "epoch": 0.04503940948329788, + "grad_norm": 51.6373176574707, + "learning_rate": 7.506255212677231e-06, + "loss": 0.8252, + "step": 900 + }, + { + "epoch": 0.04553984736644564, + "grad_norm": 36.29755783081055, + "learning_rate": 7.589658048373645e-06, + "loss": 0.758, + "step": 910 + }, + { + "epoch": 0.046040285249593395, + "grad_norm": 30.626394271850586, + "learning_rate": 7.67306088407006e-06, + "loss": 0.7653, + "step": 920 + }, + { + "epoch": 0.04654072313274115, + "grad_norm": 32.421730041503906, + "learning_rate": 7.756463719766472e-06, + "loss": 0.8319, + "step": 930 + }, + { + "epoch": 0.047041161015888906, + "grad_norm": 28.329086303710938, + "learning_rate": 7.839866555462887e-06, + "loss": 0.6061, + "step": 940 + }, + { + "epoch": 0.047541598899036655, + "grad_norm": 56.08214569091797, + "learning_rate": 7.9232693911593e-06, + "loss": 0.641, + "step": 950 + }, + { + "epoch": 0.04804203678218441, + "grad_norm": 60.818199157714844, + "learning_rate": 8.006672226855714e-06, + "loss": 0.6565, + "step": 960 + }, + { + "epoch": 0.048542474665332166, + "grad_norm": 35.1428337097168, + "learning_rate": 8.090075062552127e-06, + "loss": 0.5296, + "step": 970 + }, + { + "epoch": 0.04904291254847992, + "grad_norm": 44.58134078979492, + "learning_rate": 8.173477898248541e-06, + "loss": 0.719, + "step": 980 + }, + { + "epoch": 0.04954335043162767, + "grad_norm": 20.013525009155273, + "learning_rate": 8.256880733944954e-06, + "loss": 0.6457, + "step": 990 + }, + { + "epoch": 0.05004378831477543, + "grad_norm": 67.25614166259766, + "learning_rate": 8.340283569641367e-06, + "loss": 0.7452, + "step": 1000 + }, + { + "epoch": 0.05054422619792318, + "grad_norm": 31.597511291503906, + "learning_rate": 8.423686405337782e-06, + "loss": 0.5147, + "step": 1010 + }, + { + "epoch": 0.05104466408107094, + "grad_norm": 46.437644958496094, + "learning_rate": 8.507089241034196e-06, + "loss": 0.8789, + "step": 1020 + }, + { + "epoch": 0.051545101964218694, + "grad_norm": 11.039417266845703, + "learning_rate": 8.590492076730609e-06, + "loss": 0.7296, + "step": 1030 + }, + { + "epoch": 0.05204553984736644, + "grad_norm": 32.3238410949707, + "learning_rate": 8.673894912427023e-06, + "loss": 0.7087, + "step": 1040 + }, + { + "epoch": 0.0525459777305142, + "grad_norm": 41.09556198120117, + "learning_rate": 8.757297748123438e-06, + "loss": 0.6381, + "step": 1050 + }, + { + "epoch": 0.053046415613661954, + "grad_norm": 48.648284912109375, + "learning_rate": 8.840700583819849e-06, + "loss": 0.5852, + "step": 1060 + }, + { + "epoch": 0.05354685349680971, + "grad_norm": 44.74356460571289, + "learning_rate": 8.924103419516264e-06, + "loss": 0.7914, + "step": 1070 + }, + { + "epoch": 0.054047291379957466, + "grad_norm": 50.12004089355469, + "learning_rate": 9.007506255212678e-06, + "loss": 0.5922, + "step": 1080 + }, + { + "epoch": 0.054547729263105214, + "grad_norm": 54.37643051147461, + "learning_rate": 9.090909090909091e-06, + "loss": 0.6813, + "step": 1090 + }, + { + "epoch": 0.05504816714625297, + "grad_norm": 30.132946014404297, + "learning_rate": 9.174311926605506e-06, + "loss": 0.6456, + "step": 1100 + }, + { + "epoch": 0.055548605029400726, + "grad_norm": 15.43805980682373, + "learning_rate": 9.257714762301918e-06, + "loss": 0.5431, + "step": 1110 + }, + { + "epoch": 0.05604904291254848, + "grad_norm": 37.299373626708984, + "learning_rate": 9.341117597998331e-06, + "loss": 0.5999, + "step": 1120 + }, + { + "epoch": 0.05654948079569624, + "grad_norm": 27.385665893554688, + "learning_rate": 9.424520433694746e-06, + "loss": 0.7204, + "step": 1130 + }, + { + "epoch": 0.057049918678843986, + "grad_norm": 38.93963623046875, + "learning_rate": 9.50792326939116e-06, + "loss": 0.7412, + "step": 1140 + }, + { + "epoch": 0.05755035656199174, + "grad_norm": 30.93650245666504, + "learning_rate": 9.591326105087573e-06, + "loss": 0.5788, + "step": 1150 + }, + { + "epoch": 0.0580507944451395, + "grad_norm": 22.00652503967285, + "learning_rate": 9.674728940783988e-06, + "loss": 0.5631, + "step": 1160 + }, + { + "epoch": 0.05855123232828725, + "grad_norm": 28.145112991333008, + "learning_rate": 9.7581317764804e-06, + "loss": 0.6284, + "step": 1170 + }, + { + "epoch": 0.05905167021143501, + "grad_norm": 63.07828140258789, + "learning_rate": 9.841534612176815e-06, + "loss": 0.7319, + "step": 1180 + }, + { + "epoch": 0.05955210809458276, + "grad_norm": 43.694602966308594, + "learning_rate": 9.924937447873228e-06, + "loss": 0.6624, + "step": 1190 + }, + { + "epoch": 0.06005254597773051, + "grad_norm": 33.687313079833984, + "learning_rate": 1.0008340283569642e-05, + "loss": 0.7009, + "step": 1200 + }, + { + "epoch": 0.06055298386087827, + "grad_norm": 19.009170532226562, + "learning_rate": 1.0091743119266055e-05, + "loss": 0.5602, + "step": 1210 + }, + { + "epoch": 0.061053421744026025, + "grad_norm": 91.3699722290039, + "learning_rate": 1.0175145954962468e-05, + "loss": 0.6736, + "step": 1220 + }, + { + "epoch": 0.061553859627173774, + "grad_norm": 14.598477363586426, + "learning_rate": 1.0258548790658883e-05, + "loss": 0.6962, + "step": 1230 + }, + { + "epoch": 0.06205429751032153, + "grad_norm": 26.503759384155273, + "learning_rate": 1.0341951626355297e-05, + "loss": 0.7887, + "step": 1240 + }, + { + "epoch": 0.06255473539346929, + "grad_norm": 23.208019256591797, + "learning_rate": 1.042535446205171e-05, + "loss": 0.6895, + "step": 1250 + }, + { + "epoch": 0.06305517327661704, + "grad_norm": 36.63803482055664, + "learning_rate": 1.0508757297748125e-05, + "loss": 0.5307, + "step": 1260 + }, + { + "epoch": 0.0635556111597648, + "grad_norm": 29.233863830566406, + "learning_rate": 1.0592160133444537e-05, + "loss": 0.5297, + "step": 1270 + }, + { + "epoch": 0.06405604904291255, + "grad_norm": 70.0718765258789, + "learning_rate": 1.067556296914095e-05, + "loss": 0.5183, + "step": 1280 + }, + { + "epoch": 0.06455648692606031, + "grad_norm": 4.175058841705322, + "learning_rate": 1.0758965804837365e-05, + "loss": 0.8038, + "step": 1290 + }, + { + "epoch": 0.06505692480920806, + "grad_norm": 81.0906753540039, + "learning_rate": 1.084236864053378e-05, + "loss": 0.7165, + "step": 1300 + }, + { + "epoch": 0.0655573626923558, + "grad_norm": 46.3349723815918, + "learning_rate": 1.0925771476230192e-05, + "loss": 0.7158, + "step": 1310 + }, + { + "epoch": 0.06605780057550356, + "grad_norm": 32.00333023071289, + "learning_rate": 1.1009174311926607e-05, + "loss": 0.5551, + "step": 1320 + }, + { + "epoch": 0.06655823845865132, + "grad_norm": 11.550634384155273, + "learning_rate": 1.109257714762302e-05, + "loss": 0.4565, + "step": 1330 + }, + { + "epoch": 0.06705867634179907, + "grad_norm": 25.393407821655273, + "learning_rate": 1.1175979983319432e-05, + "loss": 0.3608, + "step": 1340 + }, + { + "epoch": 0.06755911422494683, + "grad_norm": 50.75958251953125, + "learning_rate": 1.1259382819015847e-05, + "loss": 0.5784, + "step": 1350 + }, + { + "epoch": 0.06805955210809458, + "grad_norm": 19.823833465576172, + "learning_rate": 1.1342785654712261e-05, + "loss": 0.5616, + "step": 1360 + }, + { + "epoch": 0.06855998999124234, + "grad_norm": 103.9518814086914, + "learning_rate": 1.1426188490408674e-05, + "loss": 0.519, + "step": 1370 + }, + { + "epoch": 0.0690604278743901, + "grad_norm": 48.692405700683594, + "learning_rate": 1.1509591326105087e-05, + "loss": 0.4662, + "step": 1380 + }, + { + "epoch": 0.06956086575753785, + "grad_norm": 50.593990325927734, + "learning_rate": 1.1592994161801502e-05, + "loss": 0.6349, + "step": 1390 + }, + { + "epoch": 0.0700613036406856, + "grad_norm": 17.675006866455078, + "learning_rate": 1.1676396997497916e-05, + "loss": 0.6532, + "step": 1400 + }, + { + "epoch": 0.07056174152383335, + "grad_norm": 20.002723693847656, + "learning_rate": 1.1759799833194329e-05, + "loss": 0.4825, + "step": 1410 + }, + { + "epoch": 0.0710621794069811, + "grad_norm": 15.174713134765625, + "learning_rate": 1.1843202668890744e-05, + "loss": 0.7273, + "step": 1420 + }, + { + "epoch": 0.07156261729012886, + "grad_norm": 44.3665885925293, + "learning_rate": 1.1926605504587156e-05, + "loss": 0.65, + "step": 1430 + }, + { + "epoch": 0.07206305517327662, + "grad_norm": 16.98662757873535, + "learning_rate": 1.201000834028357e-05, + "loss": 0.582, + "step": 1440 + }, + { + "epoch": 0.07256349305642437, + "grad_norm": 23.304597854614258, + "learning_rate": 1.2093411175979984e-05, + "loss": 0.745, + "step": 1450 + }, + { + "epoch": 0.07306393093957213, + "grad_norm": 54.32075500488281, + "learning_rate": 1.2176814011676398e-05, + "loss": 0.577, + "step": 1460 + }, + { + "epoch": 0.07356436882271988, + "grad_norm": 43.47785568237305, + "learning_rate": 1.2260216847372811e-05, + "loss": 0.515, + "step": 1470 + }, + { + "epoch": 0.07406480670586764, + "grad_norm": 20.0679988861084, + "learning_rate": 1.2343619683069226e-05, + "loss": 0.5149, + "step": 1480 + }, + { + "epoch": 0.0745652445890154, + "grad_norm": 38.191890716552734, + "learning_rate": 1.2427022518765639e-05, + "loss": 0.5998, + "step": 1490 + }, + { + "epoch": 0.07506568247216314, + "grad_norm": 16.479337692260742, + "learning_rate": 1.2510425354462051e-05, + "loss": 0.5817, + "step": 1500 + }, + { + "epoch": 0.07556612035531089, + "grad_norm": 20.03363037109375, + "learning_rate": 1.2593828190158466e-05, + "loss": 0.374, + "step": 1510 + }, + { + "epoch": 0.07606655823845865, + "grad_norm": 33.73834991455078, + "learning_rate": 1.267723102585488e-05, + "loss": 0.5792, + "step": 1520 + }, + { + "epoch": 0.0765669961216064, + "grad_norm": 53.7574577331543, + "learning_rate": 1.2760633861551293e-05, + "loss": 0.6099, + "step": 1530 + }, + { + "epoch": 0.07706743400475416, + "grad_norm": 32.091312408447266, + "learning_rate": 1.2844036697247708e-05, + "loss": 0.4459, + "step": 1540 + }, + { + "epoch": 0.07756787188790192, + "grad_norm": 2.990734815597534, + "learning_rate": 1.2927439532944122e-05, + "loss": 0.4994, + "step": 1550 + }, + { + "epoch": 0.07806830977104967, + "grad_norm": 30.59862518310547, + "learning_rate": 1.3010842368640535e-05, + "loss": 0.4233, + "step": 1560 + }, + { + "epoch": 0.07856874765419743, + "grad_norm": 7.853830814361572, + "learning_rate": 1.3094245204336946e-05, + "loss": 0.5665, + "step": 1570 + }, + { + "epoch": 0.07906918553734518, + "grad_norm": 22.605350494384766, + "learning_rate": 1.3177648040033361e-05, + "loss": 0.7907, + "step": 1580 + }, + { + "epoch": 0.07956962342049294, + "grad_norm": 2.0066943168640137, + "learning_rate": 1.3261050875729775e-05, + "loss": 0.4201, + "step": 1590 + }, + { + "epoch": 0.08007006130364068, + "grad_norm": 1.4775019884109497, + "learning_rate": 1.3344453711426188e-05, + "loss": 0.6052, + "step": 1600 + }, + { + "epoch": 0.08057049918678844, + "grad_norm": 39.52610397338867, + "learning_rate": 1.3427856547122603e-05, + "loss": 0.5053, + "step": 1610 + }, + { + "epoch": 0.08107093706993619, + "grad_norm": 24.951831817626953, + "learning_rate": 1.3511259382819017e-05, + "loss": 0.4889, + "step": 1620 + }, + { + "epoch": 0.08157137495308395, + "grad_norm": 14.120335578918457, + "learning_rate": 1.359466221851543e-05, + "loss": 0.3476, + "step": 1630 + }, + { + "epoch": 0.0820718128362317, + "grad_norm": 23.290159225463867, + "learning_rate": 1.3678065054211845e-05, + "loss": 0.5953, + "step": 1640 + }, + { + "epoch": 0.08257225071937946, + "grad_norm": 20.84331512451172, + "learning_rate": 1.3761467889908258e-05, + "loss": 0.5022, + "step": 1650 + }, + { + "epoch": 0.08307268860252721, + "grad_norm": 20.95245361328125, + "learning_rate": 1.3844870725604672e-05, + "loss": 0.4222, + "step": 1660 + }, + { + "epoch": 0.08357312648567497, + "grad_norm": 86.65015411376953, + "learning_rate": 1.3928273561301087e-05, + "loss": 0.4198, + "step": 1670 + }, + { + "epoch": 0.08407356436882273, + "grad_norm": 10.353116035461426, + "learning_rate": 1.4011676396997498e-05, + "loss": 0.2987, + "step": 1680 + }, + { + "epoch": 0.08457400225197047, + "grad_norm": 46.748165130615234, + "learning_rate": 1.409507923269391e-05, + "loss": 0.5628, + "step": 1690 + }, + { + "epoch": 0.08507444013511822, + "grad_norm": 42.11788558959961, + "learning_rate": 1.4178482068390325e-05, + "loss": 0.6262, + "step": 1700 + }, + { + "epoch": 0.08557487801826598, + "grad_norm": 13.839410781860352, + "learning_rate": 1.426188490408674e-05, + "loss": 0.304, + "step": 1710 + }, + { + "epoch": 0.08607531590141373, + "grad_norm": 27.951698303222656, + "learning_rate": 1.4345287739783152e-05, + "loss": 0.8102, + "step": 1720 + }, + { + "epoch": 0.08657575378456149, + "grad_norm": 18.14162254333496, + "learning_rate": 1.4428690575479567e-05, + "loss": 0.5626, + "step": 1730 + }, + { + "epoch": 0.08707619166770925, + "grad_norm": 51.93001174926758, + "learning_rate": 1.4512093411175982e-05, + "loss": 0.7349, + "step": 1740 + }, + { + "epoch": 0.087576629550857, + "grad_norm": 10.264296531677246, + "learning_rate": 1.4595496246872394e-05, + "loss": 0.3083, + "step": 1750 + }, + { + "epoch": 0.08807706743400476, + "grad_norm": 43.271217346191406, + "learning_rate": 1.4678899082568809e-05, + "loss": 0.6291, + "step": 1760 + }, + { + "epoch": 0.08857750531715251, + "grad_norm": 39.189815521240234, + "learning_rate": 1.4762301918265223e-05, + "loss": 0.6731, + "step": 1770 + }, + { + "epoch": 0.08907794320030027, + "grad_norm": 31.52585220336914, + "learning_rate": 1.4845704753961636e-05, + "loss": 0.7766, + "step": 1780 + }, + { + "epoch": 0.08957838108344801, + "grad_norm": 7.720813751220703, + "learning_rate": 1.4929107589658047e-05, + "loss": 0.3344, + "step": 1790 + }, + { + "epoch": 0.09007881896659577, + "grad_norm": 68.28572082519531, + "learning_rate": 1.5012510425354462e-05, + "loss": 0.5244, + "step": 1800 + }, + { + "epoch": 0.09057925684974352, + "grad_norm": 5.61021614074707, + "learning_rate": 1.5095913261050877e-05, + "loss": 0.4397, + "step": 1810 + }, + { + "epoch": 0.09107969473289128, + "grad_norm": 12.357000350952148, + "learning_rate": 1.517931609674729e-05, + "loss": 0.4443, + "step": 1820 + }, + { + "epoch": 0.09158013261603903, + "grad_norm": 45.38551712036133, + "learning_rate": 1.5262718932443702e-05, + "loss": 0.4153, + "step": 1830 + }, + { + "epoch": 0.09208057049918679, + "grad_norm": 35.613075256347656, + "learning_rate": 1.534612176814012e-05, + "loss": 0.5348, + "step": 1840 + }, + { + "epoch": 0.09258100838233455, + "grad_norm": 33.981529235839844, + "learning_rate": 1.542952460383653e-05, + "loss": 0.7721, + "step": 1850 + }, + { + "epoch": 0.0930814462654823, + "grad_norm": 4.8698506355285645, + "learning_rate": 1.5512927439532944e-05, + "loss": 0.6501, + "step": 1860 + }, + { + "epoch": 0.09358188414863006, + "grad_norm": 27.787235260009766, + "learning_rate": 1.559633027522936e-05, + "loss": 0.5507, + "step": 1870 + }, + { + "epoch": 0.09408232203177781, + "grad_norm": 68.6563491821289, + "learning_rate": 1.5679733110925773e-05, + "loss": 0.8338, + "step": 1880 + }, + { + "epoch": 0.09458275991492555, + "grad_norm": 11.00479793548584, + "learning_rate": 1.5763135946622186e-05, + "loss": 0.7752, + "step": 1890 + }, + { + "epoch": 0.09508319779807331, + "grad_norm": 20.79509162902832, + "learning_rate": 1.58465387823186e-05, + "loss": 0.5113, + "step": 1900 + }, + { + "epoch": 0.09558363568122107, + "grad_norm": 47.05356979370117, + "learning_rate": 1.592994161801501e-05, + "loss": 0.5846, + "step": 1910 + }, + { + "epoch": 0.09608407356436882, + "grad_norm": 34.579734802246094, + "learning_rate": 1.6013344453711428e-05, + "loss": 0.3475, + "step": 1920 + }, + { + "epoch": 0.09658451144751658, + "grad_norm": 23.423057556152344, + "learning_rate": 1.609674728940784e-05, + "loss": 0.3787, + "step": 1930 + }, + { + "epoch": 0.09708494933066433, + "grad_norm": 22.126537322998047, + "learning_rate": 1.6180150125104254e-05, + "loss": 0.3639, + "step": 1940 + }, + { + "epoch": 0.09758538721381209, + "grad_norm": 67.84209442138672, + "learning_rate": 1.626355296080067e-05, + "loss": 0.5609, + "step": 1950 + }, + { + "epoch": 0.09808582509695984, + "grad_norm": 77.43997955322266, + "learning_rate": 1.6346955796497083e-05, + "loss": 0.4568, + "step": 1960 + }, + { + "epoch": 0.0985862629801076, + "grad_norm": 18.019269943237305, + "learning_rate": 1.6430358632193495e-05, + "loss": 0.3436, + "step": 1970 + }, + { + "epoch": 0.09908670086325534, + "grad_norm": 30.68067741394043, + "learning_rate": 1.651376146788991e-05, + "loss": 0.5339, + "step": 1980 + }, + { + "epoch": 0.0995871387464031, + "grad_norm": 29.789569854736328, + "learning_rate": 1.6597164303586325e-05, + "loss": 0.8205, + "step": 1990 + }, + { + "epoch": 0.10008757662955085, + "grad_norm": 6.085501670837402, + "learning_rate": 1.6680567139282734e-05, + "loss": 0.4888, + "step": 2000 + }, + { + "epoch": 0.10058801451269861, + "grad_norm": 53.11146545410156, + "learning_rate": 1.676396997497915e-05, + "loss": 0.5237, + "step": 2010 + }, + { + "epoch": 0.10108845239584636, + "grad_norm": 1.160164713859558, + "learning_rate": 1.6847372810675563e-05, + "loss": 0.2818, + "step": 2020 + }, + { + "epoch": 0.10158889027899412, + "grad_norm": 10.790987014770508, + "learning_rate": 1.6930775646371976e-05, + "loss": 0.3423, + "step": 2030 + }, + { + "epoch": 0.10208932816214188, + "grad_norm": 22.034494400024414, + "learning_rate": 1.7014178482068392e-05, + "loss": 0.4844, + "step": 2040 + }, + { + "epoch": 0.10258976604528963, + "grad_norm": 65.7363052368164, + "learning_rate": 1.7097581317764805e-05, + "loss": 0.5817, + "step": 2050 + }, + { + "epoch": 0.10309020392843739, + "grad_norm": 67.16058349609375, + "learning_rate": 1.7180984153461218e-05, + "loss": 0.8467, + "step": 2060 + }, + { + "epoch": 0.10359064181158514, + "grad_norm": 149.21482849121094, + "learning_rate": 1.7264386989157634e-05, + "loss": 0.5965, + "step": 2070 + }, + { + "epoch": 0.10409107969473289, + "grad_norm": 45.86624526977539, + "learning_rate": 1.7347789824854047e-05, + "loss": 0.369, + "step": 2080 + }, + { + "epoch": 0.10459151757788064, + "grad_norm": 1.615970492362976, + "learning_rate": 1.743119266055046e-05, + "loss": 0.4815, + "step": 2090 + }, + { + "epoch": 0.1050919554610284, + "grad_norm": 37.556331634521484, + "learning_rate": 1.7514595496246876e-05, + "loss": 0.3871, + "step": 2100 + }, + { + "epoch": 0.10559239334417615, + "grad_norm": 22.57245635986328, + "learning_rate": 1.7597998331943285e-05, + "loss": 0.3946, + "step": 2110 + }, + { + "epoch": 0.10609283122732391, + "grad_norm": 32.86416244506836, + "learning_rate": 1.7681401167639698e-05, + "loss": 0.4811, + "step": 2120 + }, + { + "epoch": 0.10659326911047166, + "grad_norm": 47.19029998779297, + "learning_rate": 1.7764804003336114e-05, + "loss": 0.3518, + "step": 2130 + }, + { + "epoch": 0.10709370699361942, + "grad_norm": 15.307121276855469, + "learning_rate": 1.7848206839032527e-05, + "loss": 0.3572, + "step": 2140 + }, + { + "epoch": 0.10759414487676718, + "grad_norm": 31.79294204711914, + "learning_rate": 1.793160967472894e-05, + "loss": 0.5434, + "step": 2150 + }, + { + "epoch": 0.10809458275991493, + "grad_norm": 1.7724000215530396, + "learning_rate": 1.8015012510425356e-05, + "loss": 0.5165, + "step": 2160 + }, + { + "epoch": 0.10859502064306267, + "grad_norm": 48.3746452331543, + "learning_rate": 1.809841534612177e-05, + "loss": 0.9163, + "step": 2170 + }, + { + "epoch": 0.10909545852621043, + "grad_norm": 17.02535057067871, + "learning_rate": 1.8181818181818182e-05, + "loss": 0.85, + "step": 2180 + }, + { + "epoch": 0.10959589640935818, + "grad_norm": 41.27964401245117, + "learning_rate": 1.8265221017514598e-05, + "loss": 0.647, + "step": 2190 + }, + { + "epoch": 0.11009633429250594, + "grad_norm": 24.74510955810547, + "learning_rate": 1.834862385321101e-05, + "loss": 0.6705, + "step": 2200 + }, + { + "epoch": 0.1105967721756537, + "grad_norm": 32.60014724731445, + "learning_rate": 1.8432026688907424e-05, + "loss": 0.6962, + "step": 2210 + }, + { + "epoch": 0.11109721005880145, + "grad_norm": 16.24282455444336, + "learning_rate": 1.8515429524603837e-05, + "loss": 0.4545, + "step": 2220 + }, + { + "epoch": 0.11159764794194921, + "grad_norm": 61.44316101074219, + "learning_rate": 1.859883236030025e-05, + "loss": 0.719, + "step": 2230 + }, + { + "epoch": 0.11209808582509696, + "grad_norm": 21.992538452148438, + "learning_rate": 1.8682235195996663e-05, + "loss": 0.5845, + "step": 2240 + }, + { + "epoch": 0.11259852370824472, + "grad_norm": 21.817018508911133, + "learning_rate": 1.876563803169308e-05, + "loss": 0.5753, + "step": 2250 + }, + { + "epoch": 0.11309896159139247, + "grad_norm": 20.465312957763672, + "learning_rate": 1.884904086738949e-05, + "loss": 0.5552, + "step": 2260 + }, + { + "epoch": 0.11359939947454022, + "grad_norm": 12.04063892364502, + "learning_rate": 1.8932443703085904e-05, + "loss": 0.5016, + "step": 2270 + }, + { + "epoch": 0.11409983735768797, + "grad_norm": 14.254342079162598, + "learning_rate": 1.901584653878232e-05, + "loss": 0.5084, + "step": 2280 + }, + { + "epoch": 0.11460027524083573, + "grad_norm": 63.92483901977539, + "learning_rate": 1.9099249374478733e-05, + "loss": 0.7559, + "step": 2290 + }, + { + "epoch": 0.11510071312398348, + "grad_norm": 28.907207489013672, + "learning_rate": 1.9182652210175146e-05, + "loss": 0.352, + "step": 2300 + }, + { + "epoch": 0.11560115100713124, + "grad_norm": 18.336408615112305, + "learning_rate": 1.9266055045871563e-05, + "loss": 0.5463, + "step": 2310 + }, + { + "epoch": 0.116101588890279, + "grad_norm": 40.83494567871094, + "learning_rate": 1.9349457881567975e-05, + "loss": 0.6936, + "step": 2320 + }, + { + "epoch": 0.11660202677342675, + "grad_norm": 28.904827117919922, + "learning_rate": 1.9432860717264388e-05, + "loss": 0.5234, + "step": 2330 + }, + { + "epoch": 0.1171024646565745, + "grad_norm": 89.64140319824219, + "learning_rate": 1.95162635529608e-05, + "loss": 0.6567, + "step": 2340 + }, + { + "epoch": 0.11760290253972226, + "grad_norm": 32.0169792175293, + "learning_rate": 1.9599666388657214e-05, + "loss": 0.3438, + "step": 2350 + }, + { + "epoch": 0.11810334042287002, + "grad_norm": 36.8966178894043, + "learning_rate": 1.968306922435363e-05, + "loss": 0.3215, + "step": 2360 + }, + { + "epoch": 0.11860377830601776, + "grad_norm": 3.091876268386841, + "learning_rate": 1.9766472060050043e-05, + "loss": 0.3052, + "step": 2370 + }, + { + "epoch": 0.11910421618916552, + "grad_norm": 31.58437728881836, + "learning_rate": 1.9849874895746456e-05, + "loss": 0.4461, + "step": 2380 + }, + { + "epoch": 0.11960465407231327, + "grad_norm": 2.9002201557159424, + "learning_rate": 1.9933277731442872e-05, + "loss": 0.515, + "step": 2390 + }, + { + "epoch": 0.12010509195546103, + "grad_norm": 28.363834381103516, + "learning_rate": 2.0016680567139285e-05, + "loss": 0.5174, + "step": 2400 + }, + { + "epoch": 0.12060552983860878, + "grad_norm": 107.77320861816406, + "learning_rate": 2.0100083402835698e-05, + "loss": 0.4137, + "step": 2410 + }, + { + "epoch": 0.12110596772175654, + "grad_norm": 87.85038757324219, + "learning_rate": 2.018348623853211e-05, + "loss": 0.5113, + "step": 2420 + }, + { + "epoch": 0.1216064056049043, + "grad_norm": 84.22132873535156, + "learning_rate": 2.0266889074228527e-05, + "loss": 0.8378, + "step": 2430 + }, + { + "epoch": 0.12210684348805205, + "grad_norm": 36.95043182373047, + "learning_rate": 2.0350291909924936e-05, + "loss": 0.4783, + "step": 2440 + }, + { + "epoch": 0.1226072813711998, + "grad_norm": 0.8114583492279053, + "learning_rate": 2.0433694745621352e-05, + "loss": 0.7182, + "step": 2450 + }, + { + "epoch": 0.12310771925434755, + "grad_norm": 41.86635208129883, + "learning_rate": 2.0517097581317765e-05, + "loss": 0.496, + "step": 2460 + }, + { + "epoch": 0.1236081571374953, + "grad_norm": 70.68716430664062, + "learning_rate": 2.0600500417014178e-05, + "loss": 0.9629, + "step": 2470 + }, + { + "epoch": 0.12410859502064306, + "grad_norm": 68.03734588623047, + "learning_rate": 2.0683903252710594e-05, + "loss": 0.9982, + "step": 2480 + }, + { + "epoch": 0.12460903290379081, + "grad_norm": 6.974307060241699, + "learning_rate": 2.0767306088407007e-05, + "loss": 0.8146, + "step": 2490 + }, + { + "epoch": 0.12510947078693857, + "grad_norm": 18.01585578918457, + "learning_rate": 2.085070892410342e-05, + "loss": 0.3707, + "step": 2500 + }, + { + "epoch": 0.1256099086700863, + "grad_norm": 24.565736770629883, + "learning_rate": 2.0934111759799836e-05, + "loss": 0.4287, + "step": 2510 + }, + { + "epoch": 0.12611034655323408, + "grad_norm": 8.185214042663574, + "learning_rate": 2.101751459549625e-05, + "loss": 0.4095, + "step": 2520 + }, + { + "epoch": 0.12661078443638182, + "grad_norm": 14.92050552368164, + "learning_rate": 2.1100917431192662e-05, + "loss": 0.4699, + "step": 2530 + }, + { + "epoch": 0.1271112223195296, + "grad_norm": 44.73469543457031, + "learning_rate": 2.1184320266889075e-05, + "loss": 0.6895, + "step": 2540 + }, + { + "epoch": 0.12761166020267734, + "grad_norm": 122.71125030517578, + "learning_rate": 2.1267723102585488e-05, + "loss": 0.4131, + "step": 2550 + }, + { + "epoch": 0.1281120980858251, + "grad_norm": 61.24140930175781, + "learning_rate": 2.13511259382819e-05, + "loss": 0.613, + "step": 2560 + }, + { + "epoch": 0.12861253596897285, + "grad_norm": 13.875518798828125, + "learning_rate": 2.1434528773978317e-05, + "loss": 0.2627, + "step": 2570 + }, + { + "epoch": 0.12911297385212062, + "grad_norm": 39.86324691772461, + "learning_rate": 2.151793160967473e-05, + "loss": 0.5763, + "step": 2580 + }, + { + "epoch": 0.12961341173526836, + "grad_norm": 18.45928192138672, + "learning_rate": 2.1601334445371142e-05, + "loss": 0.6009, + "step": 2590 + }, + { + "epoch": 0.13011384961841613, + "grad_norm": 31.145288467407227, + "learning_rate": 2.168473728106756e-05, + "loss": 0.4974, + "step": 2600 + }, + { + "epoch": 0.13061428750156387, + "grad_norm": 75.06790924072266, + "learning_rate": 2.176814011676397e-05, + "loss": 0.4267, + "step": 2610 + }, + { + "epoch": 0.1311147253847116, + "grad_norm": 31.58403778076172, + "learning_rate": 2.1851542952460384e-05, + "loss": 0.3985, + "step": 2620 + }, + { + "epoch": 0.13161516326785938, + "grad_norm": 140.95263671875, + "learning_rate": 2.19349457881568e-05, + "loss": 0.792, + "step": 2630 + }, + { + "epoch": 0.13211560115100712, + "grad_norm": 51.8648796081543, + "learning_rate": 2.2018348623853213e-05, + "loss": 0.5607, + "step": 2640 + }, + { + "epoch": 0.1326160390341549, + "grad_norm": 64.62076568603516, + "learning_rate": 2.2101751459549623e-05, + "loss": 0.69, + "step": 2650 + }, + { + "epoch": 0.13311647691730263, + "grad_norm": 7.515187740325928, + "learning_rate": 2.218515429524604e-05, + "loss": 0.6683, + "step": 2660 + }, + { + "epoch": 0.1336169148004504, + "grad_norm": 45.4011344909668, + "learning_rate": 2.2268557130942452e-05, + "loss": 0.9029, + "step": 2670 + }, + { + "epoch": 0.13411735268359815, + "grad_norm": 73.5013656616211, + "learning_rate": 2.2351959966638865e-05, + "loss": 0.6211, + "step": 2680 + }, + { + "epoch": 0.13461779056674592, + "grad_norm": 39.82784652709961, + "learning_rate": 2.243536280233528e-05, + "loss": 0.6752, + "step": 2690 + }, + { + "epoch": 0.13511822844989366, + "grad_norm": 11.46080493927002, + "learning_rate": 2.2518765638031694e-05, + "loss": 0.5305, + "step": 2700 + }, + { + "epoch": 0.1356186663330414, + "grad_norm": 24.587520599365234, + "learning_rate": 2.2602168473728107e-05, + "loss": 0.4137, + "step": 2710 + }, + { + "epoch": 0.13611910421618917, + "grad_norm": 32.35281753540039, + "learning_rate": 2.2685571309424523e-05, + "loss": 0.5203, + "step": 2720 + }, + { + "epoch": 0.1366195420993369, + "grad_norm": 17.56407928466797, + "learning_rate": 2.2768974145120936e-05, + "loss": 0.3729, + "step": 2730 + }, + { + "epoch": 0.13711997998248468, + "grad_norm": 30.782974243164062, + "learning_rate": 2.285237698081735e-05, + "loss": 0.7633, + "step": 2740 + }, + { + "epoch": 0.13762041786563242, + "grad_norm": 30.59442710876465, + "learning_rate": 2.2935779816513765e-05, + "loss": 0.4511, + "step": 2750 + }, + { + "epoch": 0.1381208557487802, + "grad_norm": 39.69466781616211, + "learning_rate": 2.3019182652210174e-05, + "loss": 0.5177, + "step": 2760 + }, + { + "epoch": 0.13862129363192793, + "grad_norm": 13.784320831298828, + "learning_rate": 2.310258548790659e-05, + "loss": 0.3371, + "step": 2770 + }, + { + "epoch": 0.1391217315150757, + "grad_norm": 31.712787628173828, + "learning_rate": 2.3185988323603003e-05, + "loss": 0.2439, + "step": 2780 + }, + { + "epoch": 0.13962216939822344, + "grad_norm": 78.91867065429688, + "learning_rate": 2.3269391159299416e-05, + "loss": 0.7662, + "step": 2790 + }, + { + "epoch": 0.1401226072813712, + "grad_norm": 71.55599975585938, + "learning_rate": 2.3352793994995832e-05, + "loss": 1.0931, + "step": 2800 + }, + { + "epoch": 0.14062304516451896, + "grad_norm": 44.56541442871094, + "learning_rate": 2.3436196830692245e-05, + "loss": 0.4296, + "step": 2810 + }, + { + "epoch": 0.1411234830476667, + "grad_norm": 0.8325983285903931, + "learning_rate": 2.3519599666388658e-05, + "loss": 0.4102, + "step": 2820 + }, + { + "epoch": 0.14162392093081447, + "grad_norm": 42.29323959350586, + "learning_rate": 2.3603002502085074e-05, + "loss": 0.4475, + "step": 2830 + }, + { + "epoch": 0.1421243588139622, + "grad_norm": 0.9139430522918701, + "learning_rate": 2.3686405337781487e-05, + "loss": 0.4216, + "step": 2840 + }, + { + "epoch": 0.14262479669710998, + "grad_norm": 18.860942840576172, + "learning_rate": 2.37698081734779e-05, + "loss": 0.3037, + "step": 2850 + }, + { + "epoch": 0.14312523458025772, + "grad_norm": 66.13958740234375, + "learning_rate": 2.3853211009174313e-05, + "loss": 0.3762, + "step": 2860 + }, + { + "epoch": 0.1436256724634055, + "grad_norm": 1.4282374382019043, + "learning_rate": 2.3936613844870726e-05, + "loss": 0.7067, + "step": 2870 + }, + { + "epoch": 0.14412611034655323, + "grad_norm": 20.335763931274414, + "learning_rate": 2.402001668056714e-05, + "loss": 0.5328, + "step": 2880 + }, + { + "epoch": 0.144626548229701, + "grad_norm": 42.31678009033203, + "learning_rate": 2.4103419516263555e-05, + "loss": 0.6766, + "step": 2890 + }, + { + "epoch": 0.14512698611284874, + "grad_norm": 53.43568801879883, + "learning_rate": 2.4186822351959968e-05, + "loss": 0.2468, + "step": 2900 + }, + { + "epoch": 0.14562742399599649, + "grad_norm": 61.73591995239258, + "learning_rate": 2.427022518765638e-05, + "loss": 0.5827, + "step": 2910 + }, + { + "epoch": 0.14612786187914426, + "grad_norm": 5.069060802459717, + "learning_rate": 2.4353628023352797e-05, + "loss": 0.9897, + "step": 2920 + }, + { + "epoch": 0.146628299762292, + "grad_norm": 5.480781078338623, + "learning_rate": 2.443703085904921e-05, + "loss": 0.5293, + "step": 2930 + }, + { + "epoch": 0.14712873764543977, + "grad_norm": 75.21513366699219, + "learning_rate": 2.4520433694745622e-05, + "loss": 0.4037, + "step": 2940 + }, + { + "epoch": 0.1476291755285875, + "grad_norm": 114.32780456542969, + "learning_rate": 2.460383653044204e-05, + "loss": 0.6513, + "step": 2950 + }, + { + "epoch": 0.14812961341173528, + "grad_norm": 122.11763763427734, + "learning_rate": 2.468723936613845e-05, + "loss": 0.4845, + "step": 2960 + }, + { + "epoch": 0.14863005129488302, + "grad_norm": 43.1311149597168, + "learning_rate": 2.4770642201834864e-05, + "loss": 0.4416, + "step": 2970 + }, + { + "epoch": 0.1491304891780308, + "grad_norm": 32.31309509277344, + "learning_rate": 2.4854045037531277e-05, + "loss": 0.6019, + "step": 2980 + }, + { + "epoch": 0.14963092706117853, + "grad_norm": 27.17719078063965, + "learning_rate": 2.493744787322769e-05, + "loss": 0.7143, + "step": 2990 + }, + { + "epoch": 0.15013136494432627, + "grad_norm": 71.85342407226562, + "learning_rate": 2.5020850708924103e-05, + "loss": 0.4247, + "step": 3000 + }, + { + "epoch": 0.15063180282747404, + "grad_norm": 10.430753707885742, + "learning_rate": 2.5104253544620522e-05, + "loss": 0.3543, + "step": 3010 + }, + { + "epoch": 0.15113224071062178, + "grad_norm": 14.872661590576172, + "learning_rate": 2.5187656380316932e-05, + "loss": 0.2975, + "step": 3020 + }, + { + "epoch": 0.15163267859376955, + "grad_norm": 44.741233825683594, + "learning_rate": 2.5271059216013345e-05, + "loss": 0.4277, + "step": 3030 + }, + { + "epoch": 0.1521331164769173, + "grad_norm": 66.55695343017578, + "learning_rate": 2.535446205170976e-05, + "loss": 0.498, + "step": 3040 + }, + { + "epoch": 0.15263355436006507, + "grad_norm": 26.916732788085938, + "learning_rate": 2.543786488740617e-05, + "loss": 0.3849, + "step": 3050 + }, + { + "epoch": 0.1531339922432128, + "grad_norm": 1.4033081531524658, + "learning_rate": 2.5521267723102587e-05, + "loss": 0.6496, + "step": 3060 + }, + { + "epoch": 0.15363443012636058, + "grad_norm": 13.1010103225708, + "learning_rate": 2.5604670558799e-05, + "loss": 0.701, + "step": 3070 + }, + { + "epoch": 0.15413486800950832, + "grad_norm": 48.619449615478516, + "learning_rate": 2.5688073394495416e-05, + "loss": 0.5892, + "step": 3080 + }, + { + "epoch": 0.15463530589265606, + "grad_norm": 65.61621856689453, + "learning_rate": 2.5771476230191825e-05, + "loss": 0.5606, + "step": 3090 + }, + { + "epoch": 0.15513574377580383, + "grad_norm": 35.73674774169922, + "learning_rate": 2.5854879065888245e-05, + "loss": 0.5266, + "step": 3100 + }, + { + "epoch": 0.15563618165895157, + "grad_norm": 12.863226890563965, + "learning_rate": 2.5938281901584654e-05, + "loss": 0.9057, + "step": 3110 + }, + { + "epoch": 0.15613661954209934, + "grad_norm": 81.14071655273438, + "learning_rate": 2.602168473728107e-05, + "loss": 0.6145, + "step": 3120 + }, + { + "epoch": 0.15663705742524708, + "grad_norm": 0.44801774621009827, + "learning_rate": 2.6105087572977483e-05, + "loss": 0.5483, + "step": 3130 + }, + { + "epoch": 0.15713749530839485, + "grad_norm": 72.50271606445312, + "learning_rate": 2.6188490408673893e-05, + "loss": 0.5081, + "step": 3140 + }, + { + "epoch": 0.1576379331915426, + "grad_norm": 22.548009872436523, + "learning_rate": 2.627189324437031e-05, + "loss": 0.8583, + "step": 3150 + }, + { + "epoch": 0.15813837107469036, + "grad_norm": 0.22425808012485504, + "learning_rate": 2.6355296080066722e-05, + "loss": 0.5214, + "step": 3160 + }, + { + "epoch": 0.1586388089578381, + "grad_norm": 61.865840911865234, + "learning_rate": 2.6438698915763138e-05, + "loss": 0.4202, + "step": 3170 + }, + { + "epoch": 0.15913924684098588, + "grad_norm": 46.971656799316406, + "learning_rate": 2.652210175145955e-05, + "loss": 0.4391, + "step": 3180 + }, + { + "epoch": 0.15963968472413362, + "grad_norm": 1.495849609375, + "learning_rate": 2.6605504587155967e-05, + "loss": 0.3708, + "step": 3190 + }, + { + "epoch": 0.16014012260728136, + "grad_norm": 39.14710235595703, + "learning_rate": 2.6688907422852376e-05, + "loss": 0.8249, + "step": 3200 + }, + { + "epoch": 0.16064056049042913, + "grad_norm": 14.552321434020996, + "learning_rate": 2.6772310258548793e-05, + "loss": 0.206, + "step": 3210 + }, + { + "epoch": 0.16114099837357687, + "grad_norm": 46.5123176574707, + "learning_rate": 2.6855713094245206e-05, + "loss": 1.0206, + "step": 3220 + }, + { + "epoch": 0.16164143625672464, + "grad_norm": 4.732786655426025, + "learning_rate": 2.6939115929941622e-05, + "loss": 0.512, + "step": 3230 + }, + { + "epoch": 0.16214187413987238, + "grad_norm": 40.3809928894043, + "learning_rate": 2.7022518765638035e-05, + "loss": 0.3538, + "step": 3240 + }, + { + "epoch": 0.16264231202302015, + "grad_norm": 35.87002944946289, + "learning_rate": 2.7105921601334444e-05, + "loss": 0.4894, + "step": 3250 + }, + { + "epoch": 0.1631427499061679, + "grad_norm": 23.920412063598633, + "learning_rate": 2.718932443703086e-05, + "loss": 0.6194, + "step": 3260 + }, + { + "epoch": 0.16364318778931566, + "grad_norm": 13.700675010681152, + "learning_rate": 2.7272727272727273e-05, + "loss": 0.5375, + "step": 3270 + }, + { + "epoch": 0.1641436256724634, + "grad_norm": 31.935976028442383, + "learning_rate": 2.735613010842369e-05, + "loss": 0.6258, + "step": 3280 + }, + { + "epoch": 0.16464406355561115, + "grad_norm": 17.32752799987793, + "learning_rate": 2.74395329441201e-05, + "loss": 0.5131, + "step": 3290 + }, + { + "epoch": 0.16514450143875892, + "grad_norm": 19.947324752807617, + "learning_rate": 2.7522935779816515e-05, + "loss": 0.6431, + "step": 3300 + }, + { + "epoch": 0.16564493932190666, + "grad_norm": 8.365926742553711, + "learning_rate": 2.7606338615512928e-05, + "loss": 0.2434, + "step": 3310 + }, + { + "epoch": 0.16614537720505443, + "grad_norm": 37.22792434692383, + "learning_rate": 2.7689741451209344e-05, + "loss": 0.4667, + "step": 3320 + }, + { + "epoch": 0.16664581508820217, + "grad_norm": 43.424949645996094, + "learning_rate": 2.7773144286905757e-05, + "loss": 0.4392, + "step": 3330 + }, + { + "epoch": 0.16714625297134994, + "grad_norm": 31.40007972717285, + "learning_rate": 2.7856547122602173e-05, + "loss": 0.3765, + "step": 3340 + }, + { + "epoch": 0.16764669085449768, + "grad_norm": 18.484201431274414, + "learning_rate": 2.7939949958298583e-05, + "loss": 0.465, + "step": 3350 + }, + { + "epoch": 0.16814712873764545, + "grad_norm": 58.2779655456543, + "learning_rate": 2.8023352793994995e-05, + "loss": 0.6239, + "step": 3360 + }, + { + "epoch": 0.1686475666207932, + "grad_norm": 30.20337677001953, + "learning_rate": 2.810675562969141e-05, + "loss": 0.4938, + "step": 3370 + }, + { + "epoch": 0.16914800450394094, + "grad_norm": 102.58242797851562, + "learning_rate": 2.819015846538782e-05, + "loss": 0.8335, + "step": 3380 + }, + { + "epoch": 0.1696484423870887, + "grad_norm": 52.82198715209961, + "learning_rate": 2.827356130108424e-05, + "loss": 0.1389, + "step": 3390 + }, + { + "epoch": 0.17014888027023645, + "grad_norm": 55.281734466552734, + "learning_rate": 2.835696413678065e-05, + "loss": 0.5939, + "step": 3400 + }, + { + "epoch": 0.17064931815338422, + "grad_norm": 13.0058012008667, + "learning_rate": 2.8440366972477066e-05, + "loss": 0.7177, + "step": 3410 + }, + { + "epoch": 0.17114975603653196, + "grad_norm": 36.97469711303711, + "learning_rate": 2.852376980817348e-05, + "loss": 0.5026, + "step": 3420 + }, + { + "epoch": 0.17165019391967973, + "grad_norm": 1.2471802234649658, + "learning_rate": 2.8607172643869895e-05, + "loss": 0.4368, + "step": 3430 + }, + { + "epoch": 0.17215063180282747, + "grad_norm": 3.8690056800842285, + "learning_rate": 2.8690575479566305e-05, + "loss": 0.4155, + "step": 3440 + }, + { + "epoch": 0.17265106968597524, + "grad_norm": 8.443232536315918, + "learning_rate": 2.8773978315262725e-05, + "loss": 0.4481, + "step": 3450 + }, + { + "epoch": 0.17315150756912298, + "grad_norm": 4.827065944671631, + "learning_rate": 2.8857381150959134e-05, + "loss": 0.5168, + "step": 3460 + }, + { + "epoch": 0.17365194545227075, + "grad_norm": 14.029801368713379, + "learning_rate": 2.8940783986655547e-05, + "loss": 0.3654, + "step": 3470 + }, + { + "epoch": 0.1741523833354185, + "grad_norm": 4.397176742553711, + "learning_rate": 2.9024186822351963e-05, + "loss": 0.4886, + "step": 3480 + }, + { + "epoch": 0.17465282121856623, + "grad_norm": 32.087982177734375, + "learning_rate": 2.9107589658048373e-05, + "loss": 0.7024, + "step": 3490 + }, + { + "epoch": 0.175153259101714, + "grad_norm": 1.0698304176330566, + "learning_rate": 2.919099249374479e-05, + "loss": 0.3082, + "step": 3500 + }, + { + "epoch": 0.17565369698486175, + "grad_norm": 1.3044782876968384, + "learning_rate": 2.92743953294412e-05, + "loss": 0.54, + "step": 3510 + }, + { + "epoch": 0.17615413486800952, + "grad_norm": 60.2294921875, + "learning_rate": 2.9357798165137618e-05, + "loss": 0.5323, + "step": 3520 + }, + { + "epoch": 0.17665457275115726, + "grad_norm": 54.82414627075195, + "learning_rate": 2.9441201000834027e-05, + "loss": 0.5112, + "step": 3530 + }, + { + "epoch": 0.17715501063430503, + "grad_norm": 4.5659332275390625, + "learning_rate": 2.9524603836530447e-05, + "loss": 0.6121, + "step": 3540 + }, + { + "epoch": 0.17765544851745277, + "grad_norm": 0.059190597385168076, + "learning_rate": 2.9608006672226856e-05, + "loss": 0.7381, + "step": 3550 + }, + { + "epoch": 0.17815588640060054, + "grad_norm": 53.288291931152344, + "learning_rate": 2.9691409507923273e-05, + "loss": 0.6173, + "step": 3560 + }, + { + "epoch": 0.17865632428374828, + "grad_norm": 56.822933197021484, + "learning_rate": 2.9774812343619685e-05, + "loss": 0.7307, + "step": 3570 + }, + { + "epoch": 0.17915676216689602, + "grad_norm": 46.482940673828125, + "learning_rate": 2.9858215179316095e-05, + "loss": 0.7157, + "step": 3580 + }, + { + "epoch": 0.1796572000500438, + "grad_norm": 30.860034942626953, + "learning_rate": 2.994161801501251e-05, + "loss": 0.6135, + "step": 3590 + }, + { + "epoch": 0.18015763793319153, + "grad_norm": 49.109535217285156, + "learning_rate": 3.0025020850708924e-05, + "loss": 0.7534, + "step": 3600 + }, + { + "epoch": 0.1806580758163393, + "grad_norm": 15.499408721923828, + "learning_rate": 3.010842368640534e-05, + "loss": 0.2711, + "step": 3610 + }, + { + "epoch": 0.18115851369948704, + "grad_norm": 47.528907775878906, + "learning_rate": 3.0191826522101753e-05, + "loss": 0.3914, + "step": 3620 + }, + { + "epoch": 0.18165895158263481, + "grad_norm": 65.87004852294922, + "learning_rate": 3.027522935779817e-05, + "loss": 0.6514, + "step": 3630 + }, + { + "epoch": 0.18215938946578256, + "grad_norm": 58.91048812866211, + "learning_rate": 3.035863219349458e-05, + "loss": 0.5933, + "step": 3640 + }, + { + "epoch": 0.18265982734893033, + "grad_norm": 12.425536155700684, + "learning_rate": 3.0442035029190995e-05, + "loss": 0.4118, + "step": 3650 + }, + { + "epoch": 0.18316026523207807, + "grad_norm": 30.425209045410156, + "learning_rate": 3.0525437864887404e-05, + "loss": 0.2105, + "step": 3660 + }, + { + "epoch": 0.1836607031152258, + "grad_norm": 54.209354400634766, + "learning_rate": 3.0608840700583824e-05, + "loss": 0.3805, + "step": 3670 + }, + { + "epoch": 0.18416114099837358, + "grad_norm": 79.9751968383789, + "learning_rate": 3.069224353628024e-05, + "loss": 0.9611, + "step": 3680 + }, + { + "epoch": 0.18466157888152132, + "grad_norm": 7.382099151611328, + "learning_rate": 3.077564637197665e-05, + "loss": 0.613, + "step": 3690 + }, + { + "epoch": 0.1851620167646691, + "grad_norm": 55.720191955566406, + "learning_rate": 3.085904920767306e-05, + "loss": 0.7477, + "step": 3700 + }, + { + "epoch": 0.18566245464781683, + "grad_norm": 57.06126403808594, + "learning_rate": 3.0942452043369475e-05, + "loss": 0.5406, + "step": 3710 + }, + { + "epoch": 0.1861628925309646, + "grad_norm": 11.809860229492188, + "learning_rate": 3.102585487906589e-05, + "loss": 0.9808, + "step": 3720 + }, + { + "epoch": 0.18666333041411234, + "grad_norm": 19.19344139099121, + "learning_rate": 3.11092577147623e-05, + "loss": 0.553, + "step": 3730 + }, + { + "epoch": 0.1871637682972601, + "grad_norm": 44.50214385986328, + "learning_rate": 3.119266055045872e-05, + "loss": 0.5506, + "step": 3740 + }, + { + "epoch": 0.18766420618040786, + "grad_norm": 32.0246696472168, + "learning_rate": 3.127606338615513e-05, + "loss": 0.7167, + "step": 3750 + }, + { + "epoch": 0.18816464406355562, + "grad_norm": 14.153768539428711, + "learning_rate": 3.1359466221851546e-05, + "loss": 0.4655, + "step": 3760 + }, + { + "epoch": 0.18866508194670337, + "grad_norm": 27.39801597595215, + "learning_rate": 3.144286905754796e-05, + "loss": 0.5709, + "step": 3770 + }, + { + "epoch": 0.1891655198298511, + "grad_norm": 23.275848388671875, + "learning_rate": 3.152627189324437e-05, + "loss": 0.3447, + "step": 3780 + }, + { + "epoch": 0.18966595771299888, + "grad_norm": 2.238745927810669, + "learning_rate": 3.1609674728940785e-05, + "loss": 0.3407, + "step": 3790 + }, + { + "epoch": 0.19016639559614662, + "grad_norm": 1.092089295387268, + "learning_rate": 3.16930775646372e-05, + "loss": 0.3261, + "step": 3800 + }, + { + "epoch": 0.1906668334792944, + "grad_norm": 11.40645980834961, + "learning_rate": 3.177648040033361e-05, + "loss": 0.5478, + "step": 3810 + }, + { + "epoch": 0.19116727136244213, + "grad_norm": 59.40361785888672, + "learning_rate": 3.185988323603002e-05, + "loss": 0.5673, + "step": 3820 + }, + { + "epoch": 0.1916677092455899, + "grad_norm": 47.518768310546875, + "learning_rate": 3.194328607172644e-05, + "loss": 0.7593, + "step": 3830 + }, + { + "epoch": 0.19216814712873764, + "grad_norm": 28.600204467773438, + "learning_rate": 3.2026688907422856e-05, + "loss": 0.6547, + "step": 3840 + }, + { + "epoch": 0.1926685850118854, + "grad_norm": 14.341931343078613, + "learning_rate": 3.211009174311927e-05, + "loss": 0.453, + "step": 3850 + }, + { + "epoch": 0.19316902289503315, + "grad_norm": 61.003997802734375, + "learning_rate": 3.219349457881568e-05, + "loss": 0.5698, + "step": 3860 + }, + { + "epoch": 0.1936694607781809, + "grad_norm": 15.194714546203613, + "learning_rate": 3.2276897414512094e-05, + "loss": 0.8043, + "step": 3870 + }, + { + "epoch": 0.19416989866132867, + "grad_norm": 8.299256324768066, + "learning_rate": 3.236030025020851e-05, + "loss": 0.7091, + "step": 3880 + }, + { + "epoch": 0.1946703365444764, + "grad_norm": 8.289032936096191, + "learning_rate": 3.244370308590492e-05, + "loss": 0.5884, + "step": 3890 + }, + { + "epoch": 0.19517077442762418, + "grad_norm": 60.39479064941406, + "learning_rate": 3.252710592160134e-05, + "loss": 0.6642, + "step": 3900 + }, + { + "epoch": 0.19567121231077192, + "grad_norm": 0.19470028579235077, + "learning_rate": 3.2610508757297746e-05, + "loss": 0.8625, + "step": 3910 + }, + { + "epoch": 0.1961716501939197, + "grad_norm": 12.04682731628418, + "learning_rate": 3.2693911592994165e-05, + "loss": 0.5328, + "step": 3920 + }, + { + "epoch": 0.19667208807706743, + "grad_norm": 31.123306274414062, + "learning_rate": 3.277731442869058e-05, + "loss": 0.4369, + "step": 3930 + }, + { + "epoch": 0.1971725259602152, + "grad_norm": 39.618045806884766, + "learning_rate": 3.286071726438699e-05, + "loss": 0.4121, + "step": 3940 + }, + { + "epoch": 0.19767296384336294, + "grad_norm": 7.954986572265625, + "learning_rate": 3.2944120100083404e-05, + "loss": 0.7229, + "step": 3950 + }, + { + "epoch": 0.19817340172651068, + "grad_norm": 54.104888916015625, + "learning_rate": 3.302752293577982e-05, + "loss": 0.5365, + "step": 3960 + }, + { + "epoch": 0.19867383960965845, + "grad_norm": 19.549301147460938, + "learning_rate": 3.311092577147623e-05, + "loss": 0.5666, + "step": 3970 + }, + { + "epoch": 0.1991742774928062, + "grad_norm": 79.07774353027344, + "learning_rate": 3.319432860717265e-05, + "loss": 0.4428, + "step": 3980 + }, + { + "epoch": 0.19967471537595397, + "grad_norm": 82.81560516357422, + "learning_rate": 3.327773144286906e-05, + "loss": 0.3728, + "step": 3990 + }, + { + "epoch": 0.2001751532591017, + "grad_norm": 1.9170231819152832, + "learning_rate": 3.336113427856547e-05, + "loss": 0.7531, + "step": 4000 + }, + { + "epoch": 0.20067559114224948, + "grad_norm": 18.15715789794922, + "learning_rate": 3.344453711426189e-05, + "loss": 0.2653, + "step": 4010 + }, + { + "epoch": 0.20117602902539722, + "grad_norm": 46.6406135559082, + "learning_rate": 3.35279399499583e-05, + "loss": 0.8163, + "step": 4020 + }, + { + "epoch": 0.201676466908545, + "grad_norm": 25.13616180419922, + "learning_rate": 3.361134278565471e-05, + "loss": 0.7855, + "step": 4030 + }, + { + "epoch": 0.20217690479169273, + "grad_norm": 3.9452226161956787, + "learning_rate": 3.3694745621351126e-05, + "loss": 0.3608, + "step": 4040 + }, + { + "epoch": 0.20267734267484047, + "grad_norm": 44.64923095703125, + "learning_rate": 3.3778148457047546e-05, + "loss": 0.5978, + "step": 4050 + }, + { + "epoch": 0.20317778055798824, + "grad_norm": 56.39286422729492, + "learning_rate": 3.386155129274395e-05, + "loss": 0.4842, + "step": 4060 + }, + { + "epoch": 0.20367821844113598, + "grad_norm": 41.4847297668457, + "learning_rate": 3.394495412844037e-05, + "loss": 0.3737, + "step": 4070 + }, + { + "epoch": 0.20417865632428375, + "grad_norm": 34.48544692993164, + "learning_rate": 3.4028356964136784e-05, + "loss": 0.5122, + "step": 4080 + }, + { + "epoch": 0.2046790942074315, + "grad_norm": 27.515913009643555, + "learning_rate": 3.41117597998332e-05, + "loss": 0.4421, + "step": 4090 + }, + { + "epoch": 0.20517953209057926, + "grad_norm": 7.75834321975708, + "learning_rate": 3.419516263552961e-05, + "loss": 0.2932, + "step": 4100 + }, + { + "epoch": 0.205679969973727, + "grad_norm": 60.186458587646484, + "learning_rate": 3.427856547122602e-05, + "loss": 0.6163, + "step": 4110 + }, + { + "epoch": 0.20618040785687478, + "grad_norm": 0.16573095321655273, + "learning_rate": 3.4361968306922436e-05, + "loss": 0.5596, + "step": 4120 + }, + { + "epoch": 0.20668084574002252, + "grad_norm": 20.66874122619629, + "learning_rate": 3.444537114261885e-05, + "loss": 0.4817, + "step": 4130 + }, + { + "epoch": 0.2071812836231703, + "grad_norm": 31.455095291137695, + "learning_rate": 3.452877397831527e-05, + "loss": 0.4971, + "step": 4140 + }, + { + "epoch": 0.20768172150631803, + "grad_norm": 66.17622375488281, + "learning_rate": 3.4612176814011674e-05, + "loss": 1.093, + "step": 4150 + }, + { + "epoch": 0.20818215938946577, + "grad_norm": 59.76629638671875, + "learning_rate": 3.4695579649708094e-05, + "loss": 0.3948, + "step": 4160 + }, + { + "epoch": 0.20868259727261354, + "grad_norm": 42.59615707397461, + "learning_rate": 3.477898248540451e-05, + "loss": 0.5429, + "step": 4170 + }, + { + "epoch": 0.20918303515576128, + "grad_norm": 31.853050231933594, + "learning_rate": 3.486238532110092e-05, + "loss": 0.286, + "step": 4180 + }, + { + "epoch": 0.20968347303890905, + "grad_norm": 97.81887817382812, + "learning_rate": 3.494578815679733e-05, + "loss": 0.4787, + "step": 4190 + }, + { + "epoch": 0.2101839109220568, + "grad_norm": 47.924068450927734, + "learning_rate": 3.502919099249375e-05, + "loss": 0.4363, + "step": 4200 + }, + { + "epoch": 0.21068434880520456, + "grad_norm": 22.490070343017578, + "learning_rate": 3.511259382819016e-05, + "loss": 0.5923, + "step": 4210 + }, + { + "epoch": 0.2111847866883523, + "grad_norm": 27.43729591369629, + "learning_rate": 3.519599666388657e-05, + "loss": 0.5455, + "step": 4220 + }, + { + "epoch": 0.21168522457150007, + "grad_norm": 1.9464672803878784, + "learning_rate": 3.527939949958299e-05, + "loss": 0.2897, + "step": 4230 + }, + { + "epoch": 0.21218566245464782, + "grad_norm": 14.351051330566406, + "learning_rate": 3.5362802335279397e-05, + "loss": 0.577, + "step": 4240 + }, + { + "epoch": 0.21268610033779556, + "grad_norm": 58.62644577026367, + "learning_rate": 3.5446205170975816e-05, + "loss": 0.4298, + "step": 4250 + }, + { + "epoch": 0.21318653822094333, + "grad_norm": 1.5865960121154785, + "learning_rate": 3.552960800667223e-05, + "loss": 0.6141, + "step": 4260 + }, + { + "epoch": 0.21368697610409107, + "grad_norm": 105.57713317871094, + "learning_rate": 3.561301084236864e-05, + "loss": 0.7661, + "step": 4270 + }, + { + "epoch": 0.21418741398723884, + "grad_norm": 41.431732177734375, + "learning_rate": 3.5696413678065055e-05, + "loss": 0.4826, + "step": 4280 + }, + { + "epoch": 0.21468785187038658, + "grad_norm": 16.044492721557617, + "learning_rate": 3.5779816513761474e-05, + "loss": 0.2354, + "step": 4290 + }, + { + "epoch": 0.21518828975353435, + "grad_norm": 5.421501159667969, + "learning_rate": 3.586321934945788e-05, + "loss": 0.3778, + "step": 4300 + }, + { + "epoch": 0.2156887276366821, + "grad_norm": 22.81307601928711, + "learning_rate": 3.59466221851543e-05, + "loss": 0.4881, + "step": 4310 + }, + { + "epoch": 0.21618916551982986, + "grad_norm": 70.0987777709961, + "learning_rate": 3.603002502085071e-05, + "loss": 0.5456, + "step": 4320 + }, + { + "epoch": 0.2166896034029776, + "grad_norm": 59.90314865112305, + "learning_rate": 3.611342785654712e-05, + "loss": 0.6923, + "step": 4330 + }, + { + "epoch": 0.21719004128612535, + "grad_norm": 0.11245470494031906, + "learning_rate": 3.619683069224354e-05, + "loss": 0.8482, + "step": 4340 + }, + { + "epoch": 0.21769047916927312, + "grad_norm": 55.70448684692383, + "learning_rate": 3.628023352793995e-05, + "loss": 0.6245, + "step": 4350 + }, + { + "epoch": 0.21819091705242086, + "grad_norm": 157.45533752441406, + "learning_rate": 3.6363636363636364e-05, + "loss": 0.7044, + "step": 4360 + }, + { + "epoch": 0.21869135493556863, + "grad_norm": 24.453365325927734, + "learning_rate": 3.644703919933278e-05, + "loss": 0.2078, + "step": 4370 + }, + { + "epoch": 0.21919179281871637, + "grad_norm": 26.094911575317383, + "learning_rate": 3.6530442035029197e-05, + "loss": 0.462, + "step": 4380 + }, + { + "epoch": 0.21969223070186414, + "grad_norm": 22.429595947265625, + "learning_rate": 3.66138448707256e-05, + "loss": 0.8248, + "step": 4390 + }, + { + "epoch": 0.22019266858501188, + "grad_norm": 9.390774726867676, + "learning_rate": 3.669724770642202e-05, + "loss": 0.3561, + "step": 4400 + }, + { + "epoch": 0.22069310646815965, + "grad_norm": 75.94200134277344, + "learning_rate": 3.6780650542118435e-05, + "loss": 0.6203, + "step": 4410 + }, + { + "epoch": 0.2211935443513074, + "grad_norm": 0.5867464542388916, + "learning_rate": 3.686405337781485e-05, + "loss": 0.3872, + "step": 4420 + }, + { + "epoch": 0.22169398223445516, + "grad_norm": 31.979068756103516, + "learning_rate": 3.694745621351126e-05, + "loss": 0.5832, + "step": 4430 + }, + { + "epoch": 0.2221944201176029, + "grad_norm": 54.91218185424805, + "learning_rate": 3.7030859049207674e-05, + "loss": 0.6299, + "step": 4440 + }, + { + "epoch": 0.22269485800075065, + "grad_norm": 7.563572883605957, + "learning_rate": 3.7114261884904086e-05, + "loss": 0.4304, + "step": 4450 + }, + { + "epoch": 0.22319529588389841, + "grad_norm": 2.4852206707000732, + "learning_rate": 3.71976647206005e-05, + "loss": 0.2796, + "step": 4460 + }, + { + "epoch": 0.22369573376704616, + "grad_norm": 26.526731491088867, + "learning_rate": 3.728106755629692e-05, + "loss": 0.6095, + "step": 4470 + }, + { + "epoch": 0.22419617165019393, + "grad_norm": 48.31926345825195, + "learning_rate": 3.7364470391993325e-05, + "loss": 0.5502, + "step": 4480 + }, + { + "epoch": 0.22469660953334167, + "grad_norm": 119.29351806640625, + "learning_rate": 3.7447873227689745e-05, + "loss": 0.5207, + "step": 4490 + }, + { + "epoch": 0.22519704741648944, + "grad_norm": 59.480064392089844, + "learning_rate": 3.753127606338616e-05, + "loss": 0.7665, + "step": 4500 + }, + { + "epoch": 0.22569748529963718, + "grad_norm": 19.014957427978516, + "learning_rate": 3.761467889908257e-05, + "loss": 0.3892, + "step": 4510 + }, + { + "epoch": 0.22619792318278495, + "grad_norm": 11.278446197509766, + "learning_rate": 3.769808173477898e-05, + "loss": 0.8863, + "step": 4520 + }, + { + "epoch": 0.2266983610659327, + "grad_norm": 16.630571365356445, + "learning_rate": 3.77814845704754e-05, + "loss": 0.5769, + "step": 4530 + }, + { + "epoch": 0.22719879894908043, + "grad_norm": 2.4195024967193604, + "learning_rate": 3.786488740617181e-05, + "loss": 0.3267, + "step": 4540 + }, + { + "epoch": 0.2276992368322282, + "grad_norm": 50.60469436645508, + "learning_rate": 3.794829024186822e-05, + "loss": 0.3671, + "step": 4550 + }, + { + "epoch": 0.22819967471537594, + "grad_norm": 35.203067779541016, + "learning_rate": 3.803169307756464e-05, + "loss": 0.5179, + "step": 4560 + }, + { + "epoch": 0.2287001125985237, + "grad_norm": 88.9637222290039, + "learning_rate": 3.8115095913261054e-05, + "loss": 0.5074, + "step": 4570 + }, + { + "epoch": 0.22920055048167146, + "grad_norm": 29.579782485961914, + "learning_rate": 3.819849874895747e-05, + "loss": 0.4561, + "step": 4580 + }, + { + "epoch": 0.22970098836481923, + "grad_norm": 0.009941134601831436, + "learning_rate": 3.828190158465388e-05, + "loss": 0.3721, + "step": 4590 + }, + { + "epoch": 0.23020142624796697, + "grad_norm": 12.493212699890137, + "learning_rate": 3.836530442035029e-05, + "loss": 0.3414, + "step": 4600 + }, + { + "epoch": 0.23070186413111474, + "grad_norm": 1.3338897228240967, + "learning_rate": 3.8448707256046705e-05, + "loss": 0.872, + "step": 4610 + }, + { + "epoch": 0.23120230201426248, + "grad_norm": 49.78994369506836, + "learning_rate": 3.8532110091743125e-05, + "loss": 0.5949, + "step": 4620 + }, + { + "epoch": 0.23170273989741022, + "grad_norm": 20.969486236572266, + "learning_rate": 3.861551292743953e-05, + "loss": 0.3696, + "step": 4630 + }, + { + "epoch": 0.232203177780558, + "grad_norm": 5.0722270011901855, + "learning_rate": 3.869891576313595e-05, + "loss": 0.4888, + "step": 4640 + }, + { + "epoch": 0.23270361566370573, + "grad_norm": 60.42429733276367, + "learning_rate": 3.8782318598832364e-05, + "loss": 0.2457, + "step": 4650 + }, + { + "epoch": 0.2332040535468535, + "grad_norm": 0.002261493820697069, + "learning_rate": 3.8865721434528776e-05, + "loss": 0.8478, + "step": 4660 + }, + { + "epoch": 0.23370449143000124, + "grad_norm": 21.61867332458496, + "learning_rate": 3.894912427022519e-05, + "loss": 0.6572, + "step": 4670 + }, + { + "epoch": 0.234204929313149, + "grad_norm": 14.474299430847168, + "learning_rate": 3.90325271059216e-05, + "loss": 0.2516, + "step": 4680 + }, + { + "epoch": 0.23470536719629675, + "grad_norm": 61.1234016418457, + "learning_rate": 3.9115929941618015e-05, + "loss": 0.4935, + "step": 4690 + }, + { + "epoch": 0.23520580507944452, + "grad_norm": 23.632478713989258, + "learning_rate": 3.919933277731443e-05, + "loss": 0.5899, + "step": 4700 + }, + { + "epoch": 0.23570624296259227, + "grad_norm": 68.15599822998047, + "learning_rate": 3.928273561301085e-05, + "loss": 0.4668, + "step": 4710 + }, + { + "epoch": 0.23620668084574004, + "grad_norm": 2.4591064453125, + "learning_rate": 3.936613844870726e-05, + "loss": 0.5787, + "step": 4720 + }, + { + "epoch": 0.23670711872888778, + "grad_norm": 7.732545852661133, + "learning_rate": 3.944954128440367e-05, + "loss": 0.3477, + "step": 4730 + }, + { + "epoch": 0.23720755661203552, + "grad_norm": 25.070650100708008, + "learning_rate": 3.9532944120100086e-05, + "loss": 0.5985, + "step": 4740 + }, + { + "epoch": 0.2377079944951833, + "grad_norm": 33.108070373535156, + "learning_rate": 3.96163469557965e-05, + "loss": 0.4059, + "step": 4750 + }, + { + "epoch": 0.23820843237833103, + "grad_norm": 8.65207290649414, + "learning_rate": 3.969974979149291e-05, + "loss": 0.4821, + "step": 4760 + }, + { + "epoch": 0.2387088702614788, + "grad_norm": 59.94301223754883, + "learning_rate": 3.9783152627189324e-05, + "loss": 0.7141, + "step": 4770 + }, + { + "epoch": 0.23920930814462654, + "grad_norm": 36.59794616699219, + "learning_rate": 3.9866555462885744e-05, + "loss": 0.5778, + "step": 4780 + }, + { + "epoch": 0.2397097460277743, + "grad_norm": 24.40534210205078, + "learning_rate": 3.994995829858215e-05, + "loss": 0.4603, + "step": 4790 + }, + { + "epoch": 0.24021018391092205, + "grad_norm": 15.629915237426758, + "learning_rate": 4.003336113427857e-05, + "loss": 0.342, + "step": 4800 + }, + { + "epoch": 0.24071062179406982, + "grad_norm": 8.233383178710938, + "learning_rate": 4.011676396997498e-05, + "loss": 0.3649, + "step": 4810 + }, + { + "epoch": 0.24121105967721757, + "grad_norm": 0.08335284143686295, + "learning_rate": 4.0200166805671395e-05, + "loss": 0.905, + "step": 4820 + }, + { + "epoch": 0.2417114975603653, + "grad_norm": 21.82447052001953, + "learning_rate": 4.028356964136781e-05, + "loss": 0.8642, + "step": 4830 + }, + { + "epoch": 0.24221193544351308, + "grad_norm": 14.461441993713379, + "learning_rate": 4.036697247706422e-05, + "loss": 0.7305, + "step": 4840 + }, + { + "epoch": 0.24271237332666082, + "grad_norm": 17.851411819458008, + "learning_rate": 4.0450375312760634e-05, + "loss": 0.296, + "step": 4850 + }, + { + "epoch": 0.2432128112098086, + "grad_norm": 18.806398391723633, + "learning_rate": 4.0533778148457054e-05, + "loss": 0.3174, + "step": 4860 + }, + { + "epoch": 0.24371324909295633, + "grad_norm": 4.5930938720703125, + "learning_rate": 4.0617180984153466e-05, + "loss": 0.4719, + "step": 4870 + }, + { + "epoch": 0.2442136869761041, + "grad_norm": 4.100564956665039, + "learning_rate": 4.070058381984987e-05, + "loss": 0.6453, + "step": 4880 + }, + { + "epoch": 0.24471412485925184, + "grad_norm": 38.994590759277344, + "learning_rate": 4.078398665554629e-05, + "loss": 0.7049, + "step": 4890 + }, + { + "epoch": 0.2452145627423996, + "grad_norm": 1.85924232006073, + "learning_rate": 4.0867389491242705e-05, + "loss": 0.433, + "step": 4900 + }, + { + "epoch": 0.24571500062554735, + "grad_norm": 6.00260591506958, + "learning_rate": 4.095079232693912e-05, + "loss": 0.3578, + "step": 4910 + }, + { + "epoch": 0.2462154385086951, + "grad_norm": 75.64608764648438, + "learning_rate": 4.103419516263553e-05, + "loss": 0.9343, + "step": 4920 + }, + { + "epoch": 0.24671587639184286, + "grad_norm": 147.2863006591797, + "learning_rate": 4.111759799833195e-05, + "loss": 0.5555, + "step": 4930 + }, + { + "epoch": 0.2472163142749906, + "grad_norm": 272.0299987792969, + "learning_rate": 4.1201000834028356e-05, + "loss": 1.3269, + "step": 4940 + }, + { + "epoch": 0.24771675215813838, + "grad_norm": 70.04656982421875, + "learning_rate": 4.1284403669724776e-05, + "loss": 0.9894, + "step": 4950 + }, + { + "epoch": 0.24821719004128612, + "grad_norm": 43.161651611328125, + "learning_rate": 4.136780650542119e-05, + "loss": 0.4872, + "step": 4960 + }, + { + "epoch": 0.2487176279244339, + "grad_norm": 76.15550994873047, + "learning_rate": 4.1451209341117595e-05, + "loss": 0.5464, + "step": 4970 + }, + { + "epoch": 0.24921806580758163, + "grad_norm": 21.166522979736328, + "learning_rate": 4.1534612176814014e-05, + "loss": 0.4713, + "step": 4980 + }, + { + "epoch": 0.2497185036907294, + "grad_norm": 23.178564071655273, + "learning_rate": 4.161801501251043e-05, + "loss": 0.5092, + "step": 4990 + }, + { + "epoch": 0.25021894157387714, + "grad_norm": 26.96135711669922, + "learning_rate": 4.170141784820684e-05, + "loss": 0.6797, + "step": 5000 + }, + { + "epoch": 0.2507193794570249, + "grad_norm": 32.15453338623047, + "learning_rate": 4.178482068390325e-05, + "loss": 0.3598, + "step": 5010 + }, + { + "epoch": 0.2512198173401726, + "grad_norm": 18.699811935424805, + "learning_rate": 4.186822351959967e-05, + "loss": 0.431, + "step": 5020 + }, + { + "epoch": 0.2517202552233204, + "grad_norm": 6.3686065673828125, + "learning_rate": 4.195162635529608e-05, + "loss": 0.411, + "step": 5030 + }, + { + "epoch": 0.25222069310646816, + "grad_norm": 41.526119232177734, + "learning_rate": 4.20350291909925e-05, + "loss": 0.8789, + "step": 5040 + }, + { + "epoch": 0.2527211309896159, + "grad_norm": 2.0874736309051514, + "learning_rate": 4.211843202668891e-05, + "loss": 0.7421, + "step": 5050 + }, + { + "epoch": 0.25322156887276365, + "grad_norm": 24.107097625732422, + "learning_rate": 4.2201834862385324e-05, + "loss": 0.7547, + "step": 5060 + }, + { + "epoch": 0.25372200675591144, + "grad_norm": 0.017932103946805, + "learning_rate": 4.228523769808174e-05, + "loss": 0.3839, + "step": 5070 + }, + { + "epoch": 0.2542224446390592, + "grad_norm": 44.21803665161133, + "learning_rate": 4.236864053377815e-05, + "loss": 0.4289, + "step": 5080 + }, + { + "epoch": 0.25472288252220693, + "grad_norm": 64.5974349975586, + "learning_rate": 4.245204336947456e-05, + "loss": 0.3679, + "step": 5090 + }, + { + "epoch": 0.25522332040535467, + "grad_norm": 32.83170700073242, + "learning_rate": 4.2535446205170975e-05, + "loss": 0.6084, + "step": 5100 + }, + { + "epoch": 0.25572375828850247, + "grad_norm": 74.71546936035156, + "learning_rate": 4.2618849040867395e-05, + "loss": 0.5081, + "step": 5110 + }, + { + "epoch": 0.2562241961716502, + "grad_norm": 36.60004806518555, + "learning_rate": 4.27022518765638e-05, + "loss": 0.6106, + "step": 5120 + }, + { + "epoch": 0.25672463405479795, + "grad_norm": 0.16247905790805817, + "learning_rate": 4.278565471226022e-05, + "loss": 0.5219, + "step": 5130 + }, + { + "epoch": 0.2572250719379457, + "grad_norm": 11.735374450683594, + "learning_rate": 4.2869057547956633e-05, + "loss": 0.4294, + "step": 5140 + }, + { + "epoch": 0.25772550982109343, + "grad_norm": 60.48006820678711, + "learning_rate": 4.2952460383653046e-05, + "loss": 0.222, + "step": 5150 + }, + { + "epoch": 0.25822594770424123, + "grad_norm": 4.259871482849121, + "learning_rate": 4.303586321934946e-05, + "loss": 0.6497, + "step": 5160 + }, + { + "epoch": 0.258726385587389, + "grad_norm": 1.9654138088226318, + "learning_rate": 4.311926605504588e-05, + "loss": 0.2482, + "step": 5170 + }, + { + "epoch": 0.2592268234705367, + "grad_norm": 42.19381332397461, + "learning_rate": 4.3202668890742285e-05, + "loss": 0.5964, + "step": 5180 + }, + { + "epoch": 0.25972726135368446, + "grad_norm": 139.48361206054688, + "learning_rate": 4.32860717264387e-05, + "loss": 0.4643, + "step": 5190 + }, + { + "epoch": 0.26022769923683226, + "grad_norm": 95.39739227294922, + "learning_rate": 4.336947456213512e-05, + "loss": 0.7245, + "step": 5200 + }, + { + "epoch": 0.26072813711998, + "grad_norm": 23.835752487182617, + "learning_rate": 4.345287739783152e-05, + "loss": 0.469, + "step": 5210 + }, + { + "epoch": 0.26122857500312774, + "grad_norm": 18.726959228515625, + "learning_rate": 4.353628023352794e-05, + "loss": 0.2498, + "step": 5220 + }, + { + "epoch": 0.2617290128862755, + "grad_norm": 18.522645950317383, + "learning_rate": 4.3619683069224356e-05, + "loss": 0.5103, + "step": 5230 + }, + { + "epoch": 0.2622294507694232, + "grad_norm": 1.3740856647491455, + "learning_rate": 4.370308590492077e-05, + "loss": 0.7745, + "step": 5240 + }, + { + "epoch": 0.262729888652571, + "grad_norm": 2.839979887008667, + "learning_rate": 4.378648874061718e-05, + "loss": 0.4629, + "step": 5250 + }, + { + "epoch": 0.26323032653571876, + "grad_norm": 0.15321099758148193, + "learning_rate": 4.38698915763136e-05, + "loss": 0.9793, + "step": 5260 + }, + { + "epoch": 0.2637307644188665, + "grad_norm": 0.0345902293920517, + "learning_rate": 4.395329441201001e-05, + "loss": 0.5433, + "step": 5270 + }, + { + "epoch": 0.26423120230201425, + "grad_norm": 40.451744079589844, + "learning_rate": 4.403669724770643e-05, + "loss": 0.7327, + "step": 5280 + }, + { + "epoch": 0.26473164018516204, + "grad_norm": 111.32527160644531, + "learning_rate": 4.412010008340284e-05, + "loss": 0.9108, + "step": 5290 + }, + { + "epoch": 0.2652320780683098, + "grad_norm": 0.10947636514902115, + "learning_rate": 4.4203502919099246e-05, + "loss": 0.5099, + "step": 5300 + }, + { + "epoch": 0.2657325159514575, + "grad_norm": 1.2890784740447998, + "learning_rate": 4.4286905754795665e-05, + "loss": 0.3635, + "step": 5310 + }, + { + "epoch": 0.26623295383460527, + "grad_norm": 1.3132479190826416, + "learning_rate": 4.437030859049208e-05, + "loss": 0.5126, + "step": 5320 + }, + { + "epoch": 0.266733391717753, + "grad_norm": 0.004424304701387882, + "learning_rate": 4.445371142618849e-05, + "loss": 0.2837, + "step": 5330 + }, + { + "epoch": 0.2672338296009008, + "grad_norm": 19.534645080566406, + "learning_rate": 4.4537114261884904e-05, + "loss": 0.3359, + "step": 5340 + }, + { + "epoch": 0.26773426748404855, + "grad_norm": 7.430948734283447, + "learning_rate": 4.4620517097581323e-05, + "loss": 0.894, + "step": 5350 + }, + { + "epoch": 0.2682347053671963, + "grad_norm": 26.94919204711914, + "learning_rate": 4.470391993327773e-05, + "loss": 0.4149, + "step": 5360 + }, + { + "epoch": 0.26873514325034403, + "grad_norm": 44.15589904785156, + "learning_rate": 4.478732276897415e-05, + "loss": 0.7029, + "step": 5370 + }, + { + "epoch": 0.26923558113349183, + "grad_norm": 50.88851547241211, + "learning_rate": 4.487072560467056e-05, + "loss": 0.5547, + "step": 5380 + }, + { + "epoch": 0.26973601901663957, + "grad_norm": 101.4876708984375, + "learning_rate": 4.4954128440366975e-05, + "loss": 0.5644, + "step": 5390 + }, + { + "epoch": 0.2702364568997873, + "grad_norm": 34.520877838134766, + "learning_rate": 4.503753127606339e-05, + "loss": 0.3835, + "step": 5400 + }, + { + "epoch": 0.27073689478293506, + "grad_norm": 0.03332913666963577, + "learning_rate": 4.51209341117598e-05, + "loss": 0.5433, + "step": 5410 + }, + { + "epoch": 0.2712373326660828, + "grad_norm": 52.24263381958008, + "learning_rate": 4.520433694745621e-05, + "loss": 0.5364, + "step": 5420 + }, + { + "epoch": 0.2717377705492306, + "grad_norm": 62.57723617553711, + "learning_rate": 4.5287739783152626e-05, + "loss": 0.3753, + "step": 5430 + }, + { + "epoch": 0.27223820843237834, + "grad_norm": 2.9040215015411377, + "learning_rate": 4.5371142618849046e-05, + "loss": 0.405, + "step": 5440 + }, + { + "epoch": 0.2727386463155261, + "grad_norm": 33.40290451049805, + "learning_rate": 4.545454545454546e-05, + "loss": 0.3348, + "step": 5450 + }, + { + "epoch": 0.2732390841986738, + "grad_norm": 69.51597595214844, + "learning_rate": 4.553794829024187e-05, + "loss": 0.7194, + "step": 5460 + }, + { + "epoch": 0.2737395220818216, + "grad_norm": 15.624861717224121, + "learning_rate": 4.5621351125938284e-05, + "loss": 0.2684, + "step": 5470 + }, + { + "epoch": 0.27423995996496936, + "grad_norm": 43.53606033325195, + "learning_rate": 4.57047539616347e-05, + "loss": 0.4202, + "step": 5480 + }, + { + "epoch": 0.2747403978481171, + "grad_norm": 0.3256034255027771, + "learning_rate": 4.578815679733111e-05, + "loss": 0.428, + "step": 5490 + }, + { + "epoch": 0.27524083573126484, + "grad_norm": 15.836487770080566, + "learning_rate": 4.587155963302753e-05, + "loss": 0.3734, + "step": 5500 + }, + { + "epoch": 0.2757412736144126, + "grad_norm": 50.19599533081055, + "learning_rate": 4.5954962468723936e-05, + "loss": 0.4839, + "step": 5510 + }, + { + "epoch": 0.2762417114975604, + "grad_norm": 14.381315231323242, + "learning_rate": 4.603836530442035e-05, + "loss": 0.8722, + "step": 5520 + }, + { + "epoch": 0.2767421493807081, + "grad_norm": 16.872190475463867, + "learning_rate": 4.612176814011677e-05, + "loss": 0.6978, + "step": 5530 + }, + { + "epoch": 0.27724258726385587, + "grad_norm": 31.099498748779297, + "learning_rate": 4.620517097581318e-05, + "loss": 0.5356, + "step": 5540 + }, + { + "epoch": 0.2777430251470036, + "grad_norm": 41.00047302246094, + "learning_rate": 4.6288573811509594e-05, + "loss": 0.4818, + "step": 5550 + }, + { + "epoch": 0.2782434630301514, + "grad_norm": 20.849727630615234, + "learning_rate": 4.6371976647206007e-05, + "loss": 0.2611, + "step": 5560 + }, + { + "epoch": 0.27874390091329915, + "grad_norm": 34.5333366394043, + "learning_rate": 4.645537948290242e-05, + "loss": 0.5506, + "step": 5570 + }, + { + "epoch": 0.2792443387964469, + "grad_norm": 23.273820877075195, + "learning_rate": 4.653878231859883e-05, + "loss": 0.4961, + "step": 5580 + }, + { + "epoch": 0.27974477667959463, + "grad_norm": 0.001963281538337469, + "learning_rate": 4.662218515429525e-05, + "loss": 0.3465, + "step": 5590 + }, + { + "epoch": 0.2802452145627424, + "grad_norm": 73.29209899902344, + "learning_rate": 4.6705587989991665e-05, + "loss": 0.4533, + "step": 5600 + }, + { + "epoch": 0.28074565244589017, + "grad_norm": 109.50919342041016, + "learning_rate": 4.678899082568808e-05, + "loss": 0.6799, + "step": 5610 + }, + { + "epoch": 0.2812460903290379, + "grad_norm": 8.418110847473145, + "learning_rate": 4.687239366138449e-05, + "loss": 0.6139, + "step": 5620 + }, + { + "epoch": 0.28174652821218565, + "grad_norm": 4.011080265045166, + "learning_rate": 4.69557964970809e-05, + "loss": 0.8507, + "step": 5630 + }, + { + "epoch": 0.2822469660953334, + "grad_norm": 16.020362854003906, + "learning_rate": 4.7039199332777316e-05, + "loss": 0.4041, + "step": 5640 + }, + { + "epoch": 0.2827474039784812, + "grad_norm": 23.318206787109375, + "learning_rate": 4.712260216847373e-05, + "loss": 0.7662, + "step": 5650 + }, + { + "epoch": 0.28324784186162894, + "grad_norm": 14.476593017578125, + "learning_rate": 4.720600500417015e-05, + "loss": 0.3381, + "step": 5660 + }, + { + "epoch": 0.2837482797447767, + "grad_norm": 19.9843692779541, + "learning_rate": 4.7289407839866555e-05, + "loss": 0.3612, + "step": 5670 + }, + { + "epoch": 0.2842487176279244, + "grad_norm": 5.605803489685059, + "learning_rate": 4.7372810675562974e-05, + "loss": 0.4863, + "step": 5680 + }, + { + "epoch": 0.28474915551107216, + "grad_norm": 3.136143922805786, + "learning_rate": 4.745621351125939e-05, + "loss": 0.9302, + "step": 5690 + }, + { + "epoch": 0.28524959339421996, + "grad_norm": 93.13472747802734, + "learning_rate": 4.75396163469558e-05, + "loss": 0.5885, + "step": 5700 + }, + { + "epoch": 0.2857500312773677, + "grad_norm": 0.4756988286972046, + "learning_rate": 4.762301918265221e-05, + "loss": 0.6869, + "step": 5710 + }, + { + "epoch": 0.28625046916051544, + "grad_norm": 1.7234265804290771, + "learning_rate": 4.7706422018348626e-05, + "loss": 0.2499, + "step": 5720 + }, + { + "epoch": 0.2867509070436632, + "grad_norm": 14.258783340454102, + "learning_rate": 4.778982485404504e-05, + "loss": 0.7057, + "step": 5730 + }, + { + "epoch": 0.287251344926811, + "grad_norm": 0.2176150232553482, + "learning_rate": 4.787322768974145e-05, + "loss": 0.4783, + "step": 5740 + }, + { + "epoch": 0.2877517828099587, + "grad_norm": 38.104713439941406, + "learning_rate": 4.795663052543787e-05, + "loss": 0.5429, + "step": 5750 + }, + { + "epoch": 0.28825222069310646, + "grad_norm": 2.54300856590271, + "learning_rate": 4.804003336113428e-05, + "loss": 0.6737, + "step": 5760 + }, + { + "epoch": 0.2887526585762542, + "grad_norm": 24.85074806213379, + "learning_rate": 4.8123436196830697e-05, + "loss": 0.2702, + "step": 5770 + }, + { + "epoch": 0.289253096459402, + "grad_norm": 31.364971160888672, + "learning_rate": 4.820683903252711e-05, + "loss": 0.4332, + "step": 5780 + }, + { + "epoch": 0.28975353434254975, + "grad_norm": 13.392885208129883, + "learning_rate": 4.829024186822352e-05, + "loss": 0.6016, + "step": 5790 + }, + { + "epoch": 0.2902539722256975, + "grad_norm": 7.57896614074707, + "learning_rate": 4.8373644703919935e-05, + "loss": 0.3839, + "step": 5800 + }, + { + "epoch": 0.29075441010884523, + "grad_norm": 36.755062103271484, + "learning_rate": 4.8457047539616355e-05, + "loss": 0.5815, + "step": 5810 + }, + { + "epoch": 0.29125484799199297, + "grad_norm": 7.283461570739746, + "learning_rate": 4.854045037531276e-05, + "loss": 0.4559, + "step": 5820 + }, + { + "epoch": 0.29175528587514077, + "grad_norm": 15.004409790039062, + "learning_rate": 4.862385321100918e-05, + "loss": 0.7355, + "step": 5830 + }, + { + "epoch": 0.2922557237582885, + "grad_norm": 3.937574863433838, + "learning_rate": 4.870725604670559e-05, + "loss": 0.6461, + "step": 5840 + }, + { + "epoch": 0.29275616164143625, + "grad_norm": 0.7306815385818481, + "learning_rate": 4.8790658882402e-05, + "loss": 0.3832, + "step": 5850 + }, + { + "epoch": 0.293256599524584, + "grad_norm": 4.9494805335998535, + "learning_rate": 4.887406171809842e-05, + "loss": 0.35, + "step": 5860 + }, + { + "epoch": 0.2937570374077318, + "grad_norm": 1.542823314666748, + "learning_rate": 4.895746455379483e-05, + "loss": 0.245, + "step": 5870 + }, + { + "epoch": 0.29425747529087953, + "grad_norm": 5.048789024353027, + "learning_rate": 4.9040867389491245e-05, + "loss": 0.6564, + "step": 5880 + }, + { + "epoch": 0.2947579131740273, + "grad_norm": 0.36370936036109924, + "learning_rate": 4.912427022518766e-05, + "loss": 0.1381, + "step": 5890 + }, + { + "epoch": 0.295258351057175, + "grad_norm": 29.278762817382812, + "learning_rate": 4.920767306088408e-05, + "loss": 1.0081, + "step": 5900 + }, + { + "epoch": 0.29575878894032276, + "grad_norm": 0.138155996799469, + "learning_rate": 4.929107589658048e-05, + "loss": 0.3237, + "step": 5910 + }, + { + "epoch": 0.29625922682347056, + "grad_norm": 15.23071575164795, + "learning_rate": 4.93744787322769e-05, + "loss": 0.8463, + "step": 5920 + }, + { + "epoch": 0.2967596647066183, + "grad_norm": 2.8561720848083496, + "learning_rate": 4.9457881567973316e-05, + "loss": 0.2846, + "step": 5930 + }, + { + "epoch": 0.29726010258976604, + "grad_norm": 11.707077980041504, + "learning_rate": 4.954128440366973e-05, + "loss": 0.4318, + "step": 5940 + }, + { + "epoch": 0.2977605404729138, + "grad_norm": 23.929033279418945, + "learning_rate": 4.962468723936614e-05, + "loss": 0.3635, + "step": 5950 + }, + { + "epoch": 0.2982609783560616, + "grad_norm": 65.90140533447266, + "learning_rate": 4.9708090075062554e-05, + "loss": 0.6694, + "step": 5960 + }, + { + "epoch": 0.2987614162392093, + "grad_norm": 41.4105339050293, + "learning_rate": 4.979149291075897e-05, + "loss": 0.535, + "step": 5970 + }, + { + "epoch": 0.29926185412235706, + "grad_norm": 45.041587829589844, + "learning_rate": 4.987489574645538e-05, + "loss": 0.7274, + "step": 5980 + }, + { + "epoch": 0.2997622920055048, + "grad_norm": 123.51217651367188, + "learning_rate": 4.99582985821518e-05, + "loss": 0.4181, + "step": 5990 + }, + { + "epoch": 0.30026272988865255, + "grad_norm": 10.284601211547852, + "learning_rate": 4.999536616559471e-05, + "loss": 0.3717, + "step": 6000 + }, + { + "epoch": 0.30076316777180034, + "grad_norm": 4.650702953338623, + "learning_rate": 4.9986098496784126e-05, + "loss": 0.3783, + "step": 6010 + }, + { + "epoch": 0.3012636056549481, + "grad_norm": 24.62755012512207, + "learning_rate": 4.9976830827973535e-05, + "loss": 0.5603, + "step": 6020 + }, + { + "epoch": 0.3017640435380958, + "grad_norm": 64.3497314453125, + "learning_rate": 4.9967563159162944e-05, + "loss": 0.96, + "step": 6030 + }, + { + "epoch": 0.30226448142124357, + "grad_norm": 38.60239028930664, + "learning_rate": 4.995829549035236e-05, + "loss": 0.6857, + "step": 6040 + }, + { + "epoch": 0.30276491930439137, + "grad_norm": 53.29572677612305, + "learning_rate": 4.9949027821541775e-05, + "loss": 0.4281, + "step": 6050 + }, + { + "epoch": 0.3032653571875391, + "grad_norm": 66.73191833496094, + "learning_rate": 4.9939760152731184e-05, + "loss": 0.3386, + "step": 6060 + }, + { + "epoch": 0.30376579507068685, + "grad_norm": 32.48128890991211, + "learning_rate": 4.99304924839206e-05, + "loss": 0.6126, + "step": 6070 + }, + { + "epoch": 0.3042662329538346, + "grad_norm": 25.98414421081543, + "learning_rate": 4.992122481511001e-05, + "loss": 0.2882, + "step": 6080 + }, + { + "epoch": 0.30476667083698233, + "grad_norm": 0.7517930269241333, + "learning_rate": 4.9911957146299424e-05, + "loss": 0.5424, + "step": 6090 + }, + { + "epoch": 0.30526710872013013, + "grad_norm": 8.64201545715332, + "learning_rate": 4.990268947748883e-05, + "loss": 0.5444, + "step": 6100 + }, + { + "epoch": 0.3057675466032779, + "grad_norm": 0.33267584443092346, + "learning_rate": 4.989342180867825e-05, + "loss": 0.7034, + "step": 6110 + }, + { + "epoch": 0.3062679844864256, + "grad_norm": 27.781362533569336, + "learning_rate": 4.9884154139867664e-05, + "loss": 0.4851, + "step": 6120 + }, + { + "epoch": 0.30676842236957336, + "grad_norm": 56.12293243408203, + "learning_rate": 4.987488647105707e-05, + "loss": 0.7229, + "step": 6130 + }, + { + "epoch": 0.30726886025272115, + "grad_norm": 4.40660285949707, + "learning_rate": 4.986561880224648e-05, + "loss": 0.3026, + "step": 6140 + }, + { + "epoch": 0.3077692981358689, + "grad_norm": 22.652729034423828, + "learning_rate": 4.98563511334359e-05, + "loss": 0.9387, + "step": 6150 + }, + { + "epoch": 0.30826973601901664, + "grad_norm": 1.3265471458435059, + "learning_rate": 4.984708346462531e-05, + "loss": 0.7199, + "step": 6160 + }, + { + "epoch": 0.3087701739021644, + "grad_norm": 1.7437008619308472, + "learning_rate": 4.983781579581472e-05, + "loss": 0.6116, + "step": 6170 + }, + { + "epoch": 0.3092706117853121, + "grad_norm": 1.7282524108886719, + "learning_rate": 4.982854812700414e-05, + "loss": 0.9018, + "step": 6180 + }, + { + "epoch": 0.3097710496684599, + "grad_norm": 14.887154579162598, + "learning_rate": 4.9819280458193546e-05, + "loss": 0.3707, + "step": 6190 + }, + { + "epoch": 0.31027148755160766, + "grad_norm": 43.594017028808594, + "learning_rate": 4.981001278938296e-05, + "loss": 0.2491, + "step": 6200 + }, + { + "epoch": 0.3107719254347554, + "grad_norm": 23.61724090576172, + "learning_rate": 4.980074512057237e-05, + "loss": 0.8867, + "step": 6210 + }, + { + "epoch": 0.31127236331790314, + "grad_norm": 47.816673278808594, + "learning_rate": 4.9791477451761786e-05, + "loss": 0.4937, + "step": 6220 + }, + { + "epoch": 0.31177280120105094, + "grad_norm": 52.2849006652832, + "learning_rate": 4.97822097829512e-05, + "loss": 0.6106, + "step": 6230 + }, + { + "epoch": 0.3122732390841987, + "grad_norm": 47.56501388549805, + "learning_rate": 4.977294211414061e-05, + "loss": 0.3114, + "step": 6240 + }, + { + "epoch": 0.3127736769673464, + "grad_norm": 11.660497665405273, + "learning_rate": 4.976367444533002e-05, + "loss": 0.5937, + "step": 6250 + }, + { + "epoch": 0.31327411485049417, + "grad_norm": 19.47123908996582, + "learning_rate": 4.9754406776519435e-05, + "loss": 0.5888, + "step": 6260 + }, + { + "epoch": 0.3137745527336419, + "grad_norm": 9.19548225402832, + "learning_rate": 4.974513910770885e-05, + "loss": 0.3769, + "step": 6270 + }, + { + "epoch": 0.3142749906167897, + "grad_norm": 14.6609525680542, + "learning_rate": 4.973587143889826e-05, + "loss": 0.493, + "step": 6280 + }, + { + "epoch": 0.31477542849993745, + "grad_norm": 18.433792114257812, + "learning_rate": 4.9726603770087676e-05, + "loss": 0.1996, + "step": 6290 + }, + { + "epoch": 0.3152758663830852, + "grad_norm": 0.6270882487297058, + "learning_rate": 4.9717336101277084e-05, + "loss": 0.899, + "step": 6300 + }, + { + "epoch": 0.31577630426623293, + "grad_norm": 23.794586181640625, + "learning_rate": 4.97080684324665e-05, + "loss": 0.54, + "step": 6310 + }, + { + "epoch": 0.31627674214938073, + "grad_norm": 1.039951205253601, + "learning_rate": 4.969880076365591e-05, + "loss": 0.3716, + "step": 6320 + }, + { + "epoch": 0.31677718003252847, + "grad_norm": 16.250816345214844, + "learning_rate": 4.9689533094845324e-05, + "loss": 0.8382, + "step": 6330 + }, + { + "epoch": 0.3172776179156762, + "grad_norm": 0.7573362588882446, + "learning_rate": 4.968026542603474e-05, + "loss": 0.7764, + "step": 6340 + }, + { + "epoch": 0.31777805579882396, + "grad_norm": 104.06665802001953, + "learning_rate": 4.9670997757224156e-05, + "loss": 0.8312, + "step": 6350 + }, + { + "epoch": 0.31827849368197175, + "grad_norm": 10.37952995300293, + "learning_rate": 4.966173008841356e-05, + "loss": 0.648, + "step": 6360 + }, + { + "epoch": 0.3187789315651195, + "grad_norm": 14.41804027557373, + "learning_rate": 4.9652462419602973e-05, + "loss": 0.4216, + "step": 6370 + }, + { + "epoch": 0.31927936944826724, + "grad_norm": 45.219383239746094, + "learning_rate": 4.964319475079239e-05, + "loss": 0.8012, + "step": 6380 + }, + { + "epoch": 0.319779807331415, + "grad_norm": 1.9420487880706787, + "learning_rate": 4.9633927081981805e-05, + "loss": 0.5221, + "step": 6390 + }, + { + "epoch": 0.3202802452145627, + "grad_norm": 35.528907775878906, + "learning_rate": 4.9624659413171214e-05, + "loss": 0.3094, + "step": 6400 + }, + { + "epoch": 0.3207806830977105, + "grad_norm": 43.37843704223633, + "learning_rate": 4.961539174436063e-05, + "loss": 0.3745, + "step": 6410 + }, + { + "epoch": 0.32128112098085826, + "grad_norm": 56.60279083251953, + "learning_rate": 4.960612407555004e-05, + "loss": 0.8886, + "step": 6420 + }, + { + "epoch": 0.321781558864006, + "grad_norm": 9.306534767150879, + "learning_rate": 4.959685640673945e-05, + "loss": 0.7597, + "step": 6430 + }, + { + "epoch": 0.32228199674715374, + "grad_norm": 10.197308540344238, + "learning_rate": 4.958758873792886e-05, + "loss": 0.8718, + "step": 6440 + }, + { + "epoch": 0.32278243463030154, + "grad_norm": 17.126554489135742, + "learning_rate": 4.957832106911828e-05, + "loss": 0.4675, + "step": 6450 + }, + { + "epoch": 0.3232828725134493, + "grad_norm": 25.480859756469727, + "learning_rate": 4.9569053400307694e-05, + "loss": 0.5063, + "step": 6460 + }, + { + "epoch": 0.323783310396597, + "grad_norm": 18.25585174560547, + "learning_rate": 4.9559785731497096e-05, + "loss": 0.3212, + "step": 6470 + }, + { + "epoch": 0.32428374827974477, + "grad_norm": 42.45457458496094, + "learning_rate": 4.955051806268651e-05, + "loss": 0.5771, + "step": 6480 + }, + { + "epoch": 0.3247841861628925, + "grad_norm": 90.06767272949219, + "learning_rate": 4.954125039387593e-05, + "loss": 0.6319, + "step": 6490 + }, + { + "epoch": 0.3252846240460403, + "grad_norm": 54.73312759399414, + "learning_rate": 4.953198272506534e-05, + "loss": 0.4241, + "step": 6500 + }, + { + "epoch": 0.32578506192918805, + "grad_norm": 11.47776985168457, + "learning_rate": 4.952271505625475e-05, + "loss": 0.6681, + "step": 6510 + }, + { + "epoch": 0.3262854998123358, + "grad_norm": 83.90605926513672, + "learning_rate": 4.951344738744417e-05, + "loss": 0.351, + "step": 6520 + }, + { + "epoch": 0.32678593769548353, + "grad_norm": 23.471220016479492, + "learning_rate": 4.9504179718633576e-05, + "loss": 0.5022, + "step": 6530 + }, + { + "epoch": 0.32728637557863133, + "grad_norm": 17.77142333984375, + "learning_rate": 4.949491204982299e-05, + "loss": 0.6412, + "step": 6540 + }, + { + "epoch": 0.32778681346177907, + "grad_norm": 12.178726196289062, + "learning_rate": 4.94856443810124e-05, + "loss": 0.5388, + "step": 6550 + }, + { + "epoch": 0.3282872513449268, + "grad_norm": 20.522762298583984, + "learning_rate": 4.9476376712201816e-05, + "loss": 0.3232, + "step": 6560 + }, + { + "epoch": 0.32878768922807455, + "grad_norm": 73.15330505371094, + "learning_rate": 4.946710904339123e-05, + "loss": 0.4342, + "step": 6570 + }, + { + "epoch": 0.3292881271112223, + "grad_norm": 30.652236938476562, + "learning_rate": 4.945784137458064e-05, + "loss": 0.6115, + "step": 6580 + }, + { + "epoch": 0.3297885649943701, + "grad_norm": 106.81671905517578, + "learning_rate": 4.944857370577005e-05, + "loss": 0.3994, + "step": 6590 + }, + { + "epoch": 0.33028900287751783, + "grad_norm": 61.435821533203125, + "learning_rate": 4.9439306036959465e-05, + "loss": 0.6535, + "step": 6600 + }, + { + "epoch": 0.3307894407606656, + "grad_norm": 2.866544723510742, + "learning_rate": 4.943003836814888e-05, + "loss": 0.366, + "step": 6610 + }, + { + "epoch": 0.3312898786438133, + "grad_norm": 36.22461700439453, + "learning_rate": 4.942077069933829e-05, + "loss": 0.3098, + "step": 6620 + }, + { + "epoch": 0.3317903165269611, + "grad_norm": 74.52408599853516, + "learning_rate": 4.9411503030527705e-05, + "loss": 1.05, + "step": 6630 + }, + { + "epoch": 0.33229075441010886, + "grad_norm": 64.95677947998047, + "learning_rate": 4.9402235361717114e-05, + "loss": 0.4895, + "step": 6640 + }, + { + "epoch": 0.3327911922932566, + "grad_norm": 7.573716640472412, + "learning_rate": 4.939296769290653e-05, + "loss": 0.4911, + "step": 6650 + }, + { + "epoch": 0.33329163017640434, + "grad_norm": 34.00478744506836, + "learning_rate": 4.938370002409594e-05, + "loss": 0.6266, + "step": 6660 + }, + { + "epoch": 0.3337920680595521, + "grad_norm": 36.888572692871094, + "learning_rate": 4.9374432355285354e-05, + "loss": 0.6151, + "step": 6670 + }, + { + "epoch": 0.3342925059426999, + "grad_norm": 9.386990547180176, + "learning_rate": 4.936516468647477e-05, + "loss": 0.6382, + "step": 6680 + }, + { + "epoch": 0.3347929438258476, + "grad_norm": 0.5393758416175842, + "learning_rate": 4.935589701766418e-05, + "loss": 0.3746, + "step": 6690 + }, + { + "epoch": 0.33529338170899536, + "grad_norm": 10.450437545776367, + "learning_rate": 4.934662934885359e-05, + "loss": 0.43, + "step": 6700 + }, + { + "epoch": 0.3357938195921431, + "grad_norm": 25.651687622070312, + "learning_rate": 4.9337361680043003e-05, + "loss": 0.5194, + "step": 6710 + }, + { + "epoch": 0.3362942574752909, + "grad_norm": 11.495452880859375, + "learning_rate": 4.932809401123242e-05, + "loss": 0.4864, + "step": 6720 + }, + { + "epoch": 0.33679469535843864, + "grad_norm": 3.7020294666290283, + "learning_rate": 4.931882634242183e-05, + "loss": 0.7354, + "step": 6730 + }, + { + "epoch": 0.3372951332415864, + "grad_norm": 2.2048463821411133, + "learning_rate": 4.9309558673611244e-05, + "loss": 0.209, + "step": 6740 + }, + { + "epoch": 0.33779557112473413, + "grad_norm": 30.37478256225586, + "learning_rate": 4.930029100480065e-05, + "loss": 0.6331, + "step": 6750 + }, + { + "epoch": 0.33829600900788187, + "grad_norm": 56.73005294799805, + "learning_rate": 4.929102333599007e-05, + "loss": 0.8081, + "step": 6760 + }, + { + "epoch": 0.33879644689102967, + "grad_norm": 32.82266616821289, + "learning_rate": 4.928175566717948e-05, + "loss": 0.7522, + "step": 6770 + }, + { + "epoch": 0.3392968847741774, + "grad_norm": 38.313419342041016, + "learning_rate": 4.927248799836889e-05, + "loss": 0.6893, + "step": 6780 + }, + { + "epoch": 0.33979732265732515, + "grad_norm": 12.957073211669922, + "learning_rate": 4.926322032955831e-05, + "loss": 0.4549, + "step": 6790 + }, + { + "epoch": 0.3402977605404729, + "grad_norm": 20.804967880249023, + "learning_rate": 4.925395266074772e-05, + "loss": 0.5444, + "step": 6800 + }, + { + "epoch": 0.3407981984236207, + "grad_norm": 29.80669403076172, + "learning_rate": 4.9244684991937126e-05, + "loss": 0.4235, + "step": 6810 + }, + { + "epoch": 0.34129863630676843, + "grad_norm": 19.53624153137207, + "learning_rate": 4.923541732312654e-05, + "loss": 0.6608, + "step": 6820 + }, + { + "epoch": 0.3417990741899162, + "grad_norm": 11.329672813415527, + "learning_rate": 4.922614965431596e-05, + "loss": 0.5317, + "step": 6830 + }, + { + "epoch": 0.3422995120730639, + "grad_norm": 0.13976283371448517, + "learning_rate": 4.921688198550537e-05, + "loss": 0.3358, + "step": 6840 + }, + { + "epoch": 0.34279994995621166, + "grad_norm": 0.046013716608285904, + "learning_rate": 4.920761431669478e-05, + "loss": 0.7503, + "step": 6850 + }, + { + "epoch": 0.34330038783935946, + "grad_norm": 0.09388890862464905, + "learning_rate": 4.919834664788419e-05, + "loss": 0.8708, + "step": 6860 + }, + { + "epoch": 0.3438008257225072, + "grad_norm": 0.012222466059029102, + "learning_rate": 4.9189078979073606e-05, + "loss": 0.5868, + "step": 6870 + }, + { + "epoch": 0.34430126360565494, + "grad_norm": 24.110374450683594, + "learning_rate": 4.917981131026302e-05, + "loss": 0.6065, + "step": 6880 + }, + { + "epoch": 0.3448017014888027, + "grad_norm": 86.32856750488281, + "learning_rate": 4.917054364145243e-05, + "loss": 0.3722, + "step": 6890 + }, + { + "epoch": 0.3453021393719505, + "grad_norm": 27.401872634887695, + "learning_rate": 4.9161275972641846e-05, + "loss": 0.4248, + "step": 6900 + }, + { + "epoch": 0.3458025772550982, + "grad_norm": 4.476280689239502, + "learning_rate": 4.9152008303831255e-05, + "loss": 0.2605, + "step": 6910 + }, + { + "epoch": 0.34630301513824596, + "grad_norm": 34.40840148925781, + "learning_rate": 4.914274063502067e-05, + "loss": 0.6433, + "step": 6920 + }, + { + "epoch": 0.3468034530213937, + "grad_norm": 37.493465423583984, + "learning_rate": 4.913347296621008e-05, + "loss": 0.9055, + "step": 6930 + }, + { + "epoch": 0.3473038909045415, + "grad_norm": 2.6361935138702393, + "learning_rate": 4.9124205297399495e-05, + "loss": 0.6275, + "step": 6940 + }, + { + "epoch": 0.34780432878768924, + "grad_norm": 24.92011833190918, + "learning_rate": 4.911493762858891e-05, + "loss": 0.1248, + "step": 6950 + }, + { + "epoch": 0.348304766670837, + "grad_norm": 11.696535110473633, + "learning_rate": 4.910566995977832e-05, + "loss": 0.6394, + "step": 6960 + }, + { + "epoch": 0.3488052045539847, + "grad_norm": 1.2686972618103027, + "learning_rate": 4.909640229096773e-05, + "loss": 0.5784, + "step": 6970 + }, + { + "epoch": 0.34930564243713247, + "grad_norm": 0.7129910588264465, + "learning_rate": 4.9087134622157144e-05, + "loss": 0.3244, + "step": 6980 + }, + { + "epoch": 0.34980608032028027, + "grad_norm": 5.041841506958008, + "learning_rate": 4.907786695334656e-05, + "loss": 0.1913, + "step": 6990 + }, + { + "epoch": 0.350306518203428, + "grad_norm": 88.30072021484375, + "learning_rate": 4.906859928453597e-05, + "loss": 1.1403, + "step": 7000 + }, + { + "epoch": 0.35080695608657575, + "grad_norm": 6.836981296539307, + "learning_rate": 4.9059331615725384e-05, + "loss": 0.3484, + "step": 7010 + }, + { + "epoch": 0.3513073939697235, + "grad_norm": 30.36409568786621, + "learning_rate": 4.905006394691479e-05, + "loss": 0.6701, + "step": 7020 + }, + { + "epoch": 0.3518078318528713, + "grad_norm": 2.6202645301818848, + "learning_rate": 4.904079627810421e-05, + "loss": 0.4325, + "step": 7030 + }, + { + "epoch": 0.35230826973601903, + "grad_norm": 78.381591796875, + "learning_rate": 4.903152860929362e-05, + "loss": 0.2599, + "step": 7040 + }, + { + "epoch": 0.3528087076191668, + "grad_norm": 22.566951751708984, + "learning_rate": 4.902226094048303e-05, + "loss": 0.4269, + "step": 7050 + }, + { + "epoch": 0.3533091455023145, + "grad_norm": 13.750670433044434, + "learning_rate": 4.901299327167245e-05, + "loss": 0.8757, + "step": 7060 + }, + { + "epoch": 0.35380958338546226, + "grad_norm": 47.56636428833008, + "learning_rate": 4.900372560286186e-05, + "loss": 0.3217, + "step": 7070 + }, + { + "epoch": 0.35431002126861005, + "grad_norm": 56.46306228637695, + "learning_rate": 4.899445793405127e-05, + "loss": 0.6815, + "step": 7080 + }, + { + "epoch": 0.3548104591517578, + "grad_norm": 16.048648834228516, + "learning_rate": 4.898519026524068e-05, + "loss": 0.5136, + "step": 7090 + }, + { + "epoch": 0.35531089703490554, + "grad_norm": 87.85322570800781, + "learning_rate": 4.89759225964301e-05, + "loss": 0.8615, + "step": 7100 + }, + { + "epoch": 0.3558113349180533, + "grad_norm": 38.32938003540039, + "learning_rate": 4.896665492761951e-05, + "loss": 0.5713, + "step": 7110 + }, + { + "epoch": 0.3563117728012011, + "grad_norm": 30.651756286621094, + "learning_rate": 4.895738725880892e-05, + "loss": 0.4231, + "step": 7120 + }, + { + "epoch": 0.3568122106843488, + "grad_norm": 0.7645180225372314, + "learning_rate": 4.894811958999834e-05, + "loss": 0.4894, + "step": 7130 + }, + { + "epoch": 0.35731264856749656, + "grad_norm": 51.425926208496094, + "learning_rate": 4.893885192118775e-05, + "loss": 0.5094, + "step": 7140 + }, + { + "epoch": 0.3578130864506443, + "grad_norm": 34.11625289916992, + "learning_rate": 4.8929584252377156e-05, + "loss": 0.7538, + "step": 7150 + }, + { + "epoch": 0.35831352433379204, + "grad_norm": 30.11433219909668, + "learning_rate": 4.892031658356657e-05, + "loss": 0.4404, + "step": 7160 + }, + { + "epoch": 0.35881396221693984, + "grad_norm": 11.451240539550781, + "learning_rate": 4.891104891475599e-05, + "loss": 0.485, + "step": 7170 + }, + { + "epoch": 0.3593144001000876, + "grad_norm": 27.60335922241211, + "learning_rate": 4.89017812459454e-05, + "loss": 0.2893, + "step": 7180 + }, + { + "epoch": 0.3598148379832353, + "grad_norm": 48.55904006958008, + "learning_rate": 4.8892513577134805e-05, + "loss": 0.4074, + "step": 7190 + }, + { + "epoch": 0.36031527586638307, + "grad_norm": 14.226167678833008, + "learning_rate": 4.888324590832422e-05, + "loss": 0.4954, + "step": 7200 + }, + { + "epoch": 0.36081571374953086, + "grad_norm": 23.349485397338867, + "learning_rate": 4.8873978239513636e-05, + "loss": 0.492, + "step": 7210 + }, + { + "epoch": 0.3613161516326786, + "grad_norm": 91.44820404052734, + "learning_rate": 4.886471057070305e-05, + "loss": 0.6948, + "step": 7220 + }, + { + "epoch": 0.36181658951582635, + "grad_norm": 0.04786834865808487, + "learning_rate": 4.885544290189246e-05, + "loss": 0.5047, + "step": 7230 + }, + { + "epoch": 0.3623170273989741, + "grad_norm": 36.04766082763672, + "learning_rate": 4.8846175233081876e-05, + "loss": 0.4945, + "step": 7240 + }, + { + "epoch": 0.36281746528212183, + "grad_norm": 68.36913299560547, + "learning_rate": 4.8836907564271285e-05, + "loss": 0.5026, + "step": 7250 + }, + { + "epoch": 0.36331790316526963, + "grad_norm": 4.6468329429626465, + "learning_rate": 4.88276398954607e-05, + "loss": 0.2048, + "step": 7260 + }, + { + "epoch": 0.36381834104841737, + "grad_norm": 34.274967193603516, + "learning_rate": 4.881837222665011e-05, + "loss": 0.3419, + "step": 7270 + }, + { + "epoch": 0.3643187789315651, + "grad_norm": 1.0458893775939941, + "learning_rate": 4.8809104557839525e-05, + "loss": 0.7504, + "step": 7280 + }, + { + "epoch": 0.36481921681471285, + "grad_norm": 20.87972068786621, + "learning_rate": 4.879983688902894e-05, + "loss": 0.4922, + "step": 7290 + }, + { + "epoch": 0.36531965469786065, + "grad_norm": 3.158039093017578, + "learning_rate": 4.879056922021834e-05, + "loss": 0.3803, + "step": 7300 + }, + { + "epoch": 0.3658200925810084, + "grad_norm": 0.5562414526939392, + "learning_rate": 4.878130155140776e-05, + "loss": 0.599, + "step": 7310 + }, + { + "epoch": 0.36632053046415614, + "grad_norm": 9.293462753295898, + "learning_rate": 4.8772033882597174e-05, + "loss": 0.5058, + "step": 7320 + }, + { + "epoch": 0.3668209683473039, + "grad_norm": 0.04835820943117142, + "learning_rate": 4.876276621378659e-05, + "loss": 0.248, + "step": 7330 + }, + { + "epoch": 0.3673214062304516, + "grad_norm": 64.75221252441406, + "learning_rate": 4.8753498544976e-05, + "loss": 0.5167, + "step": 7340 + }, + { + "epoch": 0.3678218441135994, + "grad_norm": 6.7504706382751465, + "learning_rate": 4.8744230876165414e-05, + "loss": 0.6467, + "step": 7350 + }, + { + "epoch": 0.36832228199674716, + "grad_norm": 6.879955768585205, + "learning_rate": 4.873496320735482e-05, + "loss": 0.4869, + "step": 7360 + }, + { + "epoch": 0.3688227198798949, + "grad_norm": 32.34907531738281, + "learning_rate": 4.872569553854424e-05, + "loss": 0.3564, + "step": 7370 + }, + { + "epoch": 0.36932315776304264, + "grad_norm": 46.09339141845703, + "learning_rate": 4.871642786973365e-05, + "loss": 0.4434, + "step": 7380 + }, + { + "epoch": 0.36982359564619044, + "grad_norm": 44.977569580078125, + "learning_rate": 4.870716020092306e-05, + "loss": 0.5531, + "step": 7390 + }, + { + "epoch": 0.3703240335293382, + "grad_norm": 39.56232452392578, + "learning_rate": 4.869789253211248e-05, + "loss": 0.4697, + "step": 7400 + }, + { + "epoch": 0.3708244714124859, + "grad_norm": 27.89874267578125, + "learning_rate": 4.868862486330189e-05, + "loss": 0.749, + "step": 7410 + }, + { + "epoch": 0.37132490929563366, + "grad_norm": 93.8678970336914, + "learning_rate": 4.8679357194491297e-05, + "loss": 0.5323, + "step": 7420 + }, + { + "epoch": 0.3718253471787814, + "grad_norm": 36.62017822265625, + "learning_rate": 4.867008952568071e-05, + "loss": 0.7628, + "step": 7430 + }, + { + "epoch": 0.3723257850619292, + "grad_norm": 14.321237564086914, + "learning_rate": 4.866082185687013e-05, + "loss": 0.3953, + "step": 7440 + }, + { + "epoch": 0.37282622294507695, + "grad_norm": 6.287460803985596, + "learning_rate": 4.865155418805954e-05, + "loss": 0.4048, + "step": 7450 + }, + { + "epoch": 0.3733266608282247, + "grad_norm": 78.92737579345703, + "learning_rate": 4.864228651924895e-05, + "loss": 0.2257, + "step": 7460 + }, + { + "epoch": 0.37382709871137243, + "grad_norm": 24.679553985595703, + "learning_rate": 4.863301885043836e-05, + "loss": 0.2347, + "step": 7470 + }, + { + "epoch": 0.3743275365945202, + "grad_norm": 28.05343246459961, + "learning_rate": 4.862375118162778e-05, + "loss": 0.8897, + "step": 7480 + }, + { + "epoch": 0.37482797447766797, + "grad_norm": 16.718219757080078, + "learning_rate": 4.8614483512817186e-05, + "loss": 0.3193, + "step": 7490 + }, + { + "epoch": 0.3753284123608157, + "grad_norm": 6.819782733917236, + "learning_rate": 4.86052158440066e-05, + "loss": 0.392, + "step": 7500 + }, + { + "epoch": 0.37582885024396345, + "grad_norm": 0.014201158657670021, + "learning_rate": 4.859594817519602e-05, + "loss": 0.7052, + "step": 7510 + }, + { + "epoch": 0.37632928812711125, + "grad_norm": 22.267745971679688, + "learning_rate": 4.8586680506385426e-05, + "loss": 0.3143, + "step": 7520 + }, + { + "epoch": 0.376829726010259, + "grad_norm": 0.058933962136507034, + "learning_rate": 4.8577412837574835e-05, + "loss": 0.7308, + "step": 7530 + }, + { + "epoch": 0.37733016389340673, + "grad_norm": 0.630791425704956, + "learning_rate": 4.856814516876425e-05, + "loss": 0.5858, + "step": 7540 + }, + { + "epoch": 0.3778306017765545, + "grad_norm": 68.37287139892578, + "learning_rate": 4.8558877499953666e-05, + "loss": 0.435, + "step": 7550 + }, + { + "epoch": 0.3783310396597022, + "grad_norm": 170.09713745117188, + "learning_rate": 4.8549609831143075e-05, + "loss": 0.7447, + "step": 7560 + }, + { + "epoch": 0.37883147754285, + "grad_norm": 13.705775260925293, + "learning_rate": 4.854034216233249e-05, + "loss": 0.4809, + "step": 7570 + }, + { + "epoch": 0.37933191542599776, + "grad_norm": 8.5336275100708, + "learning_rate": 4.85310744935219e-05, + "loss": 0.6589, + "step": 7580 + }, + { + "epoch": 0.3798323533091455, + "grad_norm": 18.751121520996094, + "learning_rate": 4.8521806824711315e-05, + "loss": 0.4435, + "step": 7590 + }, + { + "epoch": 0.38033279119229324, + "grad_norm": 70.7726058959961, + "learning_rate": 4.8512539155900724e-05, + "loss": 0.5798, + "step": 7600 + }, + { + "epoch": 0.38083322907544104, + "grad_norm": 61.83354568481445, + "learning_rate": 4.850327148709014e-05, + "loss": 0.6288, + "step": 7610 + }, + { + "epoch": 0.3813336669585888, + "grad_norm": 115.94546508789062, + "learning_rate": 4.8494003818279555e-05, + "loss": 0.2644, + "step": 7620 + }, + { + "epoch": 0.3818341048417365, + "grad_norm": 51.835018157958984, + "learning_rate": 4.8484736149468964e-05, + "loss": 0.6293, + "step": 7630 + }, + { + "epoch": 0.38233454272488426, + "grad_norm": 17.199581146240234, + "learning_rate": 4.847546848065837e-05, + "loss": 0.2702, + "step": 7640 + }, + { + "epoch": 0.382834980608032, + "grad_norm": 50.910430908203125, + "learning_rate": 4.846620081184779e-05, + "loss": 0.3743, + "step": 7650 + }, + { + "epoch": 0.3833354184911798, + "grad_norm": 34.737056732177734, + "learning_rate": 4.8456933143037204e-05, + "loss": 0.3052, + "step": 7660 + }, + { + "epoch": 0.38383585637432754, + "grad_norm": 30.914241790771484, + "learning_rate": 4.844766547422662e-05, + "loss": 0.8248, + "step": 7670 + }, + { + "epoch": 0.3843362942574753, + "grad_norm": 24.568498611450195, + "learning_rate": 4.843839780541603e-05, + "loss": 0.654, + "step": 7680 + }, + { + "epoch": 0.38483673214062303, + "grad_norm": 0.3674372732639313, + "learning_rate": 4.842913013660544e-05, + "loss": 0.5826, + "step": 7690 + }, + { + "epoch": 0.3853371700237708, + "grad_norm": 21.260377883911133, + "learning_rate": 4.841986246779485e-05, + "loss": 0.2558, + "step": 7700 + }, + { + "epoch": 0.38583760790691857, + "grad_norm": 28.596790313720703, + "learning_rate": 4.841059479898427e-05, + "loss": 0.4601, + "step": 7710 + }, + { + "epoch": 0.3863380457900663, + "grad_norm": 37.284488677978516, + "learning_rate": 4.840132713017368e-05, + "loss": 0.8477, + "step": 7720 + }, + { + "epoch": 0.38683848367321405, + "grad_norm": 32.36518859863281, + "learning_rate": 4.839205946136309e-05, + "loss": 0.5656, + "step": 7730 + }, + { + "epoch": 0.3873389215563618, + "grad_norm": 69.78794860839844, + "learning_rate": 4.83827917925525e-05, + "loss": 0.92, + "step": 7740 + }, + { + "epoch": 0.3878393594395096, + "grad_norm": 16.729028701782227, + "learning_rate": 4.837352412374192e-05, + "loss": 0.3424, + "step": 7750 + }, + { + "epoch": 0.38833979732265733, + "grad_norm": 2.8585562705993652, + "learning_rate": 4.8364256454931326e-05, + "loss": 0.276, + "step": 7760 + }, + { + "epoch": 0.3888402352058051, + "grad_norm": 10.325372695922852, + "learning_rate": 4.835498878612074e-05, + "loss": 0.5653, + "step": 7770 + }, + { + "epoch": 0.3893406730889528, + "grad_norm": 21.951305389404297, + "learning_rate": 4.834572111731016e-05, + "loss": 0.5082, + "step": 7780 + }, + { + "epoch": 0.3898411109721006, + "grad_norm": 9.895001411437988, + "learning_rate": 4.833645344849957e-05, + "loss": 0.4686, + "step": 7790 + }, + { + "epoch": 0.39034154885524835, + "grad_norm": 37.18531799316406, + "learning_rate": 4.8327185779688975e-05, + "loss": 0.5513, + "step": 7800 + }, + { + "epoch": 0.3908419867383961, + "grad_norm": 41.870216369628906, + "learning_rate": 4.831791811087839e-05, + "loss": 0.6588, + "step": 7810 + }, + { + "epoch": 0.39134242462154384, + "grad_norm": 6.8220534324646, + "learning_rate": 4.830865044206781e-05, + "loss": 0.4663, + "step": 7820 + }, + { + "epoch": 0.3918428625046916, + "grad_norm": 58.477508544921875, + "learning_rate": 4.8299382773257216e-05, + "loss": 0.6161, + "step": 7830 + }, + { + "epoch": 0.3923433003878394, + "grad_norm": 40.91549301147461, + "learning_rate": 4.829011510444663e-05, + "loss": 0.6492, + "step": 7840 + }, + { + "epoch": 0.3928437382709871, + "grad_norm": 42.26517105102539, + "learning_rate": 4.828084743563605e-05, + "loss": 0.7754, + "step": 7850 + }, + { + "epoch": 0.39334417615413486, + "grad_norm": 43.52967834472656, + "learning_rate": 4.8271579766825456e-05, + "loss": 0.3819, + "step": 7860 + }, + { + "epoch": 0.3938446140372826, + "grad_norm": 9.116008758544922, + "learning_rate": 4.8262312098014865e-05, + "loss": 0.6246, + "step": 7870 + }, + { + "epoch": 0.3943450519204304, + "grad_norm": 4.597453594207764, + "learning_rate": 4.825304442920428e-05, + "loss": 0.5754, + "step": 7880 + }, + { + "epoch": 0.39484548980357814, + "grad_norm": 7.2050700187683105, + "learning_rate": 4.8243776760393696e-05, + "loss": 0.6797, + "step": 7890 + }, + { + "epoch": 0.3953459276867259, + "grad_norm": 14.294304847717285, + "learning_rate": 4.8234509091583105e-05, + "loss": 0.3782, + "step": 7900 + }, + { + "epoch": 0.3958463655698736, + "grad_norm": 44.5699348449707, + "learning_rate": 4.8225241422772514e-05, + "loss": 0.6546, + "step": 7910 + }, + { + "epoch": 0.39634680345302137, + "grad_norm": 25.647537231445312, + "learning_rate": 4.821597375396193e-05, + "loss": 0.3602, + "step": 7920 + }, + { + "epoch": 0.39684724133616917, + "grad_norm": 85.3887710571289, + "learning_rate": 4.8206706085151345e-05, + "loss": 0.6746, + "step": 7930 + }, + { + "epoch": 0.3973476792193169, + "grad_norm": 21.436100006103516, + "learning_rate": 4.8197438416340754e-05, + "loss": 0.4753, + "step": 7940 + }, + { + "epoch": 0.39784811710246465, + "grad_norm": 1.2359827756881714, + "learning_rate": 4.818817074753017e-05, + "loss": 0.4772, + "step": 7950 + }, + { + "epoch": 0.3983485549856124, + "grad_norm": 49.6717643737793, + "learning_rate": 4.8178903078719585e-05, + "loss": 0.2821, + "step": 7960 + }, + { + "epoch": 0.3988489928687602, + "grad_norm": 111.57734680175781, + "learning_rate": 4.8169635409908994e-05, + "loss": 0.6692, + "step": 7970 + }, + { + "epoch": 0.39934943075190793, + "grad_norm": 63.65989303588867, + "learning_rate": 4.81603677410984e-05, + "loss": 0.5731, + "step": 7980 + }, + { + "epoch": 0.39984986863505567, + "grad_norm": 97.56644439697266, + "learning_rate": 4.815110007228782e-05, + "loss": 0.7628, + "step": 7990 + }, + { + "epoch": 0.4003503065182034, + "grad_norm": 26.0208797454834, + "learning_rate": 4.8141832403477234e-05, + "loss": 0.4631, + "step": 8000 + }, + { + "epoch": 0.40085074440135116, + "grad_norm": 34.600284576416016, + "learning_rate": 4.813256473466665e-05, + "loss": 0.6117, + "step": 8010 + }, + { + "epoch": 0.40135118228449895, + "grad_norm": 1.1190848350524902, + "learning_rate": 4.812329706585605e-05, + "loss": 0.3606, + "step": 8020 + }, + { + "epoch": 0.4018516201676467, + "grad_norm": 3.9813952445983887, + "learning_rate": 4.811402939704547e-05, + "loss": 0.6105, + "step": 8030 + }, + { + "epoch": 0.40235205805079444, + "grad_norm": 8.145801544189453, + "learning_rate": 4.810476172823488e-05, + "loss": 0.3954, + "step": 8040 + }, + { + "epoch": 0.4028524959339422, + "grad_norm": 3.1184473037719727, + "learning_rate": 4.80954940594243e-05, + "loss": 0.5342, + "step": 8050 + }, + { + "epoch": 0.40335293381709, + "grad_norm": 57.58297348022461, + "learning_rate": 4.808622639061371e-05, + "loss": 0.4682, + "step": 8060 + }, + { + "epoch": 0.4038533717002377, + "grad_norm": 15.121479034423828, + "learning_rate": 4.807695872180312e-05, + "loss": 0.3199, + "step": 8070 + }, + { + "epoch": 0.40435380958338546, + "grad_norm": 1.005894422531128, + "learning_rate": 4.806769105299253e-05, + "loss": 0.8008, + "step": 8080 + }, + { + "epoch": 0.4048542474665332, + "grad_norm": 41.21805953979492, + "learning_rate": 4.805842338418195e-05, + "loss": 0.4306, + "step": 8090 + }, + { + "epoch": 0.40535468534968094, + "grad_norm": 1.2104356288909912, + "learning_rate": 4.8049155715371356e-05, + "loss": 0.6931, + "step": 8100 + }, + { + "epoch": 0.40585512323282874, + "grad_norm": 24.65207290649414, + "learning_rate": 4.803988804656077e-05, + "loss": 0.4671, + "step": 8110 + }, + { + "epoch": 0.4063555611159765, + "grad_norm": 0.6409622430801392, + "learning_rate": 4.803062037775019e-05, + "loss": 0.4616, + "step": 8120 + }, + { + "epoch": 0.4068559989991242, + "grad_norm": 23.632028579711914, + "learning_rate": 4.802135270893959e-05, + "loss": 0.4356, + "step": 8130 + }, + { + "epoch": 0.40735643688227197, + "grad_norm": 65.03801727294922, + "learning_rate": 4.8012085040129005e-05, + "loss": 0.5318, + "step": 8140 + }, + { + "epoch": 0.40785687476541976, + "grad_norm": 62.35992431640625, + "learning_rate": 4.800281737131842e-05, + "loss": 0.4288, + "step": 8150 + }, + { + "epoch": 0.4083573126485675, + "grad_norm": 98.27396392822266, + "learning_rate": 4.799354970250784e-05, + "loss": 0.5416, + "step": 8160 + }, + { + "epoch": 0.40885775053171525, + "grad_norm": 1.744025468826294, + "learning_rate": 4.7984282033697246e-05, + "loss": 0.5228, + "step": 8170 + }, + { + "epoch": 0.409358188414863, + "grad_norm": 204.17282104492188, + "learning_rate": 4.797501436488666e-05, + "loss": 0.7406, + "step": 8180 + }, + { + "epoch": 0.4098586262980108, + "grad_norm": 30.93715476989746, + "learning_rate": 4.796574669607607e-05, + "loss": 0.366, + "step": 8190 + }, + { + "epoch": 0.41035906418115853, + "grad_norm": 102.49114227294922, + "learning_rate": 4.7956479027265486e-05, + "loss": 0.5181, + "step": 8200 + }, + { + "epoch": 0.41085950206430627, + "grad_norm": 16.255374908447266, + "learning_rate": 4.7947211358454895e-05, + "loss": 0.622, + "step": 8210 + }, + { + "epoch": 0.411359939947454, + "grad_norm": 1.300433874130249, + "learning_rate": 4.793794368964431e-05, + "loss": 0.3012, + "step": 8220 + }, + { + "epoch": 0.41186037783060175, + "grad_norm": 13.411256790161133, + "learning_rate": 4.7928676020833726e-05, + "loss": 0.947, + "step": 8230 + }, + { + "epoch": 0.41236081571374955, + "grad_norm": 71.5157699584961, + "learning_rate": 4.7919408352023135e-05, + "loss": 0.6921, + "step": 8240 + }, + { + "epoch": 0.4128612535968973, + "grad_norm": 36.93513870239258, + "learning_rate": 4.7910140683212543e-05, + "loss": 0.4609, + "step": 8250 + }, + { + "epoch": 0.41336169148004503, + "grad_norm": 1.1887996196746826, + "learning_rate": 4.790087301440196e-05, + "loss": 0.7299, + "step": 8260 + }, + { + "epoch": 0.4138621293631928, + "grad_norm": 61.23517990112305, + "learning_rate": 4.7891605345591375e-05, + "loss": 0.2442, + "step": 8270 + }, + { + "epoch": 0.4143625672463406, + "grad_norm": 9.348570823669434, + "learning_rate": 4.7882337676780784e-05, + "loss": 0.5521, + "step": 8280 + }, + { + "epoch": 0.4148630051294883, + "grad_norm": 24.871320724487305, + "learning_rate": 4.78730700079702e-05, + "loss": 0.4828, + "step": 8290 + }, + { + "epoch": 0.41536344301263606, + "grad_norm": 19.84798240661621, + "learning_rate": 4.786380233915961e-05, + "loss": 0.397, + "step": 8300 + }, + { + "epoch": 0.4158638808957838, + "grad_norm": 21.86806297302246, + "learning_rate": 4.7854534670349024e-05, + "loss": 0.3601, + "step": 8310 + }, + { + "epoch": 0.41636431877893154, + "grad_norm": 56.05990982055664, + "learning_rate": 4.784526700153843e-05, + "loss": 0.8183, + "step": 8320 + }, + { + "epoch": 0.41686475666207934, + "grad_norm": 41.94434356689453, + "learning_rate": 4.783599933272785e-05, + "loss": 0.8404, + "step": 8330 + }, + { + "epoch": 0.4173651945452271, + "grad_norm": 0.15109919011592865, + "learning_rate": 4.7826731663917264e-05, + "loss": 0.543, + "step": 8340 + }, + { + "epoch": 0.4178656324283748, + "grad_norm": 43.0754508972168, + "learning_rate": 4.781746399510667e-05, + "loss": 0.7063, + "step": 8350 + }, + { + "epoch": 0.41836607031152256, + "grad_norm": 35.83393859863281, + "learning_rate": 4.780819632629608e-05, + "loss": 0.7821, + "step": 8360 + }, + { + "epoch": 0.41886650819467036, + "grad_norm": 13.206838607788086, + "learning_rate": 4.77989286574855e-05, + "loss": 0.3798, + "step": 8370 + }, + { + "epoch": 0.4193669460778181, + "grad_norm": 86.1915283203125, + "learning_rate": 4.778966098867491e-05, + "loss": 0.634, + "step": 8380 + }, + { + "epoch": 0.41986738396096585, + "grad_norm": 22.32910919189453, + "learning_rate": 4.778039331986432e-05, + "loss": 0.4809, + "step": 8390 + }, + { + "epoch": 0.4203678218441136, + "grad_norm": 17.621416091918945, + "learning_rate": 4.777112565105374e-05, + "loss": 0.3588, + "step": 8400 + }, + { + "epoch": 0.42086825972726133, + "grad_norm": 51.10501480102539, + "learning_rate": 4.7761857982243146e-05, + "loss": 0.6037, + "step": 8410 + }, + { + "epoch": 0.4213686976104091, + "grad_norm": 36.84767150878906, + "learning_rate": 4.775259031343256e-05, + "loss": 0.6141, + "step": 8420 + }, + { + "epoch": 0.42186913549355687, + "grad_norm": 0.5165908336639404, + "learning_rate": 4.774332264462197e-05, + "loss": 0.7533, + "step": 8430 + }, + { + "epoch": 0.4223695733767046, + "grad_norm": 32.501609802246094, + "learning_rate": 4.7734054975811386e-05, + "loss": 0.5696, + "step": 8440 + }, + { + "epoch": 0.42287001125985235, + "grad_norm": 33.41990280151367, + "learning_rate": 4.77247873070008e-05, + "loss": 0.3256, + "step": 8450 + }, + { + "epoch": 0.42337044914300015, + "grad_norm": 55.789161682128906, + "learning_rate": 4.771551963819021e-05, + "loss": 0.4106, + "step": 8460 + }, + { + "epoch": 0.4238708870261479, + "grad_norm": 59.46048355102539, + "learning_rate": 4.770625196937962e-05, + "loss": 0.7877, + "step": 8470 + }, + { + "epoch": 0.42437132490929563, + "grad_norm": 59.855682373046875, + "learning_rate": 4.7696984300569035e-05, + "loss": 0.6943, + "step": 8480 + }, + { + "epoch": 0.4248717627924434, + "grad_norm": 20.96187400817871, + "learning_rate": 4.768771663175845e-05, + "loss": 0.3357, + "step": 8490 + }, + { + "epoch": 0.4253722006755911, + "grad_norm": 4.456404685974121, + "learning_rate": 4.7678448962947867e-05, + "loss": 0.4805, + "step": 8500 + }, + { + "epoch": 0.4258726385587389, + "grad_norm": 23.46825408935547, + "learning_rate": 4.7669181294137275e-05, + "loss": 0.6059, + "step": 8510 + }, + { + "epoch": 0.42637307644188666, + "grad_norm": 5.344809055328369, + "learning_rate": 4.7659913625326684e-05, + "loss": 0.4898, + "step": 8520 + }, + { + "epoch": 0.4268735143250344, + "grad_norm": 45.70634078979492, + "learning_rate": 4.76506459565161e-05, + "loss": 0.413, + "step": 8530 + }, + { + "epoch": 0.42737395220818214, + "grad_norm": 2.496117115020752, + "learning_rate": 4.7641378287705516e-05, + "loss": 0.6031, + "step": 8540 + }, + { + "epoch": 0.42787439009132994, + "grad_norm": 18.412120819091797, + "learning_rate": 4.7632110618894924e-05, + "loss": 0.8433, + "step": 8550 + }, + { + "epoch": 0.4283748279744777, + "grad_norm": 17.193317413330078, + "learning_rate": 4.762284295008434e-05, + "loss": 0.4169, + "step": 8560 + }, + { + "epoch": 0.4288752658576254, + "grad_norm": 30.404123306274414, + "learning_rate": 4.7613575281273756e-05, + "loss": 0.3583, + "step": 8570 + }, + { + "epoch": 0.42937570374077316, + "grad_norm": 8.56922435760498, + "learning_rate": 4.7604307612463165e-05, + "loss": 0.4639, + "step": 8580 + }, + { + "epoch": 0.4298761416239209, + "grad_norm": 30.07659149169922, + "learning_rate": 4.7595039943652573e-05, + "loss": 0.2508, + "step": 8590 + }, + { + "epoch": 0.4303765795070687, + "grad_norm": 23.035900115966797, + "learning_rate": 4.758577227484199e-05, + "loss": 0.8127, + "step": 8600 + }, + { + "epoch": 0.43087701739021644, + "grad_norm": 34.50818634033203, + "learning_rate": 4.7576504606031405e-05, + "loss": 0.5479, + "step": 8610 + }, + { + "epoch": 0.4313774552733642, + "grad_norm": 22.202049255371094, + "learning_rate": 4.7567236937220814e-05, + "loss": 0.734, + "step": 8620 + }, + { + "epoch": 0.4318778931565119, + "grad_norm": 7.1242146492004395, + "learning_rate": 4.755796926841022e-05, + "loss": 0.3539, + "step": 8630 + }, + { + "epoch": 0.4323783310396597, + "grad_norm": 15.313249588012695, + "learning_rate": 4.754870159959964e-05, + "loss": 0.5873, + "step": 8640 + }, + { + "epoch": 0.43287876892280747, + "grad_norm": 36.42792892456055, + "learning_rate": 4.7539433930789054e-05, + "loss": 0.5712, + "step": 8650 + }, + { + "epoch": 0.4333792068059552, + "grad_norm": 34.10883331298828, + "learning_rate": 4.753016626197846e-05, + "loss": 0.5769, + "step": 8660 + }, + { + "epoch": 0.43387964468910295, + "grad_norm": 51.887603759765625, + "learning_rate": 4.752089859316788e-05, + "loss": 0.5565, + "step": 8670 + }, + { + "epoch": 0.4343800825722507, + "grad_norm": 114.74909973144531, + "learning_rate": 4.7511630924357294e-05, + "loss": 0.4919, + "step": 8680 + }, + { + "epoch": 0.4348805204553985, + "grad_norm": 38.43137741088867, + "learning_rate": 4.75023632555467e-05, + "loss": 0.5253, + "step": 8690 + }, + { + "epoch": 0.43538095833854623, + "grad_norm": 0.045635391026735306, + "learning_rate": 4.749309558673611e-05, + "loss": 0.4129, + "step": 8700 + }, + { + "epoch": 0.435881396221694, + "grad_norm": 14.743196487426758, + "learning_rate": 4.748382791792553e-05, + "loss": 0.2385, + "step": 8710 + }, + { + "epoch": 0.4363818341048417, + "grad_norm": 3.0845766067504883, + "learning_rate": 4.747456024911494e-05, + "loss": 0.3914, + "step": 8720 + }, + { + "epoch": 0.4368822719879895, + "grad_norm": 5.473989009857178, + "learning_rate": 4.746529258030435e-05, + "loss": 0.2732, + "step": 8730 + }, + { + "epoch": 0.43738270987113725, + "grad_norm": 64.6672134399414, + "learning_rate": 4.745602491149376e-05, + "loss": 0.1218, + "step": 8740 + }, + { + "epoch": 0.437883147754285, + "grad_norm": 37.79511260986328, + "learning_rate": 4.7446757242683176e-05, + "loss": 0.7965, + "step": 8750 + }, + { + "epoch": 0.43838358563743274, + "grad_norm": 10.202987670898438, + "learning_rate": 4.743748957387259e-05, + "loss": 0.5173, + "step": 8760 + }, + { + "epoch": 0.43888402352058054, + "grad_norm": 68.71057891845703, + "learning_rate": 4.7428221905062e-05, + "loss": 0.5417, + "step": 8770 + }, + { + "epoch": 0.4393844614037283, + "grad_norm": 23.784439086914062, + "learning_rate": 4.7418954236251416e-05, + "loss": 0.8463, + "step": 8780 + }, + { + "epoch": 0.439884899286876, + "grad_norm": 0.7189640998840332, + "learning_rate": 4.740968656744083e-05, + "loss": 0.2245, + "step": 8790 + }, + { + "epoch": 0.44038533717002376, + "grad_norm": 18.891815185546875, + "learning_rate": 4.740041889863024e-05, + "loss": 0.1493, + "step": 8800 + }, + { + "epoch": 0.4408857750531715, + "grad_norm": 3.1338696479797363, + "learning_rate": 4.739115122981965e-05, + "loss": 0.2329, + "step": 8810 + }, + { + "epoch": 0.4413862129363193, + "grad_norm": 6.971510887145996, + "learning_rate": 4.7381883561009065e-05, + "loss": 0.6144, + "step": 8820 + }, + { + "epoch": 0.44188665081946704, + "grad_norm": 32.47896194458008, + "learning_rate": 4.737261589219848e-05, + "loss": 0.6287, + "step": 8830 + }, + { + "epoch": 0.4423870887026148, + "grad_norm": 19.22687339782715, + "learning_rate": 4.7363348223387896e-05, + "loss": 0.3992, + "step": 8840 + }, + { + "epoch": 0.4428875265857625, + "grad_norm": 60.054710388183594, + "learning_rate": 4.73540805545773e-05, + "loss": 0.4959, + "step": 8850 + }, + { + "epoch": 0.4433879644689103, + "grad_norm": 9.602560997009277, + "learning_rate": 4.7344812885766714e-05, + "loss": 0.6127, + "step": 8860 + }, + { + "epoch": 0.44388840235205806, + "grad_norm": 23.029647827148438, + "learning_rate": 4.733554521695613e-05, + "loss": 0.5791, + "step": 8870 + }, + { + "epoch": 0.4443888402352058, + "grad_norm": 19.053245544433594, + "learning_rate": 4.7326277548145545e-05, + "loss": 0.6629, + "step": 8880 + }, + { + "epoch": 0.44488927811835355, + "grad_norm": 2.8608784675598145, + "learning_rate": 4.7317009879334954e-05, + "loss": 0.4008, + "step": 8890 + }, + { + "epoch": 0.4453897160015013, + "grad_norm": 13.415657043457031, + "learning_rate": 4.730774221052437e-05, + "loss": 0.2408, + "step": 8900 + }, + { + "epoch": 0.4458901538846491, + "grad_norm": 13.276653289794922, + "learning_rate": 4.729847454171378e-05, + "loss": 0.3785, + "step": 8910 + }, + { + "epoch": 0.44639059176779683, + "grad_norm": 68.37105560302734, + "learning_rate": 4.7289206872903194e-05, + "loss": 0.7513, + "step": 8920 + }, + { + "epoch": 0.44689102965094457, + "grad_norm": 46.02248001098633, + "learning_rate": 4.72799392040926e-05, + "loss": 0.593, + "step": 8930 + }, + { + "epoch": 0.4473914675340923, + "grad_norm": 3.238001823425293, + "learning_rate": 4.727067153528202e-05, + "loss": 0.823, + "step": 8940 + }, + { + "epoch": 0.4478919054172401, + "grad_norm": 11.21345329284668, + "learning_rate": 4.7261403866471435e-05, + "loss": 0.8661, + "step": 8950 + }, + { + "epoch": 0.44839234330038785, + "grad_norm": 6.572399616241455, + "learning_rate": 4.7252136197660843e-05, + "loss": 0.2147, + "step": 8960 + }, + { + "epoch": 0.4488927811835356, + "grad_norm": 104.5406494140625, + "learning_rate": 4.724286852885025e-05, + "loss": 0.548, + "step": 8970 + }, + { + "epoch": 0.44939321906668334, + "grad_norm": 0.04405208304524422, + "learning_rate": 4.723360086003967e-05, + "loss": 0.4883, + "step": 8980 + }, + { + "epoch": 0.4498936569498311, + "grad_norm": 1.5620454549789429, + "learning_rate": 4.7224333191229084e-05, + "loss": 0.2437, + "step": 8990 + }, + { + "epoch": 0.4503940948329789, + "grad_norm": 12.364811897277832, + "learning_rate": 4.721506552241849e-05, + "loss": 0.4734, + "step": 9000 + }, + { + "epoch": 0.4508945327161266, + "grad_norm": 30.107053756713867, + "learning_rate": 4.720579785360791e-05, + "loss": 0.5154, + "step": 9010 + }, + { + "epoch": 0.45139497059927436, + "grad_norm": 42.41006088256836, + "learning_rate": 4.719653018479732e-05, + "loss": 0.6473, + "step": 9020 + }, + { + "epoch": 0.4518954084824221, + "grad_norm": 0.15835988521575928, + "learning_rate": 4.718726251598673e-05, + "loss": 0.3704, + "step": 9030 + }, + { + "epoch": 0.4523958463655699, + "grad_norm": 39.742279052734375, + "learning_rate": 4.717799484717614e-05, + "loss": 0.3565, + "step": 9040 + }, + { + "epoch": 0.45289628424871764, + "grad_norm": 15.246678352355957, + "learning_rate": 4.716872717836556e-05, + "loss": 1.1606, + "step": 9050 + }, + { + "epoch": 0.4533967221318654, + "grad_norm": 27.0163516998291, + "learning_rate": 4.715945950955497e-05, + "loss": 0.4206, + "step": 9060 + }, + { + "epoch": 0.4538971600150131, + "grad_norm": 9.110934257507324, + "learning_rate": 4.715019184074438e-05, + "loss": 0.3902, + "step": 9070 + }, + { + "epoch": 0.45439759789816087, + "grad_norm": 16.940574645996094, + "learning_rate": 4.714092417193379e-05, + "loss": 0.3678, + "step": 9080 + }, + { + "epoch": 0.45489803578130866, + "grad_norm": 62.786407470703125, + "learning_rate": 4.7131656503123206e-05, + "loss": 0.3126, + "step": 9090 + }, + { + "epoch": 0.4553984736644564, + "grad_norm": 72.67662811279297, + "learning_rate": 4.712238883431262e-05, + "loss": 0.8075, + "step": 9100 + }, + { + "epoch": 0.45589891154760415, + "grad_norm": 86.25459289550781, + "learning_rate": 4.711312116550203e-05, + "loss": 0.7849, + "step": 9110 + }, + { + "epoch": 0.4563993494307519, + "grad_norm": 8.174720764160156, + "learning_rate": 4.7103853496691446e-05, + "loss": 0.6435, + "step": 9120 + }, + { + "epoch": 0.4568997873138997, + "grad_norm": 36.88053894042969, + "learning_rate": 4.7094585827880855e-05, + "loss": 0.8672, + "step": 9130 + }, + { + "epoch": 0.4574002251970474, + "grad_norm": 2.765181303024292, + "learning_rate": 4.708531815907027e-05, + "loss": 0.4812, + "step": 9140 + }, + { + "epoch": 0.45790066308019517, + "grad_norm": 7.532327175140381, + "learning_rate": 4.707605049025968e-05, + "loss": 0.4624, + "step": 9150 + }, + { + "epoch": 0.4584011009633429, + "grad_norm": 1.858641266822815, + "learning_rate": 4.7066782821449095e-05, + "loss": 0.3535, + "step": 9160 + }, + { + "epoch": 0.45890153884649065, + "grad_norm": 66.69048309326172, + "learning_rate": 4.705751515263851e-05, + "loss": 0.7522, + "step": 9170 + }, + { + "epoch": 0.45940197672963845, + "grad_norm": 22.506149291992188, + "learning_rate": 4.704824748382792e-05, + "loss": 0.3594, + "step": 9180 + }, + { + "epoch": 0.4599024146127862, + "grad_norm": 3.9952759742736816, + "learning_rate": 4.703897981501733e-05, + "loss": 0.3257, + "step": 9190 + }, + { + "epoch": 0.46040285249593393, + "grad_norm": 7.830325603485107, + "learning_rate": 4.7029712146206744e-05, + "loss": 0.4078, + "step": 9200 + }, + { + "epoch": 0.4609032903790817, + "grad_norm": 17.37978744506836, + "learning_rate": 4.702044447739616e-05, + "loss": 0.815, + "step": 9210 + }, + { + "epoch": 0.4614037282622295, + "grad_norm": 4.104501247406006, + "learning_rate": 4.701117680858557e-05, + "loss": 0.5996, + "step": 9220 + }, + { + "epoch": 0.4619041661453772, + "grad_norm": 13.586810111999512, + "learning_rate": 4.7001909139774984e-05, + "loss": 0.6525, + "step": 9230 + }, + { + "epoch": 0.46240460402852496, + "grad_norm": 6.0534443855285645, + "learning_rate": 4.699264147096439e-05, + "loss": 0.3982, + "step": 9240 + }, + { + "epoch": 0.4629050419116727, + "grad_norm": 1.09138023853302, + "learning_rate": 4.698337380215381e-05, + "loss": 0.3143, + "step": 9250 + }, + { + "epoch": 0.46340547979482044, + "grad_norm": 40.509037017822266, + "learning_rate": 4.697410613334322e-05, + "loss": 0.0918, + "step": 9260 + }, + { + "epoch": 0.46390591767796824, + "grad_norm": 39.248924255371094, + "learning_rate": 4.696483846453263e-05, + "loss": 0.5627, + "step": 9270 + }, + { + "epoch": 0.464406355561116, + "grad_norm": 0.018931368365883827, + "learning_rate": 4.695557079572205e-05, + "loss": 0.6386, + "step": 9280 + }, + { + "epoch": 0.4649067934442637, + "grad_norm": 19.644550323486328, + "learning_rate": 4.6946303126911465e-05, + "loss": 0.3336, + "step": 9290 + }, + { + "epoch": 0.46540723132741146, + "grad_norm": 31.11878204345703, + "learning_rate": 4.6937035458100867e-05, + "loss": 0.5478, + "step": 9300 + }, + { + "epoch": 0.46590766921055926, + "grad_norm": 27.73699188232422, + "learning_rate": 4.692776778929028e-05, + "loss": 0.3523, + "step": 9310 + }, + { + "epoch": 0.466408107093707, + "grad_norm": 34.295654296875, + "learning_rate": 4.69185001204797e-05, + "loss": 0.466, + "step": 9320 + }, + { + "epoch": 0.46690854497685474, + "grad_norm": 4.566154479980469, + "learning_rate": 4.6909232451669113e-05, + "loss": 0.5139, + "step": 9330 + }, + { + "epoch": 0.4674089828600025, + "grad_norm": 7.708838939666748, + "learning_rate": 4.689996478285852e-05, + "loss": 0.4153, + "step": 9340 + }, + { + "epoch": 0.4679094207431503, + "grad_norm": 23.678508758544922, + "learning_rate": 4.689069711404793e-05, + "loss": 0.3452, + "step": 9350 + }, + { + "epoch": 0.468409858626298, + "grad_norm": 22.9573917388916, + "learning_rate": 4.688142944523735e-05, + "loss": 0.438, + "step": 9360 + }, + { + "epoch": 0.46891029650944577, + "grad_norm": 37.041542053222656, + "learning_rate": 4.687216177642676e-05, + "loss": 0.253, + "step": 9370 + }, + { + "epoch": 0.4694107343925935, + "grad_norm": 20.078113555908203, + "learning_rate": 4.686289410761617e-05, + "loss": 0.7042, + "step": 9380 + }, + { + "epoch": 0.46991117227574125, + "grad_norm": 42.9705924987793, + "learning_rate": 4.685362643880559e-05, + "loss": 1.0307, + "step": 9390 + }, + { + "epoch": 0.47041161015888905, + "grad_norm": 0.40481218695640564, + "learning_rate": 4.6844358769995e-05, + "loss": 0.3416, + "step": 9400 + }, + { + "epoch": 0.4709120480420368, + "grad_norm": 22.261127471923828, + "learning_rate": 4.683509110118441e-05, + "loss": 0.4576, + "step": 9410 + }, + { + "epoch": 0.47141248592518453, + "grad_norm": 20.683374404907227, + "learning_rate": 4.682582343237382e-05, + "loss": 0.5748, + "step": 9420 + }, + { + "epoch": 0.4719129238083323, + "grad_norm": 0.7484374642372131, + "learning_rate": 4.6816555763563236e-05, + "loss": 0.4737, + "step": 9430 + }, + { + "epoch": 0.47241336169148007, + "grad_norm": 0.448346346616745, + "learning_rate": 4.680728809475265e-05, + "loss": 0.675, + "step": 9440 + }, + { + "epoch": 0.4729137995746278, + "grad_norm": 0.5298665761947632, + "learning_rate": 4.679802042594206e-05, + "loss": 0.3811, + "step": 9450 + }, + { + "epoch": 0.47341423745777556, + "grad_norm": 47.69575500488281, + "learning_rate": 4.678875275713147e-05, + "loss": 0.371, + "step": 9460 + }, + { + "epoch": 0.4739146753409233, + "grad_norm": 42.013824462890625, + "learning_rate": 4.6779485088320885e-05, + "loss": 0.3662, + "step": 9470 + }, + { + "epoch": 0.47441511322407104, + "grad_norm": 5.563680171966553, + "learning_rate": 4.67702174195103e-05, + "loss": 0.417, + "step": 9480 + }, + { + "epoch": 0.47491555110721884, + "grad_norm": 0.13388077914714813, + "learning_rate": 4.676094975069971e-05, + "loss": 0.6976, + "step": 9490 + }, + { + "epoch": 0.4754159889903666, + "grad_norm": 60.295562744140625, + "learning_rate": 4.6751682081889125e-05, + "loss": 0.4745, + "step": 9500 + }, + { + "epoch": 0.4759164268735143, + "grad_norm": 0.008586219511926174, + "learning_rate": 4.674241441307854e-05, + "loss": 0.3952, + "step": 9510 + }, + { + "epoch": 0.47641686475666206, + "grad_norm": 61.40500259399414, + "learning_rate": 4.673314674426795e-05, + "loss": 0.7319, + "step": 9520 + }, + { + "epoch": 0.47691730263980986, + "grad_norm": 0.0029987264424562454, + "learning_rate": 4.672387907545736e-05, + "loss": 0.3251, + "step": 9530 + }, + { + "epoch": 0.4774177405229576, + "grad_norm": 14.905648231506348, + "learning_rate": 4.6714611406646774e-05, + "loss": 0.4226, + "step": 9540 + }, + { + "epoch": 0.47791817840610534, + "grad_norm": 108.10245513916016, + "learning_rate": 4.670534373783619e-05, + "loss": 0.5822, + "step": 9550 + }, + { + "epoch": 0.4784186162892531, + "grad_norm": 0.006820970680564642, + "learning_rate": 4.66960760690256e-05, + "loss": 0.3137, + "step": 9560 + }, + { + "epoch": 0.4789190541724008, + "grad_norm": 10.264857292175293, + "learning_rate": 4.668680840021501e-05, + "loss": 0.6408, + "step": 9570 + }, + { + "epoch": 0.4794194920555486, + "grad_norm": 46.434471130371094, + "learning_rate": 4.667754073140442e-05, + "loss": 0.2851, + "step": 9580 + }, + { + "epoch": 0.47991992993869637, + "grad_norm": 1.2542067766189575, + "learning_rate": 4.666827306259384e-05, + "loss": 0.1866, + "step": 9590 + }, + { + "epoch": 0.4804203678218441, + "grad_norm": 31.199316024780273, + "learning_rate": 4.665900539378325e-05, + "loss": 0.8941, + "step": 9600 + }, + { + "epoch": 0.48092080570499185, + "grad_norm": 7.500621795654297, + "learning_rate": 4.664973772497266e-05, + "loss": 0.4205, + "step": 9610 + }, + { + "epoch": 0.48142124358813965, + "grad_norm": 32.84191131591797, + "learning_rate": 4.664047005616208e-05, + "loss": 0.3072, + "step": 9620 + }, + { + "epoch": 0.4819216814712874, + "grad_norm": 0.342899888753891, + "learning_rate": 4.663120238735149e-05, + "loss": 0.8615, + "step": 9630 + }, + { + "epoch": 0.48242211935443513, + "grad_norm": 25.965070724487305, + "learning_rate": 4.6621934718540897e-05, + "loss": 0.5207, + "step": 9640 + }, + { + "epoch": 0.4829225572375829, + "grad_norm": 1.7003529071807861, + "learning_rate": 4.661266704973031e-05, + "loss": 0.2877, + "step": 9650 + }, + { + "epoch": 0.4834229951207306, + "grad_norm": 72.5172348022461, + "learning_rate": 4.660339938091973e-05, + "loss": 0.3574, + "step": 9660 + }, + { + "epoch": 0.4839234330038784, + "grad_norm": 23.548051834106445, + "learning_rate": 4.6594131712109143e-05, + "loss": 0.5425, + "step": 9670 + }, + { + "epoch": 0.48442387088702615, + "grad_norm": 27.999448776245117, + "learning_rate": 4.6584864043298545e-05, + "loss": 0.6919, + "step": 9680 + }, + { + "epoch": 0.4849243087701739, + "grad_norm": 32.96438980102539, + "learning_rate": 4.657559637448796e-05, + "loss": 0.9679, + "step": 9690 + }, + { + "epoch": 0.48542474665332164, + "grad_norm": 34.318580627441406, + "learning_rate": 4.656632870567738e-05, + "loss": 0.4891, + "step": 9700 + }, + { + "epoch": 0.48592518453646943, + "grad_norm": 37.73475646972656, + "learning_rate": 4.655706103686679e-05, + "loss": 0.2861, + "step": 9710 + }, + { + "epoch": 0.4864256224196172, + "grad_norm": 28.22308349609375, + "learning_rate": 4.65477933680562e-05, + "loss": 0.4666, + "step": 9720 + }, + { + "epoch": 0.4869260603027649, + "grad_norm": 4.342285633087158, + "learning_rate": 4.653852569924562e-05, + "loss": 0.3359, + "step": 9730 + }, + { + "epoch": 0.48742649818591266, + "grad_norm": 52.558441162109375, + "learning_rate": 4.6529258030435026e-05, + "loss": 0.6888, + "step": 9740 + }, + { + "epoch": 0.4879269360690604, + "grad_norm": 4.615299701690674, + "learning_rate": 4.651999036162444e-05, + "loss": 0.7349, + "step": 9750 + }, + { + "epoch": 0.4884273739522082, + "grad_norm": 28.57534408569336, + "learning_rate": 4.651072269281385e-05, + "loss": 0.517, + "step": 9760 + }, + { + "epoch": 0.48892781183535594, + "grad_norm": 5.651413440704346, + "learning_rate": 4.6501455024003266e-05, + "loss": 0.2289, + "step": 9770 + }, + { + "epoch": 0.4894282497185037, + "grad_norm": 0.5669355988502502, + "learning_rate": 4.649218735519268e-05, + "loss": 0.2782, + "step": 9780 + }, + { + "epoch": 0.4899286876016514, + "grad_norm": 31.221240997314453, + "learning_rate": 4.648291968638209e-05, + "loss": 0.5454, + "step": 9790 + }, + { + "epoch": 0.4904291254847992, + "grad_norm": 47.62663650512695, + "learning_rate": 4.64736520175715e-05, + "loss": 1.0161, + "step": 9800 + }, + { + "epoch": 0.49092956336794696, + "grad_norm": 51.2740478515625, + "learning_rate": 4.6464384348760915e-05, + "loss": 0.62, + "step": 9810 + }, + { + "epoch": 0.4914300012510947, + "grad_norm": 13.20570182800293, + "learning_rate": 4.645511667995033e-05, + "loss": 0.206, + "step": 9820 + }, + { + "epoch": 0.49193043913424245, + "grad_norm": 25.61264991760254, + "learning_rate": 4.644584901113974e-05, + "loss": 0.3597, + "step": 9830 + }, + { + "epoch": 0.4924308770173902, + "grad_norm": 2.098714590072632, + "learning_rate": 4.6436581342329155e-05, + "loss": 0.5692, + "step": 9840 + }, + { + "epoch": 0.492931314900538, + "grad_norm": 17.429101943969727, + "learning_rate": 4.6427313673518564e-05, + "loss": 0.1207, + "step": 9850 + }, + { + "epoch": 0.49343175278368573, + "grad_norm": 1.1773203611373901, + "learning_rate": 4.641804600470798e-05, + "loss": 0.5998, + "step": 9860 + }, + { + "epoch": 0.49393219066683347, + "grad_norm": 110.83248901367188, + "learning_rate": 4.640877833589739e-05, + "loss": 0.744, + "step": 9870 + }, + { + "epoch": 0.4944326285499812, + "grad_norm": 86.56903839111328, + "learning_rate": 4.6399510667086804e-05, + "loss": 0.6105, + "step": 9880 + }, + { + "epoch": 0.494933066433129, + "grad_norm": 36.475345611572266, + "learning_rate": 4.639024299827622e-05, + "loss": 0.4539, + "step": 9890 + }, + { + "epoch": 0.49543350431627675, + "grad_norm": 5.966508388519287, + "learning_rate": 4.638097532946563e-05, + "loss": 0.2981, + "step": 9900 + }, + { + "epoch": 0.4959339421994245, + "grad_norm": 81.03009796142578, + "learning_rate": 4.637170766065504e-05, + "loss": 0.4828, + "step": 9910 + }, + { + "epoch": 0.49643438008257224, + "grad_norm": 15.340887069702148, + "learning_rate": 4.636243999184445e-05, + "loss": 0.5749, + "step": 9920 + }, + { + "epoch": 0.49693481796572003, + "grad_norm": 0.6524103283882141, + "learning_rate": 4.635317232303387e-05, + "loss": 0.7842, + "step": 9930 + }, + { + "epoch": 0.4974352558488678, + "grad_norm": 17.45283317565918, + "learning_rate": 4.634390465422328e-05, + "loss": 0.3505, + "step": 9940 + }, + { + "epoch": 0.4979356937320155, + "grad_norm": 2.3655765056610107, + "learning_rate": 4.633463698541269e-05, + "loss": 0.2633, + "step": 9950 + }, + { + "epoch": 0.49843613161516326, + "grad_norm": 32.0737190246582, + "learning_rate": 4.63253693166021e-05, + "loss": 0.5075, + "step": 9960 + }, + { + "epoch": 0.498936569498311, + "grad_norm": 25.002029418945312, + "learning_rate": 4.631610164779152e-05, + "loss": 0.378, + "step": 9970 + }, + { + "epoch": 0.4994370073814588, + "grad_norm": 118.2671127319336, + "learning_rate": 4.6306833978980926e-05, + "loss": 0.3498, + "step": 9980 + }, + { + "epoch": 0.49993744526460654, + "grad_norm": 54.94559860229492, + "learning_rate": 4.629756631017034e-05, + "loss": 0.7104, + "step": 9990 + }, + { + "epoch": 0.5004378831477543, + "grad_norm": 40.09208679199219, + "learning_rate": 4.628829864135976e-05, + "loss": 0.2633, + "step": 10000 + }, + { + "epoch": 0.5009383210309021, + "grad_norm": 3.7241270542144775, + "learning_rate": 4.627903097254917e-05, + "loss": 0.158, + "step": 10010 + }, + { + "epoch": 0.5014387589140498, + "grad_norm": 42.15870666503906, + "learning_rate": 4.6269763303738575e-05, + "loss": 0.8101, + "step": 10020 + }, + { + "epoch": 0.5019391967971976, + "grad_norm": 27.708194732666016, + "learning_rate": 4.626049563492799e-05, + "loss": 0.6833, + "step": 10030 + }, + { + "epoch": 0.5024396346803452, + "grad_norm": 34.67120361328125, + "learning_rate": 4.625122796611741e-05, + "loss": 0.2499, + "step": 10040 + }, + { + "epoch": 0.502940072563493, + "grad_norm": 7.182991981506348, + "learning_rate": 4.6241960297306816e-05, + "loss": 0.2825, + "step": 10050 + }, + { + "epoch": 0.5034405104466408, + "grad_norm": 5.227594375610352, + "learning_rate": 4.623269262849623e-05, + "loss": 0.613, + "step": 10060 + }, + { + "epoch": 0.5039409483297885, + "grad_norm": 3.532735824584961, + "learning_rate": 4.622342495968564e-05, + "loss": 0.4625, + "step": 10070 + }, + { + "epoch": 0.5044413862129363, + "grad_norm": 39.4844856262207, + "learning_rate": 4.6214157290875056e-05, + "loss": 0.6061, + "step": 10080 + }, + { + "epoch": 0.5049418240960841, + "grad_norm": 18.782529830932617, + "learning_rate": 4.6204889622064465e-05, + "loss": 0.3748, + "step": 10090 + }, + { + "epoch": 0.5054422619792318, + "grad_norm": 4.8212199211120605, + "learning_rate": 4.619562195325388e-05, + "loss": 0.548, + "step": 10100 + }, + { + "epoch": 0.5059426998623796, + "grad_norm": 46.056331634521484, + "learning_rate": 4.6186354284443296e-05, + "loss": 0.6526, + "step": 10110 + }, + { + "epoch": 0.5064431377455273, + "grad_norm": 6.049580097198486, + "learning_rate": 4.617708661563271e-05, + "loss": 0.2244, + "step": 10120 + }, + { + "epoch": 0.5069435756286751, + "grad_norm": 4.608890056610107, + "learning_rate": 4.6167818946822114e-05, + "loss": 0.5027, + "step": 10130 + }, + { + "epoch": 0.5074440135118229, + "grad_norm": 0.019651055335998535, + "learning_rate": 4.615855127801153e-05, + "loss": 0.6108, + "step": 10140 + }, + { + "epoch": 0.5079444513949706, + "grad_norm": 5.979386329650879, + "learning_rate": 4.6149283609200945e-05, + "loss": 0.459, + "step": 10150 + }, + { + "epoch": 0.5084448892781184, + "grad_norm": 17.36353302001953, + "learning_rate": 4.614001594039036e-05, + "loss": 0.448, + "step": 10160 + }, + { + "epoch": 0.5089453271612661, + "grad_norm": 27.129735946655273, + "learning_rate": 4.613074827157977e-05, + "loss": 0.2839, + "step": 10170 + }, + { + "epoch": 0.5094457650444139, + "grad_norm": 44.265682220458984, + "learning_rate": 4.612148060276918e-05, + "loss": 0.7886, + "step": 10180 + }, + { + "epoch": 0.5099462029275617, + "grad_norm": 7.445001125335693, + "learning_rate": 4.6112212933958594e-05, + "loss": 0.3431, + "step": 10190 + }, + { + "epoch": 0.5104466408107093, + "grad_norm": 62.159244537353516, + "learning_rate": 4.610294526514801e-05, + "loss": 1.1433, + "step": 10200 + }, + { + "epoch": 0.5109470786938571, + "grad_norm": 2.3398618698120117, + "learning_rate": 4.609367759633742e-05, + "loss": 0.2685, + "step": 10210 + }, + { + "epoch": 0.5114475165770049, + "grad_norm": 0.2915375828742981, + "learning_rate": 4.6084409927526834e-05, + "loss": 0.5148, + "step": 10220 + }, + { + "epoch": 0.5119479544601526, + "grad_norm": 0.003941272851079702, + "learning_rate": 4.607514225871625e-05, + "loss": 0.3423, + "step": 10230 + }, + { + "epoch": 0.5124483923433004, + "grad_norm": 12.489424705505371, + "learning_rate": 4.606587458990566e-05, + "loss": 0.5194, + "step": 10240 + }, + { + "epoch": 0.5129488302264481, + "grad_norm": 35.12325668334961, + "learning_rate": 4.605660692109507e-05, + "loss": 0.3077, + "step": 10250 + }, + { + "epoch": 0.5134492681095959, + "grad_norm": 36.559837341308594, + "learning_rate": 4.604733925228448e-05, + "loss": 0.615, + "step": 10260 + }, + { + "epoch": 0.5139497059927437, + "grad_norm": 25.724468231201172, + "learning_rate": 4.60380715834739e-05, + "loss": 0.2866, + "step": 10270 + }, + { + "epoch": 0.5144501438758914, + "grad_norm": 86.95427703857422, + "learning_rate": 4.602880391466331e-05, + "loss": 0.1621, + "step": 10280 + }, + { + "epoch": 0.5149505817590392, + "grad_norm": 1.7852181196212769, + "learning_rate": 4.6019536245852716e-05, + "loss": 0.4788, + "step": 10290 + }, + { + "epoch": 0.5154510196421869, + "grad_norm": 0.010143262334167957, + "learning_rate": 4.601026857704213e-05, + "loss": 0.4051, + "step": 10300 + }, + { + "epoch": 0.5159514575253347, + "grad_norm": 0.002599142724648118, + "learning_rate": 4.600100090823155e-05, + "loss": 0.1848, + "step": 10310 + }, + { + "epoch": 0.5164518954084825, + "grad_norm": 15.835326194763184, + "learning_rate": 4.5991733239420956e-05, + "loss": 0.7262, + "step": 10320 + }, + { + "epoch": 0.5169523332916302, + "grad_norm": 1.2667193412780762, + "learning_rate": 4.598246557061037e-05, + "loss": 0.5844, + "step": 10330 + }, + { + "epoch": 0.517452771174778, + "grad_norm": 2.7567906379699707, + "learning_rate": 4.597319790179979e-05, + "loss": 0.5991, + "step": 10340 + }, + { + "epoch": 0.5179532090579256, + "grad_norm": 89.98589324951172, + "learning_rate": 4.5963930232989196e-05, + "loss": 0.4443, + "step": 10350 + }, + { + "epoch": 0.5184536469410734, + "grad_norm": 55.13560104370117, + "learning_rate": 4.5954662564178605e-05, + "loss": 0.5545, + "step": 10360 + }, + { + "epoch": 0.5189540848242212, + "grad_norm": 2.207418441772461, + "learning_rate": 4.594539489536802e-05, + "loss": 0.4529, + "step": 10370 + }, + { + "epoch": 0.5194545227073689, + "grad_norm": 57.89557647705078, + "learning_rate": 4.5936127226557437e-05, + "loss": 0.6972, + "step": 10380 + }, + { + "epoch": 0.5199549605905167, + "grad_norm": 2.925941228866577, + "learning_rate": 4.5926859557746845e-05, + "loss": 0.4636, + "step": 10390 + }, + { + "epoch": 0.5204553984736645, + "grad_norm": 27.30789566040039, + "learning_rate": 4.5917591888936254e-05, + "loss": 0.2434, + "step": 10400 + }, + { + "epoch": 0.5209558363568122, + "grad_norm": 15.43437671661377, + "learning_rate": 4.590832422012567e-05, + "loss": 0.4597, + "step": 10410 + }, + { + "epoch": 0.52145627423996, + "grad_norm": 21.285364151000977, + "learning_rate": 4.5899056551315086e-05, + "loss": 0.4488, + "step": 10420 + }, + { + "epoch": 0.5219567121231077, + "grad_norm": 18.479459762573242, + "learning_rate": 4.5889788882504494e-05, + "loss": 0.3712, + "step": 10430 + }, + { + "epoch": 0.5224571500062555, + "grad_norm": 19.09623908996582, + "learning_rate": 4.588052121369391e-05, + "loss": 0.3972, + "step": 10440 + }, + { + "epoch": 0.5229575878894033, + "grad_norm": 4.9107747077941895, + "learning_rate": 4.5871253544883326e-05, + "loss": 0.5403, + "step": 10450 + }, + { + "epoch": 0.523458025772551, + "grad_norm": 0.42535021901130676, + "learning_rate": 4.5861985876072735e-05, + "loss": 0.4119, + "step": 10460 + }, + { + "epoch": 0.5239584636556988, + "grad_norm": 17.059267044067383, + "learning_rate": 4.5852718207262143e-05, + "loss": 0.3976, + "step": 10470 + }, + { + "epoch": 0.5244589015388464, + "grad_norm": 17.681806564331055, + "learning_rate": 4.584345053845156e-05, + "loss": 0.7778, + "step": 10480 + }, + { + "epoch": 0.5249593394219942, + "grad_norm": 26.89153480529785, + "learning_rate": 4.5834182869640975e-05, + "loss": 0.6357, + "step": 10490 + }, + { + "epoch": 0.525459777305142, + "grad_norm": 35.19390106201172, + "learning_rate": 4.582491520083039e-05, + "loss": 0.1856, + "step": 10500 + }, + { + "epoch": 0.5259602151882897, + "grad_norm": 56.2961540222168, + "learning_rate": 4.581564753201979e-05, + "loss": 0.7351, + "step": 10510 + }, + { + "epoch": 0.5264606530714375, + "grad_norm": 1.5538973808288574, + "learning_rate": 4.580637986320921e-05, + "loss": 0.1418, + "step": 10520 + }, + { + "epoch": 0.5269610909545852, + "grad_norm": 175.5338897705078, + "learning_rate": 4.5797112194398624e-05, + "loss": 0.4086, + "step": 10530 + }, + { + "epoch": 0.527461528837733, + "grad_norm": 11.885289192199707, + "learning_rate": 4.578784452558804e-05, + "loss": 0.6473, + "step": 10540 + }, + { + "epoch": 0.5279619667208808, + "grad_norm": 46.29182434082031, + "learning_rate": 4.577857685677745e-05, + "loss": 0.4472, + "step": 10550 + }, + { + "epoch": 0.5284624046040285, + "grad_norm": 0.013495348393917084, + "learning_rate": 4.5769309187966864e-05, + "loss": 0.4437, + "step": 10560 + }, + { + "epoch": 0.5289628424871763, + "grad_norm": 19.087261199951172, + "learning_rate": 4.576004151915627e-05, + "loss": 0.2322, + "step": 10570 + }, + { + "epoch": 0.5294632803703241, + "grad_norm": 121.56734466552734, + "learning_rate": 4.575077385034569e-05, + "loss": 0.6195, + "step": 10580 + }, + { + "epoch": 0.5299637182534718, + "grad_norm": 147.5408935546875, + "learning_rate": 4.57415061815351e-05, + "loss": 0.6804, + "step": 10590 + }, + { + "epoch": 0.5304641561366196, + "grad_norm": 10.96035099029541, + "learning_rate": 4.573223851272451e-05, + "loss": 0.1919, + "step": 10600 + }, + { + "epoch": 0.5309645940197673, + "grad_norm": 41.179710388183594, + "learning_rate": 4.572297084391393e-05, + "loss": 0.5893, + "step": 10610 + }, + { + "epoch": 0.531465031902915, + "grad_norm": 128.50051879882812, + "learning_rate": 4.571370317510334e-05, + "loss": 0.5145, + "step": 10620 + }, + { + "epoch": 0.5319654697860629, + "grad_norm": 26.072282791137695, + "learning_rate": 4.5704435506292746e-05, + "loss": 0.7127, + "step": 10630 + }, + { + "epoch": 0.5324659076692105, + "grad_norm": 0.32506632804870605, + "learning_rate": 4.569516783748216e-05, + "loss": 0.502, + "step": 10640 + }, + { + "epoch": 0.5329663455523583, + "grad_norm": 5.527798652648926, + "learning_rate": 4.568590016867158e-05, + "loss": 0.3496, + "step": 10650 + }, + { + "epoch": 0.533466783435506, + "grad_norm": 48.40260696411133, + "learning_rate": 4.5676632499860986e-05, + "loss": 0.4416, + "step": 10660 + }, + { + "epoch": 0.5339672213186538, + "grad_norm": 60.43381881713867, + "learning_rate": 4.56673648310504e-05, + "loss": 0.4384, + "step": 10670 + }, + { + "epoch": 0.5344676592018016, + "grad_norm": 11.269490242004395, + "learning_rate": 4.565809716223981e-05, + "loss": 0.599, + "step": 10680 + }, + { + "epoch": 0.5349680970849493, + "grad_norm": 15.990483283996582, + "learning_rate": 4.5648829493429226e-05, + "loss": 0.7981, + "step": 10690 + }, + { + "epoch": 0.5354685349680971, + "grad_norm": 0.05557543784379959, + "learning_rate": 4.5639561824618635e-05, + "loss": 0.3967, + "step": 10700 + }, + { + "epoch": 0.5359689728512448, + "grad_norm": 30.480085372924805, + "learning_rate": 4.563029415580805e-05, + "loss": 0.7006, + "step": 10710 + }, + { + "epoch": 0.5364694107343926, + "grad_norm": 12.990655899047852, + "learning_rate": 4.5621026486997467e-05, + "loss": 0.112, + "step": 10720 + }, + { + "epoch": 0.5369698486175404, + "grad_norm": 7.890532970428467, + "learning_rate": 4.5611758818186875e-05, + "loss": 0.643, + "step": 10730 + }, + { + "epoch": 0.5374702865006881, + "grad_norm": 174.39280700683594, + "learning_rate": 4.5602491149376284e-05, + "loss": 0.4689, + "step": 10740 + }, + { + "epoch": 0.5379707243838359, + "grad_norm": 41.44766616821289, + "learning_rate": 4.55932234805657e-05, + "loss": 0.4822, + "step": 10750 + }, + { + "epoch": 0.5384711622669837, + "grad_norm": 3.2811830043792725, + "learning_rate": 4.5583955811755115e-05, + "loss": 0.3719, + "step": 10760 + }, + { + "epoch": 0.5389716001501313, + "grad_norm": 0.8423954844474792, + "learning_rate": 4.5574688142944524e-05, + "loss": 0.3203, + "step": 10770 + }, + { + "epoch": 0.5394720380332791, + "grad_norm": 0.34487754106521606, + "learning_rate": 4.556542047413394e-05, + "loss": 0.775, + "step": 10780 + }, + { + "epoch": 0.5399724759164268, + "grad_norm": 24.394683837890625, + "learning_rate": 4.555615280532335e-05, + "loss": 0.4212, + "step": 10790 + }, + { + "epoch": 0.5404729137995746, + "grad_norm": 0.9724845290184021, + "learning_rate": 4.5546885136512764e-05, + "loss": 0.7561, + "step": 10800 + }, + { + "epoch": 0.5409733516827224, + "grad_norm": 19.232715606689453, + "learning_rate": 4.553761746770217e-05, + "loss": 0.6606, + "step": 10810 + }, + { + "epoch": 0.5414737895658701, + "grad_norm": 1.0460811853408813, + "learning_rate": 4.552834979889159e-05, + "loss": 0.7835, + "step": 10820 + }, + { + "epoch": 0.5419742274490179, + "grad_norm": 0.2055661678314209, + "learning_rate": 4.5519082130081005e-05, + "loss": 0.3087, + "step": 10830 + }, + { + "epoch": 0.5424746653321656, + "grad_norm": 0.03494323417544365, + "learning_rate": 4.550981446127042e-05, + "loss": 0.9141, + "step": 10840 + }, + { + "epoch": 0.5429751032153134, + "grad_norm": 62.368560791015625, + "learning_rate": 4.550054679245982e-05, + "loss": 0.5621, + "step": 10850 + }, + { + "epoch": 0.5434755410984612, + "grad_norm": 14.501202583312988, + "learning_rate": 4.549127912364924e-05, + "loss": 0.4725, + "step": 10860 + }, + { + "epoch": 0.5439759789816089, + "grad_norm": 2.8569278717041016, + "learning_rate": 4.5482011454838654e-05, + "loss": 0.4124, + "step": 10870 + }, + { + "epoch": 0.5444764168647567, + "grad_norm": 36.44921112060547, + "learning_rate": 4.547274378602807e-05, + "loss": 0.3596, + "step": 10880 + }, + { + "epoch": 0.5449768547479045, + "grad_norm": 1.4293357133865356, + "learning_rate": 4.546347611721748e-05, + "loss": 0.4369, + "step": 10890 + }, + { + "epoch": 0.5454772926310522, + "grad_norm": 22.33682632446289, + "learning_rate": 4.545420844840689e-05, + "loss": 0.5421, + "step": 10900 + }, + { + "epoch": 0.5459777305142, + "grad_norm": 0.6168195605278015, + "learning_rate": 4.54449407795963e-05, + "loss": 0.6414, + "step": 10910 + }, + { + "epoch": 0.5464781683973476, + "grad_norm": 2.652080535888672, + "learning_rate": 4.543567311078571e-05, + "loss": 0.6377, + "step": 10920 + }, + { + "epoch": 0.5469786062804954, + "grad_norm": 6.198836326599121, + "learning_rate": 4.542640544197513e-05, + "loss": 0.5157, + "step": 10930 + }, + { + "epoch": 0.5474790441636432, + "grad_norm": 18.381521224975586, + "learning_rate": 4.541713777316454e-05, + "loss": 0.3976, + "step": 10940 + }, + { + "epoch": 0.5479794820467909, + "grad_norm": 50.378299713134766, + "learning_rate": 4.540787010435396e-05, + "loss": 0.5123, + "step": 10950 + }, + { + "epoch": 0.5484799199299387, + "grad_norm": 0.2749289274215698, + "learning_rate": 4.539860243554336e-05, + "loss": 0.4753, + "step": 10960 + }, + { + "epoch": 0.5489803578130864, + "grad_norm": 6.972272872924805, + "learning_rate": 4.5389334766732776e-05, + "loss": 0.2903, + "step": 10970 + }, + { + "epoch": 0.5494807956962342, + "grad_norm": 51.03962707519531, + "learning_rate": 4.538006709792219e-05, + "loss": 0.5088, + "step": 10980 + }, + { + "epoch": 0.549981233579382, + "grad_norm": 0.026905428618192673, + "learning_rate": 4.537079942911161e-05, + "loss": 0.2183, + "step": 10990 + }, + { + "epoch": 0.5504816714625297, + "grad_norm": 0.01262497529387474, + "learning_rate": 4.5361531760301016e-05, + "loss": 0.539, + "step": 11000 + }, + { + "epoch": 0.5509821093456775, + "grad_norm": 0.08450735360383987, + "learning_rate": 4.5352264091490425e-05, + "loss": 0.4202, + "step": 11010 + }, + { + "epoch": 0.5514825472288252, + "grad_norm": 0.6751739978790283, + "learning_rate": 4.534299642267984e-05, + "loss": 0.4026, + "step": 11020 + }, + { + "epoch": 0.551982985111973, + "grad_norm": 44.14516830444336, + "learning_rate": 4.5333728753869256e-05, + "loss": 0.5524, + "step": 11030 + }, + { + "epoch": 0.5524834229951208, + "grad_norm": 16.249679565429688, + "learning_rate": 4.5324461085058665e-05, + "loss": 0.3114, + "step": 11040 + }, + { + "epoch": 0.5529838608782685, + "grad_norm": 21.560251235961914, + "learning_rate": 4.531519341624808e-05, + "loss": 0.8484, + "step": 11050 + }, + { + "epoch": 0.5534842987614162, + "grad_norm": 20.47876739501953, + "learning_rate": 4.5305925747437496e-05, + "loss": 0.2111, + "step": 11060 + }, + { + "epoch": 0.553984736644564, + "grad_norm": 5.75532865524292, + "learning_rate": 4.5296658078626905e-05, + "loss": 0.357, + "step": 11070 + }, + { + "epoch": 0.5544851745277117, + "grad_norm": 2.9991495609283447, + "learning_rate": 4.5287390409816314e-05, + "loss": 0.5475, + "step": 11080 + }, + { + "epoch": 0.5549856124108595, + "grad_norm": 0.056779857724905014, + "learning_rate": 4.527812274100573e-05, + "loss": 0.319, + "step": 11090 + }, + { + "epoch": 0.5554860502940072, + "grad_norm": 0.028560467064380646, + "learning_rate": 4.5268855072195145e-05, + "loss": 0.6748, + "step": 11100 + }, + { + "epoch": 0.555986488177155, + "grad_norm": 0.11975672841072083, + "learning_rate": 4.5259587403384554e-05, + "loss": 0.8272, + "step": 11110 + }, + { + "epoch": 0.5564869260603028, + "grad_norm": 17.769289016723633, + "learning_rate": 4.525031973457396e-05, + "loss": 0.3396, + "step": 11120 + }, + { + "epoch": 0.5569873639434505, + "grad_norm": 22.71141815185547, + "learning_rate": 4.524105206576338e-05, + "loss": 0.1371, + "step": 11130 + }, + { + "epoch": 0.5574878018265983, + "grad_norm": 0.11717333644628525, + "learning_rate": 4.5231784396952794e-05, + "loss": 0.224, + "step": 11140 + }, + { + "epoch": 0.557988239709746, + "grad_norm": 13.162442207336426, + "learning_rate": 4.52225167281422e-05, + "loss": 0.2301, + "step": 11150 + }, + { + "epoch": 0.5584886775928938, + "grad_norm": 1.6001412868499756, + "learning_rate": 4.521324905933162e-05, + "loss": 0.3571, + "step": 11160 + }, + { + "epoch": 0.5589891154760416, + "grad_norm": 42.987850189208984, + "learning_rate": 4.5203981390521035e-05, + "loss": 0.5947, + "step": 11170 + }, + { + "epoch": 0.5594895533591893, + "grad_norm": 0.5706591606140137, + "learning_rate": 4.519471372171044e-05, + "loss": 0.2096, + "step": 11180 + }, + { + "epoch": 0.5599899912423371, + "grad_norm": 20.607173919677734, + "learning_rate": 4.518544605289985e-05, + "loss": 0.2474, + "step": 11190 + }, + { + "epoch": 0.5604904291254847, + "grad_norm": 68.06755065917969, + "learning_rate": 4.517617838408927e-05, + "loss": 0.3365, + "step": 11200 + }, + { + "epoch": 0.5609908670086325, + "grad_norm": 11.860542297363281, + "learning_rate": 4.5166910715278684e-05, + "loss": 0.4272, + "step": 11210 + }, + { + "epoch": 0.5614913048917803, + "grad_norm": 25.990684509277344, + "learning_rate": 4.515764304646809e-05, + "loss": 0.7865, + "step": 11220 + }, + { + "epoch": 0.561991742774928, + "grad_norm": 0.7066521048545837, + "learning_rate": 4.51483753776575e-05, + "loss": 0.3194, + "step": 11230 + }, + { + "epoch": 0.5624921806580758, + "grad_norm": 63.827659606933594, + "learning_rate": 4.513910770884692e-05, + "loss": 0.5185, + "step": 11240 + }, + { + "epoch": 0.5629926185412236, + "grad_norm": 0.034407839179039, + "learning_rate": 4.512984004003633e-05, + "loss": 0.6436, + "step": 11250 + }, + { + "epoch": 0.5634930564243713, + "grad_norm": 8.385162353515625, + "learning_rate": 4.512057237122574e-05, + "loss": 0.5092, + "step": 11260 + }, + { + "epoch": 0.5639934943075191, + "grad_norm": 33.49809646606445, + "learning_rate": 4.511130470241516e-05, + "loss": 0.349, + "step": 11270 + }, + { + "epoch": 0.5644939321906668, + "grad_norm": 34.45490646362305, + "learning_rate": 4.510203703360457e-05, + "loss": 0.4576, + "step": 11280 + }, + { + "epoch": 0.5649943700738146, + "grad_norm": 0.184963196516037, + "learning_rate": 4.509276936479398e-05, + "loss": 0.5034, + "step": 11290 + }, + { + "epoch": 0.5654948079569624, + "grad_norm": 5.765860080718994, + "learning_rate": 4.508350169598339e-05, + "loss": 0.2665, + "step": 11300 + }, + { + "epoch": 0.5659952458401101, + "grad_norm": 0.6031643748283386, + "learning_rate": 4.5074234027172806e-05, + "loss": 0.3011, + "step": 11310 + }, + { + "epoch": 0.5664956837232579, + "grad_norm": 0.1271948218345642, + "learning_rate": 4.506496635836222e-05, + "loss": 0.2716, + "step": 11320 + }, + { + "epoch": 0.5669961216064056, + "grad_norm": 40.394081115722656, + "learning_rate": 4.505569868955164e-05, + "loss": 0.5936, + "step": 11330 + }, + { + "epoch": 0.5674965594895534, + "grad_norm": 0.9489890336990356, + "learning_rate": 4.504643102074104e-05, + "loss": 0.2095, + "step": 11340 + }, + { + "epoch": 0.5679969973727012, + "grad_norm": 3.354494094848633, + "learning_rate": 4.5037163351930455e-05, + "loss": 0.3194, + "step": 11350 + }, + { + "epoch": 0.5684974352558488, + "grad_norm": 70.32083892822266, + "learning_rate": 4.502789568311987e-05, + "loss": 0.164, + "step": 11360 + }, + { + "epoch": 0.5689978731389966, + "grad_norm": 61.579612731933594, + "learning_rate": 4.5018628014309286e-05, + "loss": 0.4589, + "step": 11370 + }, + { + "epoch": 0.5694983110221443, + "grad_norm": 173.04611206054688, + "learning_rate": 4.5009360345498695e-05, + "loss": 0.865, + "step": 11380 + }, + { + "epoch": 0.5699987489052921, + "grad_norm": 38.097042083740234, + "learning_rate": 4.500009267668811e-05, + "loss": 0.9198, + "step": 11390 + }, + { + "epoch": 0.5704991867884399, + "grad_norm": 13.03411865234375, + "learning_rate": 4.499082500787752e-05, + "loss": 0.3213, + "step": 11400 + }, + { + "epoch": 0.5709996246715876, + "grad_norm": 14.476287841796875, + "learning_rate": 4.4981557339066935e-05, + "loss": 0.7727, + "step": 11410 + }, + { + "epoch": 0.5715000625547354, + "grad_norm": 44.3306884765625, + "learning_rate": 4.4972289670256344e-05, + "loss": 0.6719, + "step": 11420 + }, + { + "epoch": 0.5720005004378832, + "grad_norm": 57.547088623046875, + "learning_rate": 4.496302200144576e-05, + "loss": 0.8491, + "step": 11430 + }, + { + "epoch": 0.5725009383210309, + "grad_norm": 15.582108497619629, + "learning_rate": 4.4953754332635175e-05, + "loss": 0.4515, + "step": 11440 + }, + { + "epoch": 0.5730013762041787, + "grad_norm": 13.405743598937988, + "learning_rate": 4.4944486663824584e-05, + "loss": 0.1896, + "step": 11450 + }, + { + "epoch": 0.5735018140873264, + "grad_norm": 42.134803771972656, + "learning_rate": 4.493521899501399e-05, + "loss": 0.5366, + "step": 11460 + }, + { + "epoch": 0.5740022519704742, + "grad_norm": 27.484821319580078, + "learning_rate": 4.492595132620341e-05, + "loss": 0.1877, + "step": 11470 + }, + { + "epoch": 0.574502689853622, + "grad_norm": 51.77997970581055, + "learning_rate": 4.4916683657392824e-05, + "loss": 0.5335, + "step": 11480 + }, + { + "epoch": 0.5750031277367696, + "grad_norm": 5.7292609214782715, + "learning_rate": 4.490741598858223e-05, + "loss": 0.3967, + "step": 11490 + }, + { + "epoch": 0.5755035656199174, + "grad_norm": 10.836797714233398, + "learning_rate": 4.489814831977165e-05, + "loss": 0.2481, + "step": 11500 + }, + { + "epoch": 0.5760040035030651, + "grad_norm": 0.10754628479480743, + "learning_rate": 4.488888065096106e-05, + "loss": 0.4441, + "step": 11510 + }, + { + "epoch": 0.5765044413862129, + "grad_norm": 46.4285888671875, + "learning_rate": 4.487961298215047e-05, + "loss": 0.5151, + "step": 11520 + }, + { + "epoch": 0.5770048792693607, + "grad_norm": 2.001530170440674, + "learning_rate": 4.487034531333988e-05, + "loss": 0.7035, + "step": 11530 + }, + { + "epoch": 0.5775053171525084, + "grad_norm": 88.77485656738281, + "learning_rate": 4.48610776445293e-05, + "loss": 0.7737, + "step": 11540 + }, + { + "epoch": 0.5780057550356562, + "grad_norm": 100.44953155517578, + "learning_rate": 4.4851809975718713e-05, + "loss": 0.3204, + "step": 11550 + }, + { + "epoch": 0.578506192918804, + "grad_norm": 0.10426267236471176, + "learning_rate": 4.484254230690812e-05, + "loss": 1.0199, + "step": 11560 + }, + { + "epoch": 0.5790066308019517, + "grad_norm": 21.0239200592041, + "learning_rate": 4.483327463809753e-05, + "loss": 0.9083, + "step": 11570 + }, + { + "epoch": 0.5795070686850995, + "grad_norm": 89.91841888427734, + "learning_rate": 4.482400696928695e-05, + "loss": 0.5531, + "step": 11580 + }, + { + "epoch": 0.5800075065682472, + "grad_norm": 9.191157341003418, + "learning_rate": 4.481473930047636e-05, + "loss": 0.2767, + "step": 11590 + }, + { + "epoch": 0.580507944451395, + "grad_norm": 78.03228759765625, + "learning_rate": 4.480547163166577e-05, + "loss": 0.6383, + "step": 11600 + }, + { + "epoch": 0.5810083823345428, + "grad_norm": 58.03935241699219, + "learning_rate": 4.479620396285519e-05, + "loss": 0.8378, + "step": 11610 + }, + { + "epoch": 0.5815088202176905, + "grad_norm": 31.795927047729492, + "learning_rate": 4.4786936294044596e-05, + "loss": 0.1676, + "step": 11620 + }, + { + "epoch": 0.5820092581008383, + "grad_norm": 0.11397746205329895, + "learning_rate": 4.477766862523401e-05, + "loss": 0.4106, + "step": 11630 + }, + { + "epoch": 0.5825096959839859, + "grad_norm": 34.302974700927734, + "learning_rate": 4.476840095642342e-05, + "loss": 0.4292, + "step": 11640 + }, + { + "epoch": 0.5830101338671337, + "grad_norm": 58.253578186035156, + "learning_rate": 4.4759133287612836e-05, + "loss": 0.7257, + "step": 11650 + }, + { + "epoch": 0.5835105717502815, + "grad_norm": 74.53492736816406, + "learning_rate": 4.474986561880225e-05, + "loss": 0.5337, + "step": 11660 + }, + { + "epoch": 0.5840110096334292, + "grad_norm": 39.35128402709961, + "learning_rate": 4.474059794999167e-05, + "loss": 0.4974, + "step": 11670 + }, + { + "epoch": 0.584511447516577, + "grad_norm": 42.16452407836914, + "learning_rate": 4.473133028118107e-05, + "loss": 0.349, + "step": 11680 + }, + { + "epoch": 0.5850118853997247, + "grad_norm": 28.788043975830078, + "learning_rate": 4.4722062612370485e-05, + "loss": 0.3917, + "step": 11690 + }, + { + "epoch": 0.5855123232828725, + "grad_norm": 10.282878875732422, + "learning_rate": 4.47127949435599e-05, + "loss": 0.802, + "step": 11700 + }, + { + "epoch": 0.5860127611660203, + "grad_norm": 13.873482704162598, + "learning_rate": 4.4703527274749316e-05, + "loss": 0.3612, + "step": 11710 + }, + { + "epoch": 0.586513199049168, + "grad_norm": 23.393587112426758, + "learning_rate": 4.4694259605938725e-05, + "loss": 0.2178, + "step": 11720 + }, + { + "epoch": 0.5870136369323158, + "grad_norm": 13.195243835449219, + "learning_rate": 4.4684991937128134e-05, + "loss": 0.5717, + "step": 11730 + }, + { + "epoch": 0.5875140748154636, + "grad_norm": 2.705378532409668, + "learning_rate": 4.467572426831755e-05, + "loss": 0.5463, + "step": 11740 + }, + { + "epoch": 0.5880145126986113, + "grad_norm": 2.5775368213653564, + "learning_rate": 4.466645659950696e-05, + "loss": 0.4533, + "step": 11750 + }, + { + "epoch": 0.5885149505817591, + "grad_norm": 30.79420280456543, + "learning_rate": 4.4657188930696374e-05, + "loss": 0.375, + "step": 11760 + }, + { + "epoch": 0.5890153884649068, + "grad_norm": 13.138578414916992, + "learning_rate": 4.464792126188579e-05, + "loss": 0.5999, + "step": 11770 + }, + { + "epoch": 0.5895158263480546, + "grad_norm": 33.08952713012695, + "learning_rate": 4.4638653593075205e-05, + "loss": 0.4416, + "step": 11780 + }, + { + "epoch": 0.5900162642312023, + "grad_norm": 3.6036877632141113, + "learning_rate": 4.462938592426461e-05, + "loss": 0.4372, + "step": 11790 + }, + { + "epoch": 0.59051670211435, + "grad_norm": 24.8314266204834, + "learning_rate": 4.462011825545402e-05, + "loss": 0.5763, + "step": 11800 + }, + { + "epoch": 0.5910171399974978, + "grad_norm": 1.3072457313537598, + "learning_rate": 4.461085058664344e-05, + "loss": 0.2385, + "step": 11810 + }, + { + "epoch": 0.5915175778806455, + "grad_norm": 32.64002990722656, + "learning_rate": 4.4601582917832854e-05, + "loss": 0.6766, + "step": 11820 + }, + { + "epoch": 0.5920180157637933, + "grad_norm": 1.2684423923492432, + "learning_rate": 4.459231524902226e-05, + "loss": 0.4433, + "step": 11830 + }, + { + "epoch": 0.5925184536469411, + "grad_norm": 88.63096618652344, + "learning_rate": 4.458304758021167e-05, + "loss": 0.4587, + "step": 11840 + }, + { + "epoch": 0.5930188915300888, + "grad_norm": 49.49354934692383, + "learning_rate": 4.457377991140109e-05, + "loss": 0.7712, + "step": 11850 + }, + { + "epoch": 0.5935193294132366, + "grad_norm": 42.366031646728516, + "learning_rate": 4.45645122425905e-05, + "loss": 0.9175, + "step": 11860 + }, + { + "epoch": 0.5940197672963843, + "grad_norm": 60.983062744140625, + "learning_rate": 4.455524457377991e-05, + "loss": 0.4365, + "step": 11870 + }, + { + "epoch": 0.5945202051795321, + "grad_norm": 92.1437759399414, + "learning_rate": 4.454597690496933e-05, + "loss": 0.2704, + "step": 11880 + }, + { + "epoch": 0.5950206430626799, + "grad_norm": 91.60601806640625, + "learning_rate": 4.453670923615874e-05, + "loss": 0.5537, + "step": 11890 + }, + { + "epoch": 0.5955210809458276, + "grad_norm": 0.023919852450489998, + "learning_rate": 4.452744156734815e-05, + "loss": 0.2391, + "step": 11900 + }, + { + "epoch": 0.5960215188289754, + "grad_norm": 0.9158336520195007, + "learning_rate": 4.451817389853756e-05, + "loss": 1.2987, + "step": 11910 + }, + { + "epoch": 0.5965219567121232, + "grad_norm": 45.872711181640625, + "learning_rate": 4.450890622972698e-05, + "loss": 0.8764, + "step": 11920 + }, + { + "epoch": 0.5970223945952708, + "grad_norm": 2.7229268550872803, + "learning_rate": 4.449963856091639e-05, + "loss": 0.3885, + "step": 11930 + }, + { + "epoch": 0.5975228324784186, + "grad_norm": 22.1697998046875, + "learning_rate": 4.44903708921058e-05, + "loss": 0.1862, + "step": 11940 + }, + { + "epoch": 0.5980232703615663, + "grad_norm": 30.726591110229492, + "learning_rate": 4.448110322329521e-05, + "loss": 0.3997, + "step": 11950 + }, + { + "epoch": 0.5985237082447141, + "grad_norm": 32.98790740966797, + "learning_rate": 4.4471835554484626e-05, + "loss": 0.6639, + "step": 11960 + }, + { + "epoch": 0.5990241461278619, + "grad_norm": 70.47290802001953, + "learning_rate": 4.446256788567404e-05, + "loss": 0.8417, + "step": 11970 + }, + { + "epoch": 0.5995245840110096, + "grad_norm": 0.05688260495662689, + "learning_rate": 4.445330021686345e-05, + "loss": 0.5657, + "step": 11980 + }, + { + "epoch": 0.6000250218941574, + "grad_norm": 57.06572723388672, + "learning_rate": 4.4444032548052866e-05, + "loss": 0.2091, + "step": 11990 + }, + { + "epoch": 0.6005254597773051, + "grad_norm": 20.745426177978516, + "learning_rate": 4.443476487924228e-05, + "loss": 0.4621, + "step": 12000 + }, + { + "epoch": 0.6010258976604529, + "grad_norm": 4.331947326660156, + "learning_rate": 4.442549721043169e-05, + "loss": 0.9854, + "step": 12010 + }, + { + "epoch": 0.6015263355436007, + "grad_norm": 82.78685760498047, + "learning_rate": 4.44162295416211e-05, + "loss": 0.3571, + "step": 12020 + }, + { + "epoch": 0.6020267734267484, + "grad_norm": 0.6238629817962646, + "learning_rate": 4.4406961872810515e-05, + "loss": 0.1949, + "step": 12030 + }, + { + "epoch": 0.6025272113098962, + "grad_norm": 0.7811834216117859, + "learning_rate": 4.439769420399993e-05, + "loss": 0.2467, + "step": 12040 + }, + { + "epoch": 0.603027649193044, + "grad_norm": 70.15141296386719, + "learning_rate": 4.438842653518934e-05, + "loss": 0.3664, + "step": 12050 + }, + { + "epoch": 0.6035280870761917, + "grad_norm": 0.08112870156764984, + "learning_rate": 4.437915886637875e-05, + "loss": 0.427, + "step": 12060 + }, + { + "epoch": 0.6040285249593395, + "grad_norm": 4.392254829406738, + "learning_rate": 4.4369891197568164e-05, + "loss": 0.4033, + "step": 12070 + }, + { + "epoch": 0.6045289628424871, + "grad_norm": 0.10744023323059082, + "learning_rate": 4.436062352875758e-05, + "loss": 0.2757, + "step": 12080 + }, + { + "epoch": 0.6050294007256349, + "grad_norm": 0.2993903160095215, + "learning_rate": 4.435135585994699e-05, + "loss": 0.5666, + "step": 12090 + }, + { + "epoch": 0.6055298386087827, + "grad_norm": 87.45745849609375, + "learning_rate": 4.4342088191136404e-05, + "loss": 0.7896, + "step": 12100 + }, + { + "epoch": 0.6060302764919304, + "grad_norm": 1.504975438117981, + "learning_rate": 4.433282052232582e-05, + "loss": 0.6958, + "step": 12110 + }, + { + "epoch": 0.6065307143750782, + "grad_norm": 40.05162811279297, + "learning_rate": 4.432355285351523e-05, + "loss": 0.2923, + "step": 12120 + }, + { + "epoch": 0.6070311522582259, + "grad_norm": 4.808842658996582, + "learning_rate": 4.431428518470464e-05, + "loss": 0.6429, + "step": 12130 + }, + { + "epoch": 0.6075315901413737, + "grad_norm": 11.17462158203125, + "learning_rate": 4.430501751589405e-05, + "loss": 0.9197, + "step": 12140 + }, + { + "epoch": 0.6080320280245215, + "grad_norm": 36.139705657958984, + "learning_rate": 4.429574984708347e-05, + "loss": 0.6462, + "step": 12150 + }, + { + "epoch": 0.6085324659076692, + "grad_norm": 0.6893996000289917, + "learning_rate": 4.4286482178272884e-05, + "loss": 0.426, + "step": 12160 + }, + { + "epoch": 0.609032903790817, + "grad_norm": 10.370319366455078, + "learning_rate": 4.4277214509462286e-05, + "loss": 0.498, + "step": 12170 + }, + { + "epoch": 0.6095333416739647, + "grad_norm": 0.7505565285682678, + "learning_rate": 4.42679468406517e-05, + "loss": 0.1878, + "step": 12180 + }, + { + "epoch": 0.6100337795571125, + "grad_norm": 20.08502960205078, + "learning_rate": 4.425867917184112e-05, + "loss": 0.3674, + "step": 12190 + }, + { + "epoch": 0.6105342174402603, + "grad_norm": 88.42723846435547, + "learning_rate": 4.424941150303053e-05, + "loss": 0.3131, + "step": 12200 + }, + { + "epoch": 0.611034655323408, + "grad_norm": 35.079063415527344, + "learning_rate": 4.424014383421994e-05, + "loss": 0.1549, + "step": 12210 + }, + { + "epoch": 0.6115350932065557, + "grad_norm": 34.78264236450195, + "learning_rate": 4.423087616540936e-05, + "loss": 0.5152, + "step": 12220 + }, + { + "epoch": 0.6120355310897035, + "grad_norm": 23.549707412719727, + "learning_rate": 4.4221608496598766e-05, + "loss": 0.7669, + "step": 12230 + }, + { + "epoch": 0.6125359689728512, + "grad_norm": 45.84196090698242, + "learning_rate": 4.421234082778818e-05, + "loss": 0.2708, + "step": 12240 + }, + { + "epoch": 0.613036406855999, + "grad_norm": 30.195632934570312, + "learning_rate": 4.420307315897759e-05, + "loss": 0.3372, + "step": 12250 + }, + { + "epoch": 0.6135368447391467, + "grad_norm": 23.057947158813477, + "learning_rate": 4.4193805490167007e-05, + "loss": 0.4733, + "step": 12260 + }, + { + "epoch": 0.6140372826222945, + "grad_norm": 1.8045527935028076, + "learning_rate": 4.418453782135642e-05, + "loss": 0.4554, + "step": 12270 + }, + { + "epoch": 0.6145377205054423, + "grad_norm": 0.370831698179245, + "learning_rate": 4.417527015254583e-05, + "loss": 0.4442, + "step": 12280 + }, + { + "epoch": 0.61503815838859, + "grad_norm": 1.7800496816635132, + "learning_rate": 4.416600248373524e-05, + "loss": 0.5861, + "step": 12290 + }, + { + "epoch": 0.6155385962717378, + "grad_norm": 0.24706153571605682, + "learning_rate": 4.4156734814924656e-05, + "loss": 0.1677, + "step": 12300 + }, + { + "epoch": 0.6160390341548855, + "grad_norm": 8.087127685546875, + "learning_rate": 4.414746714611407e-05, + "loss": 0.8596, + "step": 12310 + }, + { + "epoch": 0.6165394720380333, + "grad_norm": 0.3209165036678314, + "learning_rate": 4.413819947730348e-05, + "loss": 0.4341, + "step": 12320 + }, + { + "epoch": 0.6170399099211811, + "grad_norm": 78.07205963134766, + "learning_rate": 4.4128931808492896e-05, + "loss": 0.7092, + "step": 12330 + }, + { + "epoch": 0.6175403478043288, + "grad_norm": 41.46051025390625, + "learning_rate": 4.4119664139682305e-05, + "loss": 0.3335, + "step": 12340 + }, + { + "epoch": 0.6180407856874766, + "grad_norm": 35.89058303833008, + "learning_rate": 4.411039647087172e-05, + "loss": 0.1238, + "step": 12350 + }, + { + "epoch": 0.6185412235706242, + "grad_norm": 43.5078239440918, + "learning_rate": 4.410112880206113e-05, + "loss": 0.4532, + "step": 12360 + }, + { + "epoch": 0.619041661453772, + "grad_norm": 96.11925506591797, + "learning_rate": 4.4091861133250545e-05, + "loss": 1.1458, + "step": 12370 + }, + { + "epoch": 0.6195420993369198, + "grad_norm": 0.3930479884147644, + "learning_rate": 4.408259346443996e-05, + "loss": 0.8063, + "step": 12380 + }, + { + "epoch": 0.6200425372200675, + "grad_norm": 0.353585422039032, + "learning_rate": 4.407332579562937e-05, + "loss": 0.3316, + "step": 12390 + }, + { + "epoch": 0.6205429751032153, + "grad_norm": 0.06623475253582001, + "learning_rate": 4.406405812681878e-05, + "loss": 0.1954, + "step": 12400 + }, + { + "epoch": 0.6210434129863631, + "grad_norm": 46.20969772338867, + "learning_rate": 4.4054790458008194e-05, + "loss": 0.3883, + "step": 12410 + }, + { + "epoch": 0.6215438508695108, + "grad_norm": 4.702070236206055, + "learning_rate": 4.404552278919761e-05, + "loss": 0.4979, + "step": 12420 + }, + { + "epoch": 0.6220442887526586, + "grad_norm": 60.87689971923828, + "learning_rate": 4.403625512038702e-05, + "loss": 0.4639, + "step": 12430 + }, + { + "epoch": 0.6225447266358063, + "grad_norm": 35.43964767456055, + "learning_rate": 4.4026987451576434e-05, + "loss": 0.2689, + "step": 12440 + }, + { + "epoch": 0.6230451645189541, + "grad_norm": 13.349968910217285, + "learning_rate": 4.401771978276584e-05, + "loss": 0.2552, + "step": 12450 + }, + { + "epoch": 0.6235456024021019, + "grad_norm": 108.74763488769531, + "learning_rate": 4.400845211395526e-05, + "loss": 0.3843, + "step": 12460 + }, + { + "epoch": 0.6240460402852496, + "grad_norm": 2.537224769592285, + "learning_rate": 4.399918444514467e-05, + "loss": 0.4874, + "step": 12470 + }, + { + "epoch": 0.6245464781683974, + "grad_norm": 0.5688074231147766, + "learning_rate": 4.398991677633408e-05, + "loss": 0.2711, + "step": 12480 + }, + { + "epoch": 0.625046916051545, + "grad_norm": 76.73430633544922, + "learning_rate": 4.39806491075235e-05, + "loss": 0.4387, + "step": 12490 + }, + { + "epoch": 0.6255473539346929, + "grad_norm": 0.38058730959892273, + "learning_rate": 4.3971381438712914e-05, + "loss": 0.9498, + "step": 12500 + }, + { + "epoch": 0.6260477918178406, + "grad_norm": 30.155471801757812, + "learning_rate": 4.3962113769902316e-05, + "loss": 0.6381, + "step": 12510 + }, + { + "epoch": 0.6265482297009883, + "grad_norm": 25.347126007080078, + "learning_rate": 4.395284610109173e-05, + "loss": 0.4318, + "step": 12520 + }, + { + "epoch": 0.6270486675841361, + "grad_norm": 15.977556228637695, + "learning_rate": 4.394357843228115e-05, + "loss": 0.636, + "step": 12530 + }, + { + "epoch": 0.6275491054672838, + "grad_norm": 2.537062406539917, + "learning_rate": 4.393431076347056e-05, + "loss": 0.4666, + "step": 12540 + }, + { + "epoch": 0.6280495433504316, + "grad_norm": 13.399306297302246, + "learning_rate": 4.392504309465997e-05, + "loss": 0.4876, + "step": 12550 + }, + { + "epoch": 0.6285499812335794, + "grad_norm": 10.50390625, + "learning_rate": 4.391577542584938e-05, + "loss": 0.4461, + "step": 12560 + }, + { + "epoch": 0.6290504191167271, + "grad_norm": 88.41866302490234, + "learning_rate": 4.3906507757038796e-05, + "loss": 0.4295, + "step": 12570 + }, + { + "epoch": 0.6295508569998749, + "grad_norm": 0.10676202923059464, + "learning_rate": 4.389724008822821e-05, + "loss": 0.2406, + "step": 12580 + }, + { + "epoch": 0.6300512948830227, + "grad_norm": 64.36186218261719, + "learning_rate": 4.388797241941762e-05, + "loss": 0.6541, + "step": 12590 + }, + { + "epoch": 0.6305517327661704, + "grad_norm": 67.54058837890625, + "learning_rate": 4.3878704750607037e-05, + "loss": 0.488, + "step": 12600 + }, + { + "epoch": 0.6310521706493182, + "grad_norm": 0.16592150926589966, + "learning_rate": 4.386943708179645e-05, + "loss": 0.3263, + "step": 12610 + }, + { + "epoch": 0.6315526085324659, + "grad_norm": 11.627266883850098, + "learning_rate": 4.3860169412985854e-05, + "loss": 0.2492, + "step": 12620 + }, + { + "epoch": 0.6320530464156137, + "grad_norm": 41.52277755737305, + "learning_rate": 4.385090174417527e-05, + "loss": 0.3882, + "step": 12630 + }, + { + "epoch": 0.6325534842987615, + "grad_norm": 62.07949447631836, + "learning_rate": 4.3841634075364686e-05, + "loss": 0.9684, + "step": 12640 + }, + { + "epoch": 0.6330539221819091, + "grad_norm": 4.7843170166015625, + "learning_rate": 4.38323664065541e-05, + "loss": 0.5115, + "step": 12650 + }, + { + "epoch": 0.6335543600650569, + "grad_norm": 4.047616004943848, + "learning_rate": 4.382309873774351e-05, + "loss": 0.6524, + "step": 12660 + }, + { + "epoch": 0.6340547979482046, + "grad_norm": 68.84085083007812, + "learning_rate": 4.381383106893292e-05, + "loss": 0.4959, + "step": 12670 + }, + { + "epoch": 0.6345552358313524, + "grad_norm": 48.53059387207031, + "learning_rate": 4.3804563400122334e-05, + "loss": 0.492, + "step": 12680 + }, + { + "epoch": 0.6350556737145002, + "grad_norm": 45.22385787963867, + "learning_rate": 4.379529573131175e-05, + "loss": 0.5646, + "step": 12690 + }, + { + "epoch": 0.6355561115976479, + "grad_norm": 1.3907814025878906, + "learning_rate": 4.378602806250116e-05, + "loss": 0.8281, + "step": 12700 + }, + { + "epoch": 0.6360565494807957, + "grad_norm": 2.892920970916748, + "learning_rate": 4.3776760393690575e-05, + "loss": 0.4868, + "step": 12710 + }, + { + "epoch": 0.6365569873639435, + "grad_norm": 42.31459426879883, + "learning_rate": 4.376749272487999e-05, + "loss": 0.5821, + "step": 12720 + }, + { + "epoch": 0.6370574252470912, + "grad_norm": 69.1933822631836, + "learning_rate": 4.37582250560694e-05, + "loss": 0.9324, + "step": 12730 + }, + { + "epoch": 0.637557863130239, + "grad_norm": 12.492966651916504, + "learning_rate": 4.374895738725881e-05, + "loss": 0.6777, + "step": 12740 + }, + { + "epoch": 0.6380583010133867, + "grad_norm": 22.121509552001953, + "learning_rate": 4.3739689718448224e-05, + "loss": 0.3458, + "step": 12750 + }, + { + "epoch": 0.6385587388965345, + "grad_norm": 0.1293872445821762, + "learning_rate": 4.373042204963764e-05, + "loss": 0.3746, + "step": 12760 + }, + { + "epoch": 0.6390591767796823, + "grad_norm": 36.02532958984375, + "learning_rate": 4.372115438082705e-05, + "loss": 0.3614, + "step": 12770 + }, + { + "epoch": 0.63955961466283, + "grad_norm": 20.271940231323242, + "learning_rate": 4.371188671201646e-05, + "loss": 0.2803, + "step": 12780 + }, + { + "epoch": 0.6400600525459778, + "grad_norm": 38.700565338134766, + "learning_rate": 4.370261904320587e-05, + "loss": 0.7959, + "step": 12790 + }, + { + "epoch": 0.6405604904291254, + "grad_norm": 83.61766815185547, + "learning_rate": 4.369335137439529e-05, + "loss": 0.266, + "step": 12800 + }, + { + "epoch": 0.6410609283122732, + "grad_norm": 30.643518447875977, + "learning_rate": 4.36840837055847e-05, + "loss": 0.8255, + "step": 12810 + }, + { + "epoch": 0.641561366195421, + "grad_norm": 0.16006582975387573, + "learning_rate": 4.367481603677411e-05, + "loss": 0.5889, + "step": 12820 + }, + { + "epoch": 0.6420618040785687, + "grad_norm": 38.548675537109375, + "learning_rate": 4.366554836796353e-05, + "loss": 0.4962, + "step": 12830 + }, + { + "epoch": 0.6425622419617165, + "grad_norm": 41.70745849609375, + "learning_rate": 4.365628069915294e-05, + "loss": 0.4417, + "step": 12840 + }, + { + "epoch": 0.6430626798448642, + "grad_norm": 38.1447639465332, + "learning_rate": 4.3647013030342346e-05, + "loss": 0.4133, + "step": 12850 + }, + { + "epoch": 0.643563117728012, + "grad_norm": 41.826629638671875, + "learning_rate": 4.363774536153176e-05, + "loss": 0.4948, + "step": 12860 + }, + { + "epoch": 0.6440635556111598, + "grad_norm": 22.964672088623047, + "learning_rate": 4.362847769272118e-05, + "loss": 0.3367, + "step": 12870 + }, + { + "epoch": 0.6445639934943075, + "grad_norm": 30.520267486572266, + "learning_rate": 4.3619210023910586e-05, + "loss": 0.6434, + "step": 12880 + }, + { + "epoch": 0.6450644313774553, + "grad_norm": 40.79505157470703, + "learning_rate": 4.3609942355099995e-05, + "loss": 0.8192, + "step": 12890 + }, + { + "epoch": 0.6455648692606031, + "grad_norm": 9.273140907287598, + "learning_rate": 4.360067468628941e-05, + "loss": 0.7578, + "step": 12900 + }, + { + "epoch": 0.6460653071437508, + "grad_norm": 101.35041046142578, + "learning_rate": 4.3591407017478826e-05, + "loss": 0.683, + "step": 12910 + }, + { + "epoch": 0.6465657450268986, + "grad_norm": 1.1539467573165894, + "learning_rate": 4.3582139348668235e-05, + "loss": 0.3086, + "step": 12920 + }, + { + "epoch": 0.6470661829100463, + "grad_norm": 69.45064544677734, + "learning_rate": 4.357287167985765e-05, + "loss": 0.2986, + "step": 12930 + }, + { + "epoch": 0.647566620793194, + "grad_norm": 121.3116455078125, + "learning_rate": 4.3563604011047066e-05, + "loss": 0.4625, + "step": 12940 + }, + { + "epoch": 0.6480670586763418, + "grad_norm": 5.156425476074219, + "learning_rate": 4.3554336342236475e-05, + "loss": 0.4717, + "step": 12950 + }, + { + "epoch": 0.6485674965594895, + "grad_norm": 4.5928802490234375, + "learning_rate": 4.3545068673425884e-05, + "loss": 0.6426, + "step": 12960 + }, + { + "epoch": 0.6490679344426373, + "grad_norm": 41.28687286376953, + "learning_rate": 4.35358010046153e-05, + "loss": 0.3102, + "step": 12970 + }, + { + "epoch": 0.649568372325785, + "grad_norm": 2.9305872917175293, + "learning_rate": 4.3526533335804715e-05, + "loss": 0.4235, + "step": 12980 + }, + { + "epoch": 0.6500688102089328, + "grad_norm": 2.347858428955078, + "learning_rate": 4.351726566699413e-05, + "loss": 0.604, + "step": 12990 + }, + { + "epoch": 0.6505692480920806, + "grad_norm": 4.900949954986572, + "learning_rate": 4.350799799818354e-05, + "loss": 0.2951, + "step": 13000 + }, + { + "epoch": 0.6510696859752283, + "grad_norm": 6.875156879425049, + "learning_rate": 4.349873032937295e-05, + "loss": 0.4101, + "step": 13010 + }, + { + "epoch": 0.6515701238583761, + "grad_norm": 21.502370834350586, + "learning_rate": 4.3489462660562364e-05, + "loss": 0.2022, + "step": 13020 + }, + { + "epoch": 0.6520705617415238, + "grad_norm": 4.570702075958252, + "learning_rate": 4.348019499175178e-05, + "loss": 0.4162, + "step": 13030 + }, + { + "epoch": 0.6525709996246716, + "grad_norm": 45.23558044433594, + "learning_rate": 4.347092732294119e-05, + "loss": 0.7985, + "step": 13040 + }, + { + "epoch": 0.6530714375078194, + "grad_norm": 32.66095733642578, + "learning_rate": 4.3461659654130605e-05, + "loss": 0.9972, + "step": 13050 + }, + { + "epoch": 0.6535718753909671, + "grad_norm": 2.27333402633667, + "learning_rate": 4.3452391985320013e-05, + "loss": 0.1217, + "step": 13060 + }, + { + "epoch": 0.6540723132741149, + "grad_norm": 81.01744842529297, + "learning_rate": 4.344312431650943e-05, + "loss": 0.547, + "step": 13070 + }, + { + "epoch": 0.6545727511572627, + "grad_norm": 12.637042045593262, + "learning_rate": 4.343385664769884e-05, + "loss": 0.5, + "step": 13080 + }, + { + "epoch": 0.6550731890404103, + "grad_norm": 87.50300598144531, + "learning_rate": 4.3424588978888254e-05, + "loss": 0.3496, + "step": 13090 + }, + { + "epoch": 0.6555736269235581, + "grad_norm": 0.33247339725494385, + "learning_rate": 4.341532131007767e-05, + "loss": 0.3627, + "step": 13100 + }, + { + "epoch": 0.6560740648067058, + "grad_norm": 11.752845764160156, + "learning_rate": 4.340605364126708e-05, + "loss": 0.1378, + "step": 13110 + }, + { + "epoch": 0.6565745026898536, + "grad_norm": 0.7599644064903259, + "learning_rate": 4.339678597245649e-05, + "loss": 0.5112, + "step": 13120 + }, + { + "epoch": 0.6570749405730014, + "grad_norm": 0.06856774538755417, + "learning_rate": 4.33875183036459e-05, + "loss": 0.4593, + "step": 13130 + }, + { + "epoch": 0.6575753784561491, + "grad_norm": 41.78406524658203, + "learning_rate": 4.337825063483532e-05, + "loss": 0.3289, + "step": 13140 + }, + { + "epoch": 0.6580758163392969, + "grad_norm": 30.118253707885742, + "learning_rate": 4.336898296602473e-05, + "loss": 0.5388, + "step": 13150 + }, + { + "epoch": 0.6585762542224446, + "grad_norm": 32.68709182739258, + "learning_rate": 4.335971529721414e-05, + "loss": 0.2583, + "step": 13160 + }, + { + "epoch": 0.6590766921055924, + "grad_norm": 1.1538360118865967, + "learning_rate": 4.335044762840355e-05, + "loss": 0.4172, + "step": 13170 + }, + { + "epoch": 0.6595771299887402, + "grad_norm": 0.43954408168792725, + "learning_rate": 4.334117995959297e-05, + "loss": 0.2143, + "step": 13180 + }, + { + "epoch": 0.6600775678718879, + "grad_norm": 56.845863342285156, + "learning_rate": 4.3331912290782376e-05, + "loss": 0.6051, + "step": 13190 + }, + { + "epoch": 0.6605780057550357, + "grad_norm": 0.002370188245549798, + "learning_rate": 4.332264462197179e-05, + "loss": 0.1061, + "step": 13200 + }, + { + "epoch": 0.6610784436381834, + "grad_norm": 21.583030700683594, + "learning_rate": 4.331337695316121e-05, + "loss": 0.5814, + "step": 13210 + }, + { + "epoch": 0.6615788815213312, + "grad_norm": 18.485780715942383, + "learning_rate": 4.3304109284350616e-05, + "loss": 0.7994, + "step": 13220 + }, + { + "epoch": 0.662079319404479, + "grad_norm": 57.831302642822266, + "learning_rate": 4.3294841615540025e-05, + "loss": 0.6335, + "step": 13230 + }, + { + "epoch": 0.6625797572876266, + "grad_norm": 21.705793380737305, + "learning_rate": 4.328557394672944e-05, + "loss": 0.553, + "step": 13240 + }, + { + "epoch": 0.6630801951707744, + "grad_norm": 9.71451473236084, + "learning_rate": 4.3276306277918856e-05, + "loss": 0.2502, + "step": 13250 + }, + { + "epoch": 0.6635806330539222, + "grad_norm": 1.4677419662475586, + "learning_rate": 4.3267038609108265e-05, + "loss": 0.529, + "step": 13260 + }, + { + "epoch": 0.6640810709370699, + "grad_norm": 14.69130802154541, + "learning_rate": 4.325777094029768e-05, + "loss": 0.3579, + "step": 13270 + }, + { + "epoch": 0.6645815088202177, + "grad_norm": 16.951833724975586, + "learning_rate": 4.324850327148709e-05, + "loss": 0.3637, + "step": 13280 + }, + { + "epoch": 0.6650819467033654, + "grad_norm": 23.755945205688477, + "learning_rate": 4.3239235602676505e-05, + "loss": 0.3857, + "step": 13290 + }, + { + "epoch": 0.6655823845865132, + "grad_norm": 0.003661619033664465, + "learning_rate": 4.3229967933865914e-05, + "loss": 0.2741, + "step": 13300 + }, + { + "epoch": 0.666082822469661, + "grad_norm": 26.75663185119629, + "learning_rate": 4.322070026505533e-05, + "loss": 0.1307, + "step": 13310 + }, + { + "epoch": 0.6665832603528087, + "grad_norm": 12.504801750183105, + "learning_rate": 4.3211432596244745e-05, + "loss": 0.7663, + "step": 13320 + }, + { + "epoch": 0.6670836982359565, + "grad_norm": 38.03091812133789, + "learning_rate": 4.320216492743416e-05, + "loss": 0.6481, + "step": 13330 + }, + { + "epoch": 0.6675841361191042, + "grad_norm": 69.92797088623047, + "learning_rate": 4.319289725862356e-05, + "loss": 0.5662, + "step": 13340 + }, + { + "epoch": 0.668084574002252, + "grad_norm": 25.577327728271484, + "learning_rate": 4.318362958981298e-05, + "loss": 0.2139, + "step": 13350 + }, + { + "epoch": 0.6685850118853998, + "grad_norm": 2.5837903022766113, + "learning_rate": 4.3174361921002394e-05, + "loss": 0.524, + "step": 13360 + }, + { + "epoch": 0.6690854497685474, + "grad_norm": 0.0263813603669405, + "learning_rate": 4.316509425219181e-05, + "loss": 0.3274, + "step": 13370 + }, + { + "epoch": 0.6695858876516952, + "grad_norm": 6.443814754486084, + "learning_rate": 4.315582658338122e-05, + "loss": 0.4608, + "step": 13380 + }, + { + "epoch": 0.670086325534843, + "grad_norm": 5.371328353881836, + "learning_rate": 4.314655891457063e-05, + "loss": 0.3108, + "step": 13390 + }, + { + "epoch": 0.6705867634179907, + "grad_norm": 1.3910088539123535, + "learning_rate": 4.313729124576004e-05, + "loss": 0.4204, + "step": 13400 + }, + { + "epoch": 0.6710872013011385, + "grad_norm": 0.4572153091430664, + "learning_rate": 4.312802357694946e-05, + "loss": 0.3136, + "step": 13410 + }, + { + "epoch": 0.6715876391842862, + "grad_norm": 50.683807373046875, + "learning_rate": 4.311875590813887e-05, + "loss": 0.6663, + "step": 13420 + }, + { + "epoch": 0.672088077067434, + "grad_norm": 0.10915312170982361, + "learning_rate": 4.3109488239328283e-05, + "loss": 0.4265, + "step": 13430 + }, + { + "epoch": 0.6725885149505818, + "grad_norm": 49.27344512939453, + "learning_rate": 4.31002205705177e-05, + "loss": 0.1599, + "step": 13440 + }, + { + "epoch": 0.6730889528337295, + "grad_norm": 0.1435779333114624, + "learning_rate": 4.30909529017071e-05, + "loss": 0.5345, + "step": 13450 + }, + { + "epoch": 0.6735893907168773, + "grad_norm": 46.57610321044922, + "learning_rate": 4.308168523289652e-05, + "loss": 0.5841, + "step": 13460 + }, + { + "epoch": 0.674089828600025, + "grad_norm": 3.1308705806732178, + "learning_rate": 4.307241756408593e-05, + "loss": 0.0963, + "step": 13470 + }, + { + "epoch": 0.6745902664831728, + "grad_norm": 5.851505756378174, + "learning_rate": 4.306314989527535e-05, + "loss": 0.2977, + "step": 13480 + }, + { + "epoch": 0.6750907043663206, + "grad_norm": 0.8626621961593628, + "learning_rate": 4.305388222646476e-05, + "loss": 0.6514, + "step": 13490 + }, + { + "epoch": 0.6755911422494683, + "grad_norm": 33.702064514160156, + "learning_rate": 4.3044614557654166e-05, + "loss": 0.553, + "step": 13500 + }, + { + "epoch": 0.676091580132616, + "grad_norm": 0.05740151181817055, + "learning_rate": 4.303534688884358e-05, + "loss": 0.6035, + "step": 13510 + }, + { + "epoch": 0.6765920180157637, + "grad_norm": 1.8221735954284668, + "learning_rate": 4.3026079220033e-05, + "loss": 0.2447, + "step": 13520 + }, + { + "epoch": 0.6770924558989115, + "grad_norm": 21.70306396484375, + "learning_rate": 4.3016811551222406e-05, + "loss": 0.739, + "step": 13530 + }, + { + "epoch": 0.6775928937820593, + "grad_norm": 7.033788681030273, + "learning_rate": 4.300754388241182e-05, + "loss": 0.4467, + "step": 13540 + }, + { + "epoch": 0.678093331665207, + "grad_norm": 0.288972407579422, + "learning_rate": 4.299827621360124e-05, + "loss": 0.518, + "step": 13550 + }, + { + "epoch": 0.6785937695483548, + "grad_norm": 13.525001525878906, + "learning_rate": 4.2989008544790646e-05, + "loss": 0.2559, + "step": 13560 + }, + { + "epoch": 0.6790942074315026, + "grad_norm": 10.41710090637207, + "learning_rate": 4.2979740875980055e-05, + "loss": 0.5768, + "step": 13570 + }, + { + "epoch": 0.6795946453146503, + "grad_norm": 41.054664611816406, + "learning_rate": 4.297047320716947e-05, + "loss": 0.28, + "step": 13580 + }, + { + "epoch": 0.6800950831977981, + "grad_norm": 0.2954779863357544, + "learning_rate": 4.2961205538358886e-05, + "loss": 0.28, + "step": 13590 + }, + { + "epoch": 0.6805955210809458, + "grad_norm": 24.913869857788086, + "learning_rate": 4.2951937869548295e-05, + "loss": 0.6562, + "step": 13600 + }, + { + "epoch": 0.6810959589640936, + "grad_norm": 38.60606002807617, + "learning_rate": 4.2942670200737704e-05, + "loss": 0.3826, + "step": 13610 + }, + { + "epoch": 0.6815963968472414, + "grad_norm": 59.34792709350586, + "learning_rate": 4.293340253192712e-05, + "loss": 0.5746, + "step": 13620 + }, + { + "epoch": 0.6820968347303891, + "grad_norm": 0.9588309526443481, + "learning_rate": 4.2924134863116535e-05, + "loss": 0.4799, + "step": 13630 + }, + { + "epoch": 0.6825972726135369, + "grad_norm": 30.7623233795166, + "learning_rate": 4.2914867194305944e-05, + "loss": 0.2004, + "step": 13640 + }, + { + "epoch": 0.6830977104966846, + "grad_norm": 14.235272407531738, + "learning_rate": 4.290559952549536e-05, + "loss": 0.2564, + "step": 13650 + }, + { + "epoch": 0.6835981483798323, + "grad_norm": 32.18157196044922, + "learning_rate": 4.2896331856684775e-05, + "loss": 0.2256, + "step": 13660 + }, + { + "epoch": 0.6840985862629801, + "grad_norm": 55.49310302734375, + "learning_rate": 4.2887064187874184e-05, + "loss": 0.7366, + "step": 13670 + }, + { + "epoch": 0.6845990241461278, + "grad_norm": 27.80856704711914, + "learning_rate": 4.287779651906359e-05, + "loss": 0.8093, + "step": 13680 + }, + { + "epoch": 0.6850994620292756, + "grad_norm": 0.5038571953773499, + "learning_rate": 4.286852885025301e-05, + "loss": 0.3812, + "step": 13690 + }, + { + "epoch": 0.6855998999124233, + "grad_norm": 21.244699478149414, + "learning_rate": 4.2859261181442424e-05, + "loss": 0.4553, + "step": 13700 + }, + { + "epoch": 0.6861003377955711, + "grad_norm": 25.410696029663086, + "learning_rate": 4.284999351263183e-05, + "loss": 0.8983, + "step": 13710 + }, + { + "epoch": 0.6866007756787189, + "grad_norm": 0.8634992241859436, + "learning_rate": 4.284072584382125e-05, + "loss": 0.4568, + "step": 13720 + }, + { + "epoch": 0.6871012135618666, + "grad_norm": 46.07569885253906, + "learning_rate": 4.283145817501066e-05, + "loss": 0.4227, + "step": 13730 + }, + { + "epoch": 0.6876016514450144, + "grad_norm": 22.690187454223633, + "learning_rate": 4.282219050620007e-05, + "loss": 0.3164, + "step": 13740 + }, + { + "epoch": 0.6881020893281622, + "grad_norm": 9.276498794555664, + "learning_rate": 4.281292283738948e-05, + "loss": 0.3643, + "step": 13750 + }, + { + "epoch": 0.6886025272113099, + "grad_norm": 2.1595253944396973, + "learning_rate": 4.28036551685789e-05, + "loss": 0.3083, + "step": 13760 + }, + { + "epoch": 0.6891029650944577, + "grad_norm": 46.576324462890625, + "learning_rate": 4.279438749976831e-05, + "loss": 0.4126, + "step": 13770 + }, + { + "epoch": 0.6896034029776054, + "grad_norm": 61.02355194091797, + "learning_rate": 4.278511983095772e-05, + "loss": 0.3175, + "step": 13780 + }, + { + "epoch": 0.6901038408607532, + "grad_norm": 0.03991875797510147, + "learning_rate": 4.277585216214713e-05, + "loss": 0.315, + "step": 13790 + }, + { + "epoch": 0.690604278743901, + "grad_norm": 6.622135639190674, + "learning_rate": 4.276658449333655e-05, + "loss": 0.2551, + "step": 13800 + }, + { + "epoch": 0.6911047166270486, + "grad_norm": 0.4558590054512024, + "learning_rate": 4.275731682452596e-05, + "loss": 0.4233, + "step": 13810 + }, + { + "epoch": 0.6916051545101964, + "grad_norm": 73.31211853027344, + "learning_rate": 4.274804915571538e-05, + "loss": 0.6503, + "step": 13820 + }, + { + "epoch": 0.6921055923933441, + "grad_norm": 13.738667488098145, + "learning_rate": 4.273878148690479e-05, + "loss": 0.7209, + "step": 13830 + }, + { + "epoch": 0.6926060302764919, + "grad_norm": 11.688087463378906, + "learning_rate": 4.2729513818094196e-05, + "loss": 0.367, + "step": 13840 + }, + { + "epoch": 0.6931064681596397, + "grad_norm": 0.05844360589981079, + "learning_rate": 4.272024614928361e-05, + "loss": 0.671, + "step": 13850 + }, + { + "epoch": 0.6936069060427874, + "grad_norm": 0.25387221574783325, + "learning_rate": 4.271097848047303e-05, + "loss": 0.482, + "step": 13860 + }, + { + "epoch": 0.6941073439259352, + "grad_norm": 43.337196350097656, + "learning_rate": 4.2701710811662436e-05, + "loss": 0.2926, + "step": 13870 + }, + { + "epoch": 0.694607781809083, + "grad_norm": 0.5394840240478516, + "learning_rate": 4.269244314285185e-05, + "loss": 0.8513, + "step": 13880 + }, + { + "epoch": 0.6951082196922307, + "grad_norm": 30.476741790771484, + "learning_rate": 4.268317547404126e-05, + "loss": 0.6977, + "step": 13890 + }, + { + "epoch": 0.6956086575753785, + "grad_norm": 44.68612289428711, + "learning_rate": 4.2673907805230676e-05, + "loss": 0.3352, + "step": 13900 + }, + { + "epoch": 0.6961090954585262, + "grad_norm": 50.103271484375, + "learning_rate": 4.2664640136420085e-05, + "loss": 0.1776, + "step": 13910 + }, + { + "epoch": 0.696609533341674, + "grad_norm": 39.67720413208008, + "learning_rate": 4.26553724676095e-05, + "loss": 0.3745, + "step": 13920 + }, + { + "epoch": 0.6971099712248218, + "grad_norm": 45.413536071777344, + "learning_rate": 4.2646104798798916e-05, + "loss": 0.3887, + "step": 13930 + }, + { + "epoch": 0.6976104091079695, + "grad_norm": 29.445186614990234, + "learning_rate": 4.2636837129988325e-05, + "loss": 0.3362, + "step": 13940 + }, + { + "epoch": 0.6981108469911173, + "grad_norm": 12.860486030578613, + "learning_rate": 4.2627569461177734e-05, + "loss": 0.758, + "step": 13950 + }, + { + "epoch": 0.6986112848742649, + "grad_norm": 0.04202011227607727, + "learning_rate": 4.261830179236715e-05, + "loss": 0.9141, + "step": 13960 + }, + { + "epoch": 0.6991117227574127, + "grad_norm": 30.548669815063477, + "learning_rate": 4.2609034123556565e-05, + "loss": 0.5134, + "step": 13970 + }, + { + "epoch": 0.6996121606405605, + "grad_norm": 29.87716293334961, + "learning_rate": 4.2599766454745974e-05, + "loss": 0.4134, + "step": 13980 + }, + { + "epoch": 0.7001125985237082, + "grad_norm": 41.03196334838867, + "learning_rate": 4.259049878593539e-05, + "loss": 0.3048, + "step": 13990 + }, + { + "epoch": 0.700613036406856, + "grad_norm": 59.244773864746094, + "learning_rate": 4.25812311171248e-05, + "loss": 0.1517, + "step": 14000 + }, + { + "epoch": 0.7011134742900037, + "grad_norm": 48.09068298339844, + "learning_rate": 4.2571963448314214e-05, + "loss": 0.5351, + "step": 14010 + }, + { + "epoch": 0.7016139121731515, + "grad_norm": 0.01765144057571888, + "learning_rate": 4.256269577950362e-05, + "loss": 0.3207, + "step": 14020 + }, + { + "epoch": 0.7021143500562993, + "grad_norm": 0.08661237359046936, + "learning_rate": 4.255342811069304e-05, + "loss": 0.7264, + "step": 14030 + }, + { + "epoch": 0.702614787939447, + "grad_norm": 4.669400691986084, + "learning_rate": 4.2544160441882454e-05, + "loss": 0.4748, + "step": 14040 + }, + { + "epoch": 0.7031152258225948, + "grad_norm": 31.052345275878906, + "learning_rate": 4.253489277307186e-05, + "loss": 0.1003, + "step": 14050 + }, + { + "epoch": 0.7036156637057426, + "grad_norm": 43.32801055908203, + "learning_rate": 4.252562510426127e-05, + "loss": 0.265, + "step": 14060 + }, + { + "epoch": 0.7041161015888903, + "grad_norm": 96.66951751708984, + "learning_rate": 4.251635743545069e-05, + "loss": 0.4036, + "step": 14070 + }, + { + "epoch": 0.7046165394720381, + "grad_norm": 150.3602752685547, + "learning_rate": 4.25070897666401e-05, + "loss": 0.488, + "step": 14080 + }, + { + "epoch": 0.7051169773551857, + "grad_norm": 81.32795715332031, + "learning_rate": 4.249782209782951e-05, + "loss": 0.5722, + "step": 14090 + }, + { + "epoch": 0.7056174152383335, + "grad_norm": 4.7703166007995605, + "learning_rate": 4.248855442901893e-05, + "loss": 0.4588, + "step": 14100 + }, + { + "epoch": 0.7061178531214813, + "grad_norm": 81.8752212524414, + "learning_rate": 4.2479286760208336e-05, + "loss": 0.6622, + "step": 14110 + }, + { + "epoch": 0.706618291004629, + "grad_norm": 23.119911193847656, + "learning_rate": 4.247001909139775e-05, + "loss": 0.1906, + "step": 14120 + }, + { + "epoch": 0.7071187288877768, + "grad_norm": 44.2313232421875, + "learning_rate": 4.246075142258716e-05, + "loss": 0.9726, + "step": 14130 + }, + { + "epoch": 0.7076191667709245, + "grad_norm": 1.8263012170791626, + "learning_rate": 4.245148375377658e-05, + "loss": 0.9667, + "step": 14140 + }, + { + "epoch": 0.7081196046540723, + "grad_norm": 0.9914520382881165, + "learning_rate": 4.244221608496599e-05, + "loss": 0.3422, + "step": 14150 + }, + { + "epoch": 0.7086200425372201, + "grad_norm": 49.161407470703125, + "learning_rate": 4.243294841615541e-05, + "loss": 0.3924, + "step": 14160 + }, + { + "epoch": 0.7091204804203678, + "grad_norm": 18.951435089111328, + "learning_rate": 4.242368074734481e-05, + "loss": 0.107, + "step": 14170 + }, + { + "epoch": 0.7096209183035156, + "grad_norm": 16.878496170043945, + "learning_rate": 4.2414413078534226e-05, + "loss": 0.375, + "step": 14180 + }, + { + "epoch": 0.7101213561866633, + "grad_norm": 50.18949508666992, + "learning_rate": 4.240514540972364e-05, + "loss": 0.4868, + "step": 14190 + }, + { + "epoch": 0.7106217940698111, + "grad_norm": 25.944473266601562, + "learning_rate": 4.239587774091306e-05, + "loss": 0.3421, + "step": 14200 + }, + { + "epoch": 0.7111222319529589, + "grad_norm": 5.764090538024902, + "learning_rate": 4.2386610072102466e-05, + "loss": 0.306, + "step": 14210 + }, + { + "epoch": 0.7116226698361066, + "grad_norm": 11.66156005859375, + "learning_rate": 4.2377342403291875e-05, + "loss": 0.4259, + "step": 14220 + }, + { + "epoch": 0.7121231077192544, + "grad_norm": 83.23188781738281, + "learning_rate": 4.236807473448129e-05, + "loss": 0.3582, + "step": 14230 + }, + { + "epoch": 0.7126235456024022, + "grad_norm": 36.141170501708984, + "learning_rate": 4.2358807065670706e-05, + "loss": 0.7086, + "step": 14240 + }, + { + "epoch": 0.7131239834855498, + "grad_norm": 0.02628188394010067, + "learning_rate": 4.2349539396860115e-05, + "loss": 0.1458, + "step": 14250 + }, + { + "epoch": 0.7136244213686976, + "grad_norm": 0.7163134813308716, + "learning_rate": 4.234027172804953e-05, + "loss": 0.5666, + "step": 14260 + }, + { + "epoch": 0.7141248592518453, + "grad_norm": 68.46546173095703, + "learning_rate": 4.2331004059238946e-05, + "loss": 0.7173, + "step": 14270 + }, + { + "epoch": 0.7146252971349931, + "grad_norm": 69.5422592163086, + "learning_rate": 4.2321736390428355e-05, + "loss": 0.579, + "step": 14280 + }, + { + "epoch": 0.7151257350181409, + "grad_norm": 3.8768393993377686, + "learning_rate": 4.2312468721617764e-05, + "loss": 0.211, + "step": 14290 + }, + { + "epoch": 0.7156261729012886, + "grad_norm": 46.80589294433594, + "learning_rate": 4.230320105280718e-05, + "loss": 0.4025, + "step": 14300 + }, + { + "epoch": 0.7161266107844364, + "grad_norm": 14.097702980041504, + "learning_rate": 4.2293933383996595e-05, + "loss": 0.2544, + "step": 14310 + }, + { + "epoch": 0.7166270486675841, + "grad_norm": 57.645748138427734, + "learning_rate": 4.2284665715186004e-05, + "loss": 0.4011, + "step": 14320 + }, + { + "epoch": 0.7171274865507319, + "grad_norm": 0.1314251869916916, + "learning_rate": 4.227539804637541e-05, + "loss": 0.557, + "step": 14330 + }, + { + "epoch": 0.7176279244338797, + "grad_norm": 0.020194852724671364, + "learning_rate": 4.226613037756483e-05, + "loss": 0.3643, + "step": 14340 + }, + { + "epoch": 0.7181283623170274, + "grad_norm": 15.923261642456055, + "learning_rate": 4.2256862708754244e-05, + "loss": 0.7754, + "step": 14350 + }, + { + "epoch": 0.7186288002001752, + "grad_norm": 20.176340103149414, + "learning_rate": 4.224759503994365e-05, + "loss": 0.2085, + "step": 14360 + }, + { + "epoch": 0.7191292380833229, + "grad_norm": 11.479991912841797, + "learning_rate": 4.223832737113307e-05, + "loss": 0.4613, + "step": 14370 + }, + { + "epoch": 0.7196296759664706, + "grad_norm": 131.46539306640625, + "learning_rate": 4.2229059702322484e-05, + "loss": 0.386, + "step": 14380 + }, + { + "epoch": 0.7201301138496184, + "grad_norm": 44.794193267822266, + "learning_rate": 4.221979203351189e-05, + "loss": 0.6129, + "step": 14390 + }, + { + "epoch": 0.7206305517327661, + "grad_norm": 82.8419189453125, + "learning_rate": 4.22105243647013e-05, + "loss": 0.4174, + "step": 14400 + }, + { + "epoch": 0.7211309896159139, + "grad_norm": 0.0060513997450470924, + "learning_rate": 4.220125669589072e-05, + "loss": 0.4292, + "step": 14410 + }, + { + "epoch": 0.7216314274990617, + "grad_norm": 0.002232886152341962, + "learning_rate": 4.219198902708013e-05, + "loss": 0.6089, + "step": 14420 + }, + { + "epoch": 0.7221318653822094, + "grad_norm": 45.40835189819336, + "learning_rate": 4.218272135826954e-05, + "loss": 0.6882, + "step": 14430 + }, + { + "epoch": 0.7226323032653572, + "grad_norm": 2.1850554943084717, + "learning_rate": 4.217345368945896e-05, + "loss": 1.0119, + "step": 14440 + }, + { + "epoch": 0.7231327411485049, + "grad_norm": 48.842628479003906, + "learning_rate": 4.2164186020648366e-05, + "loss": 0.586, + "step": 14450 + }, + { + "epoch": 0.7236331790316527, + "grad_norm": 0.07250064611434937, + "learning_rate": 4.215491835183778e-05, + "loss": 0.4624, + "step": 14460 + }, + { + "epoch": 0.7241336169148005, + "grad_norm": 2.3354506492614746, + "learning_rate": 4.214565068302719e-05, + "loss": 0.1817, + "step": 14470 + }, + { + "epoch": 0.7246340547979482, + "grad_norm": 13.649331092834473, + "learning_rate": 4.2136383014216607e-05, + "loss": 0.5256, + "step": 14480 + }, + { + "epoch": 0.725134492681096, + "grad_norm": 39.29145812988281, + "learning_rate": 4.212711534540602e-05, + "loss": 0.6828, + "step": 14490 + }, + { + "epoch": 0.7256349305642437, + "grad_norm": 2.8405909538269043, + "learning_rate": 4.211784767659543e-05, + "loss": 0.3051, + "step": 14500 + }, + { + "epoch": 0.7261353684473915, + "grad_norm": 39.68887710571289, + "learning_rate": 4.210858000778484e-05, + "loss": 0.2897, + "step": 14510 + }, + { + "epoch": 0.7266358063305393, + "grad_norm": 0.6870024800300598, + "learning_rate": 4.2099312338974256e-05, + "loss": 0.5268, + "step": 14520 + }, + { + "epoch": 0.7271362442136869, + "grad_norm": 8.980630874633789, + "learning_rate": 4.209004467016367e-05, + "loss": 0.0889, + "step": 14530 + }, + { + "epoch": 0.7276366820968347, + "grad_norm": 4.673574447631836, + "learning_rate": 4.208077700135308e-05, + "loss": 0.325, + "step": 14540 + }, + { + "epoch": 0.7281371199799825, + "grad_norm": 5.025202751159668, + "learning_rate": 4.2071509332542496e-05, + "loss": 0.5751, + "step": 14550 + }, + { + "epoch": 0.7286375578631302, + "grad_norm": 1.0925787687301636, + "learning_rate": 4.2062241663731905e-05, + "loss": 0.3826, + "step": 14560 + }, + { + "epoch": 0.729137995746278, + "grad_norm": 1.6476620435714722, + "learning_rate": 4.205297399492132e-05, + "loss": 0.1881, + "step": 14570 + }, + { + "epoch": 0.7296384336294257, + "grad_norm": 6.3111796379089355, + "learning_rate": 4.204370632611073e-05, + "loss": 0.4017, + "step": 14580 + }, + { + "epoch": 0.7301388715125735, + "grad_norm": 5.512221813201904, + "learning_rate": 4.2034438657300145e-05, + "loss": 0.2436, + "step": 14590 + }, + { + "epoch": 0.7306393093957213, + "grad_norm": 61.3672981262207, + "learning_rate": 4.202517098848956e-05, + "loss": 0.4064, + "step": 14600 + }, + { + "epoch": 0.731139747278869, + "grad_norm": 1.328133225440979, + "learning_rate": 4.201590331967897e-05, + "loss": 0.3197, + "step": 14610 + }, + { + "epoch": 0.7316401851620168, + "grad_norm": 73.65680694580078, + "learning_rate": 4.200663565086838e-05, + "loss": 0.569, + "step": 14620 + }, + { + "epoch": 0.7321406230451645, + "grad_norm": 34.0671272277832, + "learning_rate": 4.1997367982057794e-05, + "loss": 0.5851, + "step": 14630 + }, + { + "epoch": 0.7326410609283123, + "grad_norm": 6.032987117767334, + "learning_rate": 4.198810031324721e-05, + "loss": 0.6055, + "step": 14640 + }, + { + "epoch": 0.7331414988114601, + "grad_norm": 74.66957092285156, + "learning_rate": 4.1978832644436625e-05, + "loss": 0.6359, + "step": 14650 + }, + { + "epoch": 0.7336419366946078, + "grad_norm": 12.543702125549316, + "learning_rate": 4.1969564975626034e-05, + "loss": 0.545, + "step": 14660 + }, + { + "epoch": 0.7341423745777556, + "grad_norm": 0.1353859156370163, + "learning_rate": 4.196029730681544e-05, + "loss": 0.4052, + "step": 14670 + }, + { + "epoch": 0.7346428124609032, + "grad_norm": 6.267245769500732, + "learning_rate": 4.195102963800486e-05, + "loss": 0.2702, + "step": 14680 + }, + { + "epoch": 0.735143250344051, + "grad_norm": 26.58687400817871, + "learning_rate": 4.1941761969194274e-05, + "loss": 0.3411, + "step": 14690 + }, + { + "epoch": 0.7356436882271988, + "grad_norm": 4.332326889038086, + "learning_rate": 4.193249430038368e-05, + "loss": 0.576, + "step": 14700 + }, + { + "epoch": 0.7361441261103465, + "grad_norm": 50.26930236816406, + "learning_rate": 4.19232266315731e-05, + "loss": 0.5838, + "step": 14710 + }, + { + "epoch": 0.7366445639934943, + "grad_norm": 54.90720748901367, + "learning_rate": 4.191395896276251e-05, + "loss": 0.4773, + "step": 14720 + }, + { + "epoch": 0.7371450018766421, + "grad_norm": 0.39147645235061646, + "learning_rate": 4.190469129395192e-05, + "loss": 0.3676, + "step": 14730 + }, + { + "epoch": 0.7376454397597898, + "grad_norm": 31.429611206054688, + "learning_rate": 4.189542362514133e-05, + "loss": 0.2738, + "step": 14740 + }, + { + "epoch": 0.7381458776429376, + "grad_norm": 45.17353820800781, + "learning_rate": 4.188615595633075e-05, + "loss": 0.5534, + "step": 14750 + }, + { + "epoch": 0.7386463155260853, + "grad_norm": 24.499059677124023, + "learning_rate": 4.187688828752016e-05, + "loss": 0.4139, + "step": 14760 + }, + { + "epoch": 0.7391467534092331, + "grad_norm": 67.31401824951172, + "learning_rate": 4.186762061870957e-05, + "loss": 0.4541, + "step": 14770 + }, + { + "epoch": 0.7396471912923809, + "grad_norm": 38.11884307861328, + "learning_rate": 4.185835294989898e-05, + "loss": 0.6176, + "step": 14780 + }, + { + "epoch": 0.7401476291755286, + "grad_norm": 24.02495002746582, + "learning_rate": 4.1849085281088396e-05, + "loss": 0.4559, + "step": 14790 + }, + { + "epoch": 0.7406480670586764, + "grad_norm": 48.03171157836914, + "learning_rate": 4.183981761227781e-05, + "loss": 0.8179, + "step": 14800 + }, + { + "epoch": 0.741148504941824, + "grad_norm": 0.0355466790497303, + "learning_rate": 4.183054994346722e-05, + "loss": 0.3762, + "step": 14810 + }, + { + "epoch": 0.7416489428249718, + "grad_norm": 12.92244815826416, + "learning_rate": 4.1821282274656636e-05, + "loss": 0.0785, + "step": 14820 + }, + { + "epoch": 0.7421493807081196, + "grad_norm": 98.0162582397461, + "learning_rate": 4.1812014605846045e-05, + "loss": 0.59, + "step": 14830 + }, + { + "epoch": 0.7426498185912673, + "grad_norm": 16.032211303710938, + "learning_rate": 4.180274693703546e-05, + "loss": 0.3197, + "step": 14840 + }, + { + "epoch": 0.7431502564744151, + "grad_norm": 0.03126853331923485, + "learning_rate": 4.179347926822487e-05, + "loss": 0.583, + "step": 14850 + }, + { + "epoch": 0.7436506943575628, + "grad_norm": 107.36351013183594, + "learning_rate": 4.1784211599414285e-05, + "loss": 0.4015, + "step": 14860 + }, + { + "epoch": 0.7441511322407106, + "grad_norm": 14.245692253112793, + "learning_rate": 4.17749439306037e-05, + "loss": 0.5054, + "step": 14870 + }, + { + "epoch": 0.7446515701238584, + "grad_norm": 21.451507568359375, + "learning_rate": 4.176567626179311e-05, + "loss": 0.3773, + "step": 14880 + }, + { + "epoch": 0.7451520080070061, + "grad_norm": 21.6558895111084, + "learning_rate": 4.175640859298252e-05, + "loss": 0.3123, + "step": 14890 + }, + { + "epoch": 0.7456524458901539, + "grad_norm": 1.08310067653656, + "learning_rate": 4.1747140924171934e-05, + "loss": 0.5505, + "step": 14900 + }, + { + "epoch": 0.7461528837733017, + "grad_norm": 0.5352813601493835, + "learning_rate": 4.173787325536135e-05, + "loss": 0.3621, + "step": 14910 + }, + { + "epoch": 0.7466533216564494, + "grad_norm": 32.10720443725586, + "learning_rate": 4.172860558655076e-05, + "loss": 0.183, + "step": 14920 + }, + { + "epoch": 0.7471537595395972, + "grad_norm": 122.14913940429688, + "learning_rate": 4.1719337917740175e-05, + "loss": 0.1711, + "step": 14930 + }, + { + "epoch": 0.7476541974227449, + "grad_norm": 0.12918618321418762, + "learning_rate": 4.1710070248929583e-05, + "loss": 0.7808, + "step": 14940 + }, + { + "epoch": 0.7481546353058927, + "grad_norm": 71.5500259399414, + "learning_rate": 4.1700802580119e-05, + "loss": 0.3173, + "step": 14950 + }, + { + "epoch": 0.7486550731890405, + "grad_norm": 33.110633850097656, + "learning_rate": 4.169153491130841e-05, + "loss": 0.8937, + "step": 14960 + }, + { + "epoch": 0.7491555110721881, + "grad_norm": 0.04187580943107605, + "learning_rate": 4.1682267242497824e-05, + "loss": 0.2847, + "step": 14970 + }, + { + "epoch": 0.7496559489553359, + "grad_norm": 0.05794061720371246, + "learning_rate": 4.167299957368724e-05, + "loss": 0.2313, + "step": 14980 + }, + { + "epoch": 0.7501563868384836, + "grad_norm": 17.874805450439453, + "learning_rate": 4.1663731904876655e-05, + "loss": 0.4084, + "step": 14990 + }, + { + "epoch": 0.7506568247216314, + "grad_norm": 98.95409393310547, + "learning_rate": 4.165446423606606e-05, + "loss": 0.9205, + "step": 15000 + }, + { + "epoch": 0.7511572626047792, + "grad_norm": 54.174774169921875, + "learning_rate": 4.164519656725547e-05, + "loss": 0.7778, + "step": 15010 + }, + { + "epoch": 0.7516577004879269, + "grad_norm": 22.058998107910156, + "learning_rate": 4.163592889844489e-05, + "loss": 0.3533, + "step": 15020 + }, + { + "epoch": 0.7521581383710747, + "grad_norm": 27.155378341674805, + "learning_rate": 4.1626661229634304e-05, + "loss": 0.4926, + "step": 15030 + }, + { + "epoch": 0.7526585762542225, + "grad_norm": 35.10145568847656, + "learning_rate": 4.161739356082371e-05, + "loss": 0.3909, + "step": 15040 + }, + { + "epoch": 0.7531590141373702, + "grad_norm": 31.568357467651367, + "learning_rate": 4.160812589201312e-05, + "loss": 0.4653, + "step": 15050 + }, + { + "epoch": 0.753659452020518, + "grad_norm": 31.466777801513672, + "learning_rate": 4.159885822320254e-05, + "loss": 0.483, + "step": 15060 + }, + { + "epoch": 0.7541598899036657, + "grad_norm": 29.819765090942383, + "learning_rate": 4.158959055439195e-05, + "loss": 0.6868, + "step": 15070 + }, + { + "epoch": 0.7546603277868135, + "grad_norm": 3.0357441902160645, + "learning_rate": 4.158032288558136e-05, + "loss": 0.5949, + "step": 15080 + }, + { + "epoch": 0.7551607656699613, + "grad_norm": 42.93642807006836, + "learning_rate": 4.157105521677078e-05, + "loss": 0.2277, + "step": 15090 + }, + { + "epoch": 0.755661203553109, + "grad_norm": 78.4280014038086, + "learning_rate": 4.156178754796019e-05, + "loss": 0.1924, + "step": 15100 + }, + { + "epoch": 0.7561616414362567, + "grad_norm": 9.002673149108887, + "learning_rate": 4.15525198791496e-05, + "loss": 0.3718, + "step": 15110 + }, + { + "epoch": 0.7566620793194044, + "grad_norm": 2.1876254081726074, + "learning_rate": 4.154325221033901e-05, + "loss": 0.5641, + "step": 15120 + }, + { + "epoch": 0.7571625172025522, + "grad_norm": 12.456345558166504, + "learning_rate": 4.1533984541528426e-05, + "loss": 0.6522, + "step": 15130 + }, + { + "epoch": 0.7576629550857, + "grad_norm": 37.57898712158203, + "learning_rate": 4.152471687271784e-05, + "loss": 0.3494, + "step": 15140 + }, + { + "epoch": 0.7581633929688477, + "grad_norm": 0.5105221271514893, + "learning_rate": 4.151544920390725e-05, + "loss": 0.5435, + "step": 15150 + }, + { + "epoch": 0.7586638308519955, + "grad_norm": 84.68693542480469, + "learning_rate": 4.1506181535096666e-05, + "loss": 0.5143, + "step": 15160 + }, + { + "epoch": 0.7591642687351432, + "grad_norm": 62.81201934814453, + "learning_rate": 4.1496913866286075e-05, + "loss": 0.4901, + "step": 15170 + }, + { + "epoch": 0.759664706618291, + "grad_norm": 15.753273010253906, + "learning_rate": 4.148764619747549e-05, + "loss": 0.4036, + "step": 15180 + }, + { + "epoch": 0.7601651445014388, + "grad_norm": 0.15633216500282288, + "learning_rate": 4.14783785286649e-05, + "loss": 0.3403, + "step": 15190 + }, + { + "epoch": 0.7606655823845865, + "grad_norm": 57.0444221496582, + "learning_rate": 4.1469110859854315e-05, + "loss": 0.3828, + "step": 15200 + }, + { + "epoch": 0.7611660202677343, + "grad_norm": 63.26517105102539, + "learning_rate": 4.145984319104373e-05, + "loss": 0.7876, + "step": 15210 + }, + { + "epoch": 0.7616664581508821, + "grad_norm": 49.08970260620117, + "learning_rate": 4.145057552223314e-05, + "loss": 0.5335, + "step": 15220 + }, + { + "epoch": 0.7621668960340298, + "grad_norm": 9.076170921325684, + "learning_rate": 4.144130785342255e-05, + "loss": 0.2028, + "step": 15230 + }, + { + "epoch": 0.7626673339171776, + "grad_norm": 74.81166076660156, + "learning_rate": 4.1432040184611964e-05, + "loss": 0.5949, + "step": 15240 + }, + { + "epoch": 0.7631677718003252, + "grad_norm": 0.02976149506866932, + "learning_rate": 4.142277251580138e-05, + "loss": 0.344, + "step": 15250 + }, + { + "epoch": 0.763668209683473, + "grad_norm": 0.22763121128082275, + "learning_rate": 4.141350484699079e-05, + "loss": 0.318, + "step": 15260 + }, + { + "epoch": 0.7641686475666208, + "grad_norm": 0.13137078285217285, + "learning_rate": 4.1404237178180204e-05, + "loss": 0.6001, + "step": 15270 + }, + { + "epoch": 0.7646690854497685, + "grad_norm": 0.011728361248970032, + "learning_rate": 4.139496950936961e-05, + "loss": 0.3443, + "step": 15280 + }, + { + "epoch": 0.7651695233329163, + "grad_norm": 2.6583058834075928, + "learning_rate": 4.138570184055903e-05, + "loss": 0.4105, + "step": 15290 + }, + { + "epoch": 0.765669961216064, + "grad_norm": 0.5249181389808655, + "learning_rate": 4.137643417174844e-05, + "loss": 1.0404, + "step": 15300 + }, + { + "epoch": 0.7661703990992118, + "grad_norm": 6.054710865020752, + "learning_rate": 4.1367166502937853e-05, + "loss": 0.3373, + "step": 15310 + }, + { + "epoch": 0.7666708369823596, + "grad_norm": 22.62129020690918, + "learning_rate": 4.135789883412727e-05, + "loss": 0.4001, + "step": 15320 + }, + { + "epoch": 0.7671712748655073, + "grad_norm": 1.5144822597503662, + "learning_rate": 4.134863116531668e-05, + "loss": 0.2552, + "step": 15330 + }, + { + "epoch": 0.7676717127486551, + "grad_norm": 0.19659876823425293, + "learning_rate": 4.133936349650609e-05, + "loss": 0.2802, + "step": 15340 + }, + { + "epoch": 0.7681721506318028, + "grad_norm": 16.93588638305664, + "learning_rate": 4.13300958276955e-05, + "loss": 0.8272, + "step": 15350 + }, + { + "epoch": 0.7686725885149506, + "grad_norm": 14.614128112792969, + "learning_rate": 4.132082815888492e-05, + "loss": 0.5003, + "step": 15360 + }, + { + "epoch": 0.7691730263980984, + "grad_norm": 3.054201602935791, + "learning_rate": 4.131156049007433e-05, + "loss": 0.3197, + "step": 15370 + }, + { + "epoch": 0.7696734642812461, + "grad_norm": 2.0347373485565186, + "learning_rate": 4.130229282126374e-05, + "loss": 0.377, + "step": 15380 + }, + { + "epoch": 0.7701739021643939, + "grad_norm": 35.024375915527344, + "learning_rate": 4.129302515245315e-05, + "loss": 0.7072, + "step": 15390 + }, + { + "epoch": 0.7706743400475417, + "grad_norm": 17.248470306396484, + "learning_rate": 4.128375748364257e-05, + "loss": 0.6692, + "step": 15400 + }, + { + "epoch": 0.7711747779306893, + "grad_norm": 5.745948791503906, + "learning_rate": 4.1274489814831976e-05, + "loss": 0.2145, + "step": 15410 + }, + { + "epoch": 0.7716752158138371, + "grad_norm": 23.136810302734375, + "learning_rate": 4.126522214602139e-05, + "loss": 0.2649, + "step": 15420 + }, + { + "epoch": 0.7721756536969848, + "grad_norm": 16.271100997924805, + "learning_rate": 4.125595447721081e-05, + "loss": 0.6179, + "step": 15430 + }, + { + "epoch": 0.7726760915801326, + "grad_norm": 0.43141525983810425, + "learning_rate": 4.1246686808400216e-05, + "loss": 0.461, + "step": 15440 + }, + { + "epoch": 0.7731765294632804, + "grad_norm": 47.33087158203125, + "learning_rate": 4.1237419139589625e-05, + "loss": 0.3035, + "step": 15450 + }, + { + "epoch": 0.7736769673464281, + "grad_norm": 8.36418342590332, + "learning_rate": 4.122815147077904e-05, + "loss": 0.4938, + "step": 15460 + }, + { + "epoch": 0.7741774052295759, + "grad_norm": 74.92501068115234, + "learning_rate": 4.1218883801968456e-05, + "loss": 0.7311, + "step": 15470 + }, + { + "epoch": 0.7746778431127236, + "grad_norm": 25.838459014892578, + "learning_rate": 4.120961613315787e-05, + "loss": 0.4545, + "step": 15480 + }, + { + "epoch": 0.7751782809958714, + "grad_norm": 0.8339977264404297, + "learning_rate": 4.120034846434728e-05, + "loss": 0.4056, + "step": 15490 + }, + { + "epoch": 0.7756787188790192, + "grad_norm": 48.9595832824707, + "learning_rate": 4.119108079553669e-05, + "loss": 0.3536, + "step": 15500 + }, + { + "epoch": 0.7761791567621669, + "grad_norm": 119.81086730957031, + "learning_rate": 4.1181813126726105e-05, + "loss": 0.372, + "step": 15510 + }, + { + "epoch": 0.7766795946453147, + "grad_norm": 25.493316650390625, + "learning_rate": 4.117254545791552e-05, + "loss": 0.3698, + "step": 15520 + }, + { + "epoch": 0.7771800325284624, + "grad_norm": 34.55347442626953, + "learning_rate": 4.116327778910493e-05, + "loss": 0.3666, + "step": 15530 + }, + { + "epoch": 0.7776804704116101, + "grad_norm": 73.55915832519531, + "learning_rate": 4.1154010120294345e-05, + "loss": 0.8322, + "step": 15540 + }, + { + "epoch": 0.778180908294758, + "grad_norm": 0.23045004904270172, + "learning_rate": 4.1144742451483754e-05, + "loss": 0.3956, + "step": 15550 + }, + { + "epoch": 0.7786813461779056, + "grad_norm": 0.6052928566932678, + "learning_rate": 4.113547478267317e-05, + "loss": 0.1253, + "step": 15560 + }, + { + "epoch": 0.7791817840610534, + "grad_norm": 37.25783920288086, + "learning_rate": 4.112620711386258e-05, + "loss": 0.3987, + "step": 15570 + }, + { + "epoch": 0.7796822219442012, + "grad_norm": 124.50150299072266, + "learning_rate": 4.1116939445051994e-05, + "loss": 0.5589, + "step": 15580 + }, + { + "epoch": 0.7801826598273489, + "grad_norm": 34.30792999267578, + "learning_rate": 4.110767177624141e-05, + "loss": 0.6595, + "step": 15590 + }, + { + "epoch": 0.7806830977104967, + "grad_norm": 49.536399841308594, + "learning_rate": 4.109840410743082e-05, + "loss": 0.3932, + "step": 15600 + }, + { + "epoch": 0.7811835355936444, + "grad_norm": 6.085203647613525, + "learning_rate": 4.108913643862023e-05, + "loss": 0.5903, + "step": 15610 + }, + { + "epoch": 0.7816839734767922, + "grad_norm": 19.461502075195312, + "learning_rate": 4.107986876980964e-05, + "loss": 0.2979, + "step": 15620 + }, + { + "epoch": 0.78218441135994, + "grad_norm": 5.721696376800537, + "learning_rate": 4.107060110099906e-05, + "loss": 0.2616, + "step": 15630 + }, + { + "epoch": 0.7826848492430877, + "grad_norm": 0.0460951067507267, + "learning_rate": 4.106133343218847e-05, + "loss": 0.3845, + "step": 15640 + }, + { + "epoch": 0.7831852871262355, + "grad_norm": 24.739538192749023, + "learning_rate": 4.105206576337788e-05, + "loss": 0.5664, + "step": 15650 + }, + { + "epoch": 0.7836857250093832, + "grad_norm": 40.797977447509766, + "learning_rate": 4.104279809456729e-05, + "loss": 0.2407, + "step": 15660 + }, + { + "epoch": 0.784186162892531, + "grad_norm": 0.25256386399269104, + "learning_rate": 4.103353042575671e-05, + "loss": 0.2319, + "step": 15670 + }, + { + "epoch": 0.7846866007756788, + "grad_norm": 42.876338958740234, + "learning_rate": 4.102426275694612e-05, + "loss": 0.5183, + "step": 15680 + }, + { + "epoch": 0.7851870386588264, + "grad_norm": 0.010751540772616863, + "learning_rate": 4.101499508813553e-05, + "loss": 0.7242, + "step": 15690 + }, + { + "epoch": 0.7856874765419742, + "grad_norm": 6.580509662628174, + "learning_rate": 4.100572741932495e-05, + "loss": 0.4119, + "step": 15700 + }, + { + "epoch": 0.786187914425122, + "grad_norm": 0.03643311932682991, + "learning_rate": 4.099645975051436e-05, + "loss": 0.09, + "step": 15710 + }, + { + "epoch": 0.7866883523082697, + "grad_norm": 3.0909557342529297, + "learning_rate": 4.0987192081703766e-05, + "loss": 0.2834, + "step": 15720 + }, + { + "epoch": 0.7871887901914175, + "grad_norm": 19.729032516479492, + "learning_rate": 4.097792441289318e-05, + "loss": 0.7147, + "step": 15730 + }, + { + "epoch": 0.7876892280745652, + "grad_norm": 58.28226089477539, + "learning_rate": 4.09686567440826e-05, + "loss": 0.9377, + "step": 15740 + }, + { + "epoch": 0.788189665957713, + "grad_norm": 26.73775863647461, + "learning_rate": 4.0959389075272006e-05, + "loss": 0.2882, + "step": 15750 + }, + { + "epoch": 0.7886901038408608, + "grad_norm": 29.04587745666504, + "learning_rate": 4.095012140646142e-05, + "loss": 0.4241, + "step": 15760 + }, + { + "epoch": 0.7891905417240085, + "grad_norm": 3.5517332553863525, + "learning_rate": 4.094085373765083e-05, + "loss": 0.4496, + "step": 15770 + }, + { + "epoch": 0.7896909796071563, + "grad_norm": 25.60331916809082, + "learning_rate": 4.0931586068840246e-05, + "loss": 0.5235, + "step": 15780 + }, + { + "epoch": 0.790191417490304, + "grad_norm": 8.418464660644531, + "learning_rate": 4.0922318400029655e-05, + "loss": 0.3666, + "step": 15790 + }, + { + "epoch": 0.7906918553734518, + "grad_norm": 25.79783821105957, + "learning_rate": 4.091305073121907e-05, + "loss": 0.2111, + "step": 15800 + }, + { + "epoch": 0.7911922932565996, + "grad_norm": 11.130593299865723, + "learning_rate": 4.0903783062408486e-05, + "loss": 0.1688, + "step": 15810 + }, + { + "epoch": 0.7916927311397473, + "grad_norm": 91.39788818359375, + "learning_rate": 4.08945153935979e-05, + "loss": 0.2991, + "step": 15820 + }, + { + "epoch": 0.792193169022895, + "grad_norm": 147.967529296875, + "learning_rate": 4.0885247724787304e-05, + "loss": 0.9472, + "step": 15830 + }, + { + "epoch": 0.7926936069060427, + "grad_norm": 30.3603515625, + "learning_rate": 4.087598005597672e-05, + "loss": 0.6892, + "step": 15840 + }, + { + "epoch": 0.7931940447891905, + "grad_norm": 64.333984375, + "learning_rate": 4.0866712387166135e-05, + "loss": 0.8698, + "step": 15850 + }, + { + "epoch": 0.7936944826723383, + "grad_norm": 30.5130558013916, + "learning_rate": 4.085744471835555e-05, + "loss": 0.4762, + "step": 15860 + }, + { + "epoch": 0.794194920555486, + "grad_norm": 4.171695232391357, + "learning_rate": 4.084817704954496e-05, + "loss": 0.35, + "step": 15870 + }, + { + "epoch": 0.7946953584386338, + "grad_norm": 4.42927885055542, + "learning_rate": 4.0838909380734375e-05, + "loss": 0.3771, + "step": 15880 + }, + { + "epoch": 0.7951957963217816, + "grad_norm": 2.1251113414764404, + "learning_rate": 4.0829641711923784e-05, + "loss": 0.4138, + "step": 15890 + }, + { + "epoch": 0.7956962342049293, + "grad_norm": 5.857220649719238, + "learning_rate": 4.08203740431132e-05, + "loss": 0.2993, + "step": 15900 + }, + { + "epoch": 0.7961966720880771, + "grad_norm": 8.221087455749512, + "learning_rate": 4.081110637430261e-05, + "loss": 0.4182, + "step": 15910 + }, + { + "epoch": 0.7966971099712248, + "grad_norm": 8.024846076965332, + "learning_rate": 4.0801838705492024e-05, + "loss": 0.5452, + "step": 15920 + }, + { + "epoch": 0.7971975478543726, + "grad_norm": 4.252470970153809, + "learning_rate": 4.079257103668144e-05, + "loss": 0.2246, + "step": 15930 + }, + { + "epoch": 0.7976979857375204, + "grad_norm": 4.388980388641357, + "learning_rate": 4.078330336787085e-05, + "loss": 0.5646, + "step": 15940 + }, + { + "epoch": 0.7981984236206681, + "grad_norm": 61.452964782714844, + "learning_rate": 4.077403569906026e-05, + "loss": 0.3639, + "step": 15950 + }, + { + "epoch": 0.7986988615038159, + "grad_norm": 0.36057522892951965, + "learning_rate": 4.076476803024967e-05, + "loss": 0.3753, + "step": 15960 + }, + { + "epoch": 0.7991992993869635, + "grad_norm": 53.92320251464844, + "learning_rate": 4.075550036143909e-05, + "loss": 0.6001, + "step": 15970 + }, + { + "epoch": 0.7996997372701113, + "grad_norm": 54.54835510253906, + "learning_rate": 4.07462326926285e-05, + "loss": 0.1388, + "step": 15980 + }, + { + "epoch": 0.8002001751532591, + "grad_norm": 42.236427307128906, + "learning_rate": 4.073696502381791e-05, + "loss": 0.4091, + "step": 15990 + }, + { + "epoch": 0.8007006130364068, + "grad_norm": 33.7714958190918, + "learning_rate": 4.072769735500732e-05, + "loss": 0.5456, + "step": 16000 + }, + { + "epoch": 0.8012010509195546, + "grad_norm": 39.271522521972656, + "learning_rate": 4.071842968619674e-05, + "loss": 0.3631, + "step": 16010 + }, + { + "epoch": 0.8017014888027023, + "grad_norm": 52.1532096862793, + "learning_rate": 4.070916201738615e-05, + "loss": 0.5807, + "step": 16020 + }, + { + "epoch": 0.8022019266858501, + "grad_norm": 13.527261734008789, + "learning_rate": 4.069989434857556e-05, + "loss": 0.6006, + "step": 16030 + }, + { + "epoch": 0.8027023645689979, + "grad_norm": 82.89714050292969, + "learning_rate": 4.069062667976498e-05, + "loss": 0.7233, + "step": 16040 + }, + { + "epoch": 0.8032028024521456, + "grad_norm": 4.452935695648193, + "learning_rate": 4.068135901095439e-05, + "loss": 0.2688, + "step": 16050 + }, + { + "epoch": 0.8037032403352934, + "grad_norm": 16.987585067749023, + "learning_rate": 4.0672091342143796e-05, + "loss": 0.7171, + "step": 16060 + }, + { + "epoch": 0.8042036782184412, + "grad_norm": 1.365400791168213, + "learning_rate": 4.066282367333321e-05, + "loss": 0.3058, + "step": 16070 + }, + { + "epoch": 0.8047041161015889, + "grad_norm": 1.2366137504577637, + "learning_rate": 4.065355600452263e-05, + "loss": 0.5155, + "step": 16080 + }, + { + "epoch": 0.8052045539847367, + "grad_norm": 6.112344741821289, + "learning_rate": 4.0644288335712036e-05, + "loss": 0.5684, + "step": 16090 + }, + { + "epoch": 0.8057049918678844, + "grad_norm": 14.379620552062988, + "learning_rate": 4.063502066690145e-05, + "loss": 0.834, + "step": 16100 + }, + { + "epoch": 0.8062054297510322, + "grad_norm": 6.1278815269470215, + "learning_rate": 4.062575299809086e-05, + "loss": 0.1963, + "step": 16110 + }, + { + "epoch": 0.80670586763418, + "grad_norm": 7.545967102050781, + "learning_rate": 4.0616485329280276e-05, + "loss": 0.5122, + "step": 16120 + }, + { + "epoch": 0.8072063055173276, + "grad_norm": 29.74075698852539, + "learning_rate": 4.0607217660469685e-05, + "loss": 0.3699, + "step": 16130 + }, + { + "epoch": 0.8077067434004754, + "grad_norm": 4.306427955627441, + "learning_rate": 4.05979499916591e-05, + "loss": 0.3462, + "step": 16140 + }, + { + "epoch": 0.8082071812836231, + "grad_norm": 13.021490097045898, + "learning_rate": 4.0588682322848516e-05, + "loss": 0.2424, + "step": 16150 + }, + { + "epoch": 0.8087076191667709, + "grad_norm": 22.571537017822266, + "learning_rate": 4.0579414654037925e-05, + "loss": 0.5641, + "step": 16160 + }, + { + "epoch": 0.8092080570499187, + "grad_norm": 9.754956245422363, + "learning_rate": 4.0570146985227334e-05, + "loss": 0.4244, + "step": 16170 + }, + { + "epoch": 0.8097084949330664, + "grad_norm": 0.3316529393196106, + "learning_rate": 4.056087931641675e-05, + "loss": 0.212, + "step": 16180 + }, + { + "epoch": 0.8102089328162142, + "grad_norm": 44.7596549987793, + "learning_rate": 4.0551611647606165e-05, + "loss": 0.4822, + "step": 16190 + }, + { + "epoch": 0.8107093706993619, + "grad_norm": 0.34877651929855347, + "learning_rate": 4.0542343978795574e-05, + "loss": 0.1365, + "step": 16200 + }, + { + "epoch": 0.8112098085825097, + "grad_norm": 10.845576286315918, + "learning_rate": 4.053307630998499e-05, + "loss": 0.5098, + "step": 16210 + }, + { + "epoch": 0.8117102464656575, + "grad_norm": 7.324076175689697, + "learning_rate": 4.05238086411744e-05, + "loss": 0.3946, + "step": 16220 + }, + { + "epoch": 0.8122106843488052, + "grad_norm": 2.2446508407592773, + "learning_rate": 4.0514540972363814e-05, + "loss": 0.636, + "step": 16230 + }, + { + "epoch": 0.812711122231953, + "grad_norm": 22.614198684692383, + "learning_rate": 4.050527330355322e-05, + "loss": 0.3859, + "step": 16240 + }, + { + "epoch": 0.8132115601151008, + "grad_norm": 0.802595317363739, + "learning_rate": 4.049600563474264e-05, + "loss": 0.3938, + "step": 16250 + }, + { + "epoch": 0.8137119979982484, + "grad_norm": 112.03102111816406, + "learning_rate": 4.0486737965932054e-05, + "loss": 0.3737, + "step": 16260 + }, + { + "epoch": 0.8142124358813962, + "grad_norm": 1.8384013175964355, + "learning_rate": 4.047747029712146e-05, + "loss": 0.317, + "step": 16270 + }, + { + "epoch": 0.8147128737645439, + "grad_norm": 79.17431640625, + "learning_rate": 4.046820262831087e-05, + "loss": 0.5787, + "step": 16280 + }, + { + "epoch": 0.8152133116476917, + "grad_norm": 0.27174243330955505, + "learning_rate": 4.045893495950029e-05, + "loss": 0.305, + "step": 16290 + }, + { + "epoch": 0.8157137495308395, + "grad_norm": 0.10730624943971634, + "learning_rate": 4.04496672906897e-05, + "loss": 0.2515, + "step": 16300 + }, + { + "epoch": 0.8162141874139872, + "grad_norm": 0.10946701467037201, + "learning_rate": 4.044039962187912e-05, + "loss": 0.6705, + "step": 16310 + }, + { + "epoch": 0.816714625297135, + "grad_norm": 0.04527418687939644, + "learning_rate": 4.043113195306853e-05, + "loss": 0.4904, + "step": 16320 + }, + { + "epoch": 0.8172150631802827, + "grad_norm": 36.844444274902344, + "learning_rate": 4.0421864284257936e-05, + "loss": 0.4927, + "step": 16330 + }, + { + "epoch": 0.8177155010634305, + "grad_norm": 32.35655212402344, + "learning_rate": 4.041259661544735e-05, + "loss": 0.3603, + "step": 16340 + }, + { + "epoch": 0.8182159389465783, + "grad_norm": 76.66121673583984, + "learning_rate": 4.040332894663677e-05, + "loss": 0.2638, + "step": 16350 + }, + { + "epoch": 0.818716376829726, + "grad_norm": 41.53892135620117, + "learning_rate": 4.0394061277826177e-05, + "loss": 0.2964, + "step": 16360 + }, + { + "epoch": 0.8192168147128738, + "grad_norm": 39.99506759643555, + "learning_rate": 4.038479360901559e-05, + "loss": 0.4689, + "step": 16370 + }, + { + "epoch": 0.8197172525960216, + "grad_norm": 1.552685022354126, + "learning_rate": 4.0375525940205e-05, + "loss": 0.3583, + "step": 16380 + }, + { + "epoch": 0.8202176904791693, + "grad_norm": 58.759605407714844, + "learning_rate": 4.036625827139442e-05, + "loss": 0.536, + "step": 16390 + }, + { + "epoch": 0.8207181283623171, + "grad_norm": 14.041313171386719, + "learning_rate": 4.0356990602583826e-05, + "loss": 0.4688, + "step": 16400 + }, + { + "epoch": 0.8212185662454647, + "grad_norm": 24.080554962158203, + "learning_rate": 4.034772293377324e-05, + "loss": 0.5649, + "step": 16410 + }, + { + "epoch": 0.8217190041286125, + "grad_norm": 2.166583299636841, + "learning_rate": 4.033845526496266e-05, + "loss": 0.4258, + "step": 16420 + }, + { + "epoch": 0.8222194420117603, + "grad_norm": 80.19686889648438, + "learning_rate": 4.0329187596152066e-05, + "loss": 1.115, + "step": 16430 + }, + { + "epoch": 0.822719879894908, + "grad_norm": 4.17795467376709, + "learning_rate": 4.0319919927341475e-05, + "loss": 0.4507, + "step": 16440 + }, + { + "epoch": 0.8232203177780558, + "grad_norm": 49.90502166748047, + "learning_rate": 4.031065225853089e-05, + "loss": 0.2843, + "step": 16450 + }, + { + "epoch": 0.8237207556612035, + "grad_norm": 6.073411464691162, + "learning_rate": 4.0301384589720306e-05, + "loss": 0.1628, + "step": 16460 + }, + { + "epoch": 0.8242211935443513, + "grad_norm": 0.006511304527521133, + "learning_rate": 4.0292116920909715e-05, + "loss": 0.3652, + "step": 16470 + }, + { + "epoch": 0.8247216314274991, + "grad_norm": 70.20488739013672, + "learning_rate": 4.028284925209913e-05, + "loss": 0.3405, + "step": 16480 + }, + { + "epoch": 0.8252220693106468, + "grad_norm": 0.22489358484745026, + "learning_rate": 4.027358158328854e-05, + "loss": 0.0952, + "step": 16490 + }, + { + "epoch": 0.8257225071937946, + "grad_norm": 0.03136907517910004, + "learning_rate": 4.0264313914477955e-05, + "loss": 0.0279, + "step": 16500 + }, + { + "epoch": 0.8262229450769423, + "grad_norm": 0.05037777125835419, + "learning_rate": 4.0255046245667364e-05, + "loss": 0.8892, + "step": 16510 + }, + { + "epoch": 0.8267233829600901, + "grad_norm": 31.96965789794922, + "learning_rate": 4.024577857685678e-05, + "loss": 0.6025, + "step": 16520 + }, + { + "epoch": 0.8272238208432379, + "grad_norm": 0.006646528374403715, + "learning_rate": 4.0236510908046195e-05, + "loss": 0.5462, + "step": 16530 + }, + { + "epoch": 0.8277242587263856, + "grad_norm": 72.01219177246094, + "learning_rate": 4.0227243239235604e-05, + "loss": 0.7604, + "step": 16540 + }, + { + "epoch": 0.8282246966095334, + "grad_norm": 119.09395599365234, + "learning_rate": 4.021797557042501e-05, + "loss": 0.5864, + "step": 16550 + }, + { + "epoch": 0.8287251344926811, + "grad_norm": 32.7746696472168, + "learning_rate": 4.020870790161443e-05, + "loss": 0.3184, + "step": 16560 + }, + { + "epoch": 0.8292255723758288, + "grad_norm": 53.11061477661133, + "learning_rate": 4.0199440232803844e-05, + "loss": 0.2994, + "step": 16570 + }, + { + "epoch": 0.8297260102589766, + "grad_norm": 5.666276454925537, + "learning_rate": 4.019017256399325e-05, + "loss": 0.265, + "step": 16580 + }, + { + "epoch": 0.8302264481421243, + "grad_norm": 0.07334763556718826, + "learning_rate": 4.018090489518267e-05, + "loss": 0.8281, + "step": 16590 + }, + { + "epoch": 0.8307268860252721, + "grad_norm": 213.42994689941406, + "learning_rate": 4.0171637226372084e-05, + "loss": 0.6129, + "step": 16600 + }, + { + "epoch": 0.8312273239084199, + "grad_norm": 0.05698111280798912, + "learning_rate": 4.016236955756149e-05, + "loss": 0.5913, + "step": 16610 + }, + { + "epoch": 0.8317277617915676, + "grad_norm": 68.03977966308594, + "learning_rate": 4.01531018887509e-05, + "loss": 0.3158, + "step": 16620 + }, + { + "epoch": 0.8322281996747154, + "grad_norm": 30.89476776123047, + "learning_rate": 4.014383421994032e-05, + "loss": 0.3269, + "step": 16630 + }, + { + "epoch": 0.8327286375578631, + "grad_norm": 5.205947399139404, + "learning_rate": 4.013456655112973e-05, + "loss": 0.5902, + "step": 16640 + }, + { + "epoch": 0.8332290754410109, + "grad_norm": 2.4022557735443115, + "learning_rate": 4.012529888231915e-05, + "loss": 0.5862, + "step": 16650 + }, + { + "epoch": 0.8337295133241587, + "grad_norm": 6.589649200439453, + "learning_rate": 4.011603121350855e-05, + "loss": 0.502, + "step": 16660 + }, + { + "epoch": 0.8342299512073064, + "grad_norm": 163.0982666015625, + "learning_rate": 4.0106763544697966e-05, + "loss": 0.7026, + "step": 16670 + }, + { + "epoch": 0.8347303890904542, + "grad_norm": 0.04147826135158539, + "learning_rate": 4.009749587588738e-05, + "loss": 0.4872, + "step": 16680 + }, + { + "epoch": 0.8352308269736018, + "grad_norm": 43.18259811401367, + "learning_rate": 4.00882282070768e-05, + "loss": 0.4505, + "step": 16690 + }, + { + "epoch": 0.8357312648567496, + "grad_norm": 80.27116394042969, + "learning_rate": 4.0078960538266206e-05, + "loss": 1.05, + "step": 16700 + }, + { + "epoch": 0.8362317027398974, + "grad_norm": 108.24799346923828, + "learning_rate": 4.006969286945562e-05, + "loss": 0.4955, + "step": 16710 + }, + { + "epoch": 0.8367321406230451, + "grad_norm": 10.935001373291016, + "learning_rate": 4.006042520064503e-05, + "loss": 0.1752, + "step": 16720 + }, + { + "epoch": 0.8372325785061929, + "grad_norm": 0.1761341392993927, + "learning_rate": 4.0051157531834447e-05, + "loss": 0.5444, + "step": 16730 + }, + { + "epoch": 0.8377330163893407, + "grad_norm": 4.93062686920166, + "learning_rate": 4.0041889863023855e-05, + "loss": 0.1704, + "step": 16740 + }, + { + "epoch": 0.8382334542724884, + "grad_norm": 24.301029205322266, + "learning_rate": 4.003262219421327e-05, + "loss": 0.447, + "step": 16750 + }, + { + "epoch": 0.8387338921556362, + "grad_norm": 50.108070373535156, + "learning_rate": 4.002335452540269e-05, + "loss": 0.508, + "step": 16760 + }, + { + "epoch": 0.8392343300387839, + "grad_norm": 1.02400803565979, + "learning_rate": 4.0014086856592096e-05, + "loss": 0.4074, + "step": 16770 + }, + { + "epoch": 0.8397347679219317, + "grad_norm": 32.949642181396484, + "learning_rate": 4.0004819187781504e-05, + "loss": 0.474, + "step": 16780 + }, + { + "epoch": 0.8402352058050795, + "grad_norm": 20.837987899780273, + "learning_rate": 3.999555151897092e-05, + "loss": 0.3131, + "step": 16790 + }, + { + "epoch": 0.8407356436882272, + "grad_norm": 1.5355559587478638, + "learning_rate": 3.9986283850160336e-05, + "loss": 0.2585, + "step": 16800 + }, + { + "epoch": 0.841236081571375, + "grad_norm": 9.575864791870117, + "learning_rate": 3.9977016181349745e-05, + "loss": 0.5788, + "step": 16810 + }, + { + "epoch": 0.8417365194545227, + "grad_norm": 27.067668914794922, + "learning_rate": 3.996774851253916e-05, + "loss": 0.8927, + "step": 16820 + }, + { + "epoch": 0.8422369573376705, + "grad_norm": 1.85964035987854, + "learning_rate": 3.995848084372857e-05, + "loss": 0.755, + "step": 16830 + }, + { + "epoch": 0.8427373952208183, + "grad_norm": 1.2249373197555542, + "learning_rate": 3.9949213174917985e-05, + "loss": 0.4597, + "step": 16840 + }, + { + "epoch": 0.8432378331039659, + "grad_norm": 49.059932708740234, + "learning_rate": 3.9939945506107394e-05, + "loss": 0.1432, + "step": 16850 + }, + { + "epoch": 0.8437382709871137, + "grad_norm": 0.05352775752544403, + "learning_rate": 3.993067783729681e-05, + "loss": 0.7828, + "step": 16860 + }, + { + "epoch": 0.8442387088702615, + "grad_norm": 5.228312969207764, + "learning_rate": 3.9921410168486225e-05, + "loss": 0.3105, + "step": 16870 + }, + { + "epoch": 0.8447391467534092, + "grad_norm": 39.937007904052734, + "learning_rate": 3.9912142499675634e-05, + "loss": 0.6798, + "step": 16880 + }, + { + "epoch": 0.845239584636557, + "grad_norm": 10.808924674987793, + "learning_rate": 3.990287483086504e-05, + "loss": 0.3364, + "step": 16890 + }, + { + "epoch": 0.8457400225197047, + "grad_norm": 26.609899520874023, + "learning_rate": 3.989360716205446e-05, + "loss": 0.4655, + "step": 16900 + }, + { + "epoch": 0.8462404604028525, + "grad_norm": 5.371480941772461, + "learning_rate": 3.9884339493243874e-05, + "loss": 0.1543, + "step": 16910 + }, + { + "epoch": 0.8467408982860003, + "grad_norm": 0.6826133728027344, + "learning_rate": 3.987507182443328e-05, + "loss": 0.2697, + "step": 16920 + }, + { + "epoch": 0.847241336169148, + "grad_norm": 16.817947387695312, + "learning_rate": 3.98658041556227e-05, + "loss": 0.913, + "step": 16930 + }, + { + "epoch": 0.8477417740522958, + "grad_norm": 0.5213493704795837, + "learning_rate": 3.985653648681211e-05, + "loss": 0.4878, + "step": 16940 + }, + { + "epoch": 0.8482422119354435, + "grad_norm": 0.01887083426117897, + "learning_rate": 3.984726881800152e-05, + "loss": 0.4262, + "step": 16950 + }, + { + "epoch": 0.8487426498185913, + "grad_norm": 22.137306213378906, + "learning_rate": 3.983800114919093e-05, + "loss": 0.4176, + "step": 16960 + }, + { + "epoch": 0.8492430877017391, + "grad_norm": 9.621807098388672, + "learning_rate": 3.982873348038035e-05, + "loss": 0.3325, + "step": 16970 + }, + { + "epoch": 0.8497435255848867, + "grad_norm": 2.294828176498413, + "learning_rate": 3.981946581156976e-05, + "loss": 0.0445, + "step": 16980 + }, + { + "epoch": 0.8502439634680345, + "grad_norm": 1.238515019416809, + "learning_rate": 3.981019814275917e-05, + "loss": 0.4092, + "step": 16990 + }, + { + "epoch": 0.8507444013511822, + "grad_norm": 0.9896653890609741, + "learning_rate": 3.980093047394858e-05, + "loss": 0.7306, + "step": 17000 + }, + { + "epoch": 0.85124483923433, + "grad_norm": 16.88292694091797, + "learning_rate": 3.9791662805137996e-05, + "loss": 0.361, + "step": 17010 + }, + { + "epoch": 0.8517452771174778, + "grad_norm": 1.4751477241516113, + "learning_rate": 3.978239513632741e-05, + "loss": 0.5394, + "step": 17020 + }, + { + "epoch": 0.8522457150006255, + "grad_norm": 18.517332077026367, + "learning_rate": 3.977312746751683e-05, + "loss": 0.2834, + "step": 17030 + }, + { + "epoch": 0.8527461528837733, + "grad_norm": 0.07019864767789841, + "learning_rate": 3.9763859798706236e-05, + "loss": 0.1314, + "step": 17040 + }, + { + "epoch": 0.8532465907669211, + "grad_norm": 0.38338690996170044, + "learning_rate": 3.9754592129895645e-05, + "loss": 0.9004, + "step": 17050 + }, + { + "epoch": 0.8537470286500688, + "grad_norm": 0.15481413900852203, + "learning_rate": 3.974532446108506e-05, + "loss": 0.2978, + "step": 17060 + }, + { + "epoch": 0.8542474665332166, + "grad_norm": 0.7760837078094482, + "learning_rate": 3.973605679227447e-05, + "loss": 0.2563, + "step": 17070 + }, + { + "epoch": 0.8547479044163643, + "grad_norm": 1.269622564315796, + "learning_rate": 3.9726789123463885e-05, + "loss": 0.6326, + "step": 17080 + }, + { + "epoch": 0.8552483422995121, + "grad_norm": 117.57433319091797, + "learning_rate": 3.97175214546533e-05, + "loss": 0.2517, + "step": 17090 + }, + { + "epoch": 0.8557487801826599, + "grad_norm": 25.910980224609375, + "learning_rate": 3.970825378584271e-05, + "loss": 0.5737, + "step": 17100 + }, + { + "epoch": 0.8562492180658076, + "grad_norm": 0.11069413274526596, + "learning_rate": 3.969898611703212e-05, + "loss": 0.1981, + "step": 17110 + }, + { + "epoch": 0.8567496559489554, + "grad_norm": 0.35594943165779114, + "learning_rate": 3.9689718448221534e-05, + "loss": 0.3753, + "step": 17120 + }, + { + "epoch": 0.857250093832103, + "grad_norm": 46.77696990966797, + "learning_rate": 3.968045077941095e-05, + "loss": 0.3406, + "step": 17130 + }, + { + "epoch": 0.8577505317152508, + "grad_norm": 0.04284610226750374, + "learning_rate": 3.9671183110600366e-05, + "loss": 0.4932, + "step": 17140 + }, + { + "epoch": 0.8582509695983986, + "grad_norm": 0.011355245485901833, + "learning_rate": 3.9661915441789774e-05, + "loss": 0.9787, + "step": 17150 + }, + { + "epoch": 0.8587514074815463, + "grad_norm": 41.15909194946289, + "learning_rate": 3.965264777297918e-05, + "loss": 0.77, + "step": 17160 + }, + { + "epoch": 0.8592518453646941, + "grad_norm": 9.430063247680664, + "learning_rate": 3.96433801041686e-05, + "loss": 0.0935, + "step": 17170 + }, + { + "epoch": 0.8597522832478418, + "grad_norm": 3.4157755374908447, + "learning_rate": 3.9634112435358015e-05, + "loss": 0.1607, + "step": 17180 + }, + { + "epoch": 0.8602527211309896, + "grad_norm": 3.075089693069458, + "learning_rate": 3.9624844766547423e-05, + "loss": 0.9241, + "step": 17190 + }, + { + "epoch": 0.8607531590141374, + "grad_norm": 20.55792999267578, + "learning_rate": 3.961557709773684e-05, + "loss": 0.3521, + "step": 17200 + }, + { + "epoch": 0.8612535968972851, + "grad_norm": 31.913358688354492, + "learning_rate": 3.960630942892625e-05, + "loss": 0.4556, + "step": 17210 + }, + { + "epoch": 0.8617540347804329, + "grad_norm": 7.658285140991211, + "learning_rate": 3.9597041760115664e-05, + "loss": 0.3512, + "step": 17220 + }, + { + "epoch": 0.8622544726635807, + "grad_norm": 1.6830642223358154, + "learning_rate": 3.958777409130507e-05, + "loss": 0.6357, + "step": 17230 + }, + { + "epoch": 0.8627549105467284, + "grad_norm": 5.301278591156006, + "learning_rate": 3.957850642249449e-05, + "loss": 0.3635, + "step": 17240 + }, + { + "epoch": 0.8632553484298762, + "grad_norm": 187.0818328857422, + "learning_rate": 3.9569238753683904e-05, + "loss": 0.5211, + "step": 17250 + }, + { + "epoch": 0.8637557863130239, + "grad_norm": 173.0966339111328, + "learning_rate": 3.955997108487331e-05, + "loss": 0.9383, + "step": 17260 + }, + { + "epoch": 0.8642562241961717, + "grad_norm": 9.00809383392334, + "learning_rate": 3.955070341606272e-05, + "loss": 0.3538, + "step": 17270 + }, + { + "epoch": 0.8647566620793194, + "grad_norm": 0.04498112574219704, + "learning_rate": 3.954143574725214e-05, + "loss": 0.6095, + "step": 17280 + }, + { + "epoch": 0.8652570999624671, + "grad_norm": 0.6071091890335083, + "learning_rate": 3.953216807844155e-05, + "loss": 0.231, + "step": 17290 + }, + { + "epoch": 0.8657575378456149, + "grad_norm": 0.1777132898569107, + "learning_rate": 3.952290040963096e-05, + "loss": 0.191, + "step": 17300 + }, + { + "epoch": 0.8662579757287626, + "grad_norm": 89.0108642578125, + "learning_rate": 3.951363274082038e-05, + "loss": 0.3403, + "step": 17310 + }, + { + "epoch": 0.8667584136119104, + "grad_norm": 52.168357849121094, + "learning_rate": 3.950436507200979e-05, + "loss": 0.7492, + "step": 17320 + }, + { + "epoch": 0.8672588514950582, + "grad_norm": 7.3525567054748535, + "learning_rate": 3.94950974031992e-05, + "loss": 0.2034, + "step": 17330 + }, + { + "epoch": 0.8677592893782059, + "grad_norm": 0.4828023612499237, + "learning_rate": 3.948582973438861e-05, + "loss": 0.4692, + "step": 17340 + }, + { + "epoch": 0.8682597272613537, + "grad_norm": 7.355142593383789, + "learning_rate": 3.9476562065578026e-05, + "loss": 0.737, + "step": 17350 + }, + { + "epoch": 0.8687601651445014, + "grad_norm": 14.712762832641602, + "learning_rate": 3.946729439676744e-05, + "loss": 0.3255, + "step": 17360 + }, + { + "epoch": 0.8692606030276492, + "grad_norm": 74.85032653808594, + "learning_rate": 3.945802672795685e-05, + "loss": 0.5274, + "step": 17370 + }, + { + "epoch": 0.869761040910797, + "grad_norm": 0.5020062923431396, + "learning_rate": 3.944875905914626e-05, + "loss": 0.5992, + "step": 17380 + }, + { + "epoch": 0.8702614787939447, + "grad_norm": 7.216900825500488, + "learning_rate": 3.9439491390335675e-05, + "loss": 0.5271, + "step": 17390 + }, + { + "epoch": 0.8707619166770925, + "grad_norm": 0.702370285987854, + "learning_rate": 3.943022372152509e-05, + "loss": 0.5227, + "step": 17400 + }, + { + "epoch": 0.8712623545602403, + "grad_norm": 1.637108564376831, + "learning_rate": 3.94209560527145e-05, + "loss": 0.4464, + "step": 17410 + }, + { + "epoch": 0.871762792443388, + "grad_norm": 0.534026026725769, + "learning_rate": 3.9411688383903915e-05, + "loss": 0.3018, + "step": 17420 + }, + { + "epoch": 0.8722632303265357, + "grad_norm": 79.50595092773438, + "learning_rate": 3.940242071509333e-05, + "loss": 0.8825, + "step": 17430 + }, + { + "epoch": 0.8727636682096834, + "grad_norm": 0.31877559423446655, + "learning_rate": 3.939315304628274e-05, + "loss": 0.3183, + "step": 17440 + }, + { + "epoch": 0.8732641060928312, + "grad_norm": 22.3940372467041, + "learning_rate": 3.938388537747215e-05, + "loss": 0.4271, + "step": 17450 + }, + { + "epoch": 0.873764543975979, + "grad_norm": 12.596190452575684, + "learning_rate": 3.9374617708661564e-05, + "loss": 0.5084, + "step": 17460 + }, + { + "epoch": 0.8742649818591267, + "grad_norm": 35.34507369995117, + "learning_rate": 3.936535003985098e-05, + "loss": 0.3062, + "step": 17470 + }, + { + "epoch": 0.8747654197422745, + "grad_norm": 46.012939453125, + "learning_rate": 3.9356082371040396e-05, + "loss": 0.4237, + "step": 17480 + }, + { + "epoch": 0.8752658576254222, + "grad_norm": 2.427534818649292, + "learning_rate": 3.93468147022298e-05, + "loss": 0.2328, + "step": 17490 + }, + { + "epoch": 0.87576629550857, + "grad_norm": 11.652222633361816, + "learning_rate": 3.933754703341921e-05, + "loss": 0.3638, + "step": 17500 + }, + { + "epoch": 0.8762667333917178, + "grad_norm": 4.793886184692383, + "learning_rate": 3.932827936460863e-05, + "loss": 0.4214, + "step": 17510 + }, + { + "epoch": 0.8767671712748655, + "grad_norm": 0.5347192287445068, + "learning_rate": 3.9319011695798045e-05, + "loss": 0.2485, + "step": 17520 + }, + { + "epoch": 0.8772676091580133, + "grad_norm": 113.11945343017578, + "learning_rate": 3.930974402698745e-05, + "loss": 0.6215, + "step": 17530 + }, + { + "epoch": 0.8777680470411611, + "grad_norm": 66.29792785644531, + "learning_rate": 3.930047635817687e-05, + "loss": 0.4858, + "step": 17540 + }, + { + "epoch": 0.8782684849243088, + "grad_norm": 112.1669692993164, + "learning_rate": 3.929120868936628e-05, + "loss": 0.2119, + "step": 17550 + }, + { + "epoch": 0.8787689228074566, + "grad_norm": 0.5600221157073975, + "learning_rate": 3.9281941020555694e-05, + "loss": 0.3164, + "step": 17560 + }, + { + "epoch": 0.8792693606906042, + "grad_norm": 34.927120208740234, + "learning_rate": 3.92726733517451e-05, + "loss": 0.4656, + "step": 17570 + }, + { + "epoch": 0.879769798573752, + "grad_norm": 0.020006412640213966, + "learning_rate": 3.926340568293452e-05, + "loss": 0.5706, + "step": 17580 + }, + { + "epoch": 0.8802702364568998, + "grad_norm": 2.5700716972351074, + "learning_rate": 3.9254138014123934e-05, + "loss": 0.2771, + "step": 17590 + }, + { + "epoch": 0.8807706743400475, + "grad_norm": 67.53302001953125, + "learning_rate": 3.924487034531334e-05, + "loss": 0.6647, + "step": 17600 + }, + { + "epoch": 0.8812711122231953, + "grad_norm": 47.86492156982422, + "learning_rate": 3.923560267650275e-05, + "loss": 0.1132, + "step": 17610 + }, + { + "epoch": 0.881771550106343, + "grad_norm": 66.60688781738281, + "learning_rate": 3.922633500769217e-05, + "loss": 0.3835, + "step": 17620 + }, + { + "epoch": 0.8822719879894908, + "grad_norm": 0.11549071222543716, + "learning_rate": 3.921706733888158e-05, + "loss": 0.4672, + "step": 17630 + }, + { + "epoch": 0.8827724258726386, + "grad_norm": 0.35349053144454956, + "learning_rate": 3.920779967007099e-05, + "loss": 0.3316, + "step": 17640 + }, + { + "epoch": 0.8832728637557863, + "grad_norm": 61.92650604248047, + "learning_rate": 3.919853200126041e-05, + "loss": 1.1557, + "step": 17650 + }, + { + "epoch": 0.8837733016389341, + "grad_norm": 99.66209411621094, + "learning_rate": 3.9189264332449816e-05, + "loss": 0.7987, + "step": 17660 + }, + { + "epoch": 0.8842737395220818, + "grad_norm": 4.691267967224121, + "learning_rate": 3.917999666363923e-05, + "loss": 0.2145, + "step": 17670 + }, + { + "epoch": 0.8847741774052296, + "grad_norm": 0.24810029566287994, + "learning_rate": 3.917072899482864e-05, + "loss": 0.2272, + "step": 17680 + }, + { + "epoch": 0.8852746152883774, + "grad_norm": 83.26143646240234, + "learning_rate": 3.9161461326018056e-05, + "loss": 0.4599, + "step": 17690 + }, + { + "epoch": 0.885775053171525, + "grad_norm": 43.380279541015625, + "learning_rate": 3.915219365720747e-05, + "loss": 0.7017, + "step": 17700 + }, + { + "epoch": 0.8862754910546728, + "grad_norm": 12.593765258789062, + "learning_rate": 3.914292598839688e-05, + "loss": 0.4912, + "step": 17710 + }, + { + "epoch": 0.8867759289378206, + "grad_norm": 57.27192306518555, + "learning_rate": 3.913365831958629e-05, + "loss": 0.4821, + "step": 17720 + }, + { + "epoch": 0.8872763668209683, + "grad_norm": 57.59225845336914, + "learning_rate": 3.9124390650775705e-05, + "loss": 0.5111, + "step": 17730 + }, + { + "epoch": 0.8877768047041161, + "grad_norm": 1.6165294647216797, + "learning_rate": 3.911512298196512e-05, + "loss": 0.5499, + "step": 17740 + }, + { + "epoch": 0.8882772425872638, + "grad_norm": 2.2245965003967285, + "learning_rate": 3.910585531315453e-05, + "loss": 0.4749, + "step": 17750 + }, + { + "epoch": 0.8887776804704116, + "grad_norm": 19.223173141479492, + "learning_rate": 3.9096587644343945e-05, + "loss": 0.463, + "step": 17760 + }, + { + "epoch": 0.8892781183535594, + "grad_norm": 0.047667812556028366, + "learning_rate": 3.9087319975533354e-05, + "loss": 0.5196, + "step": 17770 + }, + { + "epoch": 0.8897785562367071, + "grad_norm": 0.5452331304550171, + "learning_rate": 3.907805230672277e-05, + "loss": 0.7014, + "step": 17780 + }, + { + "epoch": 0.8902789941198549, + "grad_norm": 0.00844395998865366, + "learning_rate": 3.906878463791218e-05, + "loss": 0.635, + "step": 17790 + }, + { + "epoch": 0.8907794320030026, + "grad_norm": 0.03355895355343819, + "learning_rate": 3.9059516969101594e-05, + "loss": 0.5638, + "step": 17800 + }, + { + "epoch": 0.8912798698861504, + "grad_norm": 18.570188522338867, + "learning_rate": 3.905024930029101e-05, + "loss": 0.2627, + "step": 17810 + }, + { + "epoch": 0.8917803077692982, + "grad_norm": 40.05745315551758, + "learning_rate": 3.904098163148042e-05, + "loss": 0.382, + "step": 17820 + }, + { + "epoch": 0.8922807456524459, + "grad_norm": 10.184525489807129, + "learning_rate": 3.903171396266983e-05, + "loss": 0.4131, + "step": 17830 + }, + { + "epoch": 0.8927811835355937, + "grad_norm": 152.6305694580078, + "learning_rate": 3.902244629385924e-05, + "loss": 0.7617, + "step": 17840 + }, + { + "epoch": 0.8932816214187413, + "grad_norm": 35.8537483215332, + "learning_rate": 3.901317862504866e-05, + "loss": 0.2471, + "step": 17850 + }, + { + "epoch": 0.8937820593018891, + "grad_norm": 8.28034496307373, + "learning_rate": 3.9003910956238074e-05, + "loss": 0.1314, + "step": 17860 + }, + { + "epoch": 0.8942824971850369, + "grad_norm": 0.17339231073856354, + "learning_rate": 3.899464328742748e-05, + "loss": 0.4194, + "step": 17870 + }, + { + "epoch": 0.8947829350681846, + "grad_norm": 0.0028912739362567663, + "learning_rate": 3.898537561861689e-05, + "loss": 0.5973, + "step": 17880 + }, + { + "epoch": 0.8952833729513324, + "grad_norm": 45.775814056396484, + "learning_rate": 3.897610794980631e-05, + "loss": 0.6045, + "step": 17890 + }, + { + "epoch": 0.8957838108344802, + "grad_norm": 19.02988052368164, + "learning_rate": 3.8966840280995723e-05, + "loss": 0.666, + "step": 17900 + }, + { + "epoch": 0.8962842487176279, + "grad_norm": 55.070499420166016, + "learning_rate": 3.895757261218513e-05, + "loss": 0.4685, + "step": 17910 + }, + { + "epoch": 0.8967846866007757, + "grad_norm": 34.90407943725586, + "learning_rate": 3.894830494337455e-05, + "loss": 0.4714, + "step": 17920 + }, + { + "epoch": 0.8972851244839234, + "grad_norm": 2.2209455966949463, + "learning_rate": 3.893903727456396e-05, + "loss": 0.3787, + "step": 17930 + }, + { + "epoch": 0.8977855623670712, + "grad_norm": 74.72695922851562, + "learning_rate": 3.8929769605753366e-05, + "loss": 0.6233, + "step": 17940 + }, + { + "epoch": 0.898286000250219, + "grad_norm": 12.447969436645508, + "learning_rate": 3.892050193694278e-05, + "loss": 0.2147, + "step": 17950 + }, + { + "epoch": 0.8987864381333667, + "grad_norm": 18.637598037719727, + "learning_rate": 3.89112342681322e-05, + "loss": 0.2028, + "step": 17960 + }, + { + "epoch": 0.8992868760165145, + "grad_norm": 135.18844604492188, + "learning_rate": 3.890196659932161e-05, + "loss": 0.5318, + "step": 17970 + }, + { + "epoch": 0.8997873138996622, + "grad_norm": 33.94375228881836, + "learning_rate": 3.889269893051102e-05, + "loss": 0.2795, + "step": 17980 + }, + { + "epoch": 0.90028775178281, + "grad_norm": 20.424270629882812, + "learning_rate": 3.888343126170043e-05, + "loss": 0.2432, + "step": 17990 + }, + { + "epoch": 0.9007881896659578, + "grad_norm": 0.1556035280227661, + "learning_rate": 3.8874163592889846e-05, + "loss": 0.4377, + "step": 18000 + }, + { + "epoch": 0.9012886275491054, + "grad_norm": 0.3077274560928345, + "learning_rate": 3.886489592407926e-05, + "loss": 0.5467, + "step": 18010 + }, + { + "epoch": 0.9017890654322532, + "grad_norm": 2.9789998531341553, + "learning_rate": 3.885562825526867e-05, + "loss": 0.5634, + "step": 18020 + }, + { + "epoch": 0.902289503315401, + "grad_norm": 109.58544158935547, + "learning_rate": 3.8846360586458086e-05, + "loss": 0.7626, + "step": 18030 + }, + { + "epoch": 0.9027899411985487, + "grad_norm": 15.316466331481934, + "learning_rate": 3.88370929176475e-05, + "loss": 0.6824, + "step": 18040 + }, + { + "epoch": 0.9032903790816965, + "grad_norm": 10.224444389343262, + "learning_rate": 3.882782524883691e-05, + "loss": 0.5326, + "step": 18050 + }, + { + "epoch": 0.9037908169648442, + "grad_norm": 21.280250549316406, + "learning_rate": 3.881855758002632e-05, + "loss": 0.6242, + "step": 18060 + }, + { + "epoch": 0.904291254847992, + "grad_norm": 5.209734916687012, + "learning_rate": 3.8809289911215735e-05, + "loss": 0.4968, + "step": 18070 + }, + { + "epoch": 0.9047916927311398, + "grad_norm": 64.89057922363281, + "learning_rate": 3.880002224240515e-05, + "loss": 0.2678, + "step": 18080 + }, + { + "epoch": 0.9052921306142875, + "grad_norm": 0.11399489641189575, + "learning_rate": 3.879075457359456e-05, + "loss": 0.0834, + "step": 18090 + }, + { + "epoch": 0.9057925684974353, + "grad_norm": 0.23515290021896362, + "learning_rate": 3.878148690478397e-05, + "loss": 0.0508, + "step": 18100 + }, + { + "epoch": 0.906293006380583, + "grad_norm": 0.2473323792219162, + "learning_rate": 3.8772219235973384e-05, + "loss": 0.3471, + "step": 18110 + }, + { + "epoch": 0.9067934442637308, + "grad_norm": 22.410444259643555, + "learning_rate": 3.87629515671628e-05, + "loss": 0.6475, + "step": 18120 + }, + { + "epoch": 0.9072938821468786, + "grad_norm": 66.4444580078125, + "learning_rate": 3.875368389835221e-05, + "loss": 0.4268, + "step": 18130 + }, + { + "epoch": 0.9077943200300262, + "grad_norm": 0.020197900012135506, + "learning_rate": 3.8744416229541624e-05, + "loss": 0.3278, + "step": 18140 + }, + { + "epoch": 0.908294757913174, + "grad_norm": 28.428972244262695, + "learning_rate": 3.873514856073104e-05, + "loss": 0.2603, + "step": 18150 + }, + { + "epoch": 0.9087951957963217, + "grad_norm": 7.59912109375, + "learning_rate": 3.872588089192045e-05, + "loss": 0.3725, + "step": 18160 + }, + { + "epoch": 0.9092956336794695, + "grad_norm": 87.41991424560547, + "learning_rate": 3.871661322310986e-05, + "loss": 0.4528, + "step": 18170 + }, + { + "epoch": 0.9097960715626173, + "grad_norm": 0.11765594035387039, + "learning_rate": 3.870734555429927e-05, + "loss": 0.5319, + "step": 18180 + }, + { + "epoch": 0.910296509445765, + "grad_norm": 11.069851875305176, + "learning_rate": 3.869807788548869e-05, + "loss": 0.4705, + "step": 18190 + }, + { + "epoch": 0.9107969473289128, + "grad_norm": 0.01557129342108965, + "learning_rate": 3.86888102166781e-05, + "loss": 0.082, + "step": 18200 + }, + { + "epoch": 0.9112973852120606, + "grad_norm": 65.58367919921875, + "learning_rate": 3.8679542547867506e-05, + "loss": 0.2081, + "step": 18210 + }, + { + "epoch": 0.9117978230952083, + "grad_norm": 15.194670677185059, + "learning_rate": 3.867027487905692e-05, + "loss": 0.1886, + "step": 18220 + }, + { + "epoch": 0.9122982609783561, + "grad_norm": 0.002706536091864109, + "learning_rate": 3.866100721024634e-05, + "loss": 0.4687, + "step": 18230 + }, + { + "epoch": 0.9127986988615038, + "grad_norm": 2.613309383392334, + "learning_rate": 3.8651739541435747e-05, + "loss": 0.24, + "step": 18240 + }, + { + "epoch": 0.9132991367446516, + "grad_norm": 27.900360107421875, + "learning_rate": 3.864247187262516e-05, + "loss": 0.6095, + "step": 18250 + }, + { + "epoch": 0.9137995746277994, + "grad_norm": 77.35541534423828, + "learning_rate": 3.863320420381458e-05, + "loss": 0.605, + "step": 18260 + }, + { + "epoch": 0.9143000125109471, + "grad_norm": 5.523254871368408, + "learning_rate": 3.862393653500399e-05, + "loss": 0.4883, + "step": 18270 + }, + { + "epoch": 0.9148004503940949, + "grad_norm": 63.50475311279297, + "learning_rate": 3.8614668866193396e-05, + "loss": 0.5425, + "step": 18280 + }, + { + "epoch": 0.9153008882772425, + "grad_norm": 0.04452912509441376, + "learning_rate": 3.860540119738281e-05, + "loss": 0.2873, + "step": 18290 + }, + { + "epoch": 0.9158013261603903, + "grad_norm": 22.58211898803711, + "learning_rate": 3.859613352857223e-05, + "loss": 0.4812, + "step": 18300 + }, + { + "epoch": 0.9163017640435381, + "grad_norm": 15.958091735839844, + "learning_rate": 3.858686585976164e-05, + "loss": 0.3122, + "step": 18310 + }, + { + "epoch": 0.9168022019266858, + "grad_norm": 44.03060531616211, + "learning_rate": 3.8577598190951045e-05, + "loss": 0.6638, + "step": 18320 + }, + { + "epoch": 0.9173026398098336, + "grad_norm": 49.442752838134766, + "learning_rate": 3.856833052214046e-05, + "loss": 0.3091, + "step": 18330 + }, + { + "epoch": 0.9178030776929813, + "grad_norm": 63.63135528564453, + "learning_rate": 3.8559062853329876e-05, + "loss": 0.2639, + "step": 18340 + }, + { + "epoch": 0.9183035155761291, + "grad_norm": 0.1349039524793625, + "learning_rate": 3.854979518451929e-05, + "loss": 0.3128, + "step": 18350 + }, + { + "epoch": 0.9188039534592769, + "grad_norm": 31.40736961364746, + "learning_rate": 3.85405275157087e-05, + "loss": 0.702, + "step": 18360 + }, + { + "epoch": 0.9193043913424246, + "grad_norm": 40.76157760620117, + "learning_rate": 3.8531259846898116e-05, + "loss": 0.2715, + "step": 18370 + }, + { + "epoch": 0.9198048292255724, + "grad_norm": 12.958233833312988, + "learning_rate": 3.8521992178087525e-05, + "loss": 0.3192, + "step": 18380 + }, + { + "epoch": 0.9203052671087202, + "grad_norm": 86.82539367675781, + "learning_rate": 3.851272450927694e-05, + "loss": 0.3128, + "step": 18390 + }, + { + "epoch": 0.9208057049918679, + "grad_norm": 0.09370258450508118, + "learning_rate": 3.850345684046635e-05, + "loss": 0.4129, + "step": 18400 + }, + { + "epoch": 0.9213061428750157, + "grad_norm": 0.017874546349048615, + "learning_rate": 3.8494189171655765e-05, + "loss": 0.0826, + "step": 18410 + }, + { + "epoch": 0.9218065807581634, + "grad_norm": 26.91908836364746, + "learning_rate": 3.848492150284518e-05, + "loss": 0.2861, + "step": 18420 + }, + { + "epoch": 0.9223070186413111, + "grad_norm": 0.4923631250858307, + "learning_rate": 3.847565383403459e-05, + "loss": 0.7041, + "step": 18430 + }, + { + "epoch": 0.922807456524459, + "grad_norm": 16.95464324951172, + "learning_rate": 3.8466386165224e-05, + "loss": 0.7742, + "step": 18440 + }, + { + "epoch": 0.9233078944076066, + "grad_norm": 0.15977361798286438, + "learning_rate": 3.8457118496413414e-05, + "loss": 0.4496, + "step": 18450 + }, + { + "epoch": 0.9238083322907544, + "grad_norm": 0.11012395471334457, + "learning_rate": 3.844785082760283e-05, + "loss": 0.222, + "step": 18460 + }, + { + "epoch": 0.9243087701739021, + "grad_norm": 0.05492579564452171, + "learning_rate": 3.843858315879224e-05, + "loss": 0.3464, + "step": 18470 + }, + { + "epoch": 0.9248092080570499, + "grad_norm": 97.37149047851562, + "learning_rate": 3.8429315489981654e-05, + "loss": 0.4053, + "step": 18480 + }, + { + "epoch": 0.9253096459401977, + "grad_norm": 0.03830135613679886, + "learning_rate": 3.842004782117106e-05, + "loss": 0.5438, + "step": 18490 + }, + { + "epoch": 0.9258100838233454, + "grad_norm": 0.34452196955680847, + "learning_rate": 3.841078015236048e-05, + "loss": 0.4781, + "step": 18500 + }, + { + "epoch": 0.9263105217064932, + "grad_norm": 0.7208184599876404, + "learning_rate": 3.840151248354989e-05, + "loss": 0.5355, + "step": 18510 + }, + { + "epoch": 0.9268109595896409, + "grad_norm": 46.00852966308594, + "learning_rate": 3.83922448147393e-05, + "loss": 0.3547, + "step": 18520 + }, + { + "epoch": 0.9273113974727887, + "grad_norm": 59.23677062988281, + "learning_rate": 3.838297714592872e-05, + "loss": 0.6651, + "step": 18530 + }, + { + "epoch": 0.9278118353559365, + "grad_norm": 14.588616371154785, + "learning_rate": 3.837370947711813e-05, + "loss": 0.6622, + "step": 18540 + }, + { + "epoch": 0.9283122732390842, + "grad_norm": 1.8405739068984985, + "learning_rate": 3.8364441808307536e-05, + "loss": 0.5474, + "step": 18550 + }, + { + "epoch": 0.928812711122232, + "grad_norm": 38.38557052612305, + "learning_rate": 3.835517413949695e-05, + "loss": 0.4665, + "step": 18560 + }, + { + "epoch": 0.9293131490053798, + "grad_norm": 0.19411601126194, + "learning_rate": 3.834590647068637e-05, + "loss": 0.1242, + "step": 18570 + }, + { + "epoch": 0.9298135868885274, + "grad_norm": 20.347795486450195, + "learning_rate": 3.8336638801875776e-05, + "loss": 0.2567, + "step": 18580 + }, + { + "epoch": 0.9303140247716752, + "grad_norm": 0.11099473387002945, + "learning_rate": 3.832737113306519e-05, + "loss": 0.4223, + "step": 18590 + }, + { + "epoch": 0.9308144626548229, + "grad_norm": 0.05054068565368652, + "learning_rate": 3.83181034642546e-05, + "loss": 0.7045, + "step": 18600 + }, + { + "epoch": 0.9313149005379707, + "grad_norm": 0.024620186537504196, + "learning_rate": 3.8308835795444017e-05, + "loss": 0.1244, + "step": 18610 + }, + { + "epoch": 0.9318153384211185, + "grad_norm": 0.007812791503965855, + "learning_rate": 3.8299568126633425e-05, + "loss": 0.4479, + "step": 18620 + }, + { + "epoch": 0.9323157763042662, + "grad_norm": 26.611356735229492, + "learning_rate": 3.829030045782284e-05, + "loss": 0.4863, + "step": 18630 + }, + { + "epoch": 0.932816214187414, + "grad_norm": 1.2390875816345215, + "learning_rate": 3.828103278901226e-05, + "loss": 0.1868, + "step": 18640 + }, + { + "epoch": 0.9333166520705617, + "grad_norm": 5.04191780090332, + "learning_rate": 3.8271765120201666e-05, + "loss": 0.3003, + "step": 18650 + }, + { + "epoch": 0.9338170899537095, + "grad_norm": 5.058414459228516, + "learning_rate": 3.8262497451391074e-05, + "loss": 0.5474, + "step": 18660 + }, + { + "epoch": 0.9343175278368573, + "grad_norm": 0.4093897044658661, + "learning_rate": 3.825322978258049e-05, + "loss": 0.3208, + "step": 18670 + }, + { + "epoch": 0.934817965720005, + "grad_norm": 0.1016351655125618, + "learning_rate": 3.8243962113769906e-05, + "loss": 0.0429, + "step": 18680 + }, + { + "epoch": 0.9353184036031528, + "grad_norm": 20.529895782470703, + "learning_rate": 3.823469444495932e-05, + "loss": 0.6926, + "step": 18690 + }, + { + "epoch": 0.9358188414863006, + "grad_norm": 17.8652400970459, + "learning_rate": 3.822542677614873e-05, + "loss": 0.5029, + "step": 18700 + }, + { + "epoch": 0.9363192793694483, + "grad_norm": 2.247103691101074, + "learning_rate": 3.821615910733814e-05, + "loss": 0.3092, + "step": 18710 + }, + { + "epoch": 0.936819717252596, + "grad_norm": 0.08239307999610901, + "learning_rate": 3.8206891438527555e-05, + "loss": 0.2271, + "step": 18720 + }, + { + "epoch": 0.9373201551357437, + "grad_norm": 0.1488705575466156, + "learning_rate": 3.819762376971697e-05, + "loss": 0.1595, + "step": 18730 + }, + { + "epoch": 0.9378205930188915, + "grad_norm": 66.18177795410156, + "learning_rate": 3.818835610090638e-05, + "loss": 0.3293, + "step": 18740 + }, + { + "epoch": 0.9383210309020393, + "grad_norm": 5.99386739730835, + "learning_rate": 3.8179088432095795e-05, + "loss": 0.1843, + "step": 18750 + }, + { + "epoch": 0.938821468785187, + "grad_norm": 0.4439041018486023, + "learning_rate": 3.816982076328521e-05, + "loss": 0.2465, + "step": 18760 + }, + { + "epoch": 0.9393219066683348, + "grad_norm": 102.17914581298828, + "learning_rate": 3.816055309447461e-05, + "loss": 0.8638, + "step": 18770 + }, + { + "epoch": 0.9398223445514825, + "grad_norm": 25.816184997558594, + "learning_rate": 3.815128542566403e-05, + "loss": 0.3459, + "step": 18780 + }, + { + "epoch": 0.9403227824346303, + "grad_norm": 143.93356323242188, + "learning_rate": 3.8142017756853444e-05, + "loss": 0.1888, + "step": 18790 + }, + { + "epoch": 0.9408232203177781, + "grad_norm": 34.4100227355957, + "learning_rate": 3.813275008804286e-05, + "loss": 0.6194, + "step": 18800 + }, + { + "epoch": 0.9413236582009258, + "grad_norm": 44.61533737182617, + "learning_rate": 3.812348241923227e-05, + "loss": 0.3907, + "step": 18810 + }, + { + "epoch": 0.9418240960840736, + "grad_norm": 0.016274530440568924, + "learning_rate": 3.811421475042168e-05, + "loss": 0.2932, + "step": 18820 + }, + { + "epoch": 0.9423245339672213, + "grad_norm": 22.412281036376953, + "learning_rate": 3.810494708161109e-05, + "loss": 0.6576, + "step": 18830 + }, + { + "epoch": 0.9428249718503691, + "grad_norm": 4.509145736694336, + "learning_rate": 3.809567941280051e-05, + "loss": 0.4346, + "step": 18840 + }, + { + "epoch": 0.9433254097335169, + "grad_norm": 9.111029624938965, + "learning_rate": 3.808641174398992e-05, + "loss": 0.5181, + "step": 18850 + }, + { + "epoch": 0.9438258476166645, + "grad_norm": 8.270316123962402, + "learning_rate": 3.807714407517933e-05, + "loss": 0.2183, + "step": 18860 + }, + { + "epoch": 0.9443262854998123, + "grad_norm": 91.60276794433594, + "learning_rate": 3.806787640636875e-05, + "loss": 0.1704, + "step": 18870 + }, + { + "epoch": 0.9448267233829601, + "grad_norm": 43.75311279296875, + "learning_rate": 3.805860873755816e-05, + "loss": 0.3373, + "step": 18880 + }, + { + "epoch": 0.9453271612661078, + "grad_norm": 85.90027618408203, + "learning_rate": 3.8049341068747566e-05, + "loss": 0.2414, + "step": 18890 + }, + { + "epoch": 0.9458275991492556, + "grad_norm": 0.830310046672821, + "learning_rate": 3.804007339993698e-05, + "loss": 0.5112, + "step": 18900 + }, + { + "epoch": 0.9463280370324033, + "grad_norm": 0.6185224056243896, + "learning_rate": 3.80308057311264e-05, + "loss": 0.7104, + "step": 18910 + }, + { + "epoch": 0.9468284749155511, + "grad_norm": 80.2576904296875, + "learning_rate": 3.8021538062315806e-05, + "loss": 0.3292, + "step": 18920 + }, + { + "epoch": 0.9473289127986989, + "grad_norm": 2.1412603855133057, + "learning_rate": 3.8012270393505215e-05, + "loss": 0.3854, + "step": 18930 + }, + { + "epoch": 0.9478293506818466, + "grad_norm": 0.9597477316856384, + "learning_rate": 3.800300272469463e-05, + "loss": 0.2212, + "step": 18940 + }, + { + "epoch": 0.9483297885649944, + "grad_norm": 30.83407211303711, + "learning_rate": 3.7993735055884047e-05, + "loss": 0.6283, + "step": 18950 + }, + { + "epoch": 0.9488302264481421, + "grad_norm": 4.27736759185791, + "learning_rate": 3.7984467387073455e-05, + "loss": 0.3106, + "step": 18960 + }, + { + "epoch": 0.9493306643312899, + "grad_norm": 0.04879345744848251, + "learning_rate": 3.797519971826287e-05, + "loss": 0.2401, + "step": 18970 + }, + { + "epoch": 0.9498311022144377, + "grad_norm": 3.665548324584961, + "learning_rate": 3.796593204945229e-05, + "loss": 0.2822, + "step": 18980 + }, + { + "epoch": 0.9503315400975854, + "grad_norm": 82.6197509765625, + "learning_rate": 3.7956664380641696e-05, + "loss": 0.272, + "step": 18990 + }, + { + "epoch": 0.9508319779807332, + "grad_norm": 89.1767807006836, + "learning_rate": 3.7947396711831104e-05, + "loss": 0.3026, + "step": 19000 + }, + { + "epoch": 0.9513324158638808, + "grad_norm": 0.0018604955403134227, + "learning_rate": 3.793812904302052e-05, + "loss": 0.4722, + "step": 19010 + }, + { + "epoch": 0.9518328537470286, + "grad_norm": 0.20757070183753967, + "learning_rate": 3.7928861374209936e-05, + "loss": 0.3331, + "step": 19020 + }, + { + "epoch": 0.9523332916301764, + "grad_norm": 51.31541061401367, + "learning_rate": 3.7919593705399344e-05, + "loss": 1.067, + "step": 19030 + }, + { + "epoch": 0.9528337295133241, + "grad_norm": 35.92148971557617, + "learning_rate": 3.791032603658875e-05, + "loss": 0.7194, + "step": 19040 + }, + { + "epoch": 0.9533341673964719, + "grad_norm": 7.01915168762207, + "learning_rate": 3.790105836777817e-05, + "loss": 0.3946, + "step": 19050 + }, + { + "epoch": 0.9538346052796197, + "grad_norm": 1.1045304536819458, + "learning_rate": 3.7891790698967585e-05, + "loss": 0.3051, + "step": 19060 + }, + { + "epoch": 0.9543350431627674, + "grad_norm": 0.5044363737106323, + "learning_rate": 3.7882523030156993e-05, + "loss": 0.3754, + "step": 19070 + }, + { + "epoch": 0.9548354810459152, + "grad_norm": 0.5437859296798706, + "learning_rate": 3.787325536134641e-05, + "loss": 0.5021, + "step": 19080 + }, + { + "epoch": 0.9553359189290629, + "grad_norm": 1.540535807609558, + "learning_rate": 3.7863987692535825e-05, + "loss": 0.2055, + "step": 19090 + }, + { + "epoch": 0.9558363568122107, + "grad_norm": 0.27904394268989563, + "learning_rate": 3.7854720023725234e-05, + "loss": 0.2415, + "step": 19100 + }, + { + "epoch": 0.9563367946953585, + "grad_norm": 77.04170989990234, + "learning_rate": 3.784545235491464e-05, + "loss": 0.2296, + "step": 19110 + }, + { + "epoch": 0.9568372325785062, + "grad_norm": 1.3608944416046143, + "learning_rate": 3.783618468610406e-05, + "loss": 0.4579, + "step": 19120 + }, + { + "epoch": 0.957337670461654, + "grad_norm": 22.80614471435547, + "learning_rate": 3.7826917017293474e-05, + "loss": 1.0175, + "step": 19130 + }, + { + "epoch": 0.9578381083448017, + "grad_norm": 0.3044731020927429, + "learning_rate": 3.781764934848289e-05, + "loss": 0.1707, + "step": 19140 + }, + { + "epoch": 0.9583385462279495, + "grad_norm": 84.66175842285156, + "learning_rate": 3.780838167967229e-05, + "loss": 0.6534, + "step": 19150 + }, + { + "epoch": 0.9588389841110972, + "grad_norm": 39.51995086669922, + "learning_rate": 3.779911401086171e-05, + "loss": 0.5714, + "step": 19160 + }, + { + "epoch": 0.9593394219942449, + "grad_norm": 0.17503340542316437, + "learning_rate": 3.778984634205112e-05, + "loss": 0.562, + "step": 19170 + }, + { + "epoch": 0.9598398598773927, + "grad_norm": 45.648277282714844, + "learning_rate": 3.778057867324054e-05, + "loss": 0.4566, + "step": 19180 + }, + { + "epoch": 0.9603402977605404, + "grad_norm": 25.63664436340332, + "learning_rate": 3.777131100442995e-05, + "loss": 0.1104, + "step": 19190 + }, + { + "epoch": 0.9608407356436882, + "grad_norm": 30.292442321777344, + "learning_rate": 3.776204333561936e-05, + "loss": 0.2478, + "step": 19200 + }, + { + "epoch": 0.961341173526836, + "grad_norm": 98.79558563232422, + "learning_rate": 3.775277566680877e-05, + "loss": 0.7306, + "step": 19210 + }, + { + "epoch": 0.9618416114099837, + "grad_norm": 23.978824615478516, + "learning_rate": 3.774350799799819e-05, + "loss": 0.4329, + "step": 19220 + }, + { + "epoch": 0.9623420492931315, + "grad_norm": 3.1024115085601807, + "learning_rate": 3.7734240329187596e-05, + "loss": 0.0868, + "step": 19230 + }, + { + "epoch": 0.9628424871762793, + "grad_norm": 47.2286262512207, + "learning_rate": 3.772497266037701e-05, + "loss": 0.4602, + "step": 19240 + }, + { + "epoch": 0.963342925059427, + "grad_norm": 35.906036376953125, + "learning_rate": 3.771570499156643e-05, + "loss": 0.2888, + "step": 19250 + }, + { + "epoch": 0.9638433629425748, + "grad_norm": 0.031449105590581894, + "learning_rate": 3.7706437322755836e-05, + "loss": 0.3537, + "step": 19260 + }, + { + "epoch": 0.9643438008257225, + "grad_norm": 79.9513931274414, + "learning_rate": 3.7697169653945245e-05, + "loss": 0.3675, + "step": 19270 + }, + { + "epoch": 0.9648442387088703, + "grad_norm": 27.343460083007812, + "learning_rate": 3.768790198513466e-05, + "loss": 0.3871, + "step": 19280 + }, + { + "epoch": 0.9653446765920181, + "grad_norm": 184.4755096435547, + "learning_rate": 3.7678634316324076e-05, + "loss": 0.428, + "step": 19290 + }, + { + "epoch": 0.9658451144751657, + "grad_norm": 0.05249952897429466, + "learning_rate": 3.7669366647513485e-05, + "loss": 0.4195, + "step": 19300 + }, + { + "epoch": 0.9663455523583135, + "grad_norm": 0.07699065655469894, + "learning_rate": 3.76600989787029e-05, + "loss": 1.0414, + "step": 19310 + }, + { + "epoch": 0.9668459902414612, + "grad_norm": 0.04189734160900116, + "learning_rate": 3.765083130989231e-05, + "loss": 0.5227, + "step": 19320 + }, + { + "epoch": 0.967346428124609, + "grad_norm": 0.01382913812994957, + "learning_rate": 3.7641563641081725e-05, + "loss": 0.1435, + "step": 19330 + }, + { + "epoch": 0.9678468660077568, + "grad_norm": 42.599342346191406, + "learning_rate": 3.7632295972271134e-05, + "loss": 0.5796, + "step": 19340 + }, + { + "epoch": 0.9683473038909045, + "grad_norm": 40.111942291259766, + "learning_rate": 3.762302830346055e-05, + "loss": 0.5323, + "step": 19350 + }, + { + "epoch": 0.9688477417740523, + "grad_norm": 34.73700714111328, + "learning_rate": 3.7613760634649966e-05, + "loss": 0.1717, + "step": 19360 + }, + { + "epoch": 0.9693481796572001, + "grad_norm": 0.5157777667045593, + "learning_rate": 3.7604492965839374e-05, + "loss": 0.2502, + "step": 19370 + }, + { + "epoch": 0.9698486175403478, + "grad_norm": 2.2853753566741943, + "learning_rate": 3.759522529702878e-05, + "loss": 0.072, + "step": 19380 + }, + { + "epoch": 0.9703490554234956, + "grad_norm": 1.3188753128051758, + "learning_rate": 3.75859576282182e-05, + "loss": 1.2039, + "step": 19390 + }, + { + "epoch": 0.9708494933066433, + "grad_norm": 59.5556755065918, + "learning_rate": 3.7576689959407615e-05, + "loss": 0.3452, + "step": 19400 + }, + { + "epoch": 0.9713499311897911, + "grad_norm": 38.97928237915039, + "learning_rate": 3.756742229059702e-05, + "loss": 0.5349, + "step": 19410 + }, + { + "epoch": 0.9718503690729389, + "grad_norm": 0.08275303244590759, + "learning_rate": 3.755815462178644e-05, + "loss": 0.3486, + "step": 19420 + }, + { + "epoch": 0.9723508069560866, + "grad_norm": 1.7126134634017944, + "learning_rate": 3.754888695297585e-05, + "loss": 0.3905, + "step": 19430 + }, + { + "epoch": 0.9728512448392344, + "grad_norm": 4.6567511558532715, + "learning_rate": 3.7539619284165264e-05, + "loss": 0.2039, + "step": 19440 + }, + { + "epoch": 0.973351682722382, + "grad_norm": 15.860039710998535, + "learning_rate": 3.753035161535467e-05, + "loss": 0.1551, + "step": 19450 + }, + { + "epoch": 0.9738521206055298, + "grad_norm": 0.024703821167349815, + "learning_rate": 3.752108394654409e-05, + "loss": 0.538, + "step": 19460 + }, + { + "epoch": 0.9743525584886776, + "grad_norm": 51.23233413696289, + "learning_rate": 3.7511816277733504e-05, + "loss": 0.2103, + "step": 19470 + }, + { + "epoch": 0.9748529963718253, + "grad_norm": 24.321125030517578, + "learning_rate": 3.750254860892291e-05, + "loss": 0.2446, + "step": 19480 + }, + { + "epoch": 0.9753534342549731, + "grad_norm": 47.94352340698242, + "learning_rate": 3.749328094011232e-05, + "loss": 0.3975, + "step": 19490 + }, + { + "epoch": 0.9758538721381208, + "grad_norm": 170.8813018798828, + "learning_rate": 3.748401327130174e-05, + "loss": 0.4234, + "step": 19500 + }, + { + "epoch": 0.9763543100212686, + "grad_norm": 0.044587355107069016, + "learning_rate": 3.747474560249115e-05, + "loss": 0.5297, + "step": 19510 + }, + { + "epoch": 0.9768547479044164, + "grad_norm": 0.21074265241622925, + "learning_rate": 3.746547793368057e-05, + "loss": 0.3204, + "step": 19520 + }, + { + "epoch": 0.9773551857875641, + "grad_norm": 2.997107982635498, + "learning_rate": 3.745621026486998e-05, + "loss": 1.129, + "step": 19530 + }, + { + "epoch": 0.9778556236707119, + "grad_norm": 11.652548789978027, + "learning_rate": 3.7446942596059386e-05, + "loss": 0.1237, + "step": 19540 + }, + { + "epoch": 0.9783560615538597, + "grad_norm": 4.552567005157471, + "learning_rate": 3.74376749272488e-05, + "loss": 0.1692, + "step": 19550 + }, + { + "epoch": 0.9788564994370074, + "grad_norm": 0.030116289854049683, + "learning_rate": 3.742840725843822e-05, + "loss": 0.5077, + "step": 19560 + }, + { + "epoch": 0.9793569373201552, + "grad_norm": 0.025932997465133667, + "learning_rate": 3.7419139589627626e-05, + "loss": 0.3042, + "step": 19570 + }, + { + "epoch": 0.9798573752033028, + "grad_norm": 14.95019245147705, + "learning_rate": 3.740987192081704e-05, + "loss": 0.2868, + "step": 19580 + }, + { + "epoch": 0.9803578130864506, + "grad_norm": 2.139940023422241, + "learning_rate": 3.740060425200646e-05, + "loss": 0.1562, + "step": 19590 + }, + { + "epoch": 0.9808582509695984, + "grad_norm": 0.00541616789996624, + "learning_rate": 3.7391336583195866e-05, + "loss": 0.211, + "step": 19600 + }, + { + "epoch": 0.9813586888527461, + "grad_norm": 24.887399673461914, + "learning_rate": 3.7382068914385275e-05, + "loss": 0.2839, + "step": 19610 + }, + { + "epoch": 0.9818591267358939, + "grad_norm": 0.029818901792168617, + "learning_rate": 3.737280124557469e-05, + "loss": 0.3835, + "step": 19620 + }, + { + "epoch": 0.9823595646190416, + "grad_norm": 0.924280047416687, + "learning_rate": 3.7363533576764106e-05, + "loss": 0.4373, + "step": 19630 + }, + { + "epoch": 0.9828600025021894, + "grad_norm": 19.928794860839844, + "learning_rate": 3.7354265907953515e-05, + "loss": 0.5919, + "step": 19640 + }, + { + "epoch": 0.9833604403853372, + "grad_norm": 72.38079833984375, + "learning_rate": 3.7344998239142924e-05, + "loss": 0.2233, + "step": 19650 + }, + { + "epoch": 0.9838608782684849, + "grad_norm": 0.2221909910440445, + "learning_rate": 3.733573057033234e-05, + "loss": 0.3388, + "step": 19660 + }, + { + "epoch": 0.9843613161516327, + "grad_norm": 68.89669036865234, + "learning_rate": 3.7326462901521755e-05, + "loss": 0.9216, + "step": 19670 + }, + { + "epoch": 0.9848617540347804, + "grad_norm": 0.049460507929325104, + "learning_rate": 3.7317195232711164e-05, + "loss": 0.2536, + "step": 19680 + }, + { + "epoch": 0.9853621919179282, + "grad_norm": 181.77613830566406, + "learning_rate": 3.730792756390058e-05, + "loss": 0.6342, + "step": 19690 + }, + { + "epoch": 0.985862629801076, + "grad_norm": 0.024599306285381317, + "learning_rate": 3.7298659895089995e-05, + "loss": 0.5999, + "step": 19700 + }, + { + "epoch": 0.9863630676842237, + "grad_norm": 2.3116860389709473, + "learning_rate": 3.7289392226279404e-05, + "loss": 0.4032, + "step": 19710 + }, + { + "epoch": 0.9868635055673715, + "grad_norm": 7.086540222167969, + "learning_rate": 3.728012455746881e-05, + "loss": 0.5074, + "step": 19720 + }, + { + "epoch": 0.9873639434505193, + "grad_norm": 11.952322006225586, + "learning_rate": 3.727085688865823e-05, + "loss": 0.1937, + "step": 19730 + }, + { + "epoch": 0.9878643813336669, + "grad_norm": 0.2907423973083496, + "learning_rate": 3.7261589219847644e-05, + "loss": 0.2585, + "step": 19740 + }, + { + "epoch": 0.9883648192168147, + "grad_norm": 0.01930658146739006, + "learning_rate": 3.725232155103705e-05, + "loss": 0.3818, + "step": 19750 + }, + { + "epoch": 0.9888652570999624, + "grad_norm": 8.52457046508789, + "learning_rate": 3.724305388222646e-05, + "loss": 0.6882, + "step": 19760 + }, + { + "epoch": 0.9893656949831102, + "grad_norm": 0.5161027312278748, + "learning_rate": 3.723378621341588e-05, + "loss": 0.2128, + "step": 19770 + }, + { + "epoch": 0.989866132866258, + "grad_norm": 0.006232169456779957, + "learning_rate": 3.7224518544605293e-05, + "loss": 0.4406, + "step": 19780 + }, + { + "epoch": 0.9903665707494057, + "grad_norm": 64.90505981445312, + "learning_rate": 3.72152508757947e-05, + "loss": 0.2814, + "step": 19790 + }, + { + "epoch": 0.9908670086325535, + "grad_norm": 2.5159876346588135, + "learning_rate": 3.720598320698412e-05, + "loss": 0.3238, + "step": 19800 + }, + { + "epoch": 0.9913674465157012, + "grad_norm": 16.809404373168945, + "learning_rate": 3.7196715538173534e-05, + "loss": 0.4009, + "step": 19810 + }, + { + "epoch": 0.991867884398849, + "grad_norm": 0.11859504878520966, + "learning_rate": 3.718744786936294e-05, + "loss": 0.2869, + "step": 19820 + }, + { + "epoch": 0.9923683222819968, + "grad_norm": 0.04114588350057602, + "learning_rate": 3.717818020055235e-05, + "loss": 0.5167, + "step": 19830 + }, + { + "epoch": 0.9928687601651445, + "grad_norm": 4.334486484527588, + "learning_rate": 3.716891253174177e-05, + "loss": 0.169, + "step": 19840 + }, + { + "epoch": 0.9933691980482923, + "grad_norm": 62.102596282958984, + "learning_rate": 3.715964486293118e-05, + "loss": 0.4144, + "step": 19850 + }, + { + "epoch": 0.9938696359314401, + "grad_norm": 3.3299267292022705, + "learning_rate": 3.715037719412059e-05, + "loss": 0.833, + "step": 19860 + }, + { + "epoch": 0.9943700738145878, + "grad_norm": 0.007842205464839935, + "learning_rate": 3.714110952531e-05, + "loss": 0.2973, + "step": 19870 + }, + { + "epoch": 0.9948705116977355, + "grad_norm": 0.060196686536073685, + "learning_rate": 3.7131841856499416e-05, + "loss": 0.1792, + "step": 19880 + }, + { + "epoch": 0.9953709495808832, + "grad_norm": 26.259296417236328, + "learning_rate": 3.712257418768883e-05, + "loss": 0.484, + "step": 19890 + }, + { + "epoch": 0.995871387464031, + "grad_norm": 53.2659912109375, + "learning_rate": 3.711330651887824e-05, + "loss": 0.4735, + "step": 19900 + }, + { + "epoch": 0.9963718253471788, + "grad_norm": 13.93249225616455, + "learning_rate": 3.7104038850067656e-05, + "loss": 0.1987, + "step": 19910 + }, + { + "epoch": 0.9968722632303265, + "grad_norm": 45.78651428222656, + "learning_rate": 3.709477118125707e-05, + "loss": 0.5651, + "step": 19920 + }, + { + "epoch": 0.9973727011134743, + "grad_norm": 75.74381256103516, + "learning_rate": 3.708550351244648e-05, + "loss": 0.2648, + "step": 19930 + }, + { + "epoch": 0.997873138996622, + "grad_norm": 0.02960154041647911, + "learning_rate": 3.707623584363589e-05, + "loss": 0.2597, + "step": 19940 + }, + { + "epoch": 0.9983735768797698, + "grad_norm": 9.352913856506348, + "learning_rate": 3.7066968174825305e-05, + "loss": 0.2141, + "step": 19950 + }, + { + "epoch": 0.9988740147629176, + "grad_norm": 47.225746154785156, + "learning_rate": 3.705770050601472e-05, + "loss": 0.4026, + "step": 19960 + }, + { + "epoch": 0.9993744526460653, + "grad_norm": 0.7328864336013794, + "learning_rate": 3.7048432837204136e-05, + "loss": 0.1787, + "step": 19970 + }, + { + "epoch": 0.9998748905292131, + "grad_norm": 128.43617248535156, + "learning_rate": 3.703916516839354e-05, + "loss": 0.318, + "step": 19980 + }, + { + "epoch": 0.9999749781058426, + "eval_accuracy": 0.9417923891015537, + "eval_loss": 0.25046074390411377, + "eval_runtime": 151.6234, + "eval_samples_per_second": 58.579, + "eval_steps_per_second": 58.579, + "step": 19982 + }, + { + "epoch": 1.0003753284123609, + "grad_norm": 1.2421878576278687, + "learning_rate": 3.7029897499582954e-05, + "loss": 0.2075, + "step": 19990 + }, + { + "epoch": 1.0008757662955086, + "grad_norm": 33.96245193481445, + "learning_rate": 3.702062983077237e-05, + "loss": 0.6454, + "step": 20000 + }, + { + "epoch": 1.0013762041786562, + "grad_norm": 0.039905477315187454, + "learning_rate": 3.7011362161961785e-05, + "loss": 0.1548, + "step": 20010 + }, + { + "epoch": 1.0018766420618042, + "grad_norm": 13.83979320526123, + "learning_rate": 3.7002094493151194e-05, + "loss": 0.4257, + "step": 20020 + }, + { + "epoch": 1.0023770799449518, + "grad_norm": 31.137575149536133, + "learning_rate": 3.699282682434061e-05, + "loss": 0.3281, + "step": 20030 + }, + { + "epoch": 1.0028775178280995, + "grad_norm": 35.036842346191406, + "learning_rate": 3.698355915553002e-05, + "loss": 0.3518, + "step": 20040 + }, + { + "epoch": 1.0033779557112474, + "grad_norm": 0.07425834238529205, + "learning_rate": 3.6974291486719434e-05, + "loss": 0.3747, + "step": 20050 + }, + { + "epoch": 1.0038783935943951, + "grad_norm": 0.05867461860179901, + "learning_rate": 3.696502381790884e-05, + "loss": 0.1538, + "step": 20060 + }, + { + "epoch": 1.0043788314775428, + "grad_norm": 49.54631805419922, + "learning_rate": 3.695575614909826e-05, + "loss": 1.0513, + "step": 20070 + }, + { + "epoch": 1.0048792693606905, + "grad_norm": 56.13655471801758, + "learning_rate": 3.6946488480287674e-05, + "loss": 0.4868, + "step": 20080 + }, + { + "epoch": 1.0053797072438384, + "grad_norm": 0.005553265567868948, + "learning_rate": 3.693722081147708e-05, + "loss": 0.0428, + "step": 20090 + }, + { + "epoch": 1.005880145126986, + "grad_norm": 3.3527801036834717, + "learning_rate": 3.692795314266649e-05, + "loss": 0.2224, + "step": 20100 + }, + { + "epoch": 1.0063805830101338, + "grad_norm": 0.08514099568128586, + "learning_rate": 3.691868547385591e-05, + "loss": 0.1209, + "step": 20110 + }, + { + "epoch": 1.0068810208932817, + "grad_norm": 0.7387937903404236, + "learning_rate": 3.690941780504532e-05, + "loss": 0.5928, + "step": 20120 + }, + { + "epoch": 1.0073814587764294, + "grad_norm": 25.419692993164062, + "learning_rate": 3.690015013623473e-05, + "loss": 0.6303, + "step": 20130 + }, + { + "epoch": 1.007881896659577, + "grad_norm": 3.010554075241089, + "learning_rate": 3.689088246742415e-05, + "loss": 0.6, + "step": 20140 + }, + { + "epoch": 1.008382334542725, + "grad_norm": 63.097293853759766, + "learning_rate": 3.688161479861356e-05, + "loss": 0.4368, + "step": 20150 + }, + { + "epoch": 1.0088827724258727, + "grad_norm": 67.81033325195312, + "learning_rate": 3.687234712980297e-05, + "loss": 0.9616, + "step": 20160 + }, + { + "epoch": 1.0093832103090203, + "grad_norm": 41.367305755615234, + "learning_rate": 3.686307946099238e-05, + "loss": 0.0621, + "step": 20170 + }, + { + "epoch": 1.0098836481921682, + "grad_norm": 30.842308044433594, + "learning_rate": 3.68538117921818e-05, + "loss": 1.0226, + "step": 20180 + }, + { + "epoch": 1.010384086075316, + "grad_norm": 40.416175842285156, + "learning_rate": 3.684454412337121e-05, + "loss": 0.4497, + "step": 20190 + }, + { + "epoch": 1.0108845239584636, + "grad_norm": 25.52011489868164, + "learning_rate": 3.683527645456062e-05, + "loss": 0.4247, + "step": 20200 + }, + { + "epoch": 1.0113849618416113, + "grad_norm": 1.0441486835479736, + "learning_rate": 3.682600878575003e-05, + "loss": 0.3691, + "step": 20210 + }, + { + "epoch": 1.0118853997247592, + "grad_norm": 30.670791625976562, + "learning_rate": 3.6816741116939446e-05, + "loss": 0.2409, + "step": 20220 + }, + { + "epoch": 1.012385837607907, + "grad_norm": 26.86040687561035, + "learning_rate": 3.680747344812886e-05, + "loss": 0.1865, + "step": 20230 + }, + { + "epoch": 1.0128862754910546, + "grad_norm": 0.34046390652656555, + "learning_rate": 3.679820577931827e-05, + "loss": 0.2423, + "step": 20240 + }, + { + "epoch": 1.0133867133742025, + "grad_norm": 0.3886120021343231, + "learning_rate": 3.6788938110507686e-05, + "loss": 0.235, + "step": 20250 + }, + { + "epoch": 1.0138871512573502, + "grad_norm": 5.4685235023498535, + "learning_rate": 3.6779670441697095e-05, + "loss": 0.2532, + "step": 20260 + }, + { + "epoch": 1.0143875891404979, + "grad_norm": 10.701695442199707, + "learning_rate": 3.677040277288651e-05, + "loss": 0.2947, + "step": 20270 + }, + { + "epoch": 1.0148880270236458, + "grad_norm": 21.304523468017578, + "learning_rate": 3.676113510407592e-05, + "loss": 0.4666, + "step": 20280 + }, + { + "epoch": 1.0153884649067935, + "grad_norm": 4.725694179534912, + "learning_rate": 3.6751867435265335e-05, + "loss": 0.3963, + "step": 20290 + }, + { + "epoch": 1.0158889027899412, + "grad_norm": 134.70172119140625, + "learning_rate": 3.674259976645475e-05, + "loss": 0.5623, + "step": 20300 + }, + { + "epoch": 1.016389340673089, + "grad_norm": 4.337524890899658, + "learning_rate": 3.6733332097644166e-05, + "loss": 0.4718, + "step": 20310 + }, + { + "epoch": 1.0168897785562367, + "grad_norm": 42.056121826171875, + "learning_rate": 3.672406442883357e-05, + "loss": 0.4088, + "step": 20320 + }, + { + "epoch": 1.0173902164393844, + "grad_norm": 20.18525505065918, + "learning_rate": 3.6714796760022984e-05, + "loss": 0.3229, + "step": 20330 + }, + { + "epoch": 1.0178906543225321, + "grad_norm": 79.82831573486328, + "learning_rate": 3.67055290912124e-05, + "loss": 0.5027, + "step": 20340 + }, + { + "epoch": 1.01839109220568, + "grad_norm": 42.94245529174805, + "learning_rate": 3.6696261422401815e-05, + "loss": 0.7046, + "step": 20350 + }, + { + "epoch": 1.0188915300888277, + "grad_norm": 4.815860748291016, + "learning_rate": 3.6686993753591224e-05, + "loss": 0.1917, + "step": 20360 + }, + { + "epoch": 1.0193919679719754, + "grad_norm": 0.010738025419414043, + "learning_rate": 3.667772608478063e-05, + "loss": 0.21, + "step": 20370 + }, + { + "epoch": 1.0198924058551233, + "grad_norm": 26.172155380249023, + "learning_rate": 3.666845841597005e-05, + "loss": 0.6293, + "step": 20380 + }, + { + "epoch": 1.020392843738271, + "grad_norm": 0.02080618217587471, + "learning_rate": 3.6659190747159464e-05, + "loss": 0.6174, + "step": 20390 + }, + { + "epoch": 1.0208932816214187, + "grad_norm": 14.089874267578125, + "learning_rate": 3.664992307834887e-05, + "loss": 0.5254, + "step": 20400 + }, + { + "epoch": 1.0213937195045666, + "grad_norm": 49.68323516845703, + "learning_rate": 3.664065540953829e-05, + "loss": 0.4612, + "step": 20410 + }, + { + "epoch": 1.0218941573877143, + "grad_norm": 59.15285873413086, + "learning_rate": 3.6631387740727704e-05, + "loss": 0.4797, + "step": 20420 + }, + { + "epoch": 1.022394595270862, + "grad_norm": 91.16770935058594, + "learning_rate": 3.662212007191711e-05, + "loss": 0.4567, + "step": 20430 + }, + { + "epoch": 1.0228950331540099, + "grad_norm": 1.485797643661499, + "learning_rate": 3.661285240310652e-05, + "loss": 0.4529, + "step": 20440 + }, + { + "epoch": 1.0233954710371576, + "grad_norm": 0.7067705392837524, + "learning_rate": 3.660358473429594e-05, + "loss": 0.1632, + "step": 20450 + }, + { + "epoch": 1.0238959089203052, + "grad_norm": 25.58391761779785, + "learning_rate": 3.659431706548535e-05, + "loss": 0.2372, + "step": 20460 + }, + { + "epoch": 1.024396346803453, + "grad_norm": 0.7052942514419556, + "learning_rate": 3.658504939667476e-05, + "loss": 0.3437, + "step": 20470 + }, + { + "epoch": 1.0248967846866008, + "grad_norm": 21.142230987548828, + "learning_rate": 3.657578172786417e-05, + "loss": 0.1222, + "step": 20480 + }, + { + "epoch": 1.0253972225697485, + "grad_norm": 7.73401403427124, + "learning_rate": 3.656651405905359e-05, + "loss": 0.2138, + "step": 20490 + }, + { + "epoch": 1.0258976604528962, + "grad_norm": 0.6693430542945862, + "learning_rate": 3.6557246390243e-05, + "loss": 0.3185, + "step": 20500 + }, + { + "epoch": 1.0263980983360441, + "grad_norm": 5.498892784118652, + "learning_rate": 3.654797872143241e-05, + "loss": 0.6349, + "step": 20510 + }, + { + "epoch": 1.0268985362191918, + "grad_norm": 14.990209579467773, + "learning_rate": 3.653871105262183e-05, + "loss": 0.2301, + "step": 20520 + }, + { + "epoch": 1.0273989741023395, + "grad_norm": 0.07038919627666473, + "learning_rate": 3.652944338381124e-05, + "loss": 0.5791, + "step": 20530 + }, + { + "epoch": 1.0278994119854874, + "grad_norm": 49.250858306884766, + "learning_rate": 3.652017571500065e-05, + "loss": 0.5697, + "step": 20540 + }, + { + "epoch": 1.028399849868635, + "grad_norm": 0.07979491353034973, + "learning_rate": 3.651090804619006e-05, + "loss": 0.2617, + "step": 20550 + }, + { + "epoch": 1.0289002877517828, + "grad_norm": 0.30981624126434326, + "learning_rate": 3.6501640377379476e-05, + "loss": 0.3933, + "step": 20560 + }, + { + "epoch": 1.0294007256349305, + "grad_norm": 70.79492950439453, + "learning_rate": 3.649237270856889e-05, + "loss": 0.5371, + "step": 20570 + }, + { + "epoch": 1.0299011635180784, + "grad_norm": 13.74642276763916, + "learning_rate": 3.64831050397583e-05, + "loss": 0.2953, + "step": 20580 + }, + { + "epoch": 1.030401601401226, + "grad_norm": 15.66797924041748, + "learning_rate": 3.647383737094771e-05, + "loss": 0.4134, + "step": 20590 + }, + { + "epoch": 1.0309020392843737, + "grad_norm": 0.03969704732298851, + "learning_rate": 3.6464569702137125e-05, + "loss": 0.4355, + "step": 20600 + }, + { + "epoch": 1.0314024771675216, + "grad_norm": 150.8409881591797, + "learning_rate": 3.645530203332654e-05, + "loss": 0.7991, + "step": 20610 + }, + { + "epoch": 1.0319029150506693, + "grad_norm": 0.5148462653160095, + "learning_rate": 3.644603436451595e-05, + "loss": 0.4103, + "step": 20620 + }, + { + "epoch": 1.032403352933817, + "grad_norm": 42.93437576293945, + "learning_rate": 3.6436766695705365e-05, + "loss": 0.6143, + "step": 20630 + }, + { + "epoch": 1.032903790816965, + "grad_norm": 1.8218389749526978, + "learning_rate": 3.642749902689478e-05, + "loss": 0.4016, + "step": 20640 + }, + { + "epoch": 1.0334042287001126, + "grad_norm": 0.03079613670706749, + "learning_rate": 3.641823135808419e-05, + "loss": 0.1233, + "step": 20650 + }, + { + "epoch": 1.0339046665832603, + "grad_norm": 0.08825269341468811, + "learning_rate": 3.64089636892736e-05, + "loss": 0.7803, + "step": 20660 + }, + { + "epoch": 1.0344051044664082, + "grad_norm": 0.25848686695098877, + "learning_rate": 3.6399696020463014e-05, + "loss": 0.0848, + "step": 20670 + }, + { + "epoch": 1.034905542349556, + "grad_norm": 4.972283363342285, + "learning_rate": 3.639042835165243e-05, + "loss": 0.2907, + "step": 20680 + }, + { + "epoch": 1.0354059802327036, + "grad_norm": 0.09531142562627792, + "learning_rate": 3.638116068284184e-05, + "loss": 0.7756, + "step": 20690 + }, + { + "epoch": 1.0359064181158513, + "grad_norm": 1.3940627574920654, + "learning_rate": 3.637189301403125e-05, + "loss": 0.4732, + "step": 20700 + }, + { + "epoch": 1.0364068559989992, + "grad_norm": 97.75857543945312, + "learning_rate": 3.636262534522066e-05, + "loss": 0.2429, + "step": 20710 + }, + { + "epoch": 1.0369072938821469, + "grad_norm": 10.40120792388916, + "learning_rate": 3.635335767641008e-05, + "loss": 0.1889, + "step": 20720 + }, + { + "epoch": 1.0374077317652945, + "grad_norm": 163.02935791015625, + "learning_rate": 3.634409000759949e-05, + "loss": 0.6571, + "step": 20730 + }, + { + "epoch": 1.0379081696484425, + "grad_norm": 0.3793541491031647, + "learning_rate": 3.63348223387889e-05, + "loss": 0.3889, + "step": 20740 + }, + { + "epoch": 1.0384086075315901, + "grad_norm": 0.24115069210529327, + "learning_rate": 3.632555466997832e-05, + "loss": 0.8141, + "step": 20750 + }, + { + "epoch": 1.0389090454147378, + "grad_norm": 4.896650791168213, + "learning_rate": 3.631628700116773e-05, + "loss": 0.2661, + "step": 20760 + }, + { + "epoch": 1.0394094832978857, + "grad_norm": 19.395082473754883, + "learning_rate": 3.6307019332357136e-05, + "loss": 0.2856, + "step": 20770 + }, + { + "epoch": 1.0399099211810334, + "grad_norm": 3.1020162105560303, + "learning_rate": 3.629775166354655e-05, + "loss": 0.539, + "step": 20780 + }, + { + "epoch": 1.0404103590641811, + "grad_norm": 0.09339740127325058, + "learning_rate": 3.628848399473597e-05, + "loss": 0.2148, + "step": 20790 + }, + { + "epoch": 1.040910796947329, + "grad_norm": 0.6045913100242615, + "learning_rate": 3.627921632592538e-05, + "loss": 0.3429, + "step": 20800 + }, + { + "epoch": 1.0414112348304767, + "grad_norm": 0.07216734439134598, + "learning_rate": 3.6269948657114785e-05, + "loss": 0.5485, + "step": 20810 + }, + { + "epoch": 1.0419116727136244, + "grad_norm": 0.07610788196325302, + "learning_rate": 3.62606809883042e-05, + "loss": 0.5027, + "step": 20820 + }, + { + "epoch": 1.042412110596772, + "grad_norm": 32.65891647338867, + "learning_rate": 3.6251413319493617e-05, + "loss": 0.1086, + "step": 20830 + }, + { + "epoch": 1.04291254847992, + "grad_norm": 4.064591884613037, + "learning_rate": 3.624214565068303e-05, + "loss": 0.1803, + "step": 20840 + }, + { + "epoch": 1.0434129863630677, + "grad_norm": 0.022681767120957375, + "learning_rate": 3.623287798187244e-05, + "loss": 0.1989, + "step": 20850 + }, + { + "epoch": 1.0439134242462154, + "grad_norm": 21.771053314208984, + "learning_rate": 3.622361031306186e-05, + "loss": 0.5994, + "step": 20860 + }, + { + "epoch": 1.0444138621293633, + "grad_norm": 20.23933219909668, + "learning_rate": 3.6214342644251266e-05, + "loss": 0.2368, + "step": 20870 + }, + { + "epoch": 1.044914300012511, + "grad_norm": 0.05501705780625343, + "learning_rate": 3.620507497544068e-05, + "loss": 0.2035, + "step": 20880 + }, + { + "epoch": 1.0454147378956586, + "grad_norm": 0.2283785492181778, + "learning_rate": 3.619580730663009e-05, + "loss": 0.0271, + "step": 20890 + }, + { + "epoch": 1.0459151757788065, + "grad_norm": 7.508162975311279, + "learning_rate": 3.6186539637819506e-05, + "loss": 0.3984, + "step": 20900 + }, + { + "epoch": 1.0464156136619542, + "grad_norm": 0.2580775022506714, + "learning_rate": 3.617727196900892e-05, + "loss": 0.6762, + "step": 20910 + }, + { + "epoch": 1.046916051545102, + "grad_norm": 14.468338012695312, + "learning_rate": 3.616800430019833e-05, + "loss": 0.5989, + "step": 20920 + }, + { + "epoch": 1.0474164894282496, + "grad_norm": 1.0586098432540894, + "learning_rate": 3.615873663138774e-05, + "loss": 0.4243, + "step": 20930 + }, + { + "epoch": 1.0479169273113975, + "grad_norm": 0.6926461458206177, + "learning_rate": 3.6149468962577155e-05, + "loss": 0.4118, + "step": 20940 + }, + { + "epoch": 1.0484173651945452, + "grad_norm": 1.8944571018218994, + "learning_rate": 3.614020129376657e-05, + "loss": 0.2781, + "step": 20950 + }, + { + "epoch": 1.048917803077693, + "grad_norm": 50.04606628417969, + "learning_rate": 3.613093362495598e-05, + "loss": 0.3871, + "step": 20960 + }, + { + "epoch": 1.0494182409608408, + "grad_norm": 6.341501235961914, + "learning_rate": 3.6121665956145395e-05, + "loss": 0.4363, + "step": 20970 + }, + { + "epoch": 1.0499186788439885, + "grad_norm": 19.768993377685547, + "learning_rate": 3.6112398287334804e-05, + "loss": 0.6603, + "step": 20980 + }, + { + "epoch": 1.0504191167271362, + "grad_norm": 0.15770696103572845, + "learning_rate": 3.610313061852422e-05, + "loss": 0.4482, + "step": 20990 + }, + { + "epoch": 1.050919554610284, + "grad_norm": 133.0801544189453, + "learning_rate": 3.609386294971363e-05, + "loss": 0.2055, + "step": 21000 + }, + { + "epoch": 1.0514199924934318, + "grad_norm": 13.850976943969727, + "learning_rate": 3.6084595280903044e-05, + "loss": 0.4498, + "step": 21010 + }, + { + "epoch": 1.0519204303765795, + "grad_norm": 0.562079668045044, + "learning_rate": 3.607532761209246e-05, + "loss": 0.7804, + "step": 21020 + }, + { + "epoch": 1.0524208682597274, + "grad_norm": 0.08188023418188095, + "learning_rate": 3.606605994328187e-05, + "loss": 0.2622, + "step": 21030 + }, + { + "epoch": 1.052921306142875, + "grad_norm": 0.0739855021238327, + "learning_rate": 3.605679227447128e-05, + "loss": 0.584, + "step": 21040 + }, + { + "epoch": 1.0534217440260227, + "grad_norm": 0.4374350309371948, + "learning_rate": 3.604752460566069e-05, + "loss": 0.2362, + "step": 21050 + }, + { + "epoch": 1.0539221819091704, + "grad_norm": 9.73701286315918, + "learning_rate": 3.603825693685011e-05, + "loss": 0.5351, + "step": 21060 + }, + { + "epoch": 1.0544226197923183, + "grad_norm": 0.014404232613742352, + "learning_rate": 3.602898926803952e-05, + "loss": 0.2496, + "step": 21070 + }, + { + "epoch": 1.054923057675466, + "grad_norm": 0.031632717698812485, + "learning_rate": 3.601972159922893e-05, + "loss": 0.0977, + "step": 21080 + }, + { + "epoch": 1.0554234955586137, + "grad_norm": 0.13683444261550903, + "learning_rate": 3.601045393041834e-05, + "loss": 0.4797, + "step": 21090 + }, + { + "epoch": 1.0559239334417616, + "grad_norm": 17.20106315612793, + "learning_rate": 3.600118626160776e-05, + "loss": 0.496, + "step": 21100 + }, + { + "epoch": 1.0564243713249093, + "grad_norm": 71.53134155273438, + "learning_rate": 3.5991918592797166e-05, + "loss": 0.3303, + "step": 21110 + }, + { + "epoch": 1.056924809208057, + "grad_norm": 102.85562896728516, + "learning_rate": 3.598265092398658e-05, + "loss": 0.2739, + "step": 21120 + }, + { + "epoch": 1.057425247091205, + "grad_norm": 4.731956958770752, + "learning_rate": 3.5973383255176e-05, + "loss": 0.2526, + "step": 21130 + }, + { + "epoch": 1.0579256849743526, + "grad_norm": 0.02765670232474804, + "learning_rate": 3.596411558636541e-05, + "loss": 0.6431, + "step": 21140 + }, + { + "epoch": 1.0584261228575003, + "grad_norm": 0.005464776419103146, + "learning_rate": 3.5954847917554815e-05, + "loss": 0.1533, + "step": 21150 + }, + { + "epoch": 1.0589265607406482, + "grad_norm": 22.32105827331543, + "learning_rate": 3.594558024874423e-05, + "loss": 0.4518, + "step": 21160 + }, + { + "epoch": 1.0594269986237959, + "grad_norm": 36.74626922607422, + "learning_rate": 3.5936312579933646e-05, + "loss": 0.2624, + "step": 21170 + }, + { + "epoch": 1.0599274365069435, + "grad_norm": 14.8035249710083, + "learning_rate": 3.592704491112306e-05, + "loss": 0.8873, + "step": 21180 + }, + { + "epoch": 1.0604278743900912, + "grad_norm": 146.76019287109375, + "learning_rate": 3.591777724231247e-05, + "loss": 0.7491, + "step": 21190 + }, + { + "epoch": 1.0609283122732391, + "grad_norm": 35.775638580322266, + "learning_rate": 3.590850957350188e-05, + "loss": 0.8436, + "step": 21200 + }, + { + "epoch": 1.0614287501563868, + "grad_norm": 0.42339760065078735, + "learning_rate": 3.5899241904691295e-05, + "loss": 0.117, + "step": 21210 + }, + { + "epoch": 1.0619291880395345, + "grad_norm": 48.87562561035156, + "learning_rate": 3.588997423588071e-05, + "loss": 0.3109, + "step": 21220 + }, + { + "epoch": 1.0624296259226824, + "grad_norm": 0.08302287012338638, + "learning_rate": 3.588070656707012e-05, + "loss": 0.0643, + "step": 21230 + }, + { + "epoch": 1.06293006380583, + "grad_norm": 5.5869526863098145, + "learning_rate": 3.5871438898259536e-05, + "loss": 0.2241, + "step": 21240 + }, + { + "epoch": 1.0634305016889778, + "grad_norm": 0.09925032407045364, + "learning_rate": 3.586217122944895e-05, + "loss": 0.8337, + "step": 21250 + }, + { + "epoch": 1.0639309395721257, + "grad_norm": 11.061254501342773, + "learning_rate": 3.585290356063836e-05, + "loss": 0.0943, + "step": 21260 + }, + { + "epoch": 1.0644313774552734, + "grad_norm": 4.165212154388428, + "learning_rate": 3.584363589182777e-05, + "loss": 0.1899, + "step": 21270 + }, + { + "epoch": 1.064931815338421, + "grad_norm": 13.427473068237305, + "learning_rate": 3.5834368223017185e-05, + "loss": 0.0551, + "step": 21280 + }, + { + "epoch": 1.065432253221569, + "grad_norm": 18.070770263671875, + "learning_rate": 3.58251005542066e-05, + "loss": 0.1367, + "step": 21290 + }, + { + "epoch": 1.0659326911047167, + "grad_norm": 2.8669230937957764, + "learning_rate": 3.581583288539601e-05, + "loss": 0.597, + "step": 21300 + }, + { + "epoch": 1.0664331289878644, + "grad_norm": 45.01220703125, + "learning_rate": 3.580656521658542e-05, + "loss": 0.4071, + "step": 21310 + }, + { + "epoch": 1.066933566871012, + "grad_norm": 13.41971206665039, + "learning_rate": 3.5797297547774834e-05, + "loss": 0.5833, + "step": 21320 + }, + { + "epoch": 1.06743400475416, + "grad_norm": 109.77493286132812, + "learning_rate": 3.578802987896425e-05, + "loss": 0.1904, + "step": 21330 + }, + { + "epoch": 1.0679344426373076, + "grad_norm": 5.9450154304504395, + "learning_rate": 3.577876221015366e-05, + "loss": 0.1446, + "step": 21340 + }, + { + "epoch": 1.0684348805204553, + "grad_norm": 0.01745772361755371, + "learning_rate": 3.5769494541343074e-05, + "loss": 0.6713, + "step": 21350 + }, + { + "epoch": 1.0689353184036032, + "grad_norm": 0.1951235979795456, + "learning_rate": 3.576022687253249e-05, + "loss": 0.1577, + "step": 21360 + }, + { + "epoch": 1.069435756286751, + "grad_norm": 0.48157966136932373, + "learning_rate": 3.57509592037219e-05, + "loss": 0.2624, + "step": 21370 + }, + { + "epoch": 1.0699361941698986, + "grad_norm": 0.005256400443613529, + "learning_rate": 3.574169153491131e-05, + "loss": 0.739, + "step": 21380 + }, + { + "epoch": 1.0704366320530465, + "grad_norm": 96.07671356201172, + "learning_rate": 3.573242386610072e-05, + "loss": 0.5719, + "step": 21390 + }, + { + "epoch": 1.0709370699361942, + "grad_norm": 18.628339767456055, + "learning_rate": 3.572315619729014e-05, + "loss": 0.3338, + "step": 21400 + }, + { + "epoch": 1.0714375078193419, + "grad_norm": 0.06271278858184814, + "learning_rate": 3.571388852847955e-05, + "loss": 0.3927, + "step": 21410 + }, + { + "epoch": 1.0719379457024898, + "grad_norm": 1.6986862421035767, + "learning_rate": 3.5704620859668956e-05, + "loss": 0.5219, + "step": 21420 + }, + { + "epoch": 1.0724383835856375, + "grad_norm": 3.2349138259887695, + "learning_rate": 3.569535319085837e-05, + "loss": 0.2763, + "step": 21430 + }, + { + "epoch": 1.0729388214687852, + "grad_norm": 36.39708709716797, + "learning_rate": 3.568608552204779e-05, + "loss": 0.297, + "step": 21440 + }, + { + "epoch": 1.0734392593519329, + "grad_norm": 122.93270874023438, + "learning_rate": 3.5676817853237196e-05, + "loss": 0.1128, + "step": 21450 + }, + { + "epoch": 1.0739396972350808, + "grad_norm": 166.5091552734375, + "learning_rate": 3.566755018442661e-05, + "loss": 0.399, + "step": 21460 + }, + { + "epoch": 1.0744401351182284, + "grad_norm": 0.007320140954107046, + "learning_rate": 3.565828251561603e-05, + "loss": 0.313, + "step": 21470 + }, + { + "epoch": 1.0749405730013761, + "grad_norm": 0.016541041433811188, + "learning_rate": 3.5649014846805436e-05, + "loss": 0.5553, + "step": 21480 + }, + { + "epoch": 1.075441010884524, + "grad_norm": 27.722110748291016, + "learning_rate": 3.5639747177994845e-05, + "loss": 0.3785, + "step": 21490 + }, + { + "epoch": 1.0759414487676717, + "grad_norm": 0.03597983345389366, + "learning_rate": 3.563047950918426e-05, + "loss": 0.5582, + "step": 21500 + }, + { + "epoch": 1.0764418866508194, + "grad_norm": 20.42613410949707, + "learning_rate": 3.5621211840373676e-05, + "loss": 0.4322, + "step": 21510 + }, + { + "epoch": 1.0769423245339673, + "grad_norm": 0.13134805858135223, + "learning_rate": 3.5611944171563085e-05, + "loss": 0.4283, + "step": 21520 + }, + { + "epoch": 1.077442762417115, + "grad_norm": 44.69903564453125, + "learning_rate": 3.5602676502752494e-05, + "loss": 0.4428, + "step": 21530 + }, + { + "epoch": 1.0779432003002627, + "grad_norm": 1.526451826095581, + "learning_rate": 3.559340883394191e-05, + "loss": 0.4404, + "step": 21540 + }, + { + "epoch": 1.0784436381834104, + "grad_norm": 10.33216381072998, + "learning_rate": 3.5584141165131325e-05, + "loss": 0.4328, + "step": 21550 + }, + { + "epoch": 1.0789440760665583, + "grad_norm": 27.381757736206055, + "learning_rate": 3.5574873496320734e-05, + "loss": 0.2311, + "step": 21560 + }, + { + "epoch": 1.079444513949706, + "grad_norm": 0.4938812255859375, + "learning_rate": 3.556560582751015e-05, + "loss": 0.3099, + "step": 21570 + }, + { + "epoch": 1.0799449518328537, + "grad_norm": 2.8112266063690186, + "learning_rate": 3.5556338158699565e-05, + "loss": 0.482, + "step": 21580 + }, + { + "epoch": 1.0804453897160016, + "grad_norm": 72.15379333496094, + "learning_rate": 3.5547070489888974e-05, + "loss": 0.5372, + "step": 21590 + }, + { + "epoch": 1.0809458275991493, + "grad_norm": 59.87137985229492, + "learning_rate": 3.553780282107838e-05, + "loss": 0.1859, + "step": 21600 + }, + { + "epoch": 1.081446265482297, + "grad_norm": 7.953538417816162, + "learning_rate": 3.55285351522678e-05, + "loss": 0.2173, + "step": 21610 + }, + { + "epoch": 1.0819467033654449, + "grad_norm": 17.11985206604004, + "learning_rate": 3.5519267483457214e-05, + "loss": 0.6979, + "step": 21620 + }, + { + "epoch": 1.0824471412485925, + "grad_norm": 0.014919632114470005, + "learning_rate": 3.550999981464663e-05, + "loss": 0.8523, + "step": 21630 + }, + { + "epoch": 1.0829475791317402, + "grad_norm": 0.3965328335762024, + "learning_rate": 3.550073214583603e-05, + "loss": 0.2558, + "step": 21640 + }, + { + "epoch": 1.083448017014888, + "grad_norm": 14.03089427947998, + "learning_rate": 3.549146447702545e-05, + "loss": 0.4219, + "step": 21650 + }, + { + "epoch": 1.0839484548980358, + "grad_norm": 1.128022313117981, + "learning_rate": 3.5482196808214863e-05, + "loss": 0.3027, + "step": 21660 + }, + { + "epoch": 1.0844488927811835, + "grad_norm": 0.03420471027493477, + "learning_rate": 3.547292913940428e-05, + "loss": 0.3782, + "step": 21670 + }, + { + "epoch": 1.0849493306643312, + "grad_norm": 73.66120147705078, + "learning_rate": 3.546366147059369e-05, + "loss": 0.2668, + "step": 21680 + }, + { + "epoch": 1.085449768547479, + "grad_norm": 29.849895477294922, + "learning_rate": 3.5454393801783104e-05, + "loss": 0.4909, + "step": 21690 + }, + { + "epoch": 1.0859502064306268, + "grad_norm": 0.026815058663487434, + "learning_rate": 3.544512613297251e-05, + "loss": 0.3926, + "step": 21700 + }, + { + "epoch": 1.0864506443137745, + "grad_norm": 8.022578239440918, + "learning_rate": 3.543585846416193e-05, + "loss": 0.2922, + "step": 21710 + }, + { + "epoch": 1.0869510821969224, + "grad_norm": 90.13965606689453, + "learning_rate": 3.542659079535134e-05, + "loss": 0.2546, + "step": 21720 + }, + { + "epoch": 1.08745152008007, + "grad_norm": 4.727366924285889, + "learning_rate": 3.541732312654075e-05, + "loss": 0.2508, + "step": 21730 + }, + { + "epoch": 1.0879519579632178, + "grad_norm": 46.410499572753906, + "learning_rate": 3.540805545773017e-05, + "loss": 0.795, + "step": 21740 + }, + { + "epoch": 1.0884523958463657, + "grad_norm": 1.3100136518478394, + "learning_rate": 3.539878778891958e-05, + "loss": 0.2186, + "step": 21750 + }, + { + "epoch": 1.0889528337295133, + "grad_norm": 18.966442108154297, + "learning_rate": 3.5389520120108986e-05, + "loss": 0.1409, + "step": 21760 + }, + { + "epoch": 1.089453271612661, + "grad_norm": 30.396093368530273, + "learning_rate": 3.53802524512984e-05, + "loss": 0.1994, + "step": 21770 + }, + { + "epoch": 1.0899537094958087, + "grad_norm": 109.76573181152344, + "learning_rate": 3.537098478248782e-05, + "loss": 0.5214, + "step": 21780 + }, + { + "epoch": 1.0904541473789566, + "grad_norm": 41.38825225830078, + "learning_rate": 3.5361717113677226e-05, + "loss": 0.7516, + "step": 21790 + }, + { + "epoch": 1.0909545852621043, + "grad_norm": 91.29503631591797, + "learning_rate": 3.535244944486664e-05, + "loss": 0.2729, + "step": 21800 + }, + { + "epoch": 1.091455023145252, + "grad_norm": 7.726534366607666, + "learning_rate": 3.534318177605605e-05, + "loss": 0.18, + "step": 21810 + }, + { + "epoch": 1.0919554610284, + "grad_norm": 3.1051764488220215, + "learning_rate": 3.5333914107245466e-05, + "loss": 0.3487, + "step": 21820 + }, + { + "epoch": 1.0924558989115476, + "grad_norm": 0.30300888419151306, + "learning_rate": 3.5324646438434875e-05, + "loss": 0.7267, + "step": 21830 + }, + { + "epoch": 1.0929563367946953, + "grad_norm": 0.10431648045778275, + "learning_rate": 3.531537876962429e-05, + "loss": 0.5228, + "step": 21840 + }, + { + "epoch": 1.0934567746778432, + "grad_norm": 0.026754938066005707, + "learning_rate": 3.5306111100813706e-05, + "loss": 0.5246, + "step": 21850 + }, + { + "epoch": 1.0939572125609909, + "grad_norm": 5.9372758865356445, + "learning_rate": 3.5296843432003115e-05, + "loss": 0.1748, + "step": 21860 + }, + { + "epoch": 1.0944576504441386, + "grad_norm": 40.73536682128906, + "learning_rate": 3.5287575763192524e-05, + "loss": 0.3384, + "step": 21870 + }, + { + "epoch": 1.0949580883272865, + "grad_norm": 0.11359308660030365, + "learning_rate": 3.527830809438194e-05, + "loss": 0.2842, + "step": 21880 + }, + { + "epoch": 1.0954585262104342, + "grad_norm": 123.06280517578125, + "learning_rate": 3.5269040425571355e-05, + "loss": 0.7234, + "step": 21890 + }, + { + "epoch": 1.0959589640935818, + "grad_norm": 0.025119056925177574, + "learning_rate": 3.5259772756760764e-05, + "loss": 0.3524, + "step": 21900 + }, + { + "epoch": 1.0964594019767295, + "grad_norm": 0.5792320966720581, + "learning_rate": 3.525050508795018e-05, + "loss": 0.8565, + "step": 21910 + }, + { + "epoch": 1.0969598398598774, + "grad_norm": 77.48947143554688, + "learning_rate": 3.524123741913959e-05, + "loss": 0.5046, + "step": 21920 + }, + { + "epoch": 1.0974602777430251, + "grad_norm": 33.4445686340332, + "learning_rate": 3.5231969750329004e-05, + "loss": 0.3367, + "step": 21930 + }, + { + "epoch": 1.0979607156261728, + "grad_norm": 0.6673843264579773, + "learning_rate": 3.522270208151841e-05, + "loss": 0.0451, + "step": 21940 + }, + { + "epoch": 1.0984611535093207, + "grad_norm": 38.74599075317383, + "learning_rate": 3.521343441270783e-05, + "loss": 0.2878, + "step": 21950 + }, + { + "epoch": 1.0989615913924684, + "grad_norm": 43.314453125, + "learning_rate": 3.5204166743897244e-05, + "loss": 0.4671, + "step": 21960 + }, + { + "epoch": 1.099462029275616, + "grad_norm": 6.9864702224731445, + "learning_rate": 3.519489907508666e-05, + "loss": 0.2945, + "step": 21970 + }, + { + "epoch": 1.099962467158764, + "grad_norm": 13.790658950805664, + "learning_rate": 3.518563140627606e-05, + "loss": 0.484, + "step": 21980 + }, + { + "epoch": 1.1004629050419117, + "grad_norm": 68.52758026123047, + "learning_rate": 3.517636373746548e-05, + "loss": 0.6454, + "step": 21990 + }, + { + "epoch": 1.1009633429250594, + "grad_norm": 4.927271366119385, + "learning_rate": 3.516709606865489e-05, + "loss": 0.5965, + "step": 22000 + }, + { + "epoch": 1.1014637808082073, + "grad_norm": 84.09794616699219, + "learning_rate": 3.515782839984431e-05, + "loss": 0.2691, + "step": 22010 + }, + { + "epoch": 1.101964218691355, + "grad_norm": 0.07196968048810959, + "learning_rate": 3.514856073103372e-05, + "loss": 0.2416, + "step": 22020 + }, + { + "epoch": 1.1024646565745027, + "grad_norm": 72.51871490478516, + "learning_rate": 3.513929306222313e-05, + "loss": 0.3087, + "step": 22030 + }, + { + "epoch": 1.1029650944576503, + "grad_norm": 5.939698219299316, + "learning_rate": 3.513002539341254e-05, + "loss": 0.493, + "step": 22040 + }, + { + "epoch": 1.1034655323407982, + "grad_norm": 0.2546917498111725, + "learning_rate": 3.512075772460196e-05, + "loss": 0.4375, + "step": 22050 + }, + { + "epoch": 1.103965970223946, + "grad_norm": 31.43694496154785, + "learning_rate": 3.511149005579137e-05, + "loss": 0.2734, + "step": 22060 + }, + { + "epoch": 1.1044664081070936, + "grad_norm": 82.52342987060547, + "learning_rate": 3.510222238698078e-05, + "loss": 0.3874, + "step": 22070 + }, + { + "epoch": 1.1049668459902415, + "grad_norm": 20.49235725402832, + "learning_rate": 3.50929547181702e-05, + "loss": 0.4899, + "step": 22080 + }, + { + "epoch": 1.1054672838733892, + "grad_norm": 57.54767608642578, + "learning_rate": 3.508368704935961e-05, + "loss": 0.1008, + "step": 22090 + }, + { + "epoch": 1.105967721756537, + "grad_norm": 88.74430847167969, + "learning_rate": 3.5074419380549016e-05, + "loss": 0.2707, + "step": 22100 + }, + { + "epoch": 1.1064681596396848, + "grad_norm": 8.152602195739746, + "learning_rate": 3.506515171173843e-05, + "loss": 0.4498, + "step": 22110 + }, + { + "epoch": 1.1069685975228325, + "grad_norm": 39.84183120727539, + "learning_rate": 3.505588404292785e-05, + "loss": 0.4556, + "step": 22120 + }, + { + "epoch": 1.1074690354059802, + "grad_norm": 16.637357711791992, + "learning_rate": 3.5046616374117256e-05, + "loss": 0.3082, + "step": 22130 + }, + { + "epoch": 1.107969473289128, + "grad_norm": 103.56565856933594, + "learning_rate": 3.5037348705306665e-05, + "loss": 0.4297, + "step": 22140 + }, + { + "epoch": 1.1084699111722758, + "grad_norm": 15.27881145477295, + "learning_rate": 3.502808103649608e-05, + "loss": 0.4371, + "step": 22150 + }, + { + "epoch": 1.1089703490554235, + "grad_norm": 8.837406158447266, + "learning_rate": 3.5018813367685496e-05, + "loss": 0.8074, + "step": 22160 + }, + { + "epoch": 1.1094707869385712, + "grad_norm": 0.2058965563774109, + "learning_rate": 3.5009545698874905e-05, + "loss": 0.6548, + "step": 22170 + }, + { + "epoch": 1.109971224821719, + "grad_norm": 35.51276779174805, + "learning_rate": 3.500027803006432e-05, + "loss": 0.6747, + "step": 22180 + }, + { + "epoch": 1.1104716627048667, + "grad_norm": 4.879725933074951, + "learning_rate": 3.4991010361253736e-05, + "loss": 0.7341, + "step": 22190 + }, + { + "epoch": 1.1109721005880144, + "grad_norm": 26.494829177856445, + "learning_rate": 3.4981742692443145e-05, + "loss": 0.3125, + "step": 22200 + }, + { + "epoch": 1.1114725384711623, + "grad_norm": 0.1387578845024109, + "learning_rate": 3.4972475023632554e-05, + "loss": 0.2796, + "step": 22210 + }, + { + "epoch": 1.11197297635431, + "grad_norm": 16.484018325805664, + "learning_rate": 3.496320735482197e-05, + "loss": 0.3015, + "step": 22220 + }, + { + "epoch": 1.1124734142374577, + "grad_norm": 51.1369743347168, + "learning_rate": 3.4953939686011385e-05, + "loss": 0.4933, + "step": 22230 + }, + { + "epoch": 1.1129738521206056, + "grad_norm": 0.010395490564405918, + "learning_rate": 3.4944672017200794e-05, + "loss": 0.7002, + "step": 22240 + }, + { + "epoch": 1.1134742900037533, + "grad_norm": 0.0866616815328598, + "learning_rate": 3.49354043483902e-05, + "loss": 0.5013, + "step": 22250 + }, + { + "epoch": 1.113974727886901, + "grad_norm": 0.06712832301855087, + "learning_rate": 3.492613667957962e-05, + "loss": 0.3357, + "step": 22260 + }, + { + "epoch": 1.114475165770049, + "grad_norm": 25.021955490112305, + "learning_rate": 3.4916869010769034e-05, + "loss": 0.3199, + "step": 22270 + }, + { + "epoch": 1.1149756036531966, + "grad_norm": 30.0102596282959, + "learning_rate": 3.490760134195844e-05, + "loss": 0.232, + "step": 22280 + }, + { + "epoch": 1.1154760415363443, + "grad_norm": 2.501948595046997, + "learning_rate": 3.489833367314786e-05, + "loss": 0.292, + "step": 22290 + }, + { + "epoch": 1.115976479419492, + "grad_norm": 22.19630241394043, + "learning_rate": 3.4889066004337274e-05, + "loss": 0.3321, + "step": 22300 + }, + { + "epoch": 1.1164769173026399, + "grad_norm": 4.1296186447143555, + "learning_rate": 3.487979833552668e-05, + "loss": 0.6529, + "step": 22310 + }, + { + "epoch": 1.1169773551857876, + "grad_norm": 1.2066757678985596, + "learning_rate": 3.487053066671609e-05, + "loss": 0.4006, + "step": 22320 + }, + { + "epoch": 1.1174777930689352, + "grad_norm": 16.024152755737305, + "learning_rate": 3.486126299790551e-05, + "loss": 0.2104, + "step": 22330 + }, + { + "epoch": 1.1179782309520832, + "grad_norm": 2.209291934967041, + "learning_rate": 3.485199532909492e-05, + "loss": 0.4677, + "step": 22340 + }, + { + "epoch": 1.1184786688352308, + "grad_norm": 19.992082595825195, + "learning_rate": 3.484272766028434e-05, + "loss": 0.3672, + "step": 22350 + }, + { + "epoch": 1.1189791067183785, + "grad_norm": 0.6711990833282471, + "learning_rate": 3.483345999147374e-05, + "loss": 0.5262, + "step": 22360 + }, + { + "epoch": 1.1194795446015264, + "grad_norm": 0.31277987360954285, + "learning_rate": 3.482419232266316e-05, + "loss": 0.6118, + "step": 22370 + }, + { + "epoch": 1.1199799824846741, + "grad_norm": 0.7227023243904114, + "learning_rate": 3.481492465385257e-05, + "loss": 0.4809, + "step": 22380 + }, + { + "epoch": 1.1204804203678218, + "grad_norm": 29.628469467163086, + "learning_rate": 3.480565698504198e-05, + "loss": 0.4257, + "step": 22390 + }, + { + "epoch": 1.1209808582509697, + "grad_norm": 62.035099029541016, + "learning_rate": 3.47963893162314e-05, + "loss": 0.1474, + "step": 22400 + }, + { + "epoch": 1.1214812961341174, + "grad_norm": 0.20013684034347534, + "learning_rate": 3.478712164742081e-05, + "loss": 0.25, + "step": 22410 + }, + { + "epoch": 1.121981734017265, + "grad_norm": 0.3315155506134033, + "learning_rate": 3.477785397861022e-05, + "loss": 0.1196, + "step": 22420 + }, + { + "epoch": 1.1224821719004128, + "grad_norm": 3.080305576324463, + "learning_rate": 3.476858630979963e-05, + "loss": 0.1396, + "step": 22430 + }, + { + "epoch": 1.1229826097835607, + "grad_norm": 103.59225463867188, + "learning_rate": 3.4759318640989046e-05, + "loss": 1.004, + "step": 22440 + }, + { + "epoch": 1.1234830476667084, + "grad_norm": 45.83987045288086, + "learning_rate": 3.475005097217846e-05, + "loss": 0.5281, + "step": 22450 + }, + { + "epoch": 1.123983485549856, + "grad_norm": 0.07426069676876068, + "learning_rate": 3.474078330336788e-05, + "loss": 0.1797, + "step": 22460 + }, + { + "epoch": 1.124483923433004, + "grad_norm": 51.51866149902344, + "learning_rate": 3.4731515634557286e-05, + "loss": 0.7296, + "step": 22470 + }, + { + "epoch": 1.1249843613161516, + "grad_norm": 30.781208038330078, + "learning_rate": 3.4722247965746695e-05, + "loss": 0.5553, + "step": 22480 + }, + { + "epoch": 1.1254847991992993, + "grad_norm": 1.4409205913543701, + "learning_rate": 3.471298029693611e-05, + "loss": 0.2117, + "step": 22490 + }, + { + "epoch": 1.125985237082447, + "grad_norm": 18.483407974243164, + "learning_rate": 3.4703712628125526e-05, + "loss": 0.6964, + "step": 22500 + }, + { + "epoch": 1.126485674965595, + "grad_norm": 0.02956007421016693, + "learning_rate": 3.4694444959314935e-05, + "loss": 0.3196, + "step": 22510 + }, + { + "epoch": 1.1269861128487426, + "grad_norm": 5.70256233215332, + "learning_rate": 3.468517729050435e-05, + "loss": 0.5152, + "step": 22520 + }, + { + "epoch": 1.1274865507318905, + "grad_norm": 0.0326283723115921, + "learning_rate": 3.467590962169376e-05, + "loss": 0.3636, + "step": 22530 + }, + { + "epoch": 1.1279869886150382, + "grad_norm": 0.06480450183153152, + "learning_rate": 3.4666641952883175e-05, + "loss": 0.0566, + "step": 22540 + }, + { + "epoch": 1.128487426498186, + "grad_norm": 22.813600540161133, + "learning_rate": 3.4657374284072584e-05, + "loss": 0.0856, + "step": 22550 + }, + { + "epoch": 1.1289878643813336, + "grad_norm": 32.569862365722656, + "learning_rate": 3.4648106615262e-05, + "loss": 0.7373, + "step": 22560 + }, + { + "epoch": 1.1294883022644815, + "grad_norm": 57.25261306762695, + "learning_rate": 3.4638838946451415e-05, + "loss": 0.6047, + "step": 22570 + }, + { + "epoch": 1.1299887401476292, + "grad_norm": 74.15501403808594, + "learning_rate": 3.4629571277640824e-05, + "loss": 0.3683, + "step": 22580 + }, + { + "epoch": 1.1304891780307769, + "grad_norm": 26.791790008544922, + "learning_rate": 3.462030360883023e-05, + "loss": 0.5193, + "step": 22590 + }, + { + "epoch": 1.1309896159139248, + "grad_norm": 20.88457679748535, + "learning_rate": 3.461103594001965e-05, + "loss": 0.6109, + "step": 22600 + }, + { + "epoch": 1.1314900537970725, + "grad_norm": 71.10171508789062, + "learning_rate": 3.4601768271209064e-05, + "loss": 0.6333, + "step": 22610 + }, + { + "epoch": 1.1319904916802201, + "grad_norm": 17.746213912963867, + "learning_rate": 3.459250060239847e-05, + "loss": 0.2423, + "step": 22620 + }, + { + "epoch": 1.1324909295633678, + "grad_norm": 38.599124908447266, + "learning_rate": 3.458323293358789e-05, + "loss": 0.2847, + "step": 22630 + }, + { + "epoch": 1.1329913674465157, + "grad_norm": 0.03935597464442253, + "learning_rate": 3.45739652647773e-05, + "loss": 0.2123, + "step": 22640 + }, + { + "epoch": 1.1334918053296634, + "grad_norm": 41.86299133300781, + "learning_rate": 3.456469759596671e-05, + "loss": 0.177, + "step": 22650 + }, + { + "epoch": 1.1339922432128111, + "grad_norm": 0.003600667929276824, + "learning_rate": 3.455542992715612e-05, + "loss": 0.1302, + "step": 22660 + }, + { + "epoch": 1.134492681095959, + "grad_norm": 42.10891342163086, + "learning_rate": 3.454616225834554e-05, + "loss": 0.4515, + "step": 22670 + }, + { + "epoch": 1.1349931189791067, + "grad_norm": 0.10318410396575928, + "learning_rate": 3.453689458953495e-05, + "loss": 0.0938, + "step": 22680 + }, + { + "epoch": 1.1354935568622544, + "grad_norm": 0.025410106405615807, + "learning_rate": 3.452762692072436e-05, + "loss": 0.4155, + "step": 22690 + }, + { + "epoch": 1.1359939947454023, + "grad_norm": 17.507064819335938, + "learning_rate": 3.451835925191377e-05, + "loss": 0.0737, + "step": 22700 + }, + { + "epoch": 1.13649443262855, + "grad_norm": 85.4668197631836, + "learning_rate": 3.4509091583103187e-05, + "loss": 0.6537, + "step": 22710 + }, + { + "epoch": 1.1369948705116977, + "grad_norm": 1.9183152914047241, + "learning_rate": 3.44998239142926e-05, + "loss": 0.3256, + "step": 22720 + }, + { + "epoch": 1.1374953083948456, + "grad_norm": 17.363494873046875, + "learning_rate": 3.449055624548201e-05, + "loss": 0.6453, + "step": 22730 + }, + { + "epoch": 1.1379957462779933, + "grad_norm": 0.03710109367966652, + "learning_rate": 3.448128857667143e-05, + "loss": 0.2401, + "step": 22740 + }, + { + "epoch": 1.138496184161141, + "grad_norm": 32.63392639160156, + "learning_rate": 3.4472020907860836e-05, + "loss": 0.3764, + "step": 22750 + }, + { + "epoch": 1.1389966220442886, + "grad_norm": 0.6234535574913025, + "learning_rate": 3.446275323905025e-05, + "loss": 0.2766, + "step": 22760 + }, + { + "epoch": 1.1394970599274366, + "grad_norm": 0.14314866065979004, + "learning_rate": 3.445348557023966e-05, + "loss": 0.6499, + "step": 22770 + }, + { + "epoch": 1.1399974978105842, + "grad_norm": 4.508293151855469, + "learning_rate": 3.4444217901429076e-05, + "loss": 0.2431, + "step": 22780 + }, + { + "epoch": 1.140497935693732, + "grad_norm": 42.513099670410156, + "learning_rate": 3.443495023261849e-05, + "loss": 0.2448, + "step": 22790 + }, + { + "epoch": 1.1409983735768798, + "grad_norm": 0.3558351993560791, + "learning_rate": 3.442568256380791e-05, + "loss": 0.2887, + "step": 22800 + }, + { + "epoch": 1.1414988114600275, + "grad_norm": 0.03817274048924446, + "learning_rate": 3.441641489499731e-05, + "loss": 0.3336, + "step": 22810 + }, + { + "epoch": 1.1419992493431752, + "grad_norm": 0.1871228665113449, + "learning_rate": 3.4407147226186725e-05, + "loss": 0.4602, + "step": 22820 + }, + { + "epoch": 1.1424996872263231, + "grad_norm": 44.364925384521484, + "learning_rate": 3.439787955737614e-05, + "loss": 0.4938, + "step": 22830 + }, + { + "epoch": 1.1430001251094708, + "grad_norm": 80.73878479003906, + "learning_rate": 3.4388611888565556e-05, + "loss": 0.8285, + "step": 22840 + }, + { + "epoch": 1.1435005629926185, + "grad_norm": 112.23106384277344, + "learning_rate": 3.4379344219754965e-05, + "loss": 0.3114, + "step": 22850 + }, + { + "epoch": 1.1440010008757664, + "grad_norm": 1.3799898624420166, + "learning_rate": 3.4370076550944374e-05, + "loss": 0.4495, + "step": 22860 + }, + { + "epoch": 1.144501438758914, + "grad_norm": 0.49702420830726624, + "learning_rate": 3.436080888213379e-05, + "loss": 0.6769, + "step": 22870 + }, + { + "epoch": 1.1450018766420618, + "grad_norm": 0.08469104766845703, + "learning_rate": 3.4351541213323205e-05, + "loss": 0.7161, + "step": 22880 + }, + { + "epoch": 1.1455023145252095, + "grad_norm": 0.017032282426953316, + "learning_rate": 3.4342273544512614e-05, + "loss": 0.3533, + "step": 22890 + }, + { + "epoch": 1.1460027524083574, + "grad_norm": 0.14069178700447083, + "learning_rate": 3.433300587570203e-05, + "loss": 0.3538, + "step": 22900 + }, + { + "epoch": 1.146503190291505, + "grad_norm": 0.7046579122543335, + "learning_rate": 3.4323738206891445e-05, + "loss": 0.2816, + "step": 22910 + }, + { + "epoch": 1.1470036281746527, + "grad_norm": 0.04958682134747505, + "learning_rate": 3.4314470538080854e-05, + "loss": 0.1553, + "step": 22920 + }, + { + "epoch": 1.1475040660578006, + "grad_norm": 0.14001493155956268, + "learning_rate": 3.430520286927026e-05, + "loss": 0.3504, + "step": 22930 + }, + { + "epoch": 1.1480045039409483, + "grad_norm": 1.054066777229309, + "learning_rate": 3.429593520045968e-05, + "loss": 0.446, + "step": 22940 + }, + { + "epoch": 1.148504941824096, + "grad_norm": 9.298186302185059, + "learning_rate": 3.4286667531649094e-05, + "loss": 0.4754, + "step": 22950 + }, + { + "epoch": 1.149005379707244, + "grad_norm": 39.66640853881836, + "learning_rate": 3.42773998628385e-05, + "loss": 0.5034, + "step": 22960 + }, + { + "epoch": 1.1495058175903916, + "grad_norm": 0.05476861819624901, + "learning_rate": 3.426813219402791e-05, + "loss": 0.3263, + "step": 22970 + }, + { + "epoch": 1.1500062554735393, + "grad_norm": 34.39107894897461, + "learning_rate": 3.425886452521733e-05, + "loss": 0.5998, + "step": 22980 + }, + { + "epoch": 1.1505066933566872, + "grad_norm": 0.42905178666114807, + "learning_rate": 3.424959685640674e-05, + "loss": 0.1431, + "step": 22990 + }, + { + "epoch": 1.151007131239835, + "grad_norm": 0.020493730902671814, + "learning_rate": 3.424032918759615e-05, + "loss": 0.3223, + "step": 23000 + }, + { + "epoch": 1.1515075691229826, + "grad_norm": 0.042526740580797195, + "learning_rate": 3.423106151878557e-05, + "loss": 0.3417, + "step": 23010 + }, + { + "epoch": 1.1520080070061303, + "grad_norm": 80.45571899414062, + "learning_rate": 3.422179384997498e-05, + "loss": 0.5527, + "step": 23020 + }, + { + "epoch": 1.1525084448892782, + "grad_norm": 88.65223693847656, + "learning_rate": 3.421252618116439e-05, + "loss": 0.6209, + "step": 23030 + }, + { + "epoch": 1.1530088827724259, + "grad_norm": 1.6188738346099854, + "learning_rate": 3.42032585123538e-05, + "loss": 0.3086, + "step": 23040 + }, + { + "epoch": 1.1535093206555735, + "grad_norm": 0.1082124337553978, + "learning_rate": 3.4193990843543216e-05, + "loss": 0.0699, + "step": 23050 + }, + { + "epoch": 1.1540097585387215, + "grad_norm": 48.44356918334961, + "learning_rate": 3.418472317473263e-05, + "loss": 0.2353, + "step": 23060 + }, + { + "epoch": 1.1545101964218691, + "grad_norm": 0.05547427013516426, + "learning_rate": 3.417545550592204e-05, + "loss": 0.2553, + "step": 23070 + }, + { + "epoch": 1.1550106343050168, + "grad_norm": 73.87925720214844, + "learning_rate": 3.416618783711145e-05, + "loss": 0.6961, + "step": 23080 + }, + { + "epoch": 1.1555110721881647, + "grad_norm": 63.6040153503418, + "learning_rate": 3.4156920168300865e-05, + "loss": 0.2732, + "step": 23090 + }, + { + "epoch": 1.1560115100713124, + "grad_norm": 58.038753509521484, + "learning_rate": 3.414765249949028e-05, + "loss": 0.5316, + "step": 23100 + }, + { + "epoch": 1.15651194795446, + "grad_norm": 1.018752098083496, + "learning_rate": 3.413838483067969e-05, + "loss": 0.1893, + "step": 23110 + }, + { + "epoch": 1.157012385837608, + "grad_norm": 6.999598026275635, + "learning_rate": 3.4129117161869106e-05, + "loss": 0.3141, + "step": 23120 + }, + { + "epoch": 1.1575128237207557, + "grad_norm": 0.16376850008964539, + "learning_rate": 3.411984949305852e-05, + "loss": 0.2705, + "step": 23130 + }, + { + "epoch": 1.1580132616039034, + "grad_norm": 0.628200888633728, + "learning_rate": 3.411058182424793e-05, + "loss": 0.4862, + "step": 23140 + }, + { + "epoch": 1.158513699487051, + "grad_norm": 41.53506851196289, + "learning_rate": 3.410131415543734e-05, + "loss": 0.1693, + "step": 23150 + }, + { + "epoch": 1.159014137370199, + "grad_norm": 90.1776351928711, + "learning_rate": 3.4092046486626755e-05, + "loss": 0.4445, + "step": 23160 + }, + { + "epoch": 1.1595145752533467, + "grad_norm": 0.0426909402012825, + "learning_rate": 3.408277881781617e-05, + "loss": 0.6662, + "step": 23170 + }, + { + "epoch": 1.1600150131364944, + "grad_norm": 47.74855041503906, + "learning_rate": 3.4073511149005586e-05, + "loss": 0.2379, + "step": 23180 + }, + { + "epoch": 1.1605154510196423, + "grad_norm": 37.508880615234375, + "learning_rate": 3.4064243480194995e-05, + "loss": 0.3156, + "step": 23190 + }, + { + "epoch": 1.16101588890279, + "grad_norm": 71.3038330078125, + "learning_rate": 3.4054975811384404e-05, + "loss": 0.4477, + "step": 23200 + }, + { + "epoch": 1.1615163267859376, + "grad_norm": 0.020437020808458328, + "learning_rate": 3.404570814257382e-05, + "loss": 0.2516, + "step": 23210 + }, + { + "epoch": 1.1620167646690855, + "grad_norm": 6.835453987121582, + "learning_rate": 3.403644047376323e-05, + "loss": 0.5044, + "step": 23220 + }, + { + "epoch": 1.1625172025522332, + "grad_norm": 0.02276095375418663, + "learning_rate": 3.4027172804952644e-05, + "loss": 0.5571, + "step": 23230 + }, + { + "epoch": 1.163017640435381, + "grad_norm": 1.1695665121078491, + "learning_rate": 3.401790513614206e-05, + "loss": 0.4492, + "step": 23240 + }, + { + "epoch": 1.1635180783185288, + "grad_norm": 0.023594027385115623, + "learning_rate": 3.400863746733147e-05, + "loss": 0.6328, + "step": 23250 + }, + { + "epoch": 1.1640185162016765, + "grad_norm": 34.090335845947266, + "learning_rate": 3.399936979852088e-05, + "loss": 0.3673, + "step": 23260 + }, + { + "epoch": 1.1645189540848242, + "grad_norm": 51.26726531982422, + "learning_rate": 3.399010212971029e-05, + "loss": 0.4802, + "step": 23270 + }, + { + "epoch": 1.1650193919679719, + "grad_norm": 14.412704467773438, + "learning_rate": 3.398083446089971e-05, + "loss": 0.7101, + "step": 23280 + }, + { + "epoch": 1.1655198298511198, + "grad_norm": 35.375728607177734, + "learning_rate": 3.3971566792089124e-05, + "loss": 0.2902, + "step": 23290 + }, + { + "epoch": 1.1660202677342675, + "grad_norm": 1.2754154205322266, + "learning_rate": 3.396229912327853e-05, + "loss": 0.4022, + "step": 23300 + }, + { + "epoch": 1.1665207056174152, + "grad_norm": 6.762238502502441, + "learning_rate": 3.395303145446794e-05, + "loss": 0.5208, + "step": 23310 + }, + { + "epoch": 1.167021143500563, + "grad_norm": 0.030194994062185287, + "learning_rate": 3.394376378565736e-05, + "loss": 0.0995, + "step": 23320 + }, + { + "epoch": 1.1675215813837108, + "grad_norm": 5.233459949493408, + "learning_rate": 3.393449611684677e-05, + "loss": 0.3771, + "step": 23330 + }, + { + "epoch": 1.1680220192668584, + "grad_norm": 57.83295822143555, + "learning_rate": 3.392522844803618e-05, + "loss": 0.2395, + "step": 23340 + }, + { + "epoch": 1.1685224571500061, + "grad_norm": 0.0014709753450006247, + "learning_rate": 3.39159607792256e-05, + "loss": 0.453, + "step": 23350 + }, + { + "epoch": 1.169022895033154, + "grad_norm": 0.09619703143835068, + "learning_rate": 3.3906693110415006e-05, + "loss": 0.333, + "step": 23360 + }, + { + "epoch": 1.1695233329163017, + "grad_norm": 10.699823379516602, + "learning_rate": 3.389742544160442e-05, + "loss": 0.4437, + "step": 23370 + }, + { + "epoch": 1.1700237707994496, + "grad_norm": 13.28339958190918, + "learning_rate": 3.388815777279383e-05, + "loss": 0.104, + "step": 23380 + }, + { + "epoch": 1.1705242086825973, + "grad_norm": 0.04133359342813492, + "learning_rate": 3.3878890103983246e-05, + "loss": 0.2292, + "step": 23390 + }, + { + "epoch": 1.171024646565745, + "grad_norm": 1.2237029075622559, + "learning_rate": 3.386962243517266e-05, + "loss": 0.4732, + "step": 23400 + }, + { + "epoch": 1.1715250844488927, + "grad_norm": 190.9026641845703, + "learning_rate": 3.386035476636207e-05, + "loss": 0.2641, + "step": 23410 + }, + { + "epoch": 1.1720255223320406, + "grad_norm": 40.31526565551758, + "learning_rate": 3.385108709755148e-05, + "loss": 0.3792, + "step": 23420 + }, + { + "epoch": 1.1725259602151883, + "grad_norm": 8.893001556396484, + "learning_rate": 3.3841819428740895e-05, + "loss": 0.4959, + "step": 23430 + }, + { + "epoch": 1.173026398098336, + "grad_norm": 11.29373550415039, + "learning_rate": 3.383255175993031e-05, + "loss": 0.3891, + "step": 23440 + }, + { + "epoch": 1.1735268359814839, + "grad_norm": 13.964129447937012, + "learning_rate": 3.382328409111972e-05, + "loss": 0.2716, + "step": 23450 + }, + { + "epoch": 1.1740272738646316, + "grad_norm": 30.09575080871582, + "learning_rate": 3.3814016422309135e-05, + "loss": 0.3498, + "step": 23460 + }, + { + "epoch": 1.1745277117477793, + "grad_norm": 74.9800033569336, + "learning_rate": 3.3804748753498544e-05, + "loss": 0.3612, + "step": 23470 + }, + { + "epoch": 1.175028149630927, + "grad_norm": 109.91725158691406, + "learning_rate": 3.379548108468796e-05, + "loss": 0.4648, + "step": 23480 + }, + { + "epoch": 1.1755285875140749, + "grad_norm": 72.19371795654297, + "learning_rate": 3.378621341587737e-05, + "loss": 0.3828, + "step": 23490 + }, + { + "epoch": 1.1760290253972225, + "grad_norm": 0.5507450699806213, + "learning_rate": 3.3776945747066784e-05, + "loss": 0.5072, + "step": 23500 + }, + { + "epoch": 1.1765294632803704, + "grad_norm": 98.39134216308594, + "learning_rate": 3.37676780782562e-05, + "loss": 0.5038, + "step": 23510 + }, + { + "epoch": 1.1770299011635181, + "grad_norm": 0.01790090650320053, + "learning_rate": 3.375841040944561e-05, + "loss": 0.15, + "step": 23520 + }, + { + "epoch": 1.1775303390466658, + "grad_norm": 82.98262786865234, + "learning_rate": 3.374914274063502e-05, + "loss": 0.3508, + "step": 23530 + }, + { + "epoch": 1.1780307769298135, + "grad_norm": 36.71696853637695, + "learning_rate": 3.3739875071824433e-05, + "loss": 0.4942, + "step": 23540 + }, + { + "epoch": 1.1785312148129614, + "grad_norm": 0.00594551395624876, + "learning_rate": 3.373060740301385e-05, + "loss": 0.271, + "step": 23550 + }, + { + "epoch": 1.179031652696109, + "grad_norm": 0.5945324897766113, + "learning_rate": 3.372133973420326e-05, + "loss": 0.2851, + "step": 23560 + }, + { + "epoch": 1.1795320905792568, + "grad_norm": 98.80622100830078, + "learning_rate": 3.3712072065392674e-05, + "loss": 0.4397, + "step": 23570 + }, + { + "epoch": 1.1800325284624047, + "grad_norm": 31.47661018371582, + "learning_rate": 3.370280439658208e-05, + "loss": 0.34, + "step": 23580 + }, + { + "epoch": 1.1805329663455524, + "grad_norm": 57.496421813964844, + "learning_rate": 3.36935367277715e-05, + "loss": 0.2534, + "step": 23590 + }, + { + "epoch": 1.1810334042287, + "grad_norm": 122.57263946533203, + "learning_rate": 3.368426905896091e-05, + "loss": 0.2402, + "step": 23600 + }, + { + "epoch": 1.1815338421118478, + "grad_norm": 0.08607327938079834, + "learning_rate": 3.367500139015032e-05, + "loss": 0.5191, + "step": 23610 + }, + { + "epoch": 1.1820342799949957, + "grad_norm": 22.486021041870117, + "learning_rate": 3.366573372133974e-05, + "loss": 0.361, + "step": 23620 + }, + { + "epoch": 1.1825347178781433, + "grad_norm": 0.022272001951932907, + "learning_rate": 3.3656466052529154e-05, + "loss": 0.2035, + "step": 23630 + }, + { + "epoch": 1.183035155761291, + "grad_norm": 77.64767456054688, + "learning_rate": 3.3647198383718556e-05, + "loss": 0.1584, + "step": 23640 + }, + { + "epoch": 1.183535593644439, + "grad_norm": 0.8701883554458618, + "learning_rate": 3.363793071490797e-05, + "loss": 0.4116, + "step": 23650 + }, + { + "epoch": 1.1840360315275866, + "grad_norm": 0.06997361779212952, + "learning_rate": 3.362866304609739e-05, + "loss": 0.2551, + "step": 23660 + }, + { + "epoch": 1.1845364694107343, + "grad_norm": 0.6441417932510376, + "learning_rate": 3.36193953772868e-05, + "loss": 0.2486, + "step": 23670 + }, + { + "epoch": 1.1850369072938822, + "grad_norm": 15.765795707702637, + "learning_rate": 3.361012770847621e-05, + "loss": 0.276, + "step": 23680 + }, + { + "epoch": 1.18553734517703, + "grad_norm": 0.0035089408047497272, + "learning_rate": 3.360086003966562e-05, + "loss": 0.4384, + "step": 23690 + }, + { + "epoch": 1.1860377830601776, + "grad_norm": 53.509883880615234, + "learning_rate": 3.3591592370855036e-05, + "loss": 0.2093, + "step": 23700 + }, + { + "epoch": 1.1865382209433255, + "grad_norm": 0.14439953863620758, + "learning_rate": 3.358232470204445e-05, + "loss": 0.1904, + "step": 23710 + }, + { + "epoch": 1.1870386588264732, + "grad_norm": 0.02465715818107128, + "learning_rate": 3.357305703323386e-05, + "loss": 0.184, + "step": 23720 + }, + { + "epoch": 1.1875390967096209, + "grad_norm": 147.70169067382812, + "learning_rate": 3.3563789364423276e-05, + "loss": 0.2737, + "step": 23730 + }, + { + "epoch": 1.1880395345927686, + "grad_norm": 25.796409606933594, + "learning_rate": 3.355452169561269e-05, + "loss": 0.2897, + "step": 23740 + }, + { + "epoch": 1.1885399724759165, + "grad_norm": 1.0093647241592407, + "learning_rate": 3.35452540268021e-05, + "loss": 0.1894, + "step": 23750 + }, + { + "epoch": 1.1890404103590642, + "grad_norm": 3.1682636737823486, + "learning_rate": 3.353598635799151e-05, + "loss": 0.3595, + "step": 23760 + }, + { + "epoch": 1.1895408482422118, + "grad_norm": 0.012343388050794601, + "learning_rate": 3.3526718689180925e-05, + "loss": 0.2828, + "step": 23770 + }, + { + "epoch": 1.1900412861253598, + "grad_norm": 131.24488830566406, + "learning_rate": 3.351745102037034e-05, + "loss": 0.1989, + "step": 23780 + }, + { + "epoch": 1.1905417240085074, + "grad_norm": 73.3966293334961, + "learning_rate": 3.350818335155975e-05, + "loss": 0.2941, + "step": 23790 + }, + { + "epoch": 1.1910421618916551, + "grad_norm": 3.6541531085968018, + "learning_rate": 3.349891568274916e-05, + "loss": 0.3879, + "step": 23800 + }, + { + "epoch": 1.191542599774803, + "grad_norm": 0.0050062499940395355, + "learning_rate": 3.3489648013938574e-05, + "loss": 0.316, + "step": 23810 + }, + { + "epoch": 1.1920430376579507, + "grad_norm": 0.9953902959823608, + "learning_rate": 3.348038034512799e-05, + "loss": 0.2865, + "step": 23820 + }, + { + "epoch": 1.1925434755410984, + "grad_norm": 0.04753986746072769, + "learning_rate": 3.34711126763174e-05, + "loss": 0.3789, + "step": 23830 + }, + { + "epoch": 1.1930439134242463, + "grad_norm": 0.05159697309136391, + "learning_rate": 3.3461845007506814e-05, + "loss": 0.438, + "step": 23840 + }, + { + "epoch": 1.193544351307394, + "grad_norm": 30.442623138427734, + "learning_rate": 3.345257733869623e-05, + "loss": 0.414, + "step": 23850 + }, + { + "epoch": 1.1940447891905417, + "grad_norm": 32.14916229248047, + "learning_rate": 3.344330966988564e-05, + "loss": 0.4696, + "step": 23860 + }, + { + "epoch": 1.1945452270736894, + "grad_norm": 10.09000301361084, + "learning_rate": 3.343404200107505e-05, + "loss": 0.035, + "step": 23870 + }, + { + "epoch": 1.1950456649568373, + "grad_norm": 1.4024502038955688, + "learning_rate": 3.342477433226446e-05, + "loss": 0.3056, + "step": 23880 + }, + { + "epoch": 1.195546102839985, + "grad_norm": 0.004709731787443161, + "learning_rate": 3.341550666345388e-05, + "loss": 0.2031, + "step": 23890 + }, + { + "epoch": 1.1960465407231327, + "grad_norm": 6.022188663482666, + "learning_rate": 3.340623899464329e-05, + "loss": 0.319, + "step": 23900 + }, + { + "epoch": 1.1965469786062806, + "grad_norm": 45.89932632446289, + "learning_rate": 3.3396971325832703e-05, + "loss": 0.213, + "step": 23910 + }, + { + "epoch": 1.1970474164894283, + "grad_norm": 0.00107488501816988, + "learning_rate": 3.338770365702211e-05, + "loss": 0.2628, + "step": 23920 + }, + { + "epoch": 1.197547854372576, + "grad_norm": 2.5853688716888428, + "learning_rate": 3.337843598821153e-05, + "loss": 0.2776, + "step": 23930 + }, + { + "epoch": 1.1980482922557238, + "grad_norm": 45.90558624267578, + "learning_rate": 3.336916831940094e-05, + "loss": 0.1663, + "step": 23940 + }, + { + "epoch": 1.1985487301388715, + "grad_norm": 96.1787109375, + "learning_rate": 3.335990065059035e-05, + "loss": 0.1972, + "step": 23950 + }, + { + "epoch": 1.1990491680220192, + "grad_norm": 0.16789187490940094, + "learning_rate": 3.335063298177977e-05, + "loss": 0.4682, + "step": 23960 + }, + { + "epoch": 1.1995496059051671, + "grad_norm": 0.017088545486330986, + "learning_rate": 3.334136531296918e-05, + "loss": 0.3063, + "step": 23970 + }, + { + "epoch": 1.2000500437883148, + "grad_norm": 0.004550191108137369, + "learning_rate": 3.3332097644158586e-05, + "loss": 0.1603, + "step": 23980 + }, + { + "epoch": 1.2005504816714625, + "grad_norm": 11.551319122314453, + "learning_rate": 3.3322829975348e-05, + "loss": 0.2568, + "step": 23990 + }, + { + "epoch": 1.2010509195546102, + "grad_norm": 0.008488905616104603, + "learning_rate": 3.331356230653742e-05, + "loss": 0.2482, + "step": 24000 + }, + { + "epoch": 1.201551357437758, + "grad_norm": 7.681360721588135, + "learning_rate": 3.330429463772683e-05, + "loss": 0.4631, + "step": 24010 + }, + { + "epoch": 1.2020517953209058, + "grad_norm": 2.868546962738037, + "learning_rate": 3.329502696891624e-05, + "loss": 0.1558, + "step": 24020 + }, + { + "epoch": 1.2025522332040535, + "grad_norm": 64.14839172363281, + "learning_rate": 3.328575930010565e-05, + "loss": 0.8724, + "step": 24030 + }, + { + "epoch": 1.2030526710872014, + "grad_norm": 0.000832205347251147, + "learning_rate": 3.3276491631295066e-05, + "loss": 0.6795, + "step": 24040 + }, + { + "epoch": 1.203553108970349, + "grad_norm": 0.005569048225879669, + "learning_rate": 3.326722396248448e-05, + "loss": 0.8832, + "step": 24050 + }, + { + "epoch": 1.2040535468534967, + "grad_norm": 1.300540566444397, + "learning_rate": 3.325795629367389e-05, + "loss": 0.2951, + "step": 24060 + }, + { + "epoch": 1.2045539847366447, + "grad_norm": 13.811827659606934, + "learning_rate": 3.3248688624863306e-05, + "loss": 0.667, + "step": 24070 + }, + { + "epoch": 1.2050544226197923, + "grad_norm": 12.906695365905762, + "learning_rate": 3.3239420956052715e-05, + "loss": 0.3995, + "step": 24080 + }, + { + "epoch": 1.20555486050294, + "grad_norm": 1.4336388111114502, + "learning_rate": 3.3230153287242124e-05, + "loss": 0.1843, + "step": 24090 + }, + { + "epoch": 1.206055298386088, + "grad_norm": 122.48673248291016, + "learning_rate": 3.322088561843154e-05, + "loss": 0.7604, + "step": 24100 + }, + { + "epoch": 1.2065557362692356, + "grad_norm": 51.125762939453125, + "learning_rate": 3.3211617949620955e-05, + "loss": 0.6463, + "step": 24110 + }, + { + "epoch": 1.2070561741523833, + "grad_norm": 0.3689766228199005, + "learning_rate": 3.320235028081037e-05, + "loss": 0.3538, + "step": 24120 + }, + { + "epoch": 1.207556612035531, + "grad_norm": 3.634428024291992, + "learning_rate": 3.319308261199978e-05, + "loss": 0.5078, + "step": 24130 + }, + { + "epoch": 1.208057049918679, + "grad_norm": 22.18536376953125, + "learning_rate": 3.318381494318919e-05, + "loss": 0.2366, + "step": 24140 + }, + { + "epoch": 1.2085574878018266, + "grad_norm": 0.07821298390626907, + "learning_rate": 3.3174547274378604e-05, + "loss": 0.5579, + "step": 24150 + }, + { + "epoch": 1.2090579256849743, + "grad_norm": 2.454329490661621, + "learning_rate": 3.316527960556802e-05, + "loss": 0.258, + "step": 24160 + }, + { + "epoch": 1.2095583635681222, + "grad_norm": 0.05341324582695961, + "learning_rate": 3.315601193675743e-05, + "loss": 0.2293, + "step": 24170 + }, + { + "epoch": 1.2100588014512699, + "grad_norm": 0.8619423508644104, + "learning_rate": 3.3146744267946844e-05, + "loss": 0.2455, + "step": 24180 + }, + { + "epoch": 1.2105592393344176, + "grad_norm": 25.89310073852539, + "learning_rate": 3.313747659913625e-05, + "loss": 0.4369, + "step": 24190 + }, + { + "epoch": 1.2110596772175655, + "grad_norm": 0.7054183483123779, + "learning_rate": 3.312820893032567e-05, + "loss": 0.3943, + "step": 24200 + }, + { + "epoch": 1.2115601151007132, + "grad_norm": 16.58211898803711, + "learning_rate": 3.311894126151508e-05, + "loss": 0.2843, + "step": 24210 + }, + { + "epoch": 1.2120605529838608, + "grad_norm": 2.054788112640381, + "learning_rate": 3.310967359270449e-05, + "loss": 0.32, + "step": 24220 + }, + { + "epoch": 1.2125609908670087, + "grad_norm": 35.881343841552734, + "learning_rate": 3.310040592389391e-05, + "loss": 0.3826, + "step": 24230 + }, + { + "epoch": 1.2130614287501564, + "grad_norm": 0.12144740670919418, + "learning_rate": 3.309113825508332e-05, + "loss": 0.4933, + "step": 24240 + }, + { + "epoch": 1.2135618666333041, + "grad_norm": 2.097586154937744, + "learning_rate": 3.308187058627273e-05, + "loss": 0.3794, + "step": 24250 + }, + { + "epoch": 1.2140623045164518, + "grad_norm": 0.10838683694601059, + "learning_rate": 3.307260291746214e-05, + "loss": 0.3065, + "step": 24260 + }, + { + "epoch": 1.2145627423995997, + "grad_norm": 0.04098005220293999, + "learning_rate": 3.306333524865156e-05, + "loss": 0.4067, + "step": 24270 + }, + { + "epoch": 1.2150631802827474, + "grad_norm": 33.02265548706055, + "learning_rate": 3.305406757984097e-05, + "loss": 0.4313, + "step": 24280 + }, + { + "epoch": 1.215563618165895, + "grad_norm": 45.7121696472168, + "learning_rate": 3.304479991103038e-05, + "loss": 0.3932, + "step": 24290 + }, + { + "epoch": 1.216064056049043, + "grad_norm": 0.5925977826118469, + "learning_rate": 3.303553224221979e-05, + "loss": 0.3719, + "step": 24300 + }, + { + "epoch": 1.2165644939321907, + "grad_norm": 32.79696273803711, + "learning_rate": 3.302626457340921e-05, + "loss": 0.1403, + "step": 24310 + }, + { + "epoch": 1.2170649318153384, + "grad_norm": 55.4060173034668, + "learning_rate": 3.3016996904598616e-05, + "loss": 0.3174, + "step": 24320 + }, + { + "epoch": 1.217565369698486, + "grad_norm": 83.05979919433594, + "learning_rate": 3.300772923578803e-05, + "loss": 1.003, + "step": 24330 + }, + { + "epoch": 1.218065807581634, + "grad_norm": 55.6873779296875, + "learning_rate": 3.299846156697745e-05, + "loss": 0.8071, + "step": 24340 + }, + { + "epoch": 1.2185662454647816, + "grad_norm": 0.23284843564033508, + "learning_rate": 3.2989193898166856e-05, + "loss": 0.4159, + "step": 24350 + }, + { + "epoch": 1.2190666833479296, + "grad_norm": 0.06403537094593048, + "learning_rate": 3.2979926229356265e-05, + "loss": 0.3809, + "step": 24360 + }, + { + "epoch": 1.2195671212310772, + "grad_norm": 45.74055099487305, + "learning_rate": 3.297065856054568e-05, + "loss": 0.4803, + "step": 24370 + }, + { + "epoch": 1.220067559114225, + "grad_norm": 11.696496963500977, + "learning_rate": 3.2961390891735096e-05, + "loss": 0.5324, + "step": 24380 + }, + { + "epoch": 1.2205679969973726, + "grad_norm": 0.8479064106941223, + "learning_rate": 3.2952123222924505e-05, + "loss": 0.3921, + "step": 24390 + }, + { + "epoch": 1.2210684348805205, + "grad_norm": 39.63326644897461, + "learning_rate": 3.294285555411392e-05, + "loss": 0.365, + "step": 24400 + }, + { + "epoch": 1.2215688727636682, + "grad_norm": 113.63565063476562, + "learning_rate": 3.293358788530333e-05, + "loss": 0.4544, + "step": 24410 + }, + { + "epoch": 1.222069310646816, + "grad_norm": 49.83055877685547, + "learning_rate": 3.2924320216492745e-05, + "loss": 0.0993, + "step": 24420 + }, + { + "epoch": 1.2225697485299638, + "grad_norm": 20.296180725097656, + "learning_rate": 3.2915052547682154e-05, + "loss": 0.2527, + "step": 24430 + }, + { + "epoch": 1.2230701864131115, + "grad_norm": 33.33852767944336, + "learning_rate": 3.290578487887157e-05, + "loss": 0.4109, + "step": 24440 + }, + { + "epoch": 1.2235706242962592, + "grad_norm": 1.1051933765411377, + "learning_rate": 3.2896517210060985e-05, + "loss": 0.1447, + "step": 24450 + }, + { + "epoch": 1.2240710621794069, + "grad_norm": 55.52216720581055, + "learning_rate": 3.28872495412504e-05, + "loss": 0.132, + "step": 24460 + }, + { + "epoch": 1.2245715000625548, + "grad_norm": 26.492168426513672, + "learning_rate": 3.28779818724398e-05, + "loss": 0.5811, + "step": 24470 + }, + { + "epoch": 1.2250719379457025, + "grad_norm": 17.9793758392334, + "learning_rate": 3.286871420362922e-05, + "loss": 0.5386, + "step": 24480 + }, + { + "epoch": 1.2255723758288504, + "grad_norm": 57.881351470947266, + "learning_rate": 3.2859446534818634e-05, + "loss": 0.3336, + "step": 24490 + }, + { + "epoch": 1.226072813711998, + "grad_norm": 62.144378662109375, + "learning_rate": 3.285017886600805e-05, + "loss": 0.1627, + "step": 24500 + }, + { + "epoch": 1.2265732515951457, + "grad_norm": 0.4421941637992859, + "learning_rate": 3.284091119719746e-05, + "loss": 0.1411, + "step": 24510 + }, + { + "epoch": 1.2270736894782934, + "grad_norm": 1.0314946174621582, + "learning_rate": 3.283164352838687e-05, + "loss": 0.1976, + "step": 24520 + }, + { + "epoch": 1.2275741273614413, + "grad_norm": 0.004312742035835981, + "learning_rate": 3.282237585957628e-05, + "loss": 0.3294, + "step": 24530 + }, + { + "epoch": 1.228074565244589, + "grad_norm": 87.82038116455078, + "learning_rate": 3.28131081907657e-05, + "loss": 0.2134, + "step": 24540 + }, + { + "epoch": 1.2285750031277367, + "grad_norm": 0.27264976501464844, + "learning_rate": 3.280384052195511e-05, + "loss": 0.7046, + "step": 24550 + }, + { + "epoch": 1.2290754410108846, + "grad_norm": 21.23196029663086, + "learning_rate": 3.279457285314452e-05, + "loss": 0.3372, + "step": 24560 + }, + { + "epoch": 1.2295758788940323, + "grad_norm": 54.47974395751953, + "learning_rate": 3.278530518433394e-05, + "loss": 0.3011, + "step": 24570 + }, + { + "epoch": 1.23007631677718, + "grad_norm": 15.212749481201172, + "learning_rate": 3.277603751552335e-05, + "loss": 0.7538, + "step": 24580 + }, + { + "epoch": 1.2305767546603277, + "grad_norm": 0.4208630919456482, + "learning_rate": 3.2766769846712757e-05, + "loss": 0.5785, + "step": 24590 + }, + { + "epoch": 1.2310771925434756, + "grad_norm": 0.18668119609355927, + "learning_rate": 3.275750217790217e-05, + "loss": 0.8558, + "step": 24600 + }, + { + "epoch": 1.2315776304266233, + "grad_norm": 28.798477172851562, + "learning_rate": 3.274823450909159e-05, + "loss": 0.3142, + "step": 24610 + }, + { + "epoch": 1.232078068309771, + "grad_norm": 72.0798568725586, + "learning_rate": 3.2738966840281e-05, + "loss": 0.5155, + "step": 24620 + }, + { + "epoch": 1.2325785061929189, + "grad_norm": 6.723727226257324, + "learning_rate": 3.272969917147041e-05, + "loss": 0.2236, + "step": 24630 + }, + { + "epoch": 1.2330789440760666, + "grad_norm": 19.989120483398438, + "learning_rate": 3.272043150265982e-05, + "loss": 0.658, + "step": 24640 + }, + { + "epoch": 1.2335793819592142, + "grad_norm": 75.21302795410156, + "learning_rate": 3.271116383384924e-05, + "loss": 0.3587, + "step": 24650 + }, + { + "epoch": 1.2340798198423621, + "grad_norm": 0.03558576479554176, + "learning_rate": 3.2701896165038646e-05, + "loss": 0.267, + "step": 24660 + }, + { + "epoch": 1.2345802577255098, + "grad_norm": 5.881162643432617, + "learning_rate": 3.269262849622806e-05, + "loss": 0.0753, + "step": 24670 + }, + { + "epoch": 1.2350806956086575, + "grad_norm": 0.27301642298698425, + "learning_rate": 3.268336082741748e-05, + "loss": 0.4761, + "step": 24680 + }, + { + "epoch": 1.2355811334918054, + "grad_norm": 13.936537742614746, + "learning_rate": 3.2674093158606886e-05, + "loss": 0.4296, + "step": 24690 + }, + { + "epoch": 1.2360815713749531, + "grad_norm": 1.5583916902542114, + "learning_rate": 3.2664825489796295e-05, + "loss": 0.2392, + "step": 24700 + }, + { + "epoch": 1.2365820092581008, + "grad_norm": 7.031650066375732, + "learning_rate": 3.265555782098571e-05, + "loss": 0.1478, + "step": 24710 + }, + { + "epoch": 1.2370824471412485, + "grad_norm": 0.0663236603140831, + "learning_rate": 3.2646290152175126e-05, + "loss": 0.1859, + "step": 24720 + }, + { + "epoch": 1.2375828850243964, + "grad_norm": 0.19093948602676392, + "learning_rate": 3.2637022483364535e-05, + "loss": 0.0086, + "step": 24730 + }, + { + "epoch": 1.238083322907544, + "grad_norm": 72.94525146484375, + "learning_rate": 3.262775481455395e-05, + "loss": 0.7989, + "step": 24740 + }, + { + "epoch": 1.2385837607906918, + "grad_norm": 0.02708224393427372, + "learning_rate": 3.261848714574336e-05, + "loss": 0.1323, + "step": 24750 + }, + { + "epoch": 1.2390841986738397, + "grad_norm": 0.15634441375732422, + "learning_rate": 3.2609219476932775e-05, + "loss": 0.1778, + "step": 24760 + }, + { + "epoch": 1.2395846365569874, + "grad_norm": 1.0630221366882324, + "learning_rate": 3.2599951808122184e-05, + "loss": 0.2535, + "step": 24770 + }, + { + "epoch": 1.240085074440135, + "grad_norm": 25.829015731811523, + "learning_rate": 3.25906841393116e-05, + "loss": 0.4252, + "step": 24780 + }, + { + "epoch": 1.240585512323283, + "grad_norm": 0.311902791261673, + "learning_rate": 3.2581416470501015e-05, + "loss": 0.7684, + "step": 24790 + }, + { + "epoch": 1.2410859502064306, + "grad_norm": 0.009561363607645035, + "learning_rate": 3.2572148801690424e-05, + "loss": 0.2662, + "step": 24800 + }, + { + "epoch": 1.2415863880895783, + "grad_norm": 51.59767150878906, + "learning_rate": 3.256288113287983e-05, + "loss": 0.4675, + "step": 24810 + }, + { + "epoch": 1.2420868259727262, + "grad_norm": 0.15058039128780365, + "learning_rate": 3.255361346406925e-05, + "loss": 0.5893, + "step": 24820 + }, + { + "epoch": 1.242587263855874, + "grad_norm": 0.07743478566408157, + "learning_rate": 3.2544345795258664e-05, + "loss": 0.1853, + "step": 24830 + }, + { + "epoch": 1.2430877017390216, + "grad_norm": 22.266111373901367, + "learning_rate": 3.253507812644808e-05, + "loss": 0.5892, + "step": 24840 + }, + { + "epoch": 1.2435881396221693, + "grad_norm": 0.1926325410604477, + "learning_rate": 3.252581045763749e-05, + "loss": 0.2179, + "step": 24850 + }, + { + "epoch": 1.2440885775053172, + "grad_norm": 39.120277404785156, + "learning_rate": 3.25165427888269e-05, + "loss": 0.396, + "step": 24860 + }, + { + "epoch": 1.244589015388465, + "grad_norm": 0.10930413752794266, + "learning_rate": 3.250727512001631e-05, + "loss": 0.1728, + "step": 24870 + }, + { + "epoch": 1.2450894532716126, + "grad_norm": 4.996157169342041, + "learning_rate": 3.249800745120573e-05, + "loss": 0.3284, + "step": 24880 + }, + { + "epoch": 1.2455898911547605, + "grad_norm": 14.614059448242188, + "learning_rate": 3.248873978239514e-05, + "loss": 0.2038, + "step": 24890 + }, + { + "epoch": 1.2460903290379082, + "grad_norm": 0.27420082688331604, + "learning_rate": 3.247947211358455e-05, + "loss": 0.5101, + "step": 24900 + }, + { + "epoch": 1.2465907669210559, + "grad_norm": 42.05366134643555, + "learning_rate": 3.247020444477396e-05, + "loss": 0.8115, + "step": 24910 + }, + { + "epoch": 1.2470912048042038, + "grad_norm": 1.7178452014923096, + "learning_rate": 3.246093677596338e-05, + "loss": 0.2459, + "step": 24920 + }, + { + "epoch": 1.2475916426873515, + "grad_norm": 0.0014084530994296074, + "learning_rate": 3.2451669107152786e-05, + "loss": 0.5071, + "step": 24930 + }, + { + "epoch": 1.2480920805704991, + "grad_norm": 7.819710731506348, + "learning_rate": 3.24424014383422e-05, + "loss": 0.286, + "step": 24940 + }, + { + "epoch": 1.248592518453647, + "grad_norm": 28.961395263671875, + "learning_rate": 3.243313376953162e-05, + "loss": 0.5527, + "step": 24950 + }, + { + "epoch": 1.2490929563367947, + "grad_norm": 28.08820343017578, + "learning_rate": 3.2423866100721027e-05, + "loss": 0.8555, + "step": 24960 + }, + { + "epoch": 1.2495933942199424, + "grad_norm": 66.12683868408203, + "learning_rate": 3.2414598431910435e-05, + "loss": 0.2565, + "step": 24970 + }, + { + "epoch": 1.25009383210309, + "grad_norm": 53.01020431518555, + "learning_rate": 3.240533076309985e-05, + "loss": 0.6154, + "step": 24980 + }, + { + "epoch": 1.250594269986238, + "grad_norm": 10.719942092895508, + "learning_rate": 3.239606309428927e-05, + "loss": 0.5915, + "step": 24990 + }, + { + "epoch": 1.2510947078693857, + "grad_norm": 0.010967142879962921, + "learning_rate": 3.2386795425478676e-05, + "loss": 0.2519, + "step": 25000 + }, + { + "epoch": 1.2515951457525334, + "grad_norm": 0.7339597940444946, + "learning_rate": 3.237752775666809e-05, + "loss": 0.1487, + "step": 25010 + }, + { + "epoch": 1.2520955836356813, + "grad_norm": 25.645898818969727, + "learning_rate": 3.23682600878575e-05, + "loss": 0.6057, + "step": 25020 + }, + { + "epoch": 1.252596021518829, + "grad_norm": 0.02246260456740856, + "learning_rate": 3.2358992419046916e-05, + "loss": 0.2814, + "step": 25030 + }, + { + "epoch": 1.2530964594019767, + "grad_norm": 0.1909482777118683, + "learning_rate": 3.2349724750236325e-05, + "loss": 0.5829, + "step": 25040 + }, + { + "epoch": 1.2535968972851244, + "grad_norm": 39.65388107299805, + "learning_rate": 3.234045708142574e-05, + "loss": 0.1038, + "step": 25050 + }, + { + "epoch": 1.2540973351682723, + "grad_norm": 22.0649356842041, + "learning_rate": 3.2331189412615156e-05, + "loss": 0.2975, + "step": 25060 + }, + { + "epoch": 1.25459777305142, + "grad_norm": 0.4072197675704956, + "learning_rate": 3.2321921743804565e-05, + "loss": 0.4552, + "step": 25070 + }, + { + "epoch": 1.2550982109345679, + "grad_norm": 0.0047811586409807205, + "learning_rate": 3.2312654074993974e-05, + "loss": 0.0583, + "step": 25080 + }, + { + "epoch": 1.2555986488177155, + "grad_norm": 38.116607666015625, + "learning_rate": 3.230338640618339e-05, + "loss": 0.2544, + "step": 25090 + }, + { + "epoch": 1.2560990867008632, + "grad_norm": 21.727689743041992, + "learning_rate": 3.2294118737372805e-05, + "loss": 0.3664, + "step": 25100 + }, + { + "epoch": 1.256599524584011, + "grad_norm": 3.4558005332946777, + "learning_rate": 3.2284851068562214e-05, + "loss": 0.61, + "step": 25110 + }, + { + "epoch": 1.2570999624671588, + "grad_norm": 11.431394577026367, + "learning_rate": 3.227558339975163e-05, + "loss": 0.16, + "step": 25120 + }, + { + "epoch": 1.2576004003503065, + "grad_norm": 73.73780822753906, + "learning_rate": 3.226631573094104e-05, + "loss": 0.259, + "step": 25130 + }, + { + "epoch": 1.2581008382334542, + "grad_norm": 0.253151535987854, + "learning_rate": 3.2257048062130454e-05, + "loss": 0.1582, + "step": 25140 + }, + { + "epoch": 1.258601276116602, + "grad_norm": 38.77762222290039, + "learning_rate": 3.224778039331986e-05, + "loss": 0.2012, + "step": 25150 + }, + { + "epoch": 1.2591017139997498, + "grad_norm": 23.52592658996582, + "learning_rate": 3.223851272450928e-05, + "loss": 0.5307, + "step": 25160 + }, + { + "epoch": 1.2596021518828975, + "grad_norm": 0.02021344192326069, + "learning_rate": 3.2229245055698694e-05, + "loss": 0.1694, + "step": 25170 + }, + { + "epoch": 1.2601025897660452, + "grad_norm": 0.0002978014526888728, + "learning_rate": 3.22199773868881e-05, + "loss": 0.4582, + "step": 25180 + }, + { + "epoch": 1.260603027649193, + "grad_norm": 0.34600090980529785, + "learning_rate": 3.221070971807751e-05, + "loss": 0.5707, + "step": 25190 + }, + { + "epoch": 1.2611034655323408, + "grad_norm": 12.421102523803711, + "learning_rate": 3.220144204926693e-05, + "loss": 0.2889, + "step": 25200 + }, + { + "epoch": 1.2616039034154887, + "grad_norm": 8.929678916931152, + "learning_rate": 3.219217438045634e-05, + "loss": 0.2506, + "step": 25210 + }, + { + "epoch": 1.2621043412986364, + "grad_norm": 0.021597007289528847, + "learning_rate": 3.218290671164575e-05, + "loss": 0.2555, + "step": 25220 + }, + { + "epoch": 1.262604779181784, + "grad_norm": 1.7276452779769897, + "learning_rate": 3.217363904283517e-05, + "loss": 0.3332, + "step": 25230 + }, + { + "epoch": 1.2631052170649317, + "grad_norm": 108.20531463623047, + "learning_rate": 3.2164371374024576e-05, + "loss": 0.5053, + "step": 25240 + }, + { + "epoch": 1.2636056549480796, + "grad_norm": 0.02180132456123829, + "learning_rate": 3.215510370521399e-05, + "loss": 0.6537, + "step": 25250 + }, + { + "epoch": 1.2641060928312273, + "grad_norm": 0.002664160681888461, + "learning_rate": 3.21458360364034e-05, + "loss": 0.2038, + "step": 25260 + }, + { + "epoch": 1.264606530714375, + "grad_norm": 0.1280452460050583, + "learning_rate": 3.2136568367592816e-05, + "loss": 0.1475, + "step": 25270 + }, + { + "epoch": 1.265106968597523, + "grad_norm": 1.5695642232894897, + "learning_rate": 3.212730069878223e-05, + "loss": 0.1268, + "step": 25280 + }, + { + "epoch": 1.2656074064806706, + "grad_norm": 1.2913159132003784, + "learning_rate": 3.211803302997165e-05, + "loss": 0.5846, + "step": 25290 + }, + { + "epoch": 1.2661078443638183, + "grad_norm": 23.71942138671875, + "learning_rate": 3.210876536116105e-05, + "loss": 0.5384, + "step": 25300 + }, + { + "epoch": 1.266608282246966, + "grad_norm": 65.47146606445312, + "learning_rate": 3.2099497692350465e-05, + "loss": 0.3571, + "step": 25310 + }, + { + "epoch": 1.2671087201301139, + "grad_norm": 4.0215911865234375, + "learning_rate": 3.209023002353988e-05, + "loss": 0.4125, + "step": 25320 + }, + { + "epoch": 1.2676091580132616, + "grad_norm": 0.21253667771816254, + "learning_rate": 3.20809623547293e-05, + "loss": 0.6073, + "step": 25330 + }, + { + "epoch": 1.2681095958964095, + "grad_norm": 0.24772228300571442, + "learning_rate": 3.2071694685918705e-05, + "loss": 0.1309, + "step": 25340 + }, + { + "epoch": 1.2686100337795572, + "grad_norm": 57.81574249267578, + "learning_rate": 3.206242701710812e-05, + "loss": 0.3541, + "step": 25350 + }, + { + "epoch": 1.2691104716627049, + "grad_norm": 4.441884517669678, + "learning_rate": 3.205315934829753e-05, + "loss": 0.257, + "step": 25360 + }, + { + "epoch": 1.2696109095458525, + "grad_norm": 58.33929443359375, + "learning_rate": 3.2043891679486946e-05, + "loss": 0.7096, + "step": 25370 + }, + { + "epoch": 1.2701113474290004, + "grad_norm": 0.9736793637275696, + "learning_rate": 3.2034624010676354e-05, + "loss": 0.4658, + "step": 25380 + }, + { + "epoch": 1.2706117853121481, + "grad_norm": 0.6010904312133789, + "learning_rate": 3.202535634186577e-05, + "loss": 0.4804, + "step": 25390 + }, + { + "epoch": 1.2711122231952958, + "grad_norm": 1.768216848373413, + "learning_rate": 3.2016088673055186e-05, + "loss": 0.1953, + "step": 25400 + }, + { + "epoch": 1.2716126610784437, + "grad_norm": 63.88581085205078, + "learning_rate": 3.2006821004244595e-05, + "loss": 0.4027, + "step": 25410 + }, + { + "epoch": 1.2721130989615914, + "grad_norm": 37.68605041503906, + "learning_rate": 3.1997553335434003e-05, + "loss": 0.5081, + "step": 25420 + }, + { + "epoch": 1.272613536844739, + "grad_norm": 17.94608497619629, + "learning_rate": 3.198828566662342e-05, + "loss": 0.3103, + "step": 25430 + }, + { + "epoch": 1.2731139747278868, + "grad_norm": 2.185079336166382, + "learning_rate": 3.1979017997812835e-05, + "loss": 0.3925, + "step": 25440 + }, + { + "epoch": 1.2736144126110347, + "grad_norm": 4.930015563964844, + "learning_rate": 3.1969750329002244e-05, + "loss": 0.1894, + "step": 25450 + }, + { + "epoch": 1.2741148504941824, + "grad_norm": 27.588510513305664, + "learning_rate": 3.196048266019166e-05, + "loss": 0.078, + "step": 25460 + }, + { + "epoch": 1.2746152883773303, + "grad_norm": 65.5692367553711, + "learning_rate": 3.195121499138107e-05, + "loss": 1.1019, + "step": 25470 + }, + { + "epoch": 1.275115726260478, + "grad_norm": 0.043282803148031235, + "learning_rate": 3.1941947322570484e-05, + "loss": 0.229, + "step": 25480 + }, + { + "epoch": 1.2756161641436257, + "grad_norm": 135.05970764160156, + "learning_rate": 3.193267965375989e-05, + "loss": 0.1858, + "step": 25490 + }, + { + "epoch": 1.2761166020267734, + "grad_norm": 34.16180419921875, + "learning_rate": 3.192341198494931e-05, + "loss": 0.6933, + "step": 25500 + }, + { + "epoch": 1.2766170399099213, + "grad_norm": 6.775991439819336, + "learning_rate": 3.1914144316138724e-05, + "loss": 0.2217, + "step": 25510 + }, + { + "epoch": 1.277117477793069, + "grad_norm": 0.0945289209485054, + "learning_rate": 3.190487664732813e-05, + "loss": 0.3407, + "step": 25520 + }, + { + "epoch": 1.2776179156762166, + "grad_norm": 0.05764389410614967, + "learning_rate": 3.189560897851754e-05, + "loss": 0.4758, + "step": 25530 + }, + { + "epoch": 1.2781183535593645, + "grad_norm": 115.58191680908203, + "learning_rate": 3.188634130970696e-05, + "loss": 0.6461, + "step": 25540 + }, + { + "epoch": 1.2786187914425122, + "grad_norm": 2.2701117992401123, + "learning_rate": 3.187707364089637e-05, + "loss": 0.4329, + "step": 25550 + }, + { + "epoch": 1.27911922932566, + "grad_norm": 5.383398532867432, + "learning_rate": 3.186780597208578e-05, + "loss": 0.6231, + "step": 25560 + }, + { + "epoch": 1.2796196672088076, + "grad_norm": 0.023786313831806183, + "learning_rate": 3.18585383032752e-05, + "loss": 0.3739, + "step": 25570 + }, + { + "epoch": 1.2801201050919555, + "grad_norm": 66.8283462524414, + "learning_rate": 3.1849270634464606e-05, + "loss": 0.5043, + "step": 25580 + }, + { + "epoch": 1.2806205429751032, + "grad_norm": 43.91447448730469, + "learning_rate": 3.184000296565402e-05, + "loss": 0.177, + "step": 25590 + }, + { + "epoch": 1.281120980858251, + "grad_norm": 24.276887893676758, + "learning_rate": 3.183073529684343e-05, + "loss": 0.0637, + "step": 25600 + }, + { + "epoch": 1.2816214187413988, + "grad_norm": 66.63065338134766, + "learning_rate": 3.1821467628032846e-05, + "loss": 0.4281, + "step": 25610 + }, + { + "epoch": 1.2821218566245465, + "grad_norm": 118.58135223388672, + "learning_rate": 3.181219995922226e-05, + "loss": 0.8203, + "step": 25620 + }, + { + "epoch": 1.2826222945076942, + "grad_norm": 0.10387250781059265, + "learning_rate": 3.180293229041167e-05, + "loss": 0.1734, + "step": 25630 + }, + { + "epoch": 1.283122732390842, + "grad_norm": 13.531527519226074, + "learning_rate": 3.179366462160108e-05, + "loss": 0.2647, + "step": 25640 + }, + { + "epoch": 1.2836231702739898, + "grad_norm": 40.721248626708984, + "learning_rate": 3.1784396952790495e-05, + "loss": 0.3527, + "step": 25650 + }, + { + "epoch": 1.2841236081571374, + "grad_norm": 0.05242408812046051, + "learning_rate": 3.177512928397991e-05, + "loss": 0.5191, + "step": 25660 + }, + { + "epoch": 1.2846240460402854, + "grad_norm": 15.900094985961914, + "learning_rate": 3.1765861615169327e-05, + "loss": 0.19, + "step": 25670 + }, + { + "epoch": 1.285124483923433, + "grad_norm": 1.9080920219421387, + "learning_rate": 3.1756593946358735e-05, + "loss": 0.1639, + "step": 25680 + }, + { + "epoch": 1.2856249218065807, + "grad_norm": 0.10456342250108719, + "learning_rate": 3.1747326277548144e-05, + "loss": 0.7107, + "step": 25690 + }, + { + "epoch": 1.2861253596897284, + "grad_norm": 29.505598068237305, + "learning_rate": 3.173805860873756e-05, + "loss": 0.5394, + "step": 25700 + }, + { + "epoch": 1.2866257975728763, + "grad_norm": 0.01856047287583351, + "learning_rate": 3.1728790939926976e-05, + "loss": 0.4025, + "step": 25710 + }, + { + "epoch": 1.287126235456024, + "grad_norm": 0.08047635853290558, + "learning_rate": 3.1719523271116384e-05, + "loss": 0.1554, + "step": 25720 + }, + { + "epoch": 1.287626673339172, + "grad_norm": 35.7668342590332, + "learning_rate": 3.17102556023058e-05, + "loss": 0.3085, + "step": 25730 + }, + { + "epoch": 1.2881271112223196, + "grad_norm": 6.067139625549316, + "learning_rate": 3.170098793349521e-05, + "loss": 0.5111, + "step": 25740 + }, + { + "epoch": 1.2886275491054673, + "grad_norm": 19.321626663208008, + "learning_rate": 3.1691720264684625e-05, + "loss": 0.1238, + "step": 25750 + }, + { + "epoch": 1.289127986988615, + "grad_norm": 0.007127400953322649, + "learning_rate": 3.168245259587403e-05, + "loss": 0.3865, + "step": 25760 + }, + { + "epoch": 1.2896284248717627, + "grad_norm": 3.5871236324310303, + "learning_rate": 3.167318492706345e-05, + "loss": 0.2622, + "step": 25770 + }, + { + "epoch": 1.2901288627549106, + "grad_norm": 0.03722737729549408, + "learning_rate": 3.1663917258252865e-05, + "loss": 0.381, + "step": 25780 + }, + { + "epoch": 1.2906293006380583, + "grad_norm": 0.086258664727211, + "learning_rate": 3.1654649589442274e-05, + "loss": 0.2923, + "step": 25790 + }, + { + "epoch": 1.2911297385212062, + "grad_norm": 0.04158439859747887, + "learning_rate": 3.164538192063168e-05, + "loss": 0.646, + "step": 25800 + }, + { + "epoch": 1.2916301764043538, + "grad_norm": 46.4011344909668, + "learning_rate": 3.16361142518211e-05, + "loss": 0.6502, + "step": 25810 + }, + { + "epoch": 1.2921306142875015, + "grad_norm": 0.028365448117256165, + "learning_rate": 3.1626846583010514e-05, + "loss": 0.1068, + "step": 25820 + }, + { + "epoch": 1.2926310521706492, + "grad_norm": 97.86421203613281, + "learning_rate": 3.161757891419992e-05, + "loss": 0.3358, + "step": 25830 + }, + { + "epoch": 1.2931314900537971, + "grad_norm": 0.23127828538417816, + "learning_rate": 3.160831124538934e-05, + "loss": 0.0236, + "step": 25840 + }, + { + "epoch": 1.2936319279369448, + "grad_norm": 0.5228403806686401, + "learning_rate": 3.159904357657875e-05, + "loss": 0.3094, + "step": 25850 + }, + { + "epoch": 1.2941323658200925, + "grad_norm": 57.85523986816406, + "learning_rate": 3.158977590776816e-05, + "loss": 0.5959, + "step": 25860 + }, + { + "epoch": 1.2946328037032404, + "grad_norm": 41.963470458984375, + "learning_rate": 3.158050823895757e-05, + "loss": 0.1625, + "step": 25870 + }, + { + "epoch": 1.295133241586388, + "grad_norm": 0.034299831837415695, + "learning_rate": 3.157124057014699e-05, + "loss": 1.0904, + "step": 25880 + }, + { + "epoch": 1.2956336794695358, + "grad_norm": 74.34318542480469, + "learning_rate": 3.15619729013364e-05, + "loss": 0.1121, + "step": 25890 + }, + { + "epoch": 1.2961341173526835, + "grad_norm": 46.22909927368164, + "learning_rate": 3.155270523252581e-05, + "loss": 0.361, + "step": 25900 + }, + { + "epoch": 1.2966345552358314, + "grad_norm": 55.94325637817383, + "learning_rate": 3.154343756371522e-05, + "loss": 0.4096, + "step": 25910 + }, + { + "epoch": 1.297134993118979, + "grad_norm": 52.96874237060547, + "learning_rate": 3.1534169894904636e-05, + "loss": 0.5431, + "step": 25920 + }, + { + "epoch": 1.297635431002127, + "grad_norm": 40.5261344909668, + "learning_rate": 3.152490222609405e-05, + "loss": 0.5087, + "step": 25930 + }, + { + "epoch": 1.2981358688852747, + "grad_norm": 1.39090895652771, + "learning_rate": 3.151563455728346e-05, + "loss": 0.433, + "step": 25940 + }, + { + "epoch": 1.2986363067684223, + "grad_norm": 7.597640514373779, + "learning_rate": 3.1506366888472876e-05, + "loss": 0.238, + "step": 25950 + }, + { + "epoch": 1.29913674465157, + "grad_norm": 0.1328655332326889, + "learning_rate": 3.1497099219662285e-05, + "loss": 0.0958, + "step": 25960 + }, + { + "epoch": 1.299637182534718, + "grad_norm": 137.58163452148438, + "learning_rate": 3.14878315508517e-05, + "loss": 0.3067, + "step": 25970 + }, + { + "epoch": 1.3001376204178656, + "grad_norm": 25.347702026367188, + "learning_rate": 3.147856388204111e-05, + "loss": 0.4276, + "step": 25980 + }, + { + "epoch": 1.3006380583010133, + "grad_norm": 9.472867012023926, + "learning_rate": 3.1469296213230525e-05, + "loss": 0.4662, + "step": 25990 + }, + { + "epoch": 1.3011384961841612, + "grad_norm": 0.013485976494848728, + "learning_rate": 3.146002854441994e-05, + "loss": 0.237, + "step": 26000 + }, + { + "epoch": 1.301638934067309, + "grad_norm": 66.51757049560547, + "learning_rate": 3.145076087560935e-05, + "loss": 0.3959, + "step": 26010 + }, + { + "epoch": 1.3021393719504566, + "grad_norm": 25.359601974487305, + "learning_rate": 3.144149320679876e-05, + "loss": 0.4325, + "step": 26020 + }, + { + "epoch": 1.3026398098336043, + "grad_norm": 36.94598388671875, + "learning_rate": 3.1432225537988174e-05, + "loss": 0.7679, + "step": 26030 + }, + { + "epoch": 1.3031402477167522, + "grad_norm": 130.8651123046875, + "learning_rate": 3.142295786917759e-05, + "loss": 0.1398, + "step": 26040 + }, + { + "epoch": 1.3036406855998999, + "grad_norm": 49.948814392089844, + "learning_rate": 3.1413690200367e-05, + "loss": 0.7898, + "step": 26050 + }, + { + "epoch": 1.3041411234830478, + "grad_norm": 33.181941986083984, + "learning_rate": 3.1404422531556414e-05, + "loss": 0.8262, + "step": 26060 + }, + { + "epoch": 1.3046415613661955, + "grad_norm": 15.941652297973633, + "learning_rate": 3.139515486274583e-05, + "loss": 0.6582, + "step": 26070 + }, + { + "epoch": 1.3051419992493432, + "grad_norm": 46.93110275268555, + "learning_rate": 3.138588719393524e-05, + "loss": 0.2577, + "step": 26080 + }, + { + "epoch": 1.3056424371324908, + "grad_norm": 48.83049392700195, + "learning_rate": 3.137661952512465e-05, + "loss": 0.8236, + "step": 26090 + }, + { + "epoch": 1.3061428750156387, + "grad_norm": 1.051157832145691, + "learning_rate": 3.136735185631406e-05, + "loss": 0.3686, + "step": 26100 + }, + { + "epoch": 1.3066433128987864, + "grad_norm": 29.964841842651367, + "learning_rate": 3.135808418750348e-05, + "loss": 0.5301, + "step": 26110 + }, + { + "epoch": 1.3071437507819341, + "grad_norm": 0.4825117886066437, + "learning_rate": 3.1348816518692895e-05, + "loss": 0.4115, + "step": 26120 + }, + { + "epoch": 1.307644188665082, + "grad_norm": 0.8502637147903442, + "learning_rate": 3.13395488498823e-05, + "loss": 0.29, + "step": 26130 + }, + { + "epoch": 1.3081446265482297, + "grad_norm": 34.36320114135742, + "learning_rate": 3.133028118107171e-05, + "loss": 0.1984, + "step": 26140 + }, + { + "epoch": 1.3086450644313774, + "grad_norm": 53.06207275390625, + "learning_rate": 3.132101351226113e-05, + "loss": 0.2278, + "step": 26150 + }, + { + "epoch": 1.309145502314525, + "grad_norm": 5.197866916656494, + "learning_rate": 3.1311745843450544e-05, + "loss": 0.4818, + "step": 26160 + }, + { + "epoch": 1.309645940197673, + "grad_norm": 0.016626892611384392, + "learning_rate": 3.130247817463995e-05, + "loss": 0.2865, + "step": 26170 + }, + { + "epoch": 1.3101463780808207, + "grad_norm": 8.114641189575195, + "learning_rate": 3.129321050582937e-05, + "loss": 0.4983, + "step": 26180 + }, + { + "epoch": 1.3106468159639686, + "grad_norm": 0.010326887480914593, + "learning_rate": 3.128394283701878e-05, + "loss": 0.1819, + "step": 26190 + }, + { + "epoch": 1.3111472538471163, + "grad_norm": 58.01801300048828, + "learning_rate": 3.127467516820819e-05, + "loss": 0.2787, + "step": 26200 + }, + { + "epoch": 1.311647691730264, + "grad_norm": 11.34001350402832, + "learning_rate": 3.12654074993976e-05, + "loss": 0.4171, + "step": 26210 + }, + { + "epoch": 1.3121481296134117, + "grad_norm": 21.987085342407227, + "learning_rate": 3.125613983058702e-05, + "loss": 0.2375, + "step": 26220 + }, + { + "epoch": 1.3126485674965596, + "grad_norm": 0.5860390067100525, + "learning_rate": 3.124687216177643e-05, + "loss": 0.1305, + "step": 26230 + }, + { + "epoch": 1.3131490053797072, + "grad_norm": 97.99138641357422, + "learning_rate": 3.123760449296584e-05, + "loss": 0.611, + "step": 26240 + }, + { + "epoch": 1.313649443262855, + "grad_norm": 0.02431860938668251, + "learning_rate": 3.122833682415525e-05, + "loss": 0.1591, + "step": 26250 + }, + { + "epoch": 1.3141498811460028, + "grad_norm": 1.2348151206970215, + "learning_rate": 3.1219069155344666e-05, + "loss": 0.3853, + "step": 26260 + }, + { + "epoch": 1.3146503190291505, + "grad_norm": 0.25469303131103516, + "learning_rate": 3.120980148653408e-05, + "loss": 0.6061, + "step": 26270 + }, + { + "epoch": 1.3151507569122982, + "grad_norm": 12.190560340881348, + "learning_rate": 3.120053381772349e-05, + "loss": 0.4274, + "step": 26280 + }, + { + "epoch": 1.315651194795446, + "grad_norm": 0.7262856364250183, + "learning_rate": 3.1191266148912906e-05, + "loss": 0.4196, + "step": 26290 + }, + { + "epoch": 1.3161516326785938, + "grad_norm": 2.5230753421783447, + "learning_rate": 3.1181998480102315e-05, + "loss": 0.2905, + "step": 26300 + }, + { + "epoch": 1.3166520705617415, + "grad_norm": 67.5232162475586, + "learning_rate": 3.117273081129173e-05, + "loss": 0.3641, + "step": 26310 + }, + { + "epoch": 1.3171525084448894, + "grad_norm": 0.0027779394295066595, + "learning_rate": 3.116346314248114e-05, + "loss": 0.6973, + "step": 26320 + }, + { + "epoch": 1.317652946328037, + "grad_norm": 0.6451571583747864, + "learning_rate": 3.1154195473670555e-05, + "loss": 0.2456, + "step": 26330 + }, + { + "epoch": 1.3181533842111848, + "grad_norm": 84.83985900878906, + "learning_rate": 3.114492780485997e-05, + "loss": 0.4645, + "step": 26340 + }, + { + "epoch": 1.3186538220943325, + "grad_norm": 0.9033521413803101, + "learning_rate": 3.113566013604938e-05, + "loss": 0.2729, + "step": 26350 + }, + { + "epoch": 1.3191542599774804, + "grad_norm": 0.1514596939086914, + "learning_rate": 3.112639246723879e-05, + "loss": 0.3184, + "step": 26360 + }, + { + "epoch": 1.319654697860628, + "grad_norm": 0.042403023689985275, + "learning_rate": 3.1117124798428204e-05, + "loss": 0.3256, + "step": 26370 + }, + { + "epoch": 1.3201551357437757, + "grad_norm": 28.210206985473633, + "learning_rate": 3.110785712961762e-05, + "loss": 0.3999, + "step": 26380 + }, + { + "epoch": 1.3206555736269237, + "grad_norm": 20.6039981842041, + "learning_rate": 3.109858946080703e-05, + "loss": 0.7235, + "step": 26390 + }, + { + "epoch": 1.3211560115100713, + "grad_norm": 56.847171783447266, + "learning_rate": 3.1089321791996444e-05, + "loss": 0.2301, + "step": 26400 + }, + { + "epoch": 1.321656449393219, + "grad_norm": 13.652616500854492, + "learning_rate": 3.108005412318585e-05, + "loss": 0.3573, + "step": 26410 + }, + { + "epoch": 1.3221568872763667, + "grad_norm": 95.98921966552734, + "learning_rate": 3.107078645437527e-05, + "loss": 0.3539, + "step": 26420 + }, + { + "epoch": 1.3226573251595146, + "grad_norm": 0.010535342618823051, + "learning_rate": 3.106151878556468e-05, + "loss": 0.4446, + "step": 26430 + }, + { + "epoch": 1.3231577630426623, + "grad_norm": 0.005549049004912376, + "learning_rate": 3.105225111675409e-05, + "loss": 0.1271, + "step": 26440 + }, + { + "epoch": 1.3236582009258102, + "grad_norm": 0.07509183883666992, + "learning_rate": 3.104298344794351e-05, + "loss": 0.216, + "step": 26450 + }, + { + "epoch": 1.324158638808958, + "grad_norm": 23.756595611572266, + "learning_rate": 3.103371577913292e-05, + "loss": 0.3719, + "step": 26460 + }, + { + "epoch": 1.3246590766921056, + "grad_norm": 3.739286422729492, + "learning_rate": 3.1024448110322327e-05, + "loss": 0.2173, + "step": 26470 + }, + { + "epoch": 1.3251595145752533, + "grad_norm": 5.661739349365234, + "learning_rate": 3.101518044151174e-05, + "loss": 0.5774, + "step": 26480 + }, + { + "epoch": 1.3256599524584012, + "grad_norm": 0.023205842822790146, + "learning_rate": 3.100591277270116e-05, + "loss": 0.4018, + "step": 26490 + }, + { + "epoch": 1.3261603903415489, + "grad_norm": 41.48164367675781, + "learning_rate": 3.0996645103890573e-05, + "loss": 0.2341, + "step": 26500 + }, + { + "epoch": 1.3266608282246966, + "grad_norm": 14.449092864990234, + "learning_rate": 3.098737743507998e-05, + "loss": 0.5782, + "step": 26510 + }, + { + "epoch": 1.3271612661078445, + "grad_norm": 2.085636615753174, + "learning_rate": 3.097810976626939e-05, + "loss": 0.1931, + "step": 26520 + }, + { + "epoch": 1.3276617039909921, + "grad_norm": 3.239805221557617, + "learning_rate": 3.096884209745881e-05, + "loss": 0.1452, + "step": 26530 + }, + { + "epoch": 1.3281621418741398, + "grad_norm": 142.23074340820312, + "learning_rate": 3.095957442864822e-05, + "loss": 0.216, + "step": 26540 + }, + { + "epoch": 1.3286625797572875, + "grad_norm": 0.003358985297381878, + "learning_rate": 3.095030675983763e-05, + "loss": 0.3698, + "step": 26550 + }, + { + "epoch": 1.3291630176404354, + "grad_norm": 0.04363472759723663, + "learning_rate": 3.094103909102705e-05, + "loss": 0.6707, + "step": 26560 + }, + { + "epoch": 1.3296634555235831, + "grad_norm": 6.178335666656494, + "learning_rate": 3.0931771422216456e-05, + "loss": 0.0195, + "step": 26570 + }, + { + "epoch": 1.330163893406731, + "grad_norm": 1.4503765106201172, + "learning_rate": 3.092250375340587e-05, + "loss": 0.1179, + "step": 26580 + }, + { + "epoch": 1.3306643312898787, + "grad_norm": 0.8445348143577576, + "learning_rate": 3.091323608459528e-05, + "loss": 0.3589, + "step": 26590 + }, + { + "epoch": 1.3311647691730264, + "grad_norm": 0.003450017888098955, + "learning_rate": 3.0903968415784696e-05, + "loss": 0.1637, + "step": 26600 + }, + { + "epoch": 1.331665207056174, + "grad_norm": 0.03628704696893692, + "learning_rate": 3.089470074697411e-05, + "loss": 0.9858, + "step": 26610 + }, + { + "epoch": 1.3321656449393218, + "grad_norm": 1.4975255727767944, + "learning_rate": 3.088543307816352e-05, + "loss": 0.1177, + "step": 26620 + }, + { + "epoch": 1.3326660828224697, + "grad_norm": 15.617297172546387, + "learning_rate": 3.087616540935293e-05, + "loss": 0.501, + "step": 26630 + }, + { + "epoch": 1.3331665207056174, + "grad_norm": 5.116866588592529, + "learning_rate": 3.0866897740542345e-05, + "loss": 0.0796, + "step": 26640 + }, + { + "epoch": 1.3336669585887653, + "grad_norm": 0.07685568928718567, + "learning_rate": 3.085763007173176e-05, + "loss": 0.4227, + "step": 26650 + }, + { + "epoch": 1.334167396471913, + "grad_norm": 20.308048248291016, + "learning_rate": 3.084836240292117e-05, + "loss": 0.7015, + "step": 26660 + }, + { + "epoch": 1.3346678343550606, + "grad_norm": 6.387649059295654, + "learning_rate": 3.0839094734110585e-05, + "loss": 0.293, + "step": 26670 + }, + { + "epoch": 1.3351682722382083, + "grad_norm": 0.01580304279923439, + "learning_rate": 3.0829827065299994e-05, + "loss": 0.1158, + "step": 26680 + }, + { + "epoch": 1.3356687101213562, + "grad_norm": 130.09300231933594, + "learning_rate": 3.082055939648941e-05, + "loss": 0.3634, + "step": 26690 + }, + { + "epoch": 1.336169148004504, + "grad_norm": 0.018883459270000458, + "learning_rate": 3.081129172767882e-05, + "loss": 0.2768, + "step": 26700 + }, + { + "epoch": 1.3366695858876516, + "grad_norm": 36.090850830078125, + "learning_rate": 3.0802024058868234e-05, + "loss": 0.1781, + "step": 26710 + }, + { + "epoch": 1.3371700237707995, + "grad_norm": 116.32730102539062, + "learning_rate": 3.079275639005765e-05, + "loss": 0.4202, + "step": 26720 + }, + { + "epoch": 1.3376704616539472, + "grad_norm": 21.193140029907227, + "learning_rate": 3.078348872124706e-05, + "loss": 0.2178, + "step": 26730 + }, + { + "epoch": 1.338170899537095, + "grad_norm": 62.187225341796875, + "learning_rate": 3.077422105243647e-05, + "loss": 0.1414, + "step": 26740 + }, + { + "epoch": 1.3386713374202426, + "grad_norm": 1.7268197536468506, + "learning_rate": 3.076495338362588e-05, + "loss": 0.4169, + "step": 26750 + }, + { + "epoch": 1.3391717753033905, + "grad_norm": 0.31466394662857056, + "learning_rate": 3.07556857148153e-05, + "loss": 0.3596, + "step": 26760 + }, + { + "epoch": 1.3396722131865382, + "grad_norm": 0.008163324557244778, + "learning_rate": 3.074641804600471e-05, + "loss": 0.4229, + "step": 26770 + }, + { + "epoch": 1.340172651069686, + "grad_norm": 0.040303993970155716, + "learning_rate": 3.073715037719412e-05, + "loss": 0.4387, + "step": 26780 + }, + { + "epoch": 1.3406730889528338, + "grad_norm": 44.57462692260742, + "learning_rate": 3.072788270838354e-05, + "loss": 0.6168, + "step": 26790 + }, + { + "epoch": 1.3411735268359815, + "grad_norm": 0.38391587138175964, + "learning_rate": 3.071861503957295e-05, + "loss": 0.3945, + "step": 26800 + }, + { + "epoch": 1.3416739647191291, + "grad_norm": 0.15322722494602203, + "learning_rate": 3.0709347370762356e-05, + "loss": 0.5334, + "step": 26810 + }, + { + "epoch": 1.342174402602277, + "grad_norm": 0.14730477333068848, + "learning_rate": 3.070007970195177e-05, + "loss": 0.4167, + "step": 26820 + }, + { + "epoch": 1.3426748404854247, + "grad_norm": 42.336368560791016, + "learning_rate": 3.069081203314119e-05, + "loss": 0.3268, + "step": 26830 + }, + { + "epoch": 1.3431752783685724, + "grad_norm": 0.08134795725345612, + "learning_rate": 3.0681544364330597e-05, + "loss": 0.084, + "step": 26840 + }, + { + "epoch": 1.3436757162517203, + "grad_norm": 0.5443792343139648, + "learning_rate": 3.0672276695520005e-05, + "loss": 0.1035, + "step": 26850 + }, + { + "epoch": 1.344176154134868, + "grad_norm": 46.5866584777832, + "learning_rate": 3.066300902670942e-05, + "loss": 0.4056, + "step": 26860 + }, + { + "epoch": 1.3446765920180157, + "grad_norm": 7.487709045410156, + "learning_rate": 3.065374135789884e-05, + "loss": 0.3565, + "step": 26870 + }, + { + "epoch": 1.3451770299011634, + "grad_norm": 0.1584254503250122, + "learning_rate": 3.0644473689088246e-05, + "loss": 0.3969, + "step": 26880 + }, + { + "epoch": 1.3456774677843113, + "grad_norm": 23.94097137451172, + "learning_rate": 3.063520602027766e-05, + "loss": 0.233, + "step": 26890 + }, + { + "epoch": 1.346177905667459, + "grad_norm": 26.32032012939453, + "learning_rate": 3.062593835146708e-05, + "loss": 0.3076, + "step": 26900 + }, + { + "epoch": 1.346678343550607, + "grad_norm": 30.669004440307617, + "learning_rate": 3.0616670682656486e-05, + "loss": 0.2298, + "step": 26910 + }, + { + "epoch": 1.3471787814337546, + "grad_norm": 33.33097839355469, + "learning_rate": 3.0607403013845895e-05, + "loss": 0.511, + "step": 26920 + }, + { + "epoch": 1.3476792193169023, + "grad_norm": 0.2361687272787094, + "learning_rate": 3.059813534503531e-05, + "loss": 0.1783, + "step": 26930 + }, + { + "epoch": 1.34817965720005, + "grad_norm": 43.02874755859375, + "learning_rate": 3.0588867676224726e-05, + "loss": 0.1862, + "step": 26940 + }, + { + "epoch": 1.3486800950831979, + "grad_norm": 6.129516124725342, + "learning_rate": 3.057960000741414e-05, + "loss": 0.3583, + "step": 26950 + }, + { + "epoch": 1.3491805329663455, + "grad_norm": 0.048287443816661835, + "learning_rate": 3.0570332338603544e-05, + "loss": 0.449, + "step": 26960 + }, + { + "epoch": 1.3496809708494932, + "grad_norm": 108.03595733642578, + "learning_rate": 3.056106466979296e-05, + "loss": 0.4217, + "step": 26970 + }, + { + "epoch": 1.3501814087326411, + "grad_norm": 0.038564518094062805, + "learning_rate": 3.0551797000982375e-05, + "loss": 0.2694, + "step": 26980 + }, + { + "epoch": 1.3506818466157888, + "grad_norm": 32.8445930480957, + "learning_rate": 3.054252933217179e-05, + "loss": 0.3068, + "step": 26990 + }, + { + "epoch": 1.3511822844989365, + "grad_norm": 40.999366760253906, + "learning_rate": 3.05332616633612e-05, + "loss": 0.1881, + "step": 27000 + }, + { + "epoch": 1.3516827223820842, + "grad_norm": 0.021234657615423203, + "learning_rate": 3.0523993994550615e-05, + "loss": 0.512, + "step": 27010 + }, + { + "epoch": 1.352183160265232, + "grad_norm": 0.1568947732448578, + "learning_rate": 3.0514726325740024e-05, + "loss": 0.4223, + "step": 27020 + }, + { + "epoch": 1.3526835981483798, + "grad_norm": 115.66497039794922, + "learning_rate": 3.0505458656929436e-05, + "loss": 0.1143, + "step": 27030 + }, + { + "epoch": 1.3531840360315277, + "grad_norm": 0.018956132233142853, + "learning_rate": 3.049619098811885e-05, + "loss": 0.4501, + "step": 27040 + }, + { + "epoch": 1.3536844739146754, + "grad_norm": 32.118988037109375, + "learning_rate": 3.0486923319308264e-05, + "loss": 0.2235, + "step": 27050 + }, + { + "epoch": 1.354184911797823, + "grad_norm": 34.50407409667969, + "learning_rate": 3.0477655650497676e-05, + "loss": 0.4332, + "step": 27060 + }, + { + "epoch": 1.3546853496809708, + "grad_norm": 61.809173583984375, + "learning_rate": 3.0468387981687085e-05, + "loss": 0.4461, + "step": 27070 + }, + { + "epoch": 1.3551857875641187, + "grad_norm": 0.006437285337597132, + "learning_rate": 3.04591203128765e-05, + "loss": 0.2101, + "step": 27080 + }, + { + "epoch": 1.3556862254472664, + "grad_norm": 0.006831857841461897, + "learning_rate": 3.0449852644065913e-05, + "loss": 0.1293, + "step": 27090 + }, + { + "epoch": 1.356186663330414, + "grad_norm": 3.475320339202881, + "learning_rate": 3.0440584975255325e-05, + "loss": 0.16, + "step": 27100 + }, + { + "epoch": 1.356687101213562, + "grad_norm": 0.5577793717384338, + "learning_rate": 3.043131730644474e-05, + "loss": 0.4032, + "step": 27110 + }, + { + "epoch": 1.3571875390967096, + "grad_norm": 0.04360866919159889, + "learning_rate": 3.0422049637634153e-05, + "loss": 0.4972, + "step": 27120 + }, + { + "epoch": 1.3576879769798573, + "grad_norm": 22.410245895385742, + "learning_rate": 3.0412781968823562e-05, + "loss": 0.8098, + "step": 27130 + }, + { + "epoch": 1.358188414863005, + "grad_norm": 0.1615368127822876, + "learning_rate": 3.0403514300012974e-05, + "loss": 0.6385, + "step": 27140 + }, + { + "epoch": 1.358688852746153, + "grad_norm": 8.499972343444824, + "learning_rate": 3.039424663120239e-05, + "loss": 0.2177, + "step": 27150 + }, + { + "epoch": 1.3591892906293006, + "grad_norm": 0.02375364489853382, + "learning_rate": 3.0384978962391802e-05, + "loss": 0.185, + "step": 27160 + }, + { + "epoch": 1.3596897285124485, + "grad_norm": 232.2373504638672, + "learning_rate": 3.0375711293581218e-05, + "loss": 0.6904, + "step": 27170 + }, + { + "epoch": 1.3601901663955962, + "grad_norm": 176.88076782226562, + "learning_rate": 3.0366443624770623e-05, + "loss": 0.1989, + "step": 27180 + }, + { + "epoch": 1.3606906042787439, + "grad_norm": 51.32441329956055, + "learning_rate": 3.035717595596004e-05, + "loss": 0.4243, + "step": 27190 + }, + { + "epoch": 1.3611910421618916, + "grad_norm": 0.05249958112835884, + "learning_rate": 3.034790828714945e-05, + "loss": 0.2383, + "step": 27200 + }, + { + "epoch": 1.3616914800450395, + "grad_norm": 4.603316307067871, + "learning_rate": 3.0338640618338863e-05, + "loss": 0.3605, + "step": 27210 + }, + { + "epoch": 1.3621919179281872, + "grad_norm": 48.64521408081055, + "learning_rate": 3.032937294952828e-05, + "loss": 0.3883, + "step": 27220 + }, + { + "epoch": 1.3626923558113349, + "grad_norm": 0.00607707817107439, + "learning_rate": 3.032010528071769e-05, + "loss": 0.2452, + "step": 27230 + }, + { + "epoch": 1.3631927936944828, + "grad_norm": 13.539310455322266, + "learning_rate": 3.03108376119071e-05, + "loss": 0.4285, + "step": 27240 + }, + { + "epoch": 1.3636932315776304, + "grad_norm": 14.68066120147705, + "learning_rate": 3.0301569943096512e-05, + "loss": 0.3378, + "step": 27250 + }, + { + "epoch": 1.3641936694607781, + "grad_norm": 0.7992494702339172, + "learning_rate": 3.0292302274285928e-05, + "loss": 0.4289, + "step": 27260 + }, + { + "epoch": 1.3646941073439258, + "grad_norm": 68.95987701416016, + "learning_rate": 3.028303460547534e-05, + "loss": 0.7253, + "step": 27270 + }, + { + "epoch": 1.3651945452270737, + "grad_norm": 0.03275693953037262, + "learning_rate": 3.0273766936664756e-05, + "loss": 0.6345, + "step": 27280 + }, + { + "epoch": 1.3656949831102214, + "grad_norm": 0.06717772036790848, + "learning_rate": 3.026449926785416e-05, + "loss": 0.1385, + "step": 27290 + }, + { + "epoch": 1.3661954209933693, + "grad_norm": 56.27472686767578, + "learning_rate": 3.0255231599043577e-05, + "loss": 0.4458, + "step": 27300 + }, + { + "epoch": 1.366695858876517, + "grad_norm": 0.4582803547382355, + "learning_rate": 3.024596393023299e-05, + "loss": 0.3922, + "step": 27310 + }, + { + "epoch": 1.3671962967596647, + "grad_norm": 0.0037543121725320816, + "learning_rate": 3.0236696261422405e-05, + "loss": 0.3477, + "step": 27320 + }, + { + "epoch": 1.3676967346428124, + "grad_norm": 0.7843833565711975, + "learning_rate": 3.0227428592611817e-05, + "loss": 0.1573, + "step": 27330 + }, + { + "epoch": 1.3681971725259603, + "grad_norm": 88.82154846191406, + "learning_rate": 3.021816092380123e-05, + "loss": 0.6867, + "step": 27340 + }, + { + "epoch": 1.368697610409108, + "grad_norm": 0.05487775057554245, + "learning_rate": 3.0208893254990638e-05, + "loss": 0.524, + "step": 27350 + }, + { + "epoch": 1.3691980482922557, + "grad_norm": 18.351966857910156, + "learning_rate": 3.0199625586180054e-05, + "loss": 0.511, + "step": 27360 + }, + { + "epoch": 1.3696984861754036, + "grad_norm": 1.4974675178527832, + "learning_rate": 3.0190357917369466e-05, + "loss": 0.2025, + "step": 27370 + }, + { + "epoch": 1.3701989240585513, + "grad_norm": 83.80244445800781, + "learning_rate": 3.0181090248558878e-05, + "loss": 0.5265, + "step": 27380 + }, + { + "epoch": 1.370699361941699, + "grad_norm": 0.03301549330353737, + "learning_rate": 3.0171822579748294e-05, + "loss": 0.2723, + "step": 27390 + }, + { + "epoch": 1.3711997998248466, + "grad_norm": 66.17324829101562, + "learning_rate": 3.0162554910937703e-05, + "loss": 0.3108, + "step": 27400 + }, + { + "epoch": 1.3717002377079945, + "grad_norm": 25.363840103149414, + "learning_rate": 3.0153287242127115e-05, + "loss": 0.4931, + "step": 27410 + }, + { + "epoch": 1.3722006755911422, + "grad_norm": 0.0332595594227314, + "learning_rate": 3.0144019573316527e-05, + "loss": 0.1547, + "step": 27420 + }, + { + "epoch": 1.3727011134742901, + "grad_norm": 0.5273736715316772, + "learning_rate": 3.0134751904505943e-05, + "loss": 0.3569, + "step": 27430 + }, + { + "epoch": 1.3732015513574378, + "grad_norm": 39.88544845581055, + "learning_rate": 3.0125484235695355e-05, + "loss": 0.2344, + "step": 27440 + }, + { + "epoch": 1.3737019892405855, + "grad_norm": 49.54731750488281, + "learning_rate": 3.011621656688477e-05, + "loss": 0.5491, + "step": 27450 + }, + { + "epoch": 1.3742024271237332, + "grad_norm": 40.49800109863281, + "learning_rate": 3.0106948898074176e-05, + "loss": 0.2868, + "step": 27460 + }, + { + "epoch": 1.374702865006881, + "grad_norm": 0.1913696974515915, + "learning_rate": 3.0097681229263592e-05, + "loss": 0.3648, + "step": 27470 + }, + { + "epoch": 1.3752033028900288, + "grad_norm": 30.740549087524414, + "learning_rate": 3.0088413560453004e-05, + "loss": 0.4501, + "step": 27480 + }, + { + "epoch": 1.3757037407731765, + "grad_norm": 0.002158957999199629, + "learning_rate": 3.007914589164242e-05, + "loss": 0.5309, + "step": 27490 + }, + { + "epoch": 1.3762041786563244, + "grad_norm": 6.604252338409424, + "learning_rate": 3.0069878222831832e-05, + "loss": 0.165, + "step": 27500 + }, + { + "epoch": 1.376704616539472, + "grad_norm": 0.15444710850715637, + "learning_rate": 3.0060610554021244e-05, + "loss": 0.6595, + "step": 27510 + }, + { + "epoch": 1.3772050544226198, + "grad_norm": 0.006634024437516928, + "learning_rate": 3.0051342885210653e-05, + "loss": 0.5659, + "step": 27520 + }, + { + "epoch": 1.3777054923057674, + "grad_norm": 5.195911407470703, + "learning_rate": 3.004207521640007e-05, + "loss": 0.5027, + "step": 27530 + }, + { + "epoch": 1.3782059301889154, + "grad_norm": 0.014317233115434647, + "learning_rate": 3.003280754758948e-05, + "loss": 0.103, + "step": 27540 + }, + { + "epoch": 1.378706368072063, + "grad_norm": 86.60005950927734, + "learning_rate": 3.0023539878778893e-05, + "loss": 0.1833, + "step": 27550 + }, + { + "epoch": 1.379206805955211, + "grad_norm": 82.52273559570312, + "learning_rate": 3.001427220996831e-05, + "loss": 0.5908, + "step": 27560 + }, + { + "epoch": 1.3797072438383586, + "grad_norm": 121.24150085449219, + "learning_rate": 3.0005004541157718e-05, + "loss": 0.3022, + "step": 27570 + }, + { + "epoch": 1.3802076817215063, + "grad_norm": 0.07619506865739822, + "learning_rate": 2.999573687234713e-05, + "loss": 0.298, + "step": 27580 + }, + { + "epoch": 1.380708119604654, + "grad_norm": 42.704673767089844, + "learning_rate": 2.9986469203536542e-05, + "loss": 0.4256, + "step": 27590 + }, + { + "epoch": 1.3812085574878017, + "grad_norm": 60.81879806518555, + "learning_rate": 2.9977201534725958e-05, + "loss": 0.2508, + "step": 27600 + }, + { + "epoch": 1.3817089953709496, + "grad_norm": 4.898313045501709, + "learning_rate": 2.996793386591537e-05, + "loss": 0.2137, + "step": 27610 + }, + { + "epoch": 1.3822094332540973, + "grad_norm": 35.52041244506836, + "learning_rate": 2.9958666197104786e-05, + "loss": 0.3824, + "step": 27620 + }, + { + "epoch": 1.3827098711372452, + "grad_norm": 36.980934143066406, + "learning_rate": 2.994939852829419e-05, + "loss": 0.3735, + "step": 27630 + }, + { + "epoch": 1.3832103090203929, + "grad_norm": 16.065223693847656, + "learning_rate": 2.9940130859483607e-05, + "loss": 0.2613, + "step": 27640 + }, + { + "epoch": 1.3837107469035406, + "grad_norm": 1.1205030679702759, + "learning_rate": 2.993086319067302e-05, + "loss": 0.605, + "step": 27650 + }, + { + "epoch": 1.3842111847866883, + "grad_norm": 59.70010757446289, + "learning_rate": 2.9921595521862435e-05, + "loss": 0.4743, + "step": 27660 + }, + { + "epoch": 1.3847116226698362, + "grad_norm": 0.011567596346139908, + "learning_rate": 2.9912327853051847e-05, + "loss": 0.4357, + "step": 27670 + }, + { + "epoch": 1.3852120605529838, + "grad_norm": 0.06413094699382782, + "learning_rate": 2.9903060184241256e-05, + "loss": 0.2783, + "step": 27680 + }, + { + "epoch": 1.3857124984361315, + "grad_norm": 123.05843353271484, + "learning_rate": 2.9893792515430668e-05, + "loss": 0.2101, + "step": 27690 + }, + { + "epoch": 1.3862129363192794, + "grad_norm": 0.005530728492885828, + "learning_rate": 2.9884524846620084e-05, + "loss": 0.0324, + "step": 27700 + }, + { + "epoch": 1.3867133742024271, + "grad_norm": 30.987451553344727, + "learning_rate": 2.9875257177809496e-05, + "loss": 0.6756, + "step": 27710 + }, + { + "epoch": 1.3872138120855748, + "grad_norm": 0.02325795404613018, + "learning_rate": 2.9865989508998908e-05, + "loss": 0.4124, + "step": 27720 + }, + { + "epoch": 1.3877142499687225, + "grad_norm": 27.907869338989258, + "learning_rate": 2.9856721840188324e-05, + "loss": 0.7846, + "step": 27730 + }, + { + "epoch": 1.3882146878518704, + "grad_norm": 61.68427276611328, + "learning_rate": 2.9847454171377733e-05, + "loss": 0.4773, + "step": 27740 + }, + { + "epoch": 1.388715125735018, + "grad_norm": 59.338600158691406, + "learning_rate": 2.9838186502567145e-05, + "loss": 0.2927, + "step": 27750 + }, + { + "epoch": 1.389215563618166, + "grad_norm": 49.642974853515625, + "learning_rate": 2.9828918833756557e-05, + "loss": 0.0655, + "step": 27760 + }, + { + "epoch": 1.3897160015013137, + "grad_norm": 4.70004415512085, + "learning_rate": 2.9819651164945973e-05, + "loss": 0.059, + "step": 27770 + }, + { + "epoch": 1.3902164393844614, + "grad_norm": 21.763235092163086, + "learning_rate": 2.9810383496135385e-05, + "loss": 0.5319, + "step": 27780 + }, + { + "epoch": 1.390716877267609, + "grad_norm": 0.008271574042737484, + "learning_rate": 2.9801115827324794e-05, + "loss": 0.4262, + "step": 27790 + }, + { + "epoch": 1.391217315150757, + "grad_norm": 63.36345291137695, + "learning_rate": 2.9791848158514206e-05, + "loss": 0.2371, + "step": 27800 + }, + { + "epoch": 1.3917177530339047, + "grad_norm": 16.861736297607422, + "learning_rate": 2.9782580489703622e-05, + "loss": 0.0745, + "step": 27810 + }, + { + "epoch": 1.3922181909170523, + "grad_norm": 0.2280375063419342, + "learning_rate": 2.9773312820893034e-05, + "loss": 0.1991, + "step": 27820 + }, + { + "epoch": 1.3927186288002003, + "grad_norm": 77.36793518066406, + "learning_rate": 2.976404515208245e-05, + "loss": 0.7623, + "step": 27830 + }, + { + "epoch": 1.393219066683348, + "grad_norm": 20.263120651245117, + "learning_rate": 2.9754777483271862e-05, + "loss": 0.6161, + "step": 27840 + }, + { + "epoch": 1.3937195045664956, + "grad_norm": 28.64116096496582, + "learning_rate": 2.974550981446127e-05, + "loss": 0.191, + "step": 27850 + }, + { + "epoch": 1.3942199424496433, + "grad_norm": 41.48223114013672, + "learning_rate": 2.9736242145650683e-05, + "loss": 0.4847, + "step": 27860 + }, + { + "epoch": 1.3947203803327912, + "grad_norm": 20.577077865600586, + "learning_rate": 2.97269744768401e-05, + "loss": 0.2169, + "step": 27870 + }, + { + "epoch": 1.395220818215939, + "grad_norm": 56.006351470947266, + "learning_rate": 2.971770680802951e-05, + "loss": 0.6385, + "step": 27880 + }, + { + "epoch": 1.3957212560990868, + "grad_norm": 0.17331135272979736, + "learning_rate": 2.9708439139218923e-05, + "loss": 0.2497, + "step": 27890 + }, + { + "epoch": 1.3962216939822345, + "grad_norm": 14.623361587524414, + "learning_rate": 2.9699171470408332e-05, + "loss": 0.2656, + "step": 27900 + }, + { + "epoch": 1.3967221318653822, + "grad_norm": 0.009085774421691895, + "learning_rate": 2.9689903801597748e-05, + "loss": 0.429, + "step": 27910 + }, + { + "epoch": 1.3972225697485299, + "grad_norm": 0.6332621574401855, + "learning_rate": 2.968063613278716e-05, + "loss": 0.3247, + "step": 27920 + }, + { + "epoch": 1.3977230076316778, + "grad_norm": 0.5351153016090393, + "learning_rate": 2.9671368463976572e-05, + "loss": 0.201, + "step": 27930 + }, + { + "epoch": 1.3982234455148255, + "grad_norm": 47.20140075683594, + "learning_rate": 2.9662100795165988e-05, + "loss": 0.2539, + "step": 27940 + }, + { + "epoch": 1.3987238833979732, + "grad_norm": 62.36626434326172, + "learning_rate": 2.96528331263554e-05, + "loss": 0.6733, + "step": 27950 + }, + { + "epoch": 1.399224321281121, + "grad_norm": 23.802635192871094, + "learning_rate": 2.964356545754481e-05, + "loss": 0.4617, + "step": 27960 + }, + { + "epoch": 1.3997247591642688, + "grad_norm": 29.035470962524414, + "learning_rate": 2.963429778873422e-05, + "loss": 0.2794, + "step": 27970 + }, + { + "epoch": 1.4002251970474164, + "grad_norm": 0.1005229502916336, + "learning_rate": 2.9625030119923637e-05, + "loss": 0.1734, + "step": 27980 + }, + { + "epoch": 1.4007256349305641, + "grad_norm": 66.626953125, + "learning_rate": 2.961576245111305e-05, + "loss": 0.4563, + "step": 27990 + }, + { + "epoch": 1.401226072813712, + "grad_norm": 0.15667283535003662, + "learning_rate": 2.9606494782302465e-05, + "loss": 0.0199, + "step": 28000 + }, + { + "epoch": 1.4017265106968597, + "grad_norm": 37.95948028564453, + "learning_rate": 2.959722711349187e-05, + "loss": 0.2427, + "step": 28010 + }, + { + "epoch": 1.4022269485800076, + "grad_norm": 63.27172088623047, + "learning_rate": 2.9587959444681286e-05, + "loss": 0.8229, + "step": 28020 + }, + { + "epoch": 1.4027273864631553, + "grad_norm": 0.14672373235225677, + "learning_rate": 2.9578691775870698e-05, + "loss": 0.2496, + "step": 28030 + }, + { + "epoch": 1.403227824346303, + "grad_norm": 5.515280246734619, + "learning_rate": 2.9569424107060114e-05, + "loss": 0.5646, + "step": 28040 + }, + { + "epoch": 1.4037282622294507, + "grad_norm": 0.08112354576587677, + "learning_rate": 2.9560156438249526e-05, + "loss": 0.312, + "step": 28050 + }, + { + "epoch": 1.4042287001125986, + "grad_norm": 0.1536611020565033, + "learning_rate": 2.9550888769438938e-05, + "loss": 0.2839, + "step": 28060 + }, + { + "epoch": 1.4047291379957463, + "grad_norm": 0.21257562935352325, + "learning_rate": 2.9541621100628347e-05, + "loss": 0.3418, + "step": 28070 + }, + { + "epoch": 1.405229575878894, + "grad_norm": 59.44390869140625, + "learning_rate": 2.953235343181776e-05, + "loss": 0.4666, + "step": 28080 + }, + { + "epoch": 1.4057300137620419, + "grad_norm": 3.0662519931793213, + "learning_rate": 2.9523085763007175e-05, + "loss": 0.1807, + "step": 28090 + }, + { + "epoch": 1.4062304516451896, + "grad_norm": 0.09431280940771103, + "learning_rate": 2.9513818094196587e-05, + "loss": 0.1236, + "step": 28100 + }, + { + "epoch": 1.4067308895283372, + "grad_norm": 10.693401336669922, + "learning_rate": 2.9504550425386003e-05, + "loss": 0.2928, + "step": 28110 + }, + { + "epoch": 1.407231327411485, + "grad_norm": 0.07126208394765854, + "learning_rate": 2.9495282756575408e-05, + "loss": 0.2114, + "step": 28120 + }, + { + "epoch": 1.4077317652946328, + "grad_norm": 0.07615149021148682, + "learning_rate": 2.9486015087764824e-05, + "loss": 0.5925, + "step": 28130 + }, + { + "epoch": 1.4082322031777805, + "grad_norm": 40.16826629638672, + "learning_rate": 2.9476747418954236e-05, + "loss": 0.5241, + "step": 28140 + }, + { + "epoch": 1.4087326410609284, + "grad_norm": 46.45052719116211, + "learning_rate": 2.946747975014365e-05, + "loss": 0.2055, + "step": 28150 + }, + { + "epoch": 1.4092330789440761, + "grad_norm": 0.036779727786779404, + "learning_rate": 2.9458212081333064e-05, + "loss": 0.2759, + "step": 28160 + }, + { + "epoch": 1.4097335168272238, + "grad_norm": 84.28801727294922, + "learning_rate": 2.9448944412522476e-05, + "loss": 0.2178, + "step": 28170 + }, + { + "epoch": 1.4102339547103715, + "grad_norm": 29.614110946655273, + "learning_rate": 2.9439676743711885e-05, + "loss": 0.5461, + "step": 28180 + }, + { + "epoch": 1.4107343925935194, + "grad_norm": 0.2639738619327545, + "learning_rate": 2.94304090749013e-05, + "loss": 0.1361, + "step": 28190 + }, + { + "epoch": 1.411234830476667, + "grad_norm": 0.3320915400981903, + "learning_rate": 2.9421141406090713e-05, + "loss": 0.2937, + "step": 28200 + }, + { + "epoch": 1.4117352683598148, + "grad_norm": 0.003975837957113981, + "learning_rate": 2.9411873737280125e-05, + "loss": 0.0325, + "step": 28210 + }, + { + "epoch": 1.4122357062429627, + "grad_norm": 0.013092347420752048, + "learning_rate": 2.940260606846954e-05, + "loss": 0.9179, + "step": 28220 + }, + { + "epoch": 1.4127361441261104, + "grad_norm": 11.463540077209473, + "learning_rate": 2.939333839965895e-05, + "loss": 0.1734, + "step": 28230 + }, + { + "epoch": 1.413236582009258, + "grad_norm": 21.538944244384766, + "learning_rate": 2.9384070730848362e-05, + "loss": 0.8279, + "step": 28240 + }, + { + "epoch": 1.4137370198924057, + "grad_norm": 103.47526550292969, + "learning_rate": 2.9374803062037774e-05, + "loss": 0.2333, + "step": 28250 + }, + { + "epoch": 1.4142374577755537, + "grad_norm": 17.984619140625, + "learning_rate": 2.936553539322719e-05, + "loss": 0.4939, + "step": 28260 + }, + { + "epoch": 1.4147378956587013, + "grad_norm": 30.6973934173584, + "learning_rate": 2.9356267724416602e-05, + "loss": 0.0915, + "step": 28270 + }, + { + "epoch": 1.4152383335418492, + "grad_norm": 0.5291775465011597, + "learning_rate": 2.9347000055606018e-05, + "loss": 0.3696, + "step": 28280 + }, + { + "epoch": 1.415738771424997, + "grad_norm": 26.12749671936035, + "learning_rate": 2.9337732386795423e-05, + "loss": 0.2957, + "step": 28290 + }, + { + "epoch": 1.4162392093081446, + "grad_norm": 0.8770215511322021, + "learning_rate": 2.932846471798484e-05, + "loss": 0.4438, + "step": 28300 + }, + { + "epoch": 1.4167396471912923, + "grad_norm": 3.898805618286133, + "learning_rate": 2.931919704917425e-05, + "loss": 0.1737, + "step": 28310 + }, + { + "epoch": 1.4172400850744402, + "grad_norm": 55.540008544921875, + "learning_rate": 2.9309929380363667e-05, + "loss": 0.3838, + "step": 28320 + }, + { + "epoch": 1.417740522957588, + "grad_norm": 79.96102142333984, + "learning_rate": 2.930066171155308e-05, + "loss": 0.6491, + "step": 28330 + }, + { + "epoch": 1.4182409608407356, + "grad_norm": 82.6349868774414, + "learning_rate": 2.929139404274249e-05, + "loss": 0.1542, + "step": 28340 + }, + { + "epoch": 1.4187413987238835, + "grad_norm": 0.03665431588888168, + "learning_rate": 2.92821263739319e-05, + "loss": 0.1134, + "step": 28350 + }, + { + "epoch": 1.4192418366070312, + "grad_norm": 0.01184609904885292, + "learning_rate": 2.9272858705121316e-05, + "loss": 0.4904, + "step": 28360 + }, + { + "epoch": 1.4197422744901789, + "grad_norm": 0.2307794988155365, + "learning_rate": 2.9263591036310728e-05, + "loss": 0.4471, + "step": 28370 + }, + { + "epoch": 1.4202427123733266, + "grad_norm": 45.545799255371094, + "learning_rate": 2.925432336750014e-05, + "loss": 0.8388, + "step": 28380 + }, + { + "epoch": 1.4207431502564745, + "grad_norm": 0.41668468713760376, + "learning_rate": 2.9245055698689556e-05, + "loss": 0.1406, + "step": 28390 + }, + { + "epoch": 1.4212435881396221, + "grad_norm": 75.24083709716797, + "learning_rate": 2.9235788029878965e-05, + "loss": 0.4864, + "step": 28400 + }, + { + "epoch": 1.42174402602277, + "grad_norm": 0.023358654230833054, + "learning_rate": 2.9226520361068377e-05, + "loss": 0.4718, + "step": 28410 + }, + { + "epoch": 1.4222444639059177, + "grad_norm": 21.35106086730957, + "learning_rate": 2.921725269225779e-05, + "loss": 0.1905, + "step": 28420 + }, + { + "epoch": 1.4227449017890654, + "grad_norm": 5.379050254821777, + "learning_rate": 2.9207985023447205e-05, + "loss": 0.383, + "step": 28430 + }, + { + "epoch": 1.4232453396722131, + "grad_norm": 0.017792226746678352, + "learning_rate": 2.9198717354636617e-05, + "loss": 0.333, + "step": 28440 + }, + { + "epoch": 1.4237457775553608, + "grad_norm": 20.03049659729004, + "learning_rate": 2.9189449685826033e-05, + "loss": 0.1988, + "step": 28450 + }, + { + "epoch": 1.4242462154385087, + "grad_norm": 0.1236361563205719, + "learning_rate": 2.9180182017015438e-05, + "loss": 0.444, + "step": 28460 + }, + { + "epoch": 1.4247466533216564, + "grad_norm": 0.6416155099868774, + "learning_rate": 2.9170914348204854e-05, + "loss": 0.3289, + "step": 28470 + }, + { + "epoch": 1.4252470912048043, + "grad_norm": 89.23595428466797, + "learning_rate": 2.9161646679394266e-05, + "loss": 0.3706, + "step": 28480 + }, + { + "epoch": 1.425747529087952, + "grad_norm": 32.9511604309082, + "learning_rate": 2.915237901058368e-05, + "loss": 0.4932, + "step": 28490 + }, + { + "epoch": 1.4262479669710997, + "grad_norm": 56.693016052246094, + "learning_rate": 2.9143111341773094e-05, + "loss": 0.1759, + "step": 28500 + }, + { + "epoch": 1.4267484048542474, + "grad_norm": 0.2121482789516449, + "learning_rate": 2.9133843672962503e-05, + "loss": 0.4775, + "step": 28510 + }, + { + "epoch": 1.4272488427373953, + "grad_norm": 40.10393142700195, + "learning_rate": 2.9124576004151915e-05, + "loss": 0.3627, + "step": 28520 + }, + { + "epoch": 1.427749280620543, + "grad_norm": 54.80824661254883, + "learning_rate": 2.911530833534133e-05, + "loss": 0.2021, + "step": 28530 + }, + { + "epoch": 1.4282497185036906, + "grad_norm": 0.5331714749336243, + "learning_rate": 2.9106040666530743e-05, + "loss": 0.4195, + "step": 28540 + }, + { + "epoch": 1.4287501563868386, + "grad_norm": 7.227887153625488, + "learning_rate": 2.9096772997720155e-05, + "loss": 0.5107, + "step": 28550 + }, + { + "epoch": 1.4292505942699862, + "grad_norm": 49.64358139038086, + "learning_rate": 2.908750532890957e-05, + "loss": 0.2849, + "step": 28560 + }, + { + "epoch": 1.429751032153134, + "grad_norm": 0.07907276600599289, + "learning_rate": 2.907823766009898e-05, + "loss": 0.446, + "step": 28570 + }, + { + "epoch": 1.4302514700362816, + "grad_norm": 0.5632091164588928, + "learning_rate": 2.9068969991288392e-05, + "loss": 0.2521, + "step": 28580 + }, + { + "epoch": 1.4307519079194295, + "grad_norm": 0.0693134069442749, + "learning_rate": 2.9059702322477804e-05, + "loss": 0.276, + "step": 28590 + }, + { + "epoch": 1.4312523458025772, + "grad_norm": 45.4066047668457, + "learning_rate": 2.905043465366722e-05, + "loss": 0.2747, + "step": 28600 + }, + { + "epoch": 1.4317527836857251, + "grad_norm": 59.52333068847656, + "learning_rate": 2.9041166984856632e-05, + "loss": 0.5077, + "step": 28610 + }, + { + "epoch": 1.4322532215688728, + "grad_norm": 0.04750872030854225, + "learning_rate": 2.903189931604604e-05, + "loss": 0.4102, + "step": 28620 + }, + { + "epoch": 1.4327536594520205, + "grad_norm": 5.837104797363281, + "learning_rate": 2.9022631647235453e-05, + "loss": 0.2247, + "step": 28630 + }, + { + "epoch": 1.4332540973351682, + "grad_norm": 0.0017373302252963185, + "learning_rate": 2.901336397842487e-05, + "loss": 0.411, + "step": 28640 + }, + { + "epoch": 1.433754535218316, + "grad_norm": 1.083304524421692, + "learning_rate": 2.900409630961428e-05, + "loss": 0.3787, + "step": 28650 + }, + { + "epoch": 1.4342549731014638, + "grad_norm": 36.4754524230957, + "learning_rate": 2.8994828640803697e-05, + "loss": 0.2115, + "step": 28660 + }, + { + "epoch": 1.4347554109846115, + "grad_norm": 0.011362102814018726, + "learning_rate": 2.898556097199311e-05, + "loss": 0.6083, + "step": 28670 + }, + { + "epoch": 1.4352558488677594, + "grad_norm": 3.0572962760925293, + "learning_rate": 2.8976293303182518e-05, + "loss": 0.5343, + "step": 28680 + }, + { + "epoch": 1.435756286750907, + "grad_norm": 0.658862292766571, + "learning_rate": 2.896702563437193e-05, + "loss": 0.0518, + "step": 28690 + }, + { + "epoch": 1.4362567246340547, + "grad_norm": 78.28485870361328, + "learning_rate": 2.8957757965561346e-05, + "loss": 0.3593, + "step": 28700 + }, + { + "epoch": 1.4367571625172024, + "grad_norm": 0.002119383541867137, + "learning_rate": 2.8948490296750758e-05, + "loss": 0.0301, + "step": 28710 + }, + { + "epoch": 1.4372576004003503, + "grad_norm": 0.016796980053186417, + "learning_rate": 2.893922262794017e-05, + "loss": 0.2217, + "step": 28720 + }, + { + "epoch": 1.437758038283498, + "grad_norm": 1.739690899848938, + "learning_rate": 2.892995495912958e-05, + "loss": 0.6265, + "step": 28730 + }, + { + "epoch": 1.438258476166646, + "grad_norm": 0.02895299158990383, + "learning_rate": 2.8920687290318995e-05, + "loss": 0.2062, + "step": 28740 + }, + { + "epoch": 1.4387589140497936, + "grad_norm": 101.59785461425781, + "learning_rate": 2.8911419621508407e-05, + "loss": 0.2917, + "step": 28750 + }, + { + "epoch": 1.4392593519329413, + "grad_norm": 122.82372283935547, + "learning_rate": 2.890215195269782e-05, + "loss": 0.3785, + "step": 28760 + }, + { + "epoch": 1.439759789816089, + "grad_norm": 0.00398382218554616, + "learning_rate": 2.8892884283887235e-05, + "loss": 0.0943, + "step": 28770 + }, + { + "epoch": 1.440260227699237, + "grad_norm": 0.11476438492536545, + "learning_rate": 2.8883616615076647e-05, + "loss": 0.5114, + "step": 28780 + }, + { + "epoch": 1.4407606655823846, + "grad_norm": 30.86602210998535, + "learning_rate": 2.8874348946266056e-05, + "loss": 0.5874, + "step": 28790 + }, + { + "epoch": 1.4412611034655323, + "grad_norm": 27.416336059570312, + "learning_rate": 2.8865081277455468e-05, + "loss": 0.6131, + "step": 28800 + }, + { + "epoch": 1.4417615413486802, + "grad_norm": 0.07657050341367722, + "learning_rate": 2.8855813608644884e-05, + "loss": 0.2541, + "step": 28810 + }, + { + "epoch": 1.4422619792318279, + "grad_norm": 13.498831748962402, + "learning_rate": 2.8846545939834296e-05, + "loss": 0.4741, + "step": 28820 + }, + { + "epoch": 1.4427624171149755, + "grad_norm": 39.632904052734375, + "learning_rate": 2.883727827102371e-05, + "loss": 0.479, + "step": 28830 + }, + { + "epoch": 1.4432628549981232, + "grad_norm": 51.27903747558594, + "learning_rate": 2.8828010602213117e-05, + "loss": 0.2149, + "step": 28840 + }, + { + "epoch": 1.4437632928812711, + "grad_norm": 79.14363861083984, + "learning_rate": 2.8818742933402533e-05, + "loss": 0.4872, + "step": 28850 + }, + { + "epoch": 1.4442637307644188, + "grad_norm": 30.421512603759766, + "learning_rate": 2.8809475264591945e-05, + "loss": 0.6363, + "step": 28860 + }, + { + "epoch": 1.4447641686475667, + "grad_norm": 6.203012466430664, + "learning_rate": 2.880020759578136e-05, + "loss": 0.2149, + "step": 28870 + }, + { + "epoch": 1.4452646065307144, + "grad_norm": 7.189742565155029, + "learning_rate": 2.8790939926970773e-05, + "loss": 0.1813, + "step": 28880 + }, + { + "epoch": 1.445765044413862, + "grad_norm": 0.3045077323913574, + "learning_rate": 2.8781672258160185e-05, + "loss": 0.1734, + "step": 28890 + }, + { + "epoch": 1.4462654822970098, + "grad_norm": 0.8910033702850342, + "learning_rate": 2.8772404589349594e-05, + "loss": 0.1214, + "step": 28900 + }, + { + "epoch": 1.4467659201801577, + "grad_norm": 0.009327556006610394, + "learning_rate": 2.876313692053901e-05, + "loss": 0.3556, + "step": 28910 + }, + { + "epoch": 1.4472663580633054, + "grad_norm": 10.993637084960938, + "learning_rate": 2.8753869251728422e-05, + "loss": 0.2313, + "step": 28920 + }, + { + "epoch": 1.447766795946453, + "grad_norm": 0.015217455103993416, + "learning_rate": 2.8744601582917834e-05, + "loss": 0.3842, + "step": 28930 + }, + { + "epoch": 1.448267233829601, + "grad_norm": 40.54232406616211, + "learning_rate": 2.873533391410725e-05, + "loss": 1.0173, + "step": 28940 + }, + { + "epoch": 1.4487676717127487, + "grad_norm": 0.11980646848678589, + "learning_rate": 2.8726066245296655e-05, + "loss": 0.2297, + "step": 28950 + }, + { + "epoch": 1.4492681095958964, + "grad_norm": 0.0049742781557142735, + "learning_rate": 2.871679857648607e-05, + "loss": 0.084, + "step": 28960 + }, + { + "epoch": 1.449768547479044, + "grad_norm": 13.173548698425293, + "learning_rate": 2.8707530907675483e-05, + "loss": 0.2779, + "step": 28970 + }, + { + "epoch": 1.450268985362192, + "grad_norm": 16.326942443847656, + "learning_rate": 2.86982632388649e-05, + "loss": 0.212, + "step": 28980 + }, + { + "epoch": 1.4507694232453396, + "grad_norm": 0.3165588974952698, + "learning_rate": 2.868899557005431e-05, + "loss": 0.1863, + "step": 28990 + }, + { + "epoch": 1.4512698611284875, + "grad_norm": 0.07787943631410599, + "learning_rate": 2.8679727901243726e-05, + "loss": 0.4143, + "step": 29000 + }, + { + "epoch": 1.4517702990116352, + "grad_norm": 0.26431193947792053, + "learning_rate": 2.8670460232433132e-05, + "loss": 0.3042, + "step": 29010 + }, + { + "epoch": 1.452270736894783, + "grad_norm": 1.8013296127319336, + "learning_rate": 2.8661192563622548e-05, + "loss": 0.2642, + "step": 29020 + }, + { + "epoch": 1.4527711747779306, + "grad_norm": 46.638545989990234, + "learning_rate": 2.865192489481196e-05, + "loss": 0.0474, + "step": 29030 + }, + { + "epoch": 1.4532716126610785, + "grad_norm": 0.061007387936115265, + "learning_rate": 2.8642657226001372e-05, + "loss": 0.3027, + "step": 29040 + }, + { + "epoch": 1.4537720505442262, + "grad_norm": 21.193450927734375, + "learning_rate": 2.8633389557190788e-05, + "loss": 0.2488, + "step": 29050 + }, + { + "epoch": 1.4542724884273739, + "grad_norm": 0.011775840073823929, + "learning_rate": 2.86241218883802e-05, + "loss": 0.717, + "step": 29060 + }, + { + "epoch": 1.4547729263105218, + "grad_norm": 0.2966684401035309, + "learning_rate": 2.861485421956961e-05, + "loss": 0.2008, + "step": 29070 + }, + { + "epoch": 1.4552733641936695, + "grad_norm": 0.07977305352687836, + "learning_rate": 2.860558655075902e-05, + "loss": 0.4487, + "step": 29080 + }, + { + "epoch": 1.4557738020768172, + "grad_norm": 56.42755889892578, + "learning_rate": 2.8596318881948437e-05, + "loss": 0.343, + "step": 29090 + }, + { + "epoch": 1.4562742399599649, + "grad_norm": 0.014368810690939426, + "learning_rate": 2.858705121313785e-05, + "loss": 0.6039, + "step": 29100 + }, + { + "epoch": 1.4567746778431128, + "grad_norm": 1.2614467144012451, + "learning_rate": 2.8577783544327265e-05, + "loss": 0.4722, + "step": 29110 + }, + { + "epoch": 1.4572751157262605, + "grad_norm": 1.4770146608352661, + "learning_rate": 2.856851587551667e-05, + "loss": 0.4448, + "step": 29120 + }, + { + "epoch": 1.4577755536094084, + "grad_norm": 38.635963439941406, + "learning_rate": 2.8559248206706086e-05, + "loss": 0.3155, + "step": 29130 + }, + { + "epoch": 1.458275991492556, + "grad_norm": 3.996797800064087, + "learning_rate": 2.8549980537895498e-05, + "loss": 0.0856, + "step": 29140 + }, + { + "epoch": 1.4587764293757037, + "grad_norm": 63.6402473449707, + "learning_rate": 2.8540712869084914e-05, + "loss": 0.339, + "step": 29150 + }, + { + "epoch": 1.4592768672588514, + "grad_norm": 36.6192626953125, + "learning_rate": 2.8531445200274326e-05, + "loss": 0.2247, + "step": 29160 + }, + { + "epoch": 1.4597773051419993, + "grad_norm": 0.009710263460874557, + "learning_rate": 2.8522177531463738e-05, + "loss": 0.1824, + "step": 29170 + }, + { + "epoch": 1.460277743025147, + "grad_norm": 0.02510778233408928, + "learning_rate": 2.8512909862653147e-05, + "loss": 0.0365, + "step": 29180 + }, + { + "epoch": 1.4607781809082947, + "grad_norm": 51.05836868286133, + "learning_rate": 2.8503642193842563e-05, + "loss": 0.8193, + "step": 29190 + }, + { + "epoch": 1.4612786187914426, + "grad_norm": 5.5351033210754395, + "learning_rate": 2.8494374525031975e-05, + "loss": 0.3944, + "step": 29200 + }, + { + "epoch": 1.4617790566745903, + "grad_norm": 68.62396240234375, + "learning_rate": 2.8485106856221387e-05, + "loss": 0.6652, + "step": 29210 + }, + { + "epoch": 1.462279494557738, + "grad_norm": 25.24249267578125, + "learning_rate": 2.8475839187410803e-05, + "loss": 0.4101, + "step": 29220 + }, + { + "epoch": 1.4627799324408857, + "grad_norm": 0.7221038937568665, + "learning_rate": 2.846657151860021e-05, + "loss": 0.153, + "step": 29230 + }, + { + "epoch": 1.4632803703240336, + "grad_norm": 1.5088069438934326, + "learning_rate": 2.8457303849789624e-05, + "loss": 0.8688, + "step": 29240 + }, + { + "epoch": 1.4637808082071813, + "grad_norm": 74.59314727783203, + "learning_rate": 2.8448036180979036e-05, + "loss": 0.453, + "step": 29250 + }, + { + "epoch": 1.4642812460903292, + "grad_norm": 0.013151934370398521, + "learning_rate": 2.843876851216845e-05, + "loss": 0.2127, + "step": 29260 + }, + { + "epoch": 1.4647816839734769, + "grad_norm": 71.96598052978516, + "learning_rate": 2.8429500843357864e-05, + "loss": 0.1739, + "step": 29270 + }, + { + "epoch": 1.4652821218566245, + "grad_norm": 21.41229820251465, + "learning_rate": 2.842023317454728e-05, + "loss": 0.337, + "step": 29280 + }, + { + "epoch": 1.4657825597397722, + "grad_norm": 0.023233555257320404, + "learning_rate": 2.8410965505736685e-05, + "loss": 0.3106, + "step": 29290 + }, + { + "epoch": 1.4662829976229201, + "grad_norm": 26.97758674621582, + "learning_rate": 2.84016978369261e-05, + "loss": 0.1064, + "step": 29300 + }, + { + "epoch": 1.4667834355060678, + "grad_norm": 0.018676113337278366, + "learning_rate": 2.8392430168115513e-05, + "loss": 0.2376, + "step": 29310 + }, + { + "epoch": 1.4672838733892155, + "grad_norm": 16.751415252685547, + "learning_rate": 2.838316249930493e-05, + "loss": 0.5611, + "step": 29320 + }, + { + "epoch": 1.4677843112723634, + "grad_norm": 0.08032844960689545, + "learning_rate": 2.837389483049434e-05, + "loss": 0.3492, + "step": 29330 + }, + { + "epoch": 1.468284749155511, + "grad_norm": 114.0760726928711, + "learning_rate": 2.836462716168375e-05, + "loss": 0.2234, + "step": 29340 + }, + { + "epoch": 1.4687851870386588, + "grad_norm": 22.593347549438477, + "learning_rate": 2.8355359492873162e-05, + "loss": 0.5848, + "step": 29350 + }, + { + "epoch": 1.4692856249218065, + "grad_norm": 49.74800491333008, + "learning_rate": 2.8346091824062577e-05, + "loss": 0.4054, + "step": 29360 + }, + { + "epoch": 1.4697860628049544, + "grad_norm": 15.379765510559082, + "learning_rate": 2.833682415525199e-05, + "loss": 0.1657, + "step": 29370 + }, + { + "epoch": 1.470286500688102, + "grad_norm": 22.556856155395508, + "learning_rate": 2.8327556486441402e-05, + "loss": 0.3056, + "step": 29380 + }, + { + "epoch": 1.47078693857125, + "grad_norm": 7.6845784187316895, + "learning_rate": 2.8318288817630818e-05, + "loss": 0.2806, + "step": 29390 + }, + { + "epoch": 1.4712873764543977, + "grad_norm": 0.2937449514865875, + "learning_rate": 2.8309021148820226e-05, + "loss": 0.186, + "step": 29400 + }, + { + "epoch": 1.4717878143375454, + "grad_norm": 93.677001953125, + "learning_rate": 2.829975348000964e-05, + "loss": 0.2193, + "step": 29410 + }, + { + "epoch": 1.472288252220693, + "grad_norm": 0.007834083400666714, + "learning_rate": 2.829048581119905e-05, + "loss": 0.183, + "step": 29420 + }, + { + "epoch": 1.4727886901038407, + "grad_norm": 0.0034589492715895176, + "learning_rate": 2.8281218142388467e-05, + "loss": 0.3883, + "step": 29430 + }, + { + "epoch": 1.4732891279869886, + "grad_norm": 20.205026626586914, + "learning_rate": 2.827195047357788e-05, + "loss": 0.3232, + "step": 29440 + }, + { + "epoch": 1.4737895658701363, + "grad_norm": 0.003861786099150777, + "learning_rate": 2.8262682804767288e-05, + "loss": 0.2748, + "step": 29450 + }, + { + "epoch": 1.4742900037532842, + "grad_norm": 75.82229614257812, + "learning_rate": 2.82534151359567e-05, + "loss": 0.264, + "step": 29460 + }, + { + "epoch": 1.474790441636432, + "grad_norm": 64.73430633544922, + "learning_rate": 2.8244147467146116e-05, + "loss": 0.6519, + "step": 29470 + }, + { + "epoch": 1.4752908795195796, + "grad_norm": 0.033975496888160706, + "learning_rate": 2.8234879798335528e-05, + "loss": 0.3852, + "step": 29480 + }, + { + "epoch": 1.4757913174027273, + "grad_norm": 4.9319586753845215, + "learning_rate": 2.8225612129524943e-05, + "loss": 0.1855, + "step": 29490 + }, + { + "epoch": 1.4762917552858752, + "grad_norm": 0.011342627927660942, + "learning_rate": 2.8216344460714356e-05, + "loss": 0.6899, + "step": 29500 + }, + { + "epoch": 1.4767921931690229, + "grad_norm": 32.27150344848633, + "learning_rate": 2.8207076791903765e-05, + "loss": 0.3393, + "step": 29510 + }, + { + "epoch": 1.4772926310521706, + "grad_norm": 1.2233649492263794, + "learning_rate": 2.8197809123093177e-05, + "loss": 0.1306, + "step": 29520 + }, + { + "epoch": 1.4777930689353185, + "grad_norm": 92.91287994384766, + "learning_rate": 2.8188541454282592e-05, + "loss": 0.4403, + "step": 29530 + }, + { + "epoch": 1.4782935068184662, + "grad_norm": 1.4810423851013184, + "learning_rate": 2.8179273785472005e-05, + "loss": 0.0839, + "step": 29540 + }, + { + "epoch": 1.4787939447016138, + "grad_norm": 84.85456085205078, + "learning_rate": 2.8170006116661417e-05, + "loss": 0.2552, + "step": 29550 + }, + { + "epoch": 1.4792943825847615, + "grad_norm": 0.20953987538814545, + "learning_rate": 2.8160738447850826e-05, + "loss": 0.669, + "step": 29560 + }, + { + "epoch": 1.4797948204679094, + "grad_norm": 0.03351613134145737, + "learning_rate": 2.815147077904024e-05, + "loss": 0.5186, + "step": 29570 + }, + { + "epoch": 1.4802952583510571, + "grad_norm": 0.07889140397310257, + "learning_rate": 2.8142203110229654e-05, + "loss": 0.5972, + "step": 29580 + }, + { + "epoch": 1.480795696234205, + "grad_norm": 1.940704584121704, + "learning_rate": 2.8132935441419066e-05, + "loss": 0.0909, + "step": 29590 + }, + { + "epoch": 1.4812961341173527, + "grad_norm": 45.45945358276367, + "learning_rate": 2.812366777260848e-05, + "loss": 0.1571, + "step": 29600 + }, + { + "epoch": 1.4817965720005004, + "grad_norm": 25.644990921020508, + "learning_rate": 2.8114400103797894e-05, + "loss": 0.6352, + "step": 29610 + }, + { + "epoch": 1.482297009883648, + "grad_norm": 56.04292678833008, + "learning_rate": 2.8105132434987303e-05, + "loss": 0.7362, + "step": 29620 + }, + { + "epoch": 1.482797447766796, + "grad_norm": 26.68865394592285, + "learning_rate": 2.8095864766176715e-05, + "loss": 0.1692, + "step": 29630 + }, + { + "epoch": 1.4832978856499437, + "grad_norm": 43.5210075378418, + "learning_rate": 2.808659709736613e-05, + "loss": 0.7736, + "step": 29640 + }, + { + "epoch": 1.4837983235330914, + "grad_norm": 40.784454345703125, + "learning_rate": 2.8077329428555543e-05, + "loss": 0.2505, + "step": 29650 + }, + { + "epoch": 1.4842987614162393, + "grad_norm": 20.080739974975586, + "learning_rate": 2.806806175974496e-05, + "loss": 0.2488, + "step": 29660 + }, + { + "epoch": 1.484799199299387, + "grad_norm": 10.29134464263916, + "learning_rate": 2.8058794090934364e-05, + "loss": 0.1894, + "step": 29670 + }, + { + "epoch": 1.4852996371825347, + "grad_norm": 43.66287612915039, + "learning_rate": 2.804952642212378e-05, + "loss": 0.4468, + "step": 29680 + }, + { + "epoch": 1.4858000750656823, + "grad_norm": 0.1069248840212822, + "learning_rate": 2.8040258753313192e-05, + "loss": 0.0878, + "step": 29690 + }, + { + "epoch": 1.4863005129488303, + "grad_norm": 2.652043581008911, + "learning_rate": 2.8030991084502607e-05, + "loss": 0.3791, + "step": 29700 + }, + { + "epoch": 1.486800950831978, + "grad_norm": 0.007332124747335911, + "learning_rate": 2.802172341569202e-05, + "loss": 0.1172, + "step": 29710 + }, + { + "epoch": 1.4873013887151258, + "grad_norm": 0.011741643771529198, + "learning_rate": 2.8012455746881432e-05, + "loss": 0.3687, + "step": 29720 + }, + { + "epoch": 1.4878018265982735, + "grad_norm": 0.014883521012961864, + "learning_rate": 2.800318807807084e-05, + "loss": 0.7228, + "step": 29730 + }, + { + "epoch": 1.4883022644814212, + "grad_norm": 0.01393388956785202, + "learning_rate": 2.7993920409260256e-05, + "loss": 0.4348, + "step": 29740 + }, + { + "epoch": 1.488802702364569, + "grad_norm": 155.3873748779297, + "learning_rate": 2.798465274044967e-05, + "loss": 0.5079, + "step": 29750 + }, + { + "epoch": 1.4893031402477168, + "grad_norm": 102.2210922241211, + "learning_rate": 2.797538507163908e-05, + "loss": 0.5162, + "step": 29760 + }, + { + "epoch": 1.4898035781308645, + "grad_norm": 0.22811825573444366, + "learning_rate": 2.7966117402828496e-05, + "loss": 0.186, + "step": 29770 + }, + { + "epoch": 1.4903040160140122, + "grad_norm": 0.5159445405006409, + "learning_rate": 2.795684973401791e-05, + "loss": 0.3953, + "step": 29780 + }, + { + "epoch": 1.49080445389716, + "grad_norm": 0.6948228478431702, + "learning_rate": 2.7947582065207318e-05, + "loss": 0.1452, + "step": 29790 + }, + { + "epoch": 1.4913048917803078, + "grad_norm": 12.987117767333984, + "learning_rate": 2.793831439639673e-05, + "loss": 0.2728, + "step": 29800 + }, + { + "epoch": 1.4918053296634555, + "grad_norm": 0.061698492616415024, + "learning_rate": 2.7929046727586145e-05, + "loss": 0.4048, + "step": 29810 + }, + { + "epoch": 1.4923057675466032, + "grad_norm": 20.55242347717285, + "learning_rate": 2.7919779058775558e-05, + "loss": 0.4535, + "step": 29820 + }, + { + "epoch": 1.492806205429751, + "grad_norm": 0.5183624029159546, + "learning_rate": 2.7910511389964973e-05, + "loss": 0.0748, + "step": 29830 + }, + { + "epoch": 1.4933066433128988, + "grad_norm": 81.03374481201172, + "learning_rate": 2.790124372115438e-05, + "loss": 0.2522, + "step": 29840 + }, + { + "epoch": 1.4938070811960467, + "grad_norm": 49.05284118652344, + "learning_rate": 2.7891976052343794e-05, + "loss": 0.2928, + "step": 29850 + }, + { + "epoch": 1.4943075190791943, + "grad_norm": 8.365612983703613, + "learning_rate": 2.7882708383533207e-05, + "loss": 0.1904, + "step": 29860 + }, + { + "epoch": 1.494807956962342, + "grad_norm": 11.37111759185791, + "learning_rate": 2.787344071472262e-05, + "loss": 0.2481, + "step": 29870 + }, + { + "epoch": 1.4953083948454897, + "grad_norm": 0.01665276661515236, + "learning_rate": 2.7864173045912035e-05, + "loss": 0.1624, + "step": 29880 + }, + { + "epoch": 1.4958088327286376, + "grad_norm": 57.88682556152344, + "learning_rate": 2.7854905377101447e-05, + "loss": 0.3465, + "step": 29890 + }, + { + "epoch": 1.4963092706117853, + "grad_norm": 0.5959999561309814, + "learning_rate": 2.7845637708290856e-05, + "loss": 0.4667, + "step": 29900 + }, + { + "epoch": 1.496809708494933, + "grad_norm": 0.05532146617770195, + "learning_rate": 2.7836370039480268e-05, + "loss": 0.4051, + "step": 29910 + }, + { + "epoch": 1.497310146378081, + "grad_norm": 0.24029013514518738, + "learning_rate": 2.7827102370669684e-05, + "loss": 0.3452, + "step": 29920 + }, + { + "epoch": 1.4978105842612286, + "grad_norm": 0.45980024337768555, + "learning_rate": 2.7817834701859096e-05, + "loss": 0.3698, + "step": 29930 + }, + { + "epoch": 1.4983110221443763, + "grad_norm": 33.38716125488281, + "learning_rate": 2.780856703304851e-05, + "loss": 0.0331, + "step": 29940 + }, + { + "epoch": 1.498811460027524, + "grad_norm": 84.88690948486328, + "learning_rate": 2.7799299364237917e-05, + "loss": 0.9783, + "step": 29950 + }, + { + "epoch": 1.4993118979106719, + "grad_norm": 96.79286193847656, + "learning_rate": 2.7790031695427333e-05, + "loss": 0.2331, + "step": 29960 + }, + { + "epoch": 1.4998123357938196, + "grad_norm": 0.6820287108421326, + "learning_rate": 2.7780764026616745e-05, + "loss": 0.4836, + "step": 29970 + }, + { + "epoch": 1.5003127736769675, + "grad_norm": 37.46875762939453, + "learning_rate": 2.777149635780616e-05, + "loss": 0.3763, + "step": 29980 + }, + { + "epoch": 1.5008132115601152, + "grad_norm": 0.13901999592781067, + "learning_rate": 2.7762228688995573e-05, + "loss": 0.3377, + "step": 29990 + }, + { + "epoch": 1.5013136494432628, + "grad_norm": 0.4552798569202423, + "learning_rate": 2.7752961020184985e-05, + "loss": 0.3432, + "step": 30000 + }, + { + "epoch": 1.5018140873264105, + "grad_norm": 4.675800800323486, + "learning_rate": 2.7743693351374394e-05, + "loss": 0.4295, + "step": 30010 + }, + { + "epoch": 1.5023145252095582, + "grad_norm": 12.397222518920898, + "learning_rate": 2.773442568256381e-05, + "loss": 0.2104, + "step": 30020 + }, + { + "epoch": 1.5028149630927061, + "grad_norm": 0.03496350720524788, + "learning_rate": 2.772515801375322e-05, + "loss": 0.3076, + "step": 30030 + }, + { + "epoch": 1.503315400975854, + "grad_norm": 88.4476547241211, + "learning_rate": 2.7715890344942634e-05, + "loss": 0.2627, + "step": 30040 + }, + { + "epoch": 1.5038158388590017, + "grad_norm": 69.03022766113281, + "learning_rate": 2.770662267613205e-05, + "loss": 0.4276, + "step": 30050 + }, + { + "epoch": 1.5043162767421494, + "grad_norm": 200.00460815429688, + "learning_rate": 2.769735500732146e-05, + "loss": 0.4419, + "step": 30060 + }, + { + "epoch": 1.504816714625297, + "grad_norm": 0.058439310640096664, + "learning_rate": 2.768808733851087e-05, + "loss": 0.1831, + "step": 30070 + }, + { + "epoch": 1.5053171525084448, + "grad_norm": 23.720563888549805, + "learning_rate": 2.7678819669700283e-05, + "loss": 0.2454, + "step": 30080 + }, + { + "epoch": 1.5058175903915927, + "grad_norm": 113.98284149169922, + "learning_rate": 2.76695520008897e-05, + "loss": 0.4357, + "step": 30090 + }, + { + "epoch": 1.5063180282747404, + "grad_norm": 0.007505831308662891, + "learning_rate": 2.766028433207911e-05, + "loss": 0.039, + "step": 30100 + }, + { + "epoch": 1.5068184661578883, + "grad_norm": 0.7762181758880615, + "learning_rate": 2.7651016663268526e-05, + "loss": 0.5519, + "step": 30110 + }, + { + "epoch": 1.507318904041036, + "grad_norm": 0.010983048006892204, + "learning_rate": 2.7641748994457932e-05, + "loss": 0.4935, + "step": 30120 + }, + { + "epoch": 1.5078193419241837, + "grad_norm": 0.08401951938867569, + "learning_rate": 2.7632481325647348e-05, + "loss": 0.2419, + "step": 30130 + }, + { + "epoch": 1.5083197798073313, + "grad_norm": 0.0106881745159626, + "learning_rate": 2.762321365683676e-05, + "loss": 0.1874, + "step": 30140 + }, + { + "epoch": 1.508820217690479, + "grad_norm": 84.64973449707031, + "learning_rate": 2.7613945988026175e-05, + "loss": 0.6644, + "step": 30150 + }, + { + "epoch": 1.509320655573627, + "grad_norm": 0.21059495210647583, + "learning_rate": 2.7604678319215588e-05, + "loss": 0.2545, + "step": 30160 + }, + { + "epoch": 1.5098210934567748, + "grad_norm": 143.67526245117188, + "learning_rate": 2.7595410650404997e-05, + "loss": 0.7027, + "step": 30170 + }, + { + "epoch": 1.5103215313399225, + "grad_norm": 2.9124467372894287, + "learning_rate": 2.758614298159441e-05, + "loss": 0.3359, + "step": 30180 + }, + { + "epoch": 1.5108219692230702, + "grad_norm": 14.777406692504883, + "learning_rate": 2.7576875312783824e-05, + "loss": 0.4276, + "step": 30190 + }, + { + "epoch": 1.511322407106218, + "grad_norm": 14.837841987609863, + "learning_rate": 2.7567607643973237e-05, + "loss": 0.4374, + "step": 30200 + }, + { + "epoch": 1.5118228449893656, + "grad_norm": 17.19945526123047, + "learning_rate": 2.755833997516265e-05, + "loss": 0.6189, + "step": 30210 + }, + { + "epoch": 1.5123232828725135, + "grad_norm": 0.022332990542054176, + "learning_rate": 2.7549072306352065e-05, + "loss": 0.338, + "step": 30220 + }, + { + "epoch": 1.5128237207556612, + "grad_norm": 76.36695861816406, + "learning_rate": 2.7539804637541473e-05, + "loss": 0.408, + "step": 30230 + }, + { + "epoch": 1.513324158638809, + "grad_norm": 31.33741569519043, + "learning_rate": 2.7530536968730886e-05, + "loss": 0.2908, + "step": 30240 + }, + { + "epoch": 1.5138245965219568, + "grad_norm": 0.24871501326560974, + "learning_rate": 2.7521269299920298e-05, + "loss": 0.358, + "step": 30250 + }, + { + "epoch": 1.5143250344051045, + "grad_norm": 0.07432777434587479, + "learning_rate": 2.7512001631109713e-05, + "loss": 0.408, + "step": 30260 + }, + { + "epoch": 1.5148254722882522, + "grad_norm": 3.159221887588501, + "learning_rate": 2.7502733962299126e-05, + "loss": 0.3511, + "step": 30270 + }, + { + "epoch": 1.5153259101713998, + "grad_norm": 5.2638678550720215, + "learning_rate": 2.7493466293488535e-05, + "loss": 0.5537, + "step": 30280 + }, + { + "epoch": 1.5158263480545477, + "grad_norm": 18.343320846557617, + "learning_rate": 2.7484198624677947e-05, + "loss": 0.3564, + "step": 30290 + }, + { + "epoch": 1.5163267859376954, + "grad_norm": 0.023645449429750443, + "learning_rate": 2.7474930955867362e-05, + "loss": 0.0875, + "step": 30300 + }, + { + "epoch": 1.5168272238208433, + "grad_norm": 0.06220357120037079, + "learning_rate": 2.7465663287056775e-05, + "loss": 0.5826, + "step": 30310 + }, + { + "epoch": 1.517327661703991, + "grad_norm": 0.06881603598594666, + "learning_rate": 2.745639561824619e-05, + "loss": 0.3957, + "step": 30320 + }, + { + "epoch": 1.5178280995871387, + "grad_norm": 0.532408595085144, + "learning_rate": 2.7447127949435603e-05, + "loss": 0.2547, + "step": 30330 + }, + { + "epoch": 1.5183285374702864, + "grad_norm": 13.922101974487305, + "learning_rate": 2.743786028062501e-05, + "loss": 0.2605, + "step": 30340 + }, + { + "epoch": 1.5188289753534343, + "grad_norm": 143.68438720703125, + "learning_rate": 2.7428592611814424e-05, + "loss": 0.2245, + "step": 30350 + }, + { + "epoch": 1.519329413236582, + "grad_norm": 0.027887465432286263, + "learning_rate": 2.741932494300384e-05, + "loss": 0.1836, + "step": 30360 + }, + { + "epoch": 1.51982985111973, + "grad_norm": 0.30637139081954956, + "learning_rate": 2.741005727419325e-05, + "loss": 0.3283, + "step": 30370 + }, + { + "epoch": 1.5203302890028776, + "grad_norm": 53.38349914550781, + "learning_rate": 2.7400789605382664e-05, + "loss": 0.3216, + "step": 30380 + }, + { + "epoch": 1.5208307268860253, + "grad_norm": 0.18016959726810455, + "learning_rate": 2.7391521936572073e-05, + "loss": 0.1265, + "step": 30390 + }, + { + "epoch": 1.521331164769173, + "grad_norm": 3.7262725830078125, + "learning_rate": 2.738225426776149e-05, + "loss": 0.2152, + "step": 30400 + }, + { + "epoch": 1.5218316026523206, + "grad_norm": 0.03781423717737198, + "learning_rate": 2.73729865989509e-05, + "loss": 0.1737, + "step": 30410 + }, + { + "epoch": 1.5223320405354686, + "grad_norm": 48.56614303588867, + "learning_rate": 2.7363718930140313e-05, + "loss": 0.2758, + "step": 30420 + }, + { + "epoch": 1.5228324784186162, + "grad_norm": 26.435205459594727, + "learning_rate": 2.735445126132973e-05, + "loss": 0.0177, + "step": 30430 + }, + { + "epoch": 1.5233329163017642, + "grad_norm": 259.7550354003906, + "learning_rate": 2.734518359251914e-05, + "loss": 0.3201, + "step": 30440 + }, + { + "epoch": 1.5238333541849118, + "grad_norm": 0.3746943473815918, + "learning_rate": 2.733591592370855e-05, + "loss": 0.3177, + "step": 30450 + }, + { + "epoch": 1.5243337920680595, + "grad_norm": 0.006465022452175617, + "learning_rate": 2.7326648254897962e-05, + "loss": 0.4367, + "step": 30460 + }, + { + "epoch": 1.5248342299512072, + "grad_norm": 52.5500602722168, + "learning_rate": 2.7317380586087377e-05, + "loss": 0.3191, + "step": 30470 + }, + { + "epoch": 1.525334667834355, + "grad_norm": 0.09003955125808716, + "learning_rate": 2.730811291727679e-05, + "loss": 0.1297, + "step": 30480 + }, + { + "epoch": 1.5258351057175028, + "grad_norm": 0.003084195079281926, + "learning_rate": 2.7298845248466205e-05, + "loss": 0.5031, + "step": 30490 + }, + { + "epoch": 1.5263355436006507, + "grad_norm": 41.51118850708008, + "learning_rate": 2.7289577579655618e-05, + "loss": 0.5836, + "step": 30500 + }, + { + "epoch": 1.5268359814837984, + "grad_norm": 0.02783985063433647, + "learning_rate": 2.7280309910845026e-05, + "loss": 0.7046, + "step": 30510 + }, + { + "epoch": 1.527336419366946, + "grad_norm": 0.1910041719675064, + "learning_rate": 2.727104224203444e-05, + "loss": 0.0835, + "step": 30520 + }, + { + "epoch": 1.5278368572500938, + "grad_norm": 0.008609929122030735, + "learning_rate": 2.7261774573223854e-05, + "loss": 0.3523, + "step": 30530 + }, + { + "epoch": 1.5283372951332415, + "grad_norm": 67.78886413574219, + "learning_rate": 2.7252506904413267e-05, + "loss": 0.2112, + "step": 30540 + }, + { + "epoch": 1.5288377330163894, + "grad_norm": 41.18297576904297, + "learning_rate": 2.724323923560268e-05, + "loss": 0.475, + "step": 30550 + }, + { + "epoch": 1.529338170899537, + "grad_norm": 31.47063446044922, + "learning_rate": 2.7233971566792088e-05, + "loss": 0.5879, + "step": 30560 + }, + { + "epoch": 1.529838608782685, + "grad_norm": 6.888252258300781, + "learning_rate": 2.7224703897981503e-05, + "loss": 0.2357, + "step": 30570 + }, + { + "epoch": 1.5303390466658326, + "grad_norm": 20.344369888305664, + "learning_rate": 2.7215436229170916e-05, + "loss": 0.3169, + "step": 30580 + }, + { + "epoch": 1.5308394845489803, + "grad_norm": 0.17995403707027435, + "learning_rate": 2.7206168560360328e-05, + "loss": 0.0684, + "step": 30590 + }, + { + "epoch": 1.531339922432128, + "grad_norm": 33.51765060424805, + "learning_rate": 2.7196900891549743e-05, + "loss": 0.6871, + "step": 30600 + }, + { + "epoch": 1.5318403603152757, + "grad_norm": 29.637102127075195, + "learning_rate": 2.7187633222739156e-05, + "loss": 0.7784, + "step": 30610 + }, + { + "epoch": 1.5323407981984236, + "grad_norm": 20.285873413085938, + "learning_rate": 2.7178365553928565e-05, + "loss": 0.3616, + "step": 30620 + }, + { + "epoch": 1.5328412360815715, + "grad_norm": 10.266929626464844, + "learning_rate": 2.7169097885117977e-05, + "loss": 0.3402, + "step": 30630 + }, + { + "epoch": 1.5333416739647192, + "grad_norm": 0.04004902392625809, + "learning_rate": 2.7159830216307392e-05, + "loss": 0.428, + "step": 30640 + }, + { + "epoch": 1.533842111847867, + "grad_norm": 0.12043403834104538, + "learning_rate": 2.7150562547496805e-05, + "loss": 0.3428, + "step": 30650 + }, + { + "epoch": 1.5343425497310146, + "grad_norm": 69.50574493408203, + "learning_rate": 2.714129487868622e-05, + "loss": 0.4572, + "step": 30660 + }, + { + "epoch": 1.5348429876141623, + "grad_norm": 80.15311431884766, + "learning_rate": 2.7132027209875626e-05, + "loss": 0.2268, + "step": 30670 + }, + { + "epoch": 1.5353434254973102, + "grad_norm": 0.04792514815926552, + "learning_rate": 2.712275954106504e-05, + "loss": 0.1809, + "step": 30680 + }, + { + "epoch": 1.5358438633804579, + "grad_norm": 0.017328662797808647, + "learning_rate": 2.7113491872254454e-05, + "loss": 0.286, + "step": 30690 + }, + { + "epoch": 1.5363443012636058, + "grad_norm": 0.04663608595728874, + "learning_rate": 2.710422420344387e-05, + "loss": 0.51, + "step": 30700 + }, + { + "epoch": 1.5368447391467535, + "grad_norm": 102.59835052490234, + "learning_rate": 2.709495653463328e-05, + "loss": 0.4029, + "step": 30710 + }, + { + "epoch": 1.5373451770299011, + "grad_norm": 67.7674560546875, + "learning_rate": 2.7085688865822694e-05, + "loss": 0.1905, + "step": 30720 + }, + { + "epoch": 1.5378456149130488, + "grad_norm": 168.31411743164062, + "learning_rate": 2.7076421197012103e-05, + "loss": 0.4422, + "step": 30730 + }, + { + "epoch": 1.5383460527961965, + "grad_norm": 0.18501713871955872, + "learning_rate": 2.7067153528201515e-05, + "loss": 0.2876, + "step": 30740 + }, + { + "epoch": 1.5388464906793444, + "grad_norm": 939.5577392578125, + "learning_rate": 2.705788585939093e-05, + "loss": 0.2703, + "step": 30750 + }, + { + "epoch": 1.5393469285624923, + "grad_norm": 9.09665298461914, + "learning_rate": 2.7048618190580343e-05, + "loss": 0.9039, + "step": 30760 + }, + { + "epoch": 1.53984736644564, + "grad_norm": 0.013461408205330372, + "learning_rate": 2.703935052176976e-05, + "loss": 0.2892, + "step": 30770 + }, + { + "epoch": 1.5403478043287877, + "grad_norm": 0.8771297335624695, + "learning_rate": 2.7030082852959164e-05, + "loss": 0.1412, + "step": 30780 + }, + { + "epoch": 1.5408482422119354, + "grad_norm": 0.05178891122341156, + "learning_rate": 2.702081518414858e-05, + "loss": 0.2318, + "step": 30790 + }, + { + "epoch": 1.541348680095083, + "grad_norm": 0.18068766593933105, + "learning_rate": 2.7011547515337992e-05, + "loss": 0.464, + "step": 30800 + }, + { + "epoch": 1.541849117978231, + "grad_norm": 0.0612143874168396, + "learning_rate": 2.7002279846527407e-05, + "loss": 0.1683, + "step": 30810 + }, + { + "epoch": 1.5423495558613787, + "grad_norm": 0.4520694315433502, + "learning_rate": 2.699301217771682e-05, + "loss": 0.6133, + "step": 30820 + }, + { + "epoch": 1.5428499937445266, + "grad_norm": 52.35979461669922, + "learning_rate": 2.6983744508906232e-05, + "loss": 0.3997, + "step": 30830 + }, + { + "epoch": 1.5433504316276743, + "grad_norm": 0.012475469149649143, + "learning_rate": 2.697447684009564e-05, + "loss": 0.1904, + "step": 30840 + }, + { + "epoch": 1.543850869510822, + "grad_norm": 0.04051502048969269, + "learning_rate": 2.6965209171285056e-05, + "loss": 0.4516, + "step": 30850 + }, + { + "epoch": 1.5443513073939696, + "grad_norm": 75.82689666748047, + "learning_rate": 2.695594150247447e-05, + "loss": 0.5447, + "step": 30860 + }, + { + "epoch": 1.5448517452771173, + "grad_norm": 0.014825690537691116, + "learning_rate": 2.694667383366388e-05, + "loss": 0.0352, + "step": 30870 + }, + { + "epoch": 1.5453521831602652, + "grad_norm": 0.12960568070411682, + "learning_rate": 2.6937406164853296e-05, + "loss": 0.4071, + "step": 30880 + }, + { + "epoch": 1.5458526210434131, + "grad_norm": 0.005345550365746021, + "learning_rate": 2.6928138496042705e-05, + "loss": 0.1949, + "step": 30890 + }, + { + "epoch": 1.5463530589265608, + "grad_norm": 0.29516321420669556, + "learning_rate": 2.6918870827232118e-05, + "loss": 0.2025, + "step": 30900 + }, + { + "epoch": 1.5468534968097085, + "grad_norm": 0.4016546308994293, + "learning_rate": 2.690960315842153e-05, + "loss": 0.0013, + "step": 30910 + }, + { + "epoch": 1.5473539346928562, + "grad_norm": 0.2692883014678955, + "learning_rate": 2.6900335489610945e-05, + "loss": 0.0045, + "step": 30920 + }, + { + "epoch": 1.547854372576004, + "grad_norm": 1.7655049562454224, + "learning_rate": 2.6891067820800358e-05, + "loss": 0.0521, + "step": 30930 + }, + { + "epoch": 1.5483548104591518, + "grad_norm": 3.10745906829834, + "learning_rate": 2.6881800151989773e-05, + "loss": 0.0685, + "step": 30940 + }, + { + "epoch": 1.5488552483422995, + "grad_norm": 4.634167194366455, + "learning_rate": 2.687253248317918e-05, + "loss": 0.2353, + "step": 30950 + }, + { + "epoch": 1.5493556862254474, + "grad_norm": 0.04169250652194023, + "learning_rate": 2.6863264814368594e-05, + "loss": 0.5788, + "step": 30960 + }, + { + "epoch": 1.549856124108595, + "grad_norm": 2.864992141723633, + "learning_rate": 2.6853997145558007e-05, + "loss": 0.0191, + "step": 30970 + }, + { + "epoch": 1.5503565619917428, + "grad_norm": 48.499114990234375, + "learning_rate": 2.6844729476747422e-05, + "loss": 0.4021, + "step": 30980 + }, + { + "epoch": 1.5508569998748905, + "grad_norm": 19.008472442626953, + "learning_rate": 2.6835461807936835e-05, + "loss": 0.2377, + "step": 30990 + }, + { + "epoch": 1.5513574377580381, + "grad_norm": 0.4861891269683838, + "learning_rate": 2.6826194139126243e-05, + "loss": 0.7756, + "step": 31000 + }, + { + "epoch": 1.551857875641186, + "grad_norm": 0.04746564105153084, + "learning_rate": 2.6816926470315656e-05, + "loss": 0.0428, + "step": 31010 + }, + { + "epoch": 1.552358313524334, + "grad_norm": 0.021157721057534218, + "learning_rate": 2.680765880150507e-05, + "loss": 0.3169, + "step": 31020 + }, + { + "epoch": 1.5528587514074816, + "grad_norm": 68.8771743774414, + "learning_rate": 2.6798391132694484e-05, + "loss": 0.4557, + "step": 31030 + }, + { + "epoch": 1.5533591892906293, + "grad_norm": 34.50081253051758, + "learning_rate": 2.6789123463883896e-05, + "loss": 0.7089, + "step": 31040 + }, + { + "epoch": 1.553859627173777, + "grad_norm": 0.011299318633973598, + "learning_rate": 2.677985579507331e-05, + "loss": 0.5996, + "step": 31050 + }, + { + "epoch": 1.5543600650569247, + "grad_norm": 0.019167188555002213, + "learning_rate": 2.677058812626272e-05, + "loss": 0.3511, + "step": 31060 + }, + { + "epoch": 1.5548605029400726, + "grad_norm": 2.4802985191345215, + "learning_rate": 2.6761320457452133e-05, + "loss": 0.2136, + "step": 31070 + }, + { + "epoch": 1.5553609408232203, + "grad_norm": 100.97578430175781, + "learning_rate": 2.6752052788641545e-05, + "loss": 0.4791, + "step": 31080 + }, + { + "epoch": 1.5558613787063682, + "grad_norm": 71.9083023071289, + "learning_rate": 2.674278511983096e-05, + "loss": 0.366, + "step": 31090 + }, + { + "epoch": 1.556361816589516, + "grad_norm": 9.862528800964355, + "learning_rate": 2.6733517451020373e-05, + "loss": 0.6876, + "step": 31100 + }, + { + "epoch": 1.5568622544726636, + "grad_norm": 1.8160629272460938, + "learning_rate": 2.672424978220978e-05, + "loss": 0.5327, + "step": 31110 + }, + { + "epoch": 1.5573626923558113, + "grad_norm": 0.5882023572921753, + "learning_rate": 2.6714982113399194e-05, + "loss": 0.1083, + "step": 31120 + }, + { + "epoch": 1.557863130238959, + "grad_norm": 0.4773244261741638, + "learning_rate": 2.670571444458861e-05, + "loss": 0.5174, + "step": 31130 + }, + { + "epoch": 1.5583635681221069, + "grad_norm": 77.14653778076172, + "learning_rate": 2.669644677577802e-05, + "loss": 0.5789, + "step": 31140 + }, + { + "epoch": 1.5588640060052545, + "grad_norm": 0.10065506398677826, + "learning_rate": 2.6687179106967437e-05, + "loss": 0.3799, + "step": 31150 + }, + { + "epoch": 1.5593644438884025, + "grad_norm": 0.05134065821766853, + "learning_rate": 2.667791143815685e-05, + "loss": 0.3157, + "step": 31160 + }, + { + "epoch": 1.5598648817715501, + "grad_norm": 0.8216724395751953, + "learning_rate": 2.666864376934626e-05, + "loss": 0.3717, + "step": 31170 + }, + { + "epoch": 1.5603653196546978, + "grad_norm": 42.55986404418945, + "learning_rate": 2.665937610053567e-05, + "loss": 0.3329, + "step": 31180 + }, + { + "epoch": 1.5608657575378455, + "grad_norm": 0.5152754187583923, + "learning_rate": 2.6650108431725086e-05, + "loss": 0.3268, + "step": 31190 + }, + { + "epoch": 1.5613661954209934, + "grad_norm": 10.219765663146973, + "learning_rate": 2.66408407629145e-05, + "loss": 0.1177, + "step": 31200 + }, + { + "epoch": 1.561866633304141, + "grad_norm": 32.21329116821289, + "learning_rate": 2.663157309410391e-05, + "loss": 0.3122, + "step": 31210 + }, + { + "epoch": 1.562367071187289, + "grad_norm": 26.903533935546875, + "learning_rate": 2.6622305425293326e-05, + "loss": 0.2972, + "step": 31220 + }, + { + "epoch": 1.5628675090704367, + "grad_norm": 1.1322250366210938, + "learning_rate": 2.6613037756482735e-05, + "loss": 0.4166, + "step": 31230 + }, + { + "epoch": 1.5633679469535844, + "grad_norm": 0.00415285537019372, + "learning_rate": 2.6603770087672147e-05, + "loss": 0.3451, + "step": 31240 + }, + { + "epoch": 1.563868384836732, + "grad_norm": 0.8872724771499634, + "learning_rate": 2.659450241886156e-05, + "loss": 0.4796, + "step": 31250 + }, + { + "epoch": 1.5643688227198798, + "grad_norm": 51.05792236328125, + "learning_rate": 2.6585234750050975e-05, + "loss": 0.458, + "step": 31260 + }, + { + "epoch": 1.5648692606030277, + "grad_norm": 15.749295234680176, + "learning_rate": 2.6575967081240388e-05, + "loss": 0.0718, + "step": 31270 + }, + { + "epoch": 1.5653696984861754, + "grad_norm": 14.178160667419434, + "learning_rate": 2.6566699412429796e-05, + "loss": 0.7005, + "step": 31280 + }, + { + "epoch": 1.5658701363693233, + "grad_norm": 0.18899379670619965, + "learning_rate": 2.655743174361921e-05, + "loss": 0.2681, + "step": 31290 + }, + { + "epoch": 1.566370574252471, + "grad_norm": 52.0080451965332, + "learning_rate": 2.6548164074808624e-05, + "loss": 0.2981, + "step": 31300 + }, + { + "epoch": 1.5668710121356186, + "grad_norm": 63.819183349609375, + "learning_rate": 2.6538896405998037e-05, + "loss": 0.3102, + "step": 31310 + }, + { + "epoch": 1.5673714500187663, + "grad_norm": 0.00652950257062912, + "learning_rate": 2.6529628737187452e-05, + "loss": 0.1672, + "step": 31320 + }, + { + "epoch": 1.5678718879019142, + "grad_norm": 0.02523639239370823, + "learning_rate": 2.6520361068376864e-05, + "loss": 0.4945, + "step": 31330 + }, + { + "epoch": 1.568372325785062, + "grad_norm": 69.5326919555664, + "learning_rate": 2.6511093399566273e-05, + "loss": 0.197, + "step": 31340 + }, + { + "epoch": 1.5688727636682098, + "grad_norm": 49.80335235595703, + "learning_rate": 2.6501825730755686e-05, + "loss": 0.2225, + "step": 31350 + }, + { + "epoch": 1.5693732015513575, + "grad_norm": 121.33419799804688, + "learning_rate": 2.64925580619451e-05, + "loss": 0.3712, + "step": 31360 + }, + { + "epoch": 1.5698736394345052, + "grad_norm": 0.1943092942237854, + "learning_rate": 2.6483290393134513e-05, + "loss": 0.2425, + "step": 31370 + }, + { + "epoch": 1.5703740773176529, + "grad_norm": 3.1248579025268555, + "learning_rate": 2.6474022724323926e-05, + "loss": 0.2072, + "step": 31380 + }, + { + "epoch": 1.5708745152008006, + "grad_norm": 0.010482363402843475, + "learning_rate": 2.6464755055513335e-05, + "loss": 0.0872, + "step": 31390 + }, + { + "epoch": 1.5713749530839485, + "grad_norm": 54.139686584472656, + "learning_rate": 2.645548738670275e-05, + "loss": 0.2061, + "step": 31400 + }, + { + "epoch": 1.5718753909670962, + "grad_norm": 0.021342795342206955, + "learning_rate": 2.6446219717892162e-05, + "loss": 0.1526, + "step": 31410 + }, + { + "epoch": 1.572375828850244, + "grad_norm": 0.13634184002876282, + "learning_rate": 2.6436952049081575e-05, + "loss": 0.1297, + "step": 31420 + }, + { + "epoch": 1.5728762667333918, + "grad_norm": 5.726498603820801, + "learning_rate": 2.642768438027099e-05, + "loss": 0.1427, + "step": 31430 + }, + { + "epoch": 1.5733767046165394, + "grad_norm": 25.255191802978516, + "learning_rate": 2.6418416711460403e-05, + "loss": 0.0321, + "step": 31440 + }, + { + "epoch": 1.5738771424996871, + "grad_norm": 0.027879461646080017, + "learning_rate": 2.640914904264981e-05, + "loss": 0.6105, + "step": 31450 + }, + { + "epoch": 1.5743775803828348, + "grad_norm": 1.7521904706954956, + "learning_rate": 2.6399881373839224e-05, + "loss": 0.3015, + "step": 31460 + }, + { + "epoch": 1.5748780182659827, + "grad_norm": 217.18067932128906, + "learning_rate": 2.639061370502864e-05, + "loss": 0.7197, + "step": 31470 + }, + { + "epoch": 1.5753784561491306, + "grad_norm": 0.06598635017871857, + "learning_rate": 2.638134603621805e-05, + "loss": 0.845, + "step": 31480 + }, + { + "epoch": 1.5758788940322783, + "grad_norm": 29.13563346862793, + "learning_rate": 2.6372078367407467e-05, + "loss": 0.3394, + "step": 31490 + }, + { + "epoch": 1.576379331915426, + "grad_norm": 22.38322639465332, + "learning_rate": 2.6362810698596873e-05, + "loss": 0.4162, + "step": 31500 + }, + { + "epoch": 1.5768797697985737, + "grad_norm": 22.216062545776367, + "learning_rate": 2.6353543029786288e-05, + "loss": 0.1856, + "step": 31510 + }, + { + "epoch": 1.5773802076817214, + "grad_norm": 0.10017246007919312, + "learning_rate": 2.63442753609757e-05, + "loss": 0.5001, + "step": 31520 + }, + { + "epoch": 1.5778806455648693, + "grad_norm": 1.9049071073532104, + "learning_rate": 2.6335007692165116e-05, + "loss": 0.5022, + "step": 31530 + }, + { + "epoch": 1.578381083448017, + "grad_norm": 28.192276000976562, + "learning_rate": 2.632574002335453e-05, + "loss": 0.8791, + "step": 31540 + }, + { + "epoch": 1.5788815213311649, + "grad_norm": 69.63363647460938, + "learning_rate": 2.631647235454394e-05, + "loss": 0.33, + "step": 31550 + }, + { + "epoch": 1.5793819592143126, + "grad_norm": 4.001302719116211, + "learning_rate": 2.630720468573335e-05, + "loss": 0.4286, + "step": 31560 + }, + { + "epoch": 1.5798823970974603, + "grad_norm": 67.28568267822266, + "learning_rate": 2.6297937016922762e-05, + "loss": 0.6472, + "step": 31570 + }, + { + "epoch": 1.580382834980608, + "grad_norm": 13.990724563598633, + "learning_rate": 2.6288669348112177e-05, + "loss": 0.2628, + "step": 31580 + }, + { + "epoch": 1.5808832728637556, + "grad_norm": 58.89625549316406, + "learning_rate": 2.627940167930159e-05, + "loss": 0.4011, + "step": 31590 + }, + { + "epoch": 1.5813837107469035, + "grad_norm": 0.13952134549617767, + "learning_rate": 2.6270134010491005e-05, + "loss": 0.2504, + "step": 31600 + }, + { + "epoch": 1.5818841486300514, + "grad_norm": 162.8355255126953, + "learning_rate": 2.626086634168041e-05, + "loss": 0.3186, + "step": 31610 + }, + { + "epoch": 1.5823845865131991, + "grad_norm": 0.8061618208885193, + "learning_rate": 2.6251598672869826e-05, + "loss": 0.5014, + "step": 31620 + }, + { + "epoch": 1.5828850243963468, + "grad_norm": 24.808698654174805, + "learning_rate": 2.624233100405924e-05, + "loss": 0.3193, + "step": 31630 + }, + { + "epoch": 1.5833854622794945, + "grad_norm": 36.71739196777344, + "learning_rate": 2.6233063335248654e-05, + "loss": 0.354, + "step": 31640 + }, + { + "epoch": 1.5838859001626422, + "grad_norm": 13.136659622192383, + "learning_rate": 2.6223795666438067e-05, + "loss": 0.3452, + "step": 31650 + }, + { + "epoch": 1.58438633804579, + "grad_norm": 1.089193344116211, + "learning_rate": 2.6214527997627482e-05, + "loss": 0.1967, + "step": 31660 + }, + { + "epoch": 1.5848867759289378, + "grad_norm": 9.409536361694336, + "learning_rate": 2.6205260328816888e-05, + "loss": 0.2277, + "step": 31670 + }, + { + "epoch": 1.5853872138120857, + "grad_norm": 155.2223663330078, + "learning_rate": 2.6195992660006303e-05, + "loss": 0.3871, + "step": 31680 + }, + { + "epoch": 1.5858876516952334, + "grad_norm": 0.018478654325008392, + "learning_rate": 2.6186724991195715e-05, + "loss": 0.2885, + "step": 31690 + }, + { + "epoch": 1.586388089578381, + "grad_norm": 0.00861690379679203, + "learning_rate": 2.6177457322385128e-05, + "loss": 0.3324, + "step": 31700 + }, + { + "epoch": 1.5868885274615288, + "grad_norm": 0.039388641715049744, + "learning_rate": 2.6168189653574543e-05, + "loss": 0.4922, + "step": 31710 + }, + { + "epoch": 1.5873889653446764, + "grad_norm": 17.893936157226562, + "learning_rate": 2.6158921984763952e-05, + "loss": 0.3303, + "step": 31720 + }, + { + "epoch": 1.5878894032278243, + "grad_norm": 7.320963382720947, + "learning_rate": 2.6149654315953364e-05, + "loss": 0.2242, + "step": 31730 + }, + { + "epoch": 1.5883898411109723, + "grad_norm": 0.024082500487565994, + "learning_rate": 2.6140386647142777e-05, + "loss": 0.0034, + "step": 31740 + }, + { + "epoch": 1.58889027899412, + "grad_norm": 21.836875915527344, + "learning_rate": 2.6131118978332192e-05, + "loss": 0.2778, + "step": 31750 + }, + { + "epoch": 1.5893907168772676, + "grad_norm": 0.4840356707572937, + "learning_rate": 2.6121851309521605e-05, + "loss": 0.3417, + "step": 31760 + }, + { + "epoch": 1.5898911547604153, + "grad_norm": 0.05466405302286148, + "learning_rate": 2.611258364071102e-05, + "loss": 0.4434, + "step": 31770 + }, + { + "epoch": 1.590391592643563, + "grad_norm": 17.451688766479492, + "learning_rate": 2.6103315971900426e-05, + "loss": 0.2637, + "step": 31780 + }, + { + "epoch": 1.590892030526711, + "grad_norm": 0.01657954975962639, + "learning_rate": 2.609404830308984e-05, + "loss": 0.5545, + "step": 31790 + }, + { + "epoch": 1.5913924684098586, + "grad_norm": 0.006641357205808163, + "learning_rate": 2.6084780634279254e-05, + "loss": 0.1824, + "step": 31800 + }, + { + "epoch": 1.5918929062930065, + "grad_norm": 22.791343688964844, + "learning_rate": 2.607551296546867e-05, + "loss": 0.3765, + "step": 31810 + }, + { + "epoch": 1.5923933441761542, + "grad_norm": 6.102692604064941, + "learning_rate": 2.606624529665808e-05, + "loss": 0.1883, + "step": 31820 + }, + { + "epoch": 1.5928937820593019, + "grad_norm": 5.145735263824463, + "learning_rate": 2.605697762784749e-05, + "loss": 0.4909, + "step": 31830 + }, + { + "epoch": 1.5933942199424496, + "grad_norm": 0.04755775257945061, + "learning_rate": 2.6047709959036903e-05, + "loss": 0.2833, + "step": 31840 + }, + { + "epoch": 1.5938946578255972, + "grad_norm": 0.007472852244973183, + "learning_rate": 2.6038442290226318e-05, + "loss": 0.3225, + "step": 31850 + }, + { + "epoch": 1.5943950957087452, + "grad_norm": 0.2008572369813919, + "learning_rate": 2.602917462141573e-05, + "loss": 0.6576, + "step": 31860 + }, + { + "epoch": 1.594895533591893, + "grad_norm": 0.050308793783187866, + "learning_rate": 2.6019906952605143e-05, + "loss": 0.6048, + "step": 31870 + }, + { + "epoch": 1.5953959714750408, + "grad_norm": 0.0074388496577739716, + "learning_rate": 2.601063928379456e-05, + "loss": 0.1683, + "step": 31880 + }, + { + "epoch": 1.5958964093581884, + "grad_norm": 105.04157257080078, + "learning_rate": 2.6001371614983967e-05, + "loss": 0.3834, + "step": 31890 + }, + { + "epoch": 1.5963968472413361, + "grad_norm": 0.22336198389530182, + "learning_rate": 2.599210394617338e-05, + "loss": 0.3394, + "step": 31900 + }, + { + "epoch": 1.5968972851244838, + "grad_norm": 85.89427947998047, + "learning_rate": 2.598283627736279e-05, + "loss": 0.2858, + "step": 31910 + }, + { + "epoch": 1.5973977230076317, + "grad_norm": 8.79353141784668, + "learning_rate": 2.5973568608552207e-05, + "loss": 0.3085, + "step": 31920 + }, + { + "epoch": 1.5978981608907794, + "grad_norm": 0.050700295716524124, + "learning_rate": 2.596430093974162e-05, + "loss": 0.303, + "step": 31930 + }, + { + "epoch": 1.5983985987739273, + "grad_norm": 15.160731315612793, + "learning_rate": 2.5955033270931035e-05, + "loss": 0.1726, + "step": 31940 + }, + { + "epoch": 1.598899036657075, + "grad_norm": 0.028093351051211357, + "learning_rate": 2.594576560212044e-05, + "loss": 0.4081, + "step": 31950 + }, + { + "epoch": 1.5993994745402227, + "grad_norm": 4.603279113769531, + "learning_rate": 2.5936497933309856e-05, + "loss": 0.1088, + "step": 31960 + }, + { + "epoch": 1.5998999124233704, + "grad_norm": 0.29485711455345154, + "learning_rate": 2.592723026449927e-05, + "loss": 0.3974, + "step": 31970 + }, + { + "epoch": 1.600400350306518, + "grad_norm": 147.3387451171875, + "learning_rate": 2.5917962595688684e-05, + "loss": 0.5183, + "step": 31980 + }, + { + "epoch": 1.600900788189666, + "grad_norm": 0.21328337490558624, + "learning_rate": 2.5908694926878096e-05, + "loss": 0.179, + "step": 31990 + }, + { + "epoch": 1.6014012260728139, + "grad_norm": 25.382640838623047, + "learning_rate": 2.5899427258067505e-05, + "loss": 0.3846, + "step": 32000 + }, + { + "epoch": 1.6019016639559616, + "grad_norm": 21.308006286621094, + "learning_rate": 2.5890159589256918e-05, + "loss": 0.3147, + "step": 32010 + }, + { + "epoch": 1.6024021018391092, + "grad_norm": 0.059192463755607605, + "learning_rate": 2.5880891920446333e-05, + "loss": 0.7796, + "step": 32020 + }, + { + "epoch": 1.602902539722257, + "grad_norm": 36.21772766113281, + "learning_rate": 2.5871624251635745e-05, + "loss": 0.5936, + "step": 32030 + }, + { + "epoch": 1.6034029776054046, + "grad_norm": 18.469846725463867, + "learning_rate": 2.5862356582825158e-05, + "loss": 0.4847, + "step": 32040 + }, + { + "epoch": 1.6039034154885525, + "grad_norm": 34.78782653808594, + "learning_rate": 2.5853088914014573e-05, + "loss": 0.6535, + "step": 32050 + }, + { + "epoch": 1.6044038533717002, + "grad_norm": 0.6065393686294556, + "learning_rate": 2.5843821245203982e-05, + "loss": 0.1716, + "step": 32060 + }, + { + "epoch": 1.6049042912548481, + "grad_norm": 138.31593322753906, + "learning_rate": 2.5834553576393394e-05, + "loss": 0.355, + "step": 32070 + }, + { + "epoch": 1.6054047291379958, + "grad_norm": 2.8374390602111816, + "learning_rate": 2.5825285907582807e-05, + "loss": 0.3135, + "step": 32080 + }, + { + "epoch": 1.6059051670211435, + "grad_norm": 0.3468283414840698, + "learning_rate": 2.5816018238772222e-05, + "loss": 0.4552, + "step": 32090 + }, + { + "epoch": 1.6064056049042912, + "grad_norm": 0.24430356919765472, + "learning_rate": 2.5806750569961635e-05, + "loss": 0.2232, + "step": 32100 + }, + { + "epoch": 1.6069060427874389, + "grad_norm": 0.43783217668533325, + "learning_rate": 2.5797482901151043e-05, + "loss": 0.2169, + "step": 32110 + }, + { + "epoch": 1.6074064806705868, + "grad_norm": 2.3899261951446533, + "learning_rate": 2.5788215232340456e-05, + "loss": 0.243, + "step": 32120 + }, + { + "epoch": 1.6079069185537345, + "grad_norm": 0.24472388625144958, + "learning_rate": 2.577894756352987e-05, + "loss": 0.4491, + "step": 32130 + }, + { + "epoch": 1.6084073564368824, + "grad_norm": 26.33999252319336, + "learning_rate": 2.5769679894719284e-05, + "loss": 0.3759, + "step": 32140 + }, + { + "epoch": 1.60890779432003, + "grad_norm": 0.23131568729877472, + "learning_rate": 2.57604122259087e-05, + "loss": 0.2051, + "step": 32150 + }, + { + "epoch": 1.6094082322031777, + "grad_norm": 1.5904747247695923, + "learning_rate": 2.575114455709811e-05, + "loss": 0.1202, + "step": 32160 + }, + { + "epoch": 1.6099086700863254, + "grad_norm": 0.30862298607826233, + "learning_rate": 2.574187688828752e-05, + "loss": 0.4, + "step": 32170 + }, + { + "epoch": 1.6104091079694733, + "grad_norm": 0.02693987265229225, + "learning_rate": 2.5732609219476932e-05, + "loss": 0.2293, + "step": 32180 + }, + { + "epoch": 1.610909545852621, + "grad_norm": 0.48933711647987366, + "learning_rate": 2.5723341550666348e-05, + "loss": 0.2467, + "step": 32190 + }, + { + "epoch": 1.611409983735769, + "grad_norm": 3.9130709171295166, + "learning_rate": 2.571407388185576e-05, + "loss": 0.4979, + "step": 32200 + }, + { + "epoch": 1.6119104216189166, + "grad_norm": 0.21328066289424896, + "learning_rate": 2.5704806213045173e-05, + "loss": 0.193, + "step": 32210 + }, + { + "epoch": 1.6124108595020643, + "grad_norm": 0.1240946352481842, + "learning_rate": 2.569553854423458e-05, + "loss": 0.3522, + "step": 32220 + }, + { + "epoch": 1.612911297385212, + "grad_norm": 6.129695415496826, + "learning_rate": 2.5686270875423997e-05, + "loss": 0.5082, + "step": 32230 + }, + { + "epoch": 1.6134117352683597, + "grad_norm": 27.462677001953125, + "learning_rate": 2.567700320661341e-05, + "loss": 0.372, + "step": 32240 + }, + { + "epoch": 1.6139121731515076, + "grad_norm": 66.62861633300781, + "learning_rate": 2.566773553780282e-05, + "loss": 0.1784, + "step": 32250 + }, + { + "epoch": 1.6144126110346553, + "grad_norm": 0.20010316371917725, + "learning_rate": 2.5658467868992237e-05, + "loss": 0.1354, + "step": 32260 + }, + { + "epoch": 1.6149130489178032, + "grad_norm": 45.87023162841797, + "learning_rate": 2.564920020018165e-05, + "loss": 0.0935, + "step": 32270 + }, + { + "epoch": 1.6154134868009509, + "grad_norm": 0.04188704490661621, + "learning_rate": 2.563993253137106e-05, + "loss": 0.1624, + "step": 32280 + }, + { + "epoch": 1.6159139246840986, + "grad_norm": 0.04079123213887215, + "learning_rate": 2.563066486256047e-05, + "loss": 0.5578, + "step": 32290 + }, + { + "epoch": 1.6164143625672462, + "grad_norm": 0.040961842983961105, + "learning_rate": 2.5621397193749886e-05, + "loss": 0.2994, + "step": 32300 + }, + { + "epoch": 1.616914800450394, + "grad_norm": 0.4251742362976074, + "learning_rate": 2.56121295249393e-05, + "loss": 0.7276, + "step": 32310 + }, + { + "epoch": 1.6174152383335418, + "grad_norm": 0.013340692967176437, + "learning_rate": 2.5602861856128714e-05, + "loss": 0.4001, + "step": 32320 + }, + { + "epoch": 1.6179156762166897, + "grad_norm": 153.26670837402344, + "learning_rate": 2.559359418731812e-05, + "loss": 0.2362, + "step": 32330 + }, + { + "epoch": 1.6184161140998374, + "grad_norm": 129.0662841796875, + "learning_rate": 2.5584326518507535e-05, + "loss": 0.2967, + "step": 32340 + }, + { + "epoch": 1.6189165519829851, + "grad_norm": 0.3570416271686554, + "learning_rate": 2.5575058849696947e-05, + "loss": 0.3226, + "step": 32350 + }, + { + "epoch": 1.6194169898661328, + "grad_norm": 0.0017001839587464929, + "learning_rate": 2.5565791180886363e-05, + "loss": 0.01, + "step": 32360 + }, + { + "epoch": 1.6199174277492805, + "grad_norm": 0.15375927090644836, + "learning_rate": 2.5556523512075775e-05, + "loss": 0.0956, + "step": 32370 + }, + { + "epoch": 1.6204178656324284, + "grad_norm": 0.03849753737449646, + "learning_rate": 2.5547255843265188e-05, + "loss": 0.346, + "step": 32380 + }, + { + "epoch": 1.620918303515576, + "grad_norm": 0.23100925981998444, + "learning_rate": 2.5537988174454596e-05, + "loss": 0.2987, + "step": 32390 + }, + { + "epoch": 1.621418741398724, + "grad_norm": 73.91325378417969, + "learning_rate": 2.5528720505644012e-05, + "loss": 0.3898, + "step": 32400 + }, + { + "epoch": 1.6219191792818717, + "grad_norm": 0.05169935151934624, + "learning_rate": 2.5519452836833424e-05, + "loss": 0.3424, + "step": 32410 + }, + { + "epoch": 1.6224196171650194, + "grad_norm": 0.219170942902565, + "learning_rate": 2.5510185168022837e-05, + "loss": 0.3589, + "step": 32420 + }, + { + "epoch": 1.622920055048167, + "grad_norm": 7.765563011169434, + "learning_rate": 2.5500917499212252e-05, + "loss": 0.2776, + "step": 32430 + }, + { + "epoch": 1.6234204929313147, + "grad_norm": 13.29527759552002, + "learning_rate": 2.5491649830401658e-05, + "loss": 0.8119, + "step": 32440 + }, + { + "epoch": 1.6239209308144626, + "grad_norm": 105.3114242553711, + "learning_rate": 2.5482382161591073e-05, + "loss": 0.4691, + "step": 32450 + }, + { + "epoch": 1.6244213686976106, + "grad_norm": 0.005303649697452784, + "learning_rate": 2.5473114492780486e-05, + "loss": 0.1636, + "step": 32460 + }, + { + "epoch": 1.6249218065807582, + "grad_norm": 15.763071060180664, + "learning_rate": 2.54638468239699e-05, + "loss": 0.3451, + "step": 32470 + }, + { + "epoch": 1.625422244463906, + "grad_norm": 0.318024605512619, + "learning_rate": 2.5454579155159313e-05, + "loss": 0.6931, + "step": 32480 + }, + { + "epoch": 1.6259226823470536, + "grad_norm": 30.009841918945312, + "learning_rate": 2.544531148634873e-05, + "loss": 0.1818, + "step": 32490 + }, + { + "epoch": 1.6264231202302013, + "grad_norm": 61.55513000488281, + "learning_rate": 2.5436043817538135e-05, + "loss": 0.5402, + "step": 32500 + }, + { + "epoch": 1.6269235581133492, + "grad_norm": 1.5383881330490112, + "learning_rate": 2.542677614872755e-05, + "loss": 0.2148, + "step": 32510 + }, + { + "epoch": 1.627423995996497, + "grad_norm": 22.963939666748047, + "learning_rate": 2.5417508479916962e-05, + "loss": 0.404, + "step": 32520 + }, + { + "epoch": 1.6279244338796448, + "grad_norm": 73.84637451171875, + "learning_rate": 2.5408240811106375e-05, + "loss": 0.3408, + "step": 32530 + }, + { + "epoch": 1.6284248717627925, + "grad_norm": 0.51849365234375, + "learning_rate": 2.539897314229579e-05, + "loss": 0.2524, + "step": 32540 + }, + { + "epoch": 1.6289253096459402, + "grad_norm": 0.10199638456106186, + "learning_rate": 2.53897054734852e-05, + "loss": 0.0873, + "step": 32550 + }, + { + "epoch": 1.6294257475290879, + "grad_norm": 74.9653549194336, + "learning_rate": 2.538043780467461e-05, + "loss": 0.3456, + "step": 32560 + }, + { + "epoch": 1.6299261854122356, + "grad_norm": 32.3977165222168, + "learning_rate": 2.5371170135864024e-05, + "loss": 0.4971, + "step": 32570 + }, + { + "epoch": 1.6304266232953835, + "grad_norm": 5.196051597595215, + "learning_rate": 2.536190246705344e-05, + "loss": 0.1158, + "step": 32580 + }, + { + "epoch": 1.6309270611785314, + "grad_norm": 106.54750061035156, + "learning_rate": 2.535263479824285e-05, + "loss": 0.3226, + "step": 32590 + }, + { + "epoch": 1.631427499061679, + "grad_norm": 0.045603349804878235, + "learning_rate": 2.5343367129432267e-05, + "loss": 0.0651, + "step": 32600 + }, + { + "epoch": 1.6319279369448267, + "grad_norm": 10.787080764770508, + "learning_rate": 2.5334099460621673e-05, + "loss": 0.0445, + "step": 32610 + }, + { + "epoch": 1.6324283748279744, + "grad_norm": 49.39600372314453, + "learning_rate": 2.5324831791811088e-05, + "loss": 0.6336, + "step": 32620 + }, + { + "epoch": 1.6329288127111221, + "grad_norm": 0.14695438742637634, + "learning_rate": 2.53155641230005e-05, + "loss": 0.2386, + "step": 32630 + }, + { + "epoch": 1.63342925059427, + "grad_norm": 88.65766143798828, + "learning_rate": 2.5306296454189916e-05, + "loss": 0.3828, + "step": 32640 + }, + { + "epoch": 1.6339296884774177, + "grad_norm": 0.16197246313095093, + "learning_rate": 2.529702878537933e-05, + "loss": 0.5317, + "step": 32650 + }, + { + "epoch": 1.6344301263605656, + "grad_norm": 0.01987529918551445, + "learning_rate": 2.528776111656874e-05, + "loss": 0.1497, + "step": 32660 + }, + { + "epoch": 1.6349305642437133, + "grad_norm": 0.24809664487838745, + "learning_rate": 2.527849344775815e-05, + "loss": 0.2934, + "step": 32670 + }, + { + "epoch": 1.635431002126861, + "grad_norm": 7.151462554931641, + "learning_rate": 2.5269225778947565e-05, + "loss": 0.0071, + "step": 32680 + }, + { + "epoch": 1.6359314400100087, + "grad_norm": 0.2678244411945343, + "learning_rate": 2.5259958110136977e-05, + "loss": 0.4383, + "step": 32690 + }, + { + "epoch": 1.6364318778931564, + "grad_norm": 5.833319664001465, + "learning_rate": 2.525069044132639e-05, + "loss": 0.1213, + "step": 32700 + }, + { + "epoch": 1.6369323157763043, + "grad_norm": 51.95672607421875, + "learning_rate": 2.5241422772515805e-05, + "loss": 0.5789, + "step": 32710 + }, + { + "epoch": 1.6374327536594522, + "grad_norm": 0.012845328077673912, + "learning_rate": 2.5232155103705214e-05, + "loss": 0.7414, + "step": 32720 + }, + { + "epoch": 1.6379331915425999, + "grad_norm": 150.40444946289062, + "learning_rate": 2.5222887434894626e-05, + "loss": 0.5691, + "step": 32730 + }, + { + "epoch": 1.6384336294257476, + "grad_norm": 0.1707458347082138, + "learning_rate": 2.521361976608404e-05, + "loss": 0.3235, + "step": 32740 + }, + { + "epoch": 1.6389340673088952, + "grad_norm": 5.835041046142578, + "learning_rate": 2.5204352097273454e-05, + "loss": 0.3728, + "step": 32750 + }, + { + "epoch": 1.639434505192043, + "grad_norm": 0.045597370713949203, + "learning_rate": 2.5195084428462866e-05, + "loss": 0.3442, + "step": 32760 + }, + { + "epoch": 1.6399349430751908, + "grad_norm": 0.41897276043891907, + "learning_rate": 2.5185816759652282e-05, + "loss": 0.4011, + "step": 32770 + }, + { + "epoch": 1.6404353809583385, + "grad_norm": 15.353118896484375, + "learning_rate": 2.5176549090841688e-05, + "loss": 0.2506, + "step": 32780 + }, + { + "epoch": 1.6409358188414864, + "grad_norm": 0.006061752792447805, + "learning_rate": 2.5167281422031103e-05, + "loss": 0.5721, + "step": 32790 + }, + { + "epoch": 1.6414362567246341, + "grad_norm": 2.1482365131378174, + "learning_rate": 2.5158013753220515e-05, + "loss": 0.6174, + "step": 32800 + }, + { + "epoch": 1.6419366946077818, + "grad_norm": 33.118560791015625, + "learning_rate": 2.514874608440993e-05, + "loss": 0.4432, + "step": 32810 + }, + { + "epoch": 1.6424371324909295, + "grad_norm": 0.02001367323100567, + "learning_rate": 2.5139478415599343e-05, + "loss": 0.1526, + "step": 32820 + }, + { + "epoch": 1.6429375703740772, + "grad_norm": 117.73522186279297, + "learning_rate": 2.5130210746788752e-05, + "loss": 0.3634, + "step": 32830 + }, + { + "epoch": 1.643438008257225, + "grad_norm": 160.52394104003906, + "learning_rate": 2.5120943077978164e-05, + "loss": 0.6552, + "step": 32840 + }, + { + "epoch": 1.643938446140373, + "grad_norm": 0.030013693496584892, + "learning_rate": 2.511167540916758e-05, + "loss": 0.3087, + "step": 32850 + }, + { + "epoch": 1.6444388840235207, + "grad_norm": 0.26473480463027954, + "learning_rate": 2.5102407740356992e-05, + "loss": 0.1356, + "step": 32860 + }, + { + "epoch": 1.6449393219066684, + "grad_norm": 55.36628341674805, + "learning_rate": 2.5093140071546405e-05, + "loss": 0.2179, + "step": 32870 + }, + { + "epoch": 1.645439759789816, + "grad_norm": 0.3809390664100647, + "learning_rate": 2.508387240273582e-05, + "loss": 0.0048, + "step": 32880 + }, + { + "epoch": 1.6459401976729637, + "grad_norm": 0.0768519937992096, + "learning_rate": 2.507460473392523e-05, + "loss": 0.2045, + "step": 32890 + }, + { + "epoch": 1.6464406355561116, + "grad_norm": 0.25538572669029236, + "learning_rate": 2.506533706511464e-05, + "loss": 0.3897, + "step": 32900 + }, + { + "epoch": 1.6469410734392593, + "grad_norm": 0.017561007291078568, + "learning_rate": 2.5056069396304054e-05, + "loss": 0.3395, + "step": 32910 + }, + { + "epoch": 1.6474415113224072, + "grad_norm": 8.297914505004883, + "learning_rate": 2.504680172749347e-05, + "loss": 0.2343, + "step": 32920 + }, + { + "epoch": 1.647941949205555, + "grad_norm": 142.54869079589844, + "learning_rate": 2.503753405868288e-05, + "loss": 0.6288, + "step": 32930 + }, + { + "epoch": 1.6484423870887026, + "grad_norm": 104.26538848876953, + "learning_rate": 2.502826638987229e-05, + "loss": 0.6219, + "step": 32940 + }, + { + "epoch": 1.6489428249718503, + "grad_norm": 0.007061805110424757, + "learning_rate": 2.5018998721061703e-05, + "loss": 0.4514, + "step": 32950 + }, + { + "epoch": 1.649443262854998, + "grad_norm": 4.52899694442749, + "learning_rate": 2.5009731052251118e-05, + "loss": 0.0788, + "step": 32960 + }, + { + "epoch": 1.649943700738146, + "grad_norm": 0.08490787446498871, + "learning_rate": 2.500046338344053e-05, + "loss": 0.5702, + "step": 32970 + }, + { + "epoch": 1.6504441386212936, + "grad_norm": 0.031222615391016006, + "learning_rate": 2.4991195714629943e-05, + "loss": 0.1298, + "step": 32980 + }, + { + "epoch": 1.6509445765044415, + "grad_norm": 17.913724899291992, + "learning_rate": 2.4981928045819355e-05, + "loss": 0.49, + "step": 32990 + }, + { + "epoch": 1.6514450143875892, + "grad_norm": 118.85713958740234, + "learning_rate": 2.497266037700877e-05, + "loss": 0.2425, + "step": 33000 + }, + { + "epoch": 1.6519454522707369, + "grad_norm": 0.41494619846343994, + "learning_rate": 2.496339270819818e-05, + "loss": 0.2563, + "step": 33010 + }, + { + "epoch": 1.6524458901538845, + "grad_norm": 69.30790710449219, + "learning_rate": 2.4954125039387595e-05, + "loss": 0.5528, + "step": 33020 + }, + { + "epoch": 1.6529463280370325, + "grad_norm": 2.616309404373169, + "learning_rate": 2.4944857370577007e-05, + "loss": 0.2902, + "step": 33030 + }, + { + "epoch": 1.6534467659201801, + "grad_norm": 0.01701965369284153, + "learning_rate": 2.493558970176642e-05, + "loss": 0.4248, + "step": 33040 + }, + { + "epoch": 1.653947203803328, + "grad_norm": 55.74851608276367, + "learning_rate": 2.4926322032955832e-05, + "loss": 0.4416, + "step": 33050 + }, + { + "epoch": 1.6544476416864757, + "grad_norm": 0.21021591126918793, + "learning_rate": 2.4917054364145244e-05, + "loss": 0.2871, + "step": 33060 + }, + { + "epoch": 1.6549480795696234, + "grad_norm": 42.13960266113281, + "learning_rate": 2.4907786695334656e-05, + "loss": 0.5385, + "step": 33070 + }, + { + "epoch": 1.655448517452771, + "grad_norm": 0.07118815183639526, + "learning_rate": 2.489851902652407e-05, + "loss": 0.3192, + "step": 33080 + }, + { + "epoch": 1.6559489553359188, + "grad_norm": 0.15884804725646973, + "learning_rate": 2.488925135771348e-05, + "loss": 0.3619, + "step": 33090 + }, + { + "epoch": 1.6564493932190667, + "grad_norm": 30.499271392822266, + "learning_rate": 2.4879983688902893e-05, + "loss": 0.1389, + "step": 33100 + }, + { + "epoch": 1.6569498311022144, + "grad_norm": 0.04763022065162659, + "learning_rate": 2.487071602009231e-05, + "loss": 0.3083, + "step": 33110 + }, + { + "epoch": 1.6574502689853623, + "grad_norm": 0.026299139484763145, + "learning_rate": 2.4861448351281717e-05, + "loss": 0.2417, + "step": 33120 + }, + { + "epoch": 1.65795070686851, + "grad_norm": 2.1975297927856445, + "learning_rate": 2.4852180682471133e-05, + "loss": 0.5548, + "step": 33130 + }, + { + "epoch": 1.6584511447516577, + "grad_norm": 0.03231377154588699, + "learning_rate": 2.4842913013660545e-05, + "loss": 0.3798, + "step": 33140 + }, + { + "epoch": 1.6589515826348054, + "grad_norm": 15.513935089111328, + "learning_rate": 2.4833645344849958e-05, + "loss": 0.1479, + "step": 33150 + }, + { + "epoch": 1.6594520205179533, + "grad_norm": 0.0417209155857563, + "learning_rate": 2.482437767603937e-05, + "loss": 0.2026, + "step": 33160 + }, + { + "epoch": 1.659952458401101, + "grad_norm": 20.115890502929688, + "learning_rate": 2.4815110007228786e-05, + "loss": 0.4175, + "step": 33170 + }, + { + "epoch": 1.6604528962842489, + "grad_norm": 24.095048904418945, + "learning_rate": 2.4805842338418194e-05, + "loss": 0.4273, + "step": 33180 + }, + { + "epoch": 1.6609533341673965, + "grad_norm": 51.18112564086914, + "learning_rate": 2.479657466960761e-05, + "loss": 0.2101, + "step": 33190 + }, + { + "epoch": 1.6614537720505442, + "grad_norm": 38.6728630065918, + "learning_rate": 2.4787307000797022e-05, + "loss": 0.7157, + "step": 33200 + }, + { + "epoch": 1.661954209933692, + "grad_norm": 10.762001991271973, + "learning_rate": 2.4778039331986434e-05, + "loss": 0.3098, + "step": 33210 + }, + { + "epoch": 1.6624546478168396, + "grad_norm": 0.17561589181423187, + "learning_rate": 2.4768771663175847e-05, + "loss": 0.2074, + "step": 33220 + }, + { + "epoch": 1.6629550856999875, + "grad_norm": 0.2575613260269165, + "learning_rate": 2.475950399436526e-05, + "loss": 0.4165, + "step": 33230 + }, + { + "epoch": 1.6634555235831352, + "grad_norm": 40.566131591796875, + "learning_rate": 2.475023632555467e-05, + "loss": 0.0648, + "step": 33240 + }, + { + "epoch": 1.663955961466283, + "grad_norm": 0.03303411230444908, + "learning_rate": 2.4740968656744083e-05, + "loss": 0.6513, + "step": 33250 + }, + { + "epoch": 1.6644563993494308, + "grad_norm": 58.390655517578125, + "learning_rate": 2.4731700987933496e-05, + "loss": 0.4394, + "step": 33260 + }, + { + "epoch": 1.6649568372325785, + "grad_norm": 16.23965072631836, + "learning_rate": 2.4722433319122908e-05, + "loss": 0.5482, + "step": 33270 + }, + { + "epoch": 1.6654572751157262, + "grad_norm": 20.554044723510742, + "learning_rate": 2.4713165650312324e-05, + "loss": 0.2545, + "step": 33280 + }, + { + "epoch": 1.6659577129988739, + "grad_norm": 34.11521530151367, + "learning_rate": 2.4703897981501732e-05, + "loss": 0.2044, + "step": 33290 + }, + { + "epoch": 1.6664581508820218, + "grad_norm": 2.464559316635132, + "learning_rate": 2.4694630312691148e-05, + "loss": 0.0765, + "step": 33300 + }, + { + "epoch": 1.6669585887651697, + "grad_norm": 0.20337575674057007, + "learning_rate": 2.468536264388056e-05, + "loss": 0.3999, + "step": 33310 + }, + { + "epoch": 1.6674590266483174, + "grad_norm": 35.39899444580078, + "learning_rate": 2.4676094975069973e-05, + "loss": 0.7955, + "step": 33320 + }, + { + "epoch": 1.667959464531465, + "grad_norm": 0.3517041802406311, + "learning_rate": 2.4666827306259385e-05, + "loss": 0.247, + "step": 33330 + }, + { + "epoch": 1.6684599024146127, + "grad_norm": 0.06219504401087761, + "learning_rate": 2.4657559637448797e-05, + "loss": 0.1729, + "step": 33340 + }, + { + "epoch": 1.6689603402977604, + "grad_norm": 75.38439178466797, + "learning_rate": 2.464829196863821e-05, + "loss": 0.1843, + "step": 33350 + }, + { + "epoch": 1.6694607781809083, + "grad_norm": 2.836052894592285, + "learning_rate": 2.4639024299827625e-05, + "loss": 0.0734, + "step": 33360 + }, + { + "epoch": 1.669961216064056, + "grad_norm": 1.5327898263931274, + "learning_rate": 2.4629756631017034e-05, + "loss": 0.2289, + "step": 33370 + }, + { + "epoch": 1.670461653947204, + "grad_norm": 101.7213134765625, + "learning_rate": 2.4620488962206446e-05, + "loss": 0.2105, + "step": 33380 + }, + { + "epoch": 1.6709620918303516, + "grad_norm": 13.732851028442383, + "learning_rate": 2.461122129339586e-05, + "loss": 0.5445, + "step": 33390 + }, + { + "epoch": 1.6714625297134993, + "grad_norm": 0.020664963871240616, + "learning_rate": 2.460195362458527e-05, + "loss": 0.3688, + "step": 33400 + }, + { + "epoch": 1.671962967596647, + "grad_norm": 8.762467384338379, + "learning_rate": 2.4592685955774686e-05, + "loss": 0.005, + "step": 33410 + }, + { + "epoch": 1.6724634054797947, + "grad_norm": 210.28424072265625, + "learning_rate": 2.45834182869641e-05, + "loss": 0.8904, + "step": 33420 + }, + { + "epoch": 1.6729638433629426, + "grad_norm": 149.62831115722656, + "learning_rate": 2.457415061815351e-05, + "loss": 0.4748, + "step": 33430 + }, + { + "epoch": 1.6734642812460905, + "grad_norm": 82.04991149902344, + "learning_rate": 2.4564882949342923e-05, + "loss": 0.4741, + "step": 33440 + }, + { + "epoch": 1.6739647191292382, + "grad_norm": 21.675884246826172, + "learning_rate": 2.4555615280532335e-05, + "loss": 0.2446, + "step": 33450 + }, + { + "epoch": 1.6744651570123859, + "grad_norm": 0.01441262848675251, + "learning_rate": 2.4546347611721747e-05, + "loss": 0.2314, + "step": 33460 + }, + { + "epoch": 1.6749655948955335, + "grad_norm": 3.9823074340820312, + "learning_rate": 2.4537079942911163e-05, + "loss": 0.1844, + "step": 33470 + }, + { + "epoch": 1.6754660327786812, + "grad_norm": 0.030252959579229355, + "learning_rate": 2.4527812274100572e-05, + "loss": 0.3782, + "step": 33480 + }, + { + "epoch": 1.6759664706618291, + "grad_norm": 26.06352996826172, + "learning_rate": 2.4518544605289988e-05, + "loss": 0.4311, + "step": 33490 + }, + { + "epoch": 1.6764669085449768, + "grad_norm": 0.7316672801971436, + "learning_rate": 2.45092769364794e-05, + "loss": 0.6919, + "step": 33500 + }, + { + "epoch": 1.6769673464281247, + "grad_norm": 89.66585540771484, + "learning_rate": 2.4500009267668812e-05, + "loss": 0.016, + "step": 33510 + }, + { + "epoch": 1.6774677843112724, + "grad_norm": 22.800888061523438, + "learning_rate": 2.4490741598858224e-05, + "loss": 0.2118, + "step": 33520 + }, + { + "epoch": 1.67796822219442, + "grad_norm": 0.01770778000354767, + "learning_rate": 2.4481473930047637e-05, + "loss": 0.2046, + "step": 33530 + }, + { + "epoch": 1.6784686600775678, + "grad_norm": 35.699974060058594, + "learning_rate": 2.447220626123705e-05, + "loss": 0.3426, + "step": 33540 + }, + { + "epoch": 1.6789690979607155, + "grad_norm": 16.98150634765625, + "learning_rate": 2.446293859242646e-05, + "loss": 0.3224, + "step": 33550 + }, + { + "epoch": 1.6794695358438634, + "grad_norm": 0.13474908471107483, + "learning_rate": 2.4453670923615877e-05, + "loss": 0.2383, + "step": 33560 + }, + { + "epoch": 1.6799699737270113, + "grad_norm": 1.2196218967437744, + "learning_rate": 2.4444403254805286e-05, + "loss": 0.2664, + "step": 33570 + }, + { + "epoch": 1.680470411610159, + "grad_norm": 27.679847717285156, + "learning_rate": 2.44351355859947e-05, + "loss": 0.2231, + "step": 33580 + }, + { + "epoch": 1.6809708494933067, + "grad_norm": 0.007467067800462246, + "learning_rate": 2.442586791718411e-05, + "loss": 0.1899, + "step": 33590 + }, + { + "epoch": 1.6814712873764543, + "grad_norm": 0.006752725224941969, + "learning_rate": 2.4416600248373526e-05, + "loss": 0.4656, + "step": 33600 + }, + { + "epoch": 1.681971725259602, + "grad_norm": 65.40291595458984, + "learning_rate": 2.4407332579562938e-05, + "loss": 0.3426, + "step": 33610 + }, + { + "epoch": 1.68247216314275, + "grad_norm": 57.39335250854492, + "learning_rate": 2.439806491075235e-05, + "loss": 0.4861, + "step": 33620 + }, + { + "epoch": 1.6829726010258976, + "grad_norm": 1.4882246255874634, + "learning_rate": 2.4388797241941762e-05, + "loss": 0.1591, + "step": 33630 + }, + { + "epoch": 1.6834730389090455, + "grad_norm": 20.057844161987305, + "learning_rate": 2.4379529573131178e-05, + "loss": 0.4211, + "step": 33640 + }, + { + "epoch": 1.6839734767921932, + "grad_norm": 28.434650421142578, + "learning_rate": 2.4370261904320587e-05, + "loss": 0.2448, + "step": 33650 + }, + { + "epoch": 1.684473914675341, + "grad_norm": 10.009492874145508, + "learning_rate": 2.4360994235510002e-05, + "loss": 0.2861, + "step": 33660 + }, + { + "epoch": 1.6849743525584886, + "grad_norm": 103.63787841796875, + "learning_rate": 2.4351726566699415e-05, + "loss": 0.5554, + "step": 33670 + }, + { + "epoch": 1.6854747904416363, + "grad_norm": 0.5749062895774841, + "learning_rate": 2.4342458897888827e-05, + "loss": 0.2527, + "step": 33680 + }, + { + "epoch": 1.6859752283247842, + "grad_norm": 0.01156658586114645, + "learning_rate": 2.433319122907824e-05, + "loss": 0.2271, + "step": 33690 + }, + { + "epoch": 1.686475666207932, + "grad_norm": 4.683034896850586, + "learning_rate": 2.432392356026765e-05, + "loss": 0.1267, + "step": 33700 + }, + { + "epoch": 1.6869761040910798, + "grad_norm": 0.019376052543520927, + "learning_rate": 2.4314655891457064e-05, + "loss": 0.1648, + "step": 33710 + }, + { + "epoch": 1.6874765419742275, + "grad_norm": 84.093994140625, + "learning_rate": 2.4305388222646476e-05, + "loss": 0.1845, + "step": 33720 + }, + { + "epoch": 1.6879769798573752, + "grad_norm": 4.284000873565674, + "learning_rate": 2.4296120553835888e-05, + "loss": 0.2902, + "step": 33730 + }, + { + "epoch": 1.6884774177405228, + "grad_norm": 1.7844337224960327, + "learning_rate": 2.42868528850253e-05, + "loss": 0.3197, + "step": 33740 + }, + { + "epoch": 1.6889778556236708, + "grad_norm": 0.022838212549686432, + "learning_rate": 2.4277585216214716e-05, + "loss": 0.2397, + "step": 33750 + }, + { + "epoch": 1.6894782935068184, + "grad_norm": 0.0023198339622467756, + "learning_rate": 2.4268317547404125e-05, + "loss": 0.5994, + "step": 33760 + }, + { + "epoch": 1.6899787313899663, + "grad_norm": 0.300702303647995, + "learning_rate": 2.425904987859354e-05, + "loss": 0.5318, + "step": 33770 + }, + { + "epoch": 1.690479169273114, + "grad_norm": 16.749711990356445, + "learning_rate": 2.4249782209782953e-05, + "loss": 0.5083, + "step": 33780 + }, + { + "epoch": 1.6909796071562617, + "grad_norm": 81.97747802734375, + "learning_rate": 2.4240514540972365e-05, + "loss": 0.4281, + "step": 33790 + }, + { + "epoch": 1.6914800450394094, + "grad_norm": 0.22365516424179077, + "learning_rate": 2.4231246872161777e-05, + "loss": 0.0945, + "step": 33800 + }, + { + "epoch": 1.691980482922557, + "grad_norm": 0.10696660727262497, + "learning_rate": 2.422197920335119e-05, + "loss": 0.3208, + "step": 33810 + }, + { + "epoch": 1.692480920805705, + "grad_norm": 0.4125266373157501, + "learning_rate": 2.4212711534540602e-05, + "loss": 0.4541, + "step": 33820 + }, + { + "epoch": 1.692981358688853, + "grad_norm": 0.13574647903442383, + "learning_rate": 2.4203443865730017e-05, + "loss": 0.1534, + "step": 33830 + }, + { + "epoch": 1.6934817965720006, + "grad_norm": 0.034345593303442, + "learning_rate": 2.4194176196919426e-05, + "loss": 0.2024, + "step": 33840 + }, + { + "epoch": 1.6939822344551483, + "grad_norm": 5.123168468475342, + "learning_rate": 2.4184908528108842e-05, + "loss": 0.5884, + "step": 33850 + }, + { + "epoch": 1.694482672338296, + "grad_norm": 63.810482025146484, + "learning_rate": 2.4175640859298254e-05, + "loss": 0.7105, + "step": 33860 + }, + { + "epoch": 1.6949831102214437, + "grad_norm": 0.45672330260276794, + "learning_rate": 2.4166373190487666e-05, + "loss": 0.3641, + "step": 33870 + }, + { + "epoch": 1.6954835481045916, + "grad_norm": 0.14254313707351685, + "learning_rate": 2.415710552167708e-05, + "loss": 0.4493, + "step": 33880 + }, + { + "epoch": 1.6959839859877393, + "grad_norm": 9.60848617553711, + "learning_rate": 2.414783785286649e-05, + "loss": 0.2154, + "step": 33890 + }, + { + "epoch": 1.6964844238708872, + "grad_norm": 15.813383102416992, + "learning_rate": 2.4138570184055903e-05, + "loss": 0.3312, + "step": 33900 + }, + { + "epoch": 1.6969848617540348, + "grad_norm": 2.888733386993408, + "learning_rate": 2.4129302515245315e-05, + "loss": 0.6399, + "step": 33910 + }, + { + "epoch": 1.6974852996371825, + "grad_norm": 0.15327726304531097, + "learning_rate": 2.412003484643473e-05, + "loss": 0.1605, + "step": 33920 + }, + { + "epoch": 1.6979857375203302, + "grad_norm": 0.7046729326248169, + "learning_rate": 2.411076717762414e-05, + "loss": 0.2003, + "step": 33930 + }, + { + "epoch": 1.698486175403478, + "grad_norm": 19.76140022277832, + "learning_rate": 2.4101499508813556e-05, + "loss": 0.384, + "step": 33940 + }, + { + "epoch": 1.6989866132866258, + "grad_norm": 0.3029325306415558, + "learning_rate": 2.4092231840002964e-05, + "loss": 0.3056, + "step": 33950 + }, + { + "epoch": 1.6994870511697735, + "grad_norm": 61.91281509399414, + "learning_rate": 2.408296417119238e-05, + "loss": 0.2774, + "step": 33960 + }, + { + "epoch": 1.6999874890529214, + "grad_norm": 88.20533752441406, + "learning_rate": 2.4073696502381792e-05, + "loss": 0.1941, + "step": 33970 + }, + { + "epoch": 1.700487926936069, + "grad_norm": 3.1475045680999756, + "learning_rate": 2.4064428833571205e-05, + "loss": 0.2785, + "step": 33980 + }, + { + "epoch": 1.7009883648192168, + "grad_norm": 0.45141252875328064, + "learning_rate": 2.4055161164760617e-05, + "loss": 0.4923, + "step": 33990 + }, + { + "epoch": 1.7014888027023645, + "grad_norm": 0.05526474490761757, + "learning_rate": 2.4045893495950032e-05, + "loss": 0.2486, + "step": 34000 + }, + { + "epoch": 1.7019892405855124, + "grad_norm": 3.395172357559204, + "learning_rate": 2.403662582713944e-05, + "loss": 0.1615, + "step": 34010 + }, + { + "epoch": 1.70248967846866, + "grad_norm": 0.14037097990512848, + "learning_rate": 2.4027358158328857e-05, + "loss": 0.4285, + "step": 34020 + }, + { + "epoch": 1.702990116351808, + "grad_norm": 11.452399253845215, + "learning_rate": 2.401809048951827e-05, + "loss": 0.0957, + "step": 34030 + }, + { + "epoch": 1.7034905542349557, + "grad_norm": 12.245529174804688, + "learning_rate": 2.400882282070768e-05, + "loss": 0.7205, + "step": 34040 + }, + { + "epoch": 1.7039909921181033, + "grad_norm": 0.03370455652475357, + "learning_rate": 2.3999555151897094e-05, + "loss": 0.2827, + "step": 34050 + }, + { + "epoch": 1.704491430001251, + "grad_norm": 156.52880859375, + "learning_rate": 2.3990287483086506e-05, + "loss": 0.2874, + "step": 34060 + }, + { + "epoch": 1.7049918678843987, + "grad_norm": 0.05539505556225777, + "learning_rate": 2.3981019814275918e-05, + "loss": 0.3501, + "step": 34070 + }, + { + "epoch": 1.7054923057675466, + "grad_norm": 13.793956756591797, + "learning_rate": 2.397175214546533e-05, + "loss": 0.4213, + "step": 34080 + }, + { + "epoch": 1.7059927436506943, + "grad_norm": 10.432394981384277, + "learning_rate": 2.3962484476654743e-05, + "loss": 0.167, + "step": 34090 + }, + { + "epoch": 1.7064931815338422, + "grad_norm": 0.048320550471544266, + "learning_rate": 2.3953216807844155e-05, + "loss": 0.6703, + "step": 34100 + }, + { + "epoch": 1.70699361941699, + "grad_norm": 0.005620166659355164, + "learning_rate": 2.394394913903357e-05, + "loss": 0.1799, + "step": 34110 + }, + { + "epoch": 1.7074940573001376, + "grad_norm": 0.03323045000433922, + "learning_rate": 2.393468147022298e-05, + "loss": 0.4599, + "step": 34120 + }, + { + "epoch": 1.7079944951832853, + "grad_norm": 0.055803194642066956, + "learning_rate": 2.3925413801412395e-05, + "loss": 0.0595, + "step": 34130 + }, + { + "epoch": 1.708494933066433, + "grad_norm": 129.10511779785156, + "learning_rate": 2.3916146132601807e-05, + "loss": 0.3188, + "step": 34140 + }, + { + "epoch": 1.7089953709495809, + "grad_norm": 16.17743492126465, + "learning_rate": 2.390687846379122e-05, + "loss": 0.3162, + "step": 34150 + }, + { + "epoch": 1.7094958088327288, + "grad_norm": 43.59287643432617, + "learning_rate": 2.3897610794980632e-05, + "loss": 0.1505, + "step": 34160 + }, + { + "epoch": 1.7099962467158765, + "grad_norm": 13.427780151367188, + "learning_rate": 2.3888343126170044e-05, + "loss": 0.2895, + "step": 34170 + }, + { + "epoch": 1.7104966845990242, + "grad_norm": 9.974664688110352, + "learning_rate": 2.3879075457359456e-05, + "loss": 0.752, + "step": 34180 + }, + { + "epoch": 1.7109971224821718, + "grad_norm": 0.0025686845183372498, + "learning_rate": 2.3869807788548872e-05, + "loss": 0.1996, + "step": 34190 + }, + { + "epoch": 1.7114975603653195, + "grad_norm": 151.74639892578125, + "learning_rate": 2.386054011973828e-05, + "loss": 0.2011, + "step": 34200 + }, + { + "epoch": 1.7119979982484674, + "grad_norm": 0.018629856407642365, + "learning_rate": 2.3851272450927696e-05, + "loss": 0.2781, + "step": 34210 + }, + { + "epoch": 1.7124984361316151, + "grad_norm": 0.08719821274280548, + "learning_rate": 2.384200478211711e-05, + "loss": 0.4961, + "step": 34220 + }, + { + "epoch": 1.712998874014763, + "grad_norm": 0.019297128543257713, + "learning_rate": 2.3832737113306517e-05, + "loss": 0.3412, + "step": 34230 + }, + { + "epoch": 1.7134993118979107, + "grad_norm": 0.032778527587652206, + "learning_rate": 2.3823469444495933e-05, + "loss": 0.5163, + "step": 34240 + }, + { + "epoch": 1.7139997497810584, + "grad_norm": 40.495086669921875, + "learning_rate": 2.3814201775685345e-05, + "loss": 0.6225, + "step": 34250 + }, + { + "epoch": 1.714500187664206, + "grad_norm": 149.48243713378906, + "learning_rate": 2.3804934106874758e-05, + "loss": 0.4971, + "step": 34260 + }, + { + "epoch": 1.7150006255473538, + "grad_norm": 0.13945314288139343, + "learning_rate": 2.379566643806417e-05, + "loss": 0.0611, + "step": 34270 + }, + { + "epoch": 1.7155010634305017, + "grad_norm": 0.0259200818836689, + "learning_rate": 2.3786398769253585e-05, + "loss": 0.2177, + "step": 34280 + }, + { + "epoch": 1.7160015013136496, + "grad_norm": 55.090511322021484, + "learning_rate": 2.3777131100442994e-05, + "loss": 0.5191, + "step": 34290 + }, + { + "epoch": 1.7165019391967973, + "grad_norm": 38.47734451293945, + "learning_rate": 2.376786343163241e-05, + "loss": 0.5216, + "step": 34300 + }, + { + "epoch": 1.717002377079945, + "grad_norm": 0.10491771250963211, + "learning_rate": 2.375859576282182e-05, + "loss": 0.1507, + "step": 34310 + }, + { + "epoch": 1.7175028149630926, + "grad_norm": 97.93199920654297, + "learning_rate": 2.3749328094011234e-05, + "loss": 0.3395, + "step": 34320 + }, + { + "epoch": 1.7180032528462403, + "grad_norm": 59.53874969482422, + "learning_rate": 2.3740060425200647e-05, + "loss": 0.5343, + "step": 34330 + }, + { + "epoch": 1.7185036907293882, + "grad_norm": 0.3891690671443939, + "learning_rate": 2.373079275639006e-05, + "loss": 0.163, + "step": 34340 + }, + { + "epoch": 1.719004128612536, + "grad_norm": 0.009386297315359116, + "learning_rate": 2.372152508757947e-05, + "loss": 0.1056, + "step": 34350 + }, + { + "epoch": 1.7195045664956838, + "grad_norm": 0.3279591202735901, + "learning_rate": 2.3712257418768883e-05, + "loss": 0.5924, + "step": 34360 + }, + { + "epoch": 1.7200050043788315, + "grad_norm": 0.03682173416018486, + "learning_rate": 2.3702989749958296e-05, + "loss": 0.1276, + "step": 34370 + }, + { + "epoch": 1.7205054422619792, + "grad_norm": 127.0140609741211, + "learning_rate": 2.3693722081147708e-05, + "loss": 0.2417, + "step": 34380 + }, + { + "epoch": 1.721005880145127, + "grad_norm": 0.09992892295122147, + "learning_rate": 2.3684454412337124e-05, + "loss": 0.009, + "step": 34390 + }, + { + "epoch": 1.7215063180282746, + "grad_norm": 0.0074037788435816765, + "learning_rate": 2.3675186743526532e-05, + "loss": 0.2274, + "step": 34400 + }, + { + "epoch": 1.7220067559114225, + "grad_norm": 0.025041548535227776, + "learning_rate": 2.3665919074715948e-05, + "loss": 0.4236, + "step": 34410 + }, + { + "epoch": 1.7225071937945704, + "grad_norm": 134.14892578125, + "learning_rate": 2.3656651405905357e-05, + "loss": 0.5152, + "step": 34420 + }, + { + "epoch": 1.723007631677718, + "grad_norm": 11.445585250854492, + "learning_rate": 2.3647383737094773e-05, + "loss": 0.2552, + "step": 34430 + }, + { + "epoch": 1.7235080695608658, + "grad_norm": 0.09936672449111938, + "learning_rate": 2.3638116068284185e-05, + "loss": 0.5487, + "step": 34440 + }, + { + "epoch": 1.7240085074440135, + "grad_norm": 0.10615508258342743, + "learning_rate": 2.3628848399473597e-05, + "loss": 0.3063, + "step": 34450 + }, + { + "epoch": 1.7245089453271611, + "grad_norm": 16.438173294067383, + "learning_rate": 2.361958073066301e-05, + "loss": 0.319, + "step": 34460 + }, + { + "epoch": 1.725009383210309, + "grad_norm": 0.017015326768159866, + "learning_rate": 2.3610313061852425e-05, + "loss": 0.0159, + "step": 34470 + }, + { + "epoch": 1.7255098210934567, + "grad_norm": 12.088833808898926, + "learning_rate": 2.3601045393041834e-05, + "loss": 0.382, + "step": 34480 + }, + { + "epoch": 1.7260102589766046, + "grad_norm": 91.29315185546875, + "learning_rate": 2.359177772423125e-05, + "loss": 0.2563, + "step": 34490 + }, + { + "epoch": 1.7265106968597523, + "grad_norm": 39.824249267578125, + "learning_rate": 2.358251005542066e-05, + "loss": 0.3187, + "step": 34500 + }, + { + "epoch": 1.7270111347429, + "grad_norm": 54.97214126586914, + "learning_rate": 2.3573242386610074e-05, + "loss": 1.1336, + "step": 34510 + }, + { + "epoch": 1.7275115726260477, + "grad_norm": 86.72456359863281, + "learning_rate": 2.3563974717799486e-05, + "loss": 0.5437, + "step": 34520 + }, + { + "epoch": 1.7280120105091954, + "grad_norm": 2.3768935203552246, + "learning_rate": 2.35547070489889e-05, + "loss": 0.3195, + "step": 34530 + }, + { + "epoch": 1.7285124483923433, + "grad_norm": 53.235408782958984, + "learning_rate": 2.354543938017831e-05, + "loss": 0.5143, + "step": 34540 + }, + { + "epoch": 1.7290128862754912, + "grad_norm": 0.2628766894340515, + "learning_rate": 2.3536171711367723e-05, + "loss": 0.5664, + "step": 34550 + }, + { + "epoch": 1.729513324158639, + "grad_norm": 0.04732188582420349, + "learning_rate": 2.3526904042557135e-05, + "loss": 0.4232, + "step": 34560 + }, + { + "epoch": 1.7300137620417866, + "grad_norm": 36.19430160522461, + "learning_rate": 2.3517636373746547e-05, + "loss": 0.348, + "step": 34570 + }, + { + "epoch": 1.7305141999249343, + "grad_norm": 0.08700574189424515, + "learning_rate": 2.3508368704935963e-05, + "loss": 0.5978, + "step": 34580 + }, + { + "epoch": 1.731014637808082, + "grad_norm": 1.250132441520691, + "learning_rate": 2.3499101036125372e-05, + "loss": 0.1331, + "step": 34590 + }, + { + "epoch": 1.7315150756912299, + "grad_norm": 48.630897521972656, + "learning_rate": 2.3489833367314788e-05, + "loss": 0.5206, + "step": 34600 + }, + { + "epoch": 1.7320155135743776, + "grad_norm": 13.385117530822754, + "learning_rate": 2.34805656985042e-05, + "loss": 0.4101, + "step": 34610 + }, + { + "epoch": 1.7325159514575255, + "grad_norm": 0.053781766444444656, + "learning_rate": 2.3471298029693612e-05, + "loss": 0.4795, + "step": 34620 + }, + { + "epoch": 1.7330163893406731, + "grad_norm": 0.03821467608213425, + "learning_rate": 2.3462030360883024e-05, + "loss": 0.15, + "step": 34630 + }, + { + "epoch": 1.7335168272238208, + "grad_norm": 55.76581954956055, + "learning_rate": 2.345276269207244e-05, + "loss": 0.4028, + "step": 34640 + }, + { + "epoch": 1.7340172651069685, + "grad_norm": 0.23378488421440125, + "learning_rate": 2.344349502326185e-05, + "loss": 0.3793, + "step": 34650 + }, + { + "epoch": 1.7345177029901162, + "grad_norm": 0.2912178933620453, + "learning_rate": 2.3434227354451264e-05, + "loss": 0.3342, + "step": 34660 + }, + { + "epoch": 1.7350181408732641, + "grad_norm": 40.49432373046875, + "learning_rate": 2.3424959685640673e-05, + "loss": 0.4372, + "step": 34670 + }, + { + "epoch": 1.735518578756412, + "grad_norm": 0.3354119658470154, + "learning_rate": 2.341569201683009e-05, + "loss": 0.339, + "step": 34680 + }, + { + "epoch": 1.7360190166395597, + "grad_norm": 9.066165924072266, + "learning_rate": 2.34064243480195e-05, + "loss": 0.2463, + "step": 34690 + }, + { + "epoch": 1.7365194545227074, + "grad_norm": 0.17164452373981476, + "learning_rate": 2.3397156679208913e-05, + "loss": 0.332, + "step": 34700 + }, + { + "epoch": 1.737019892405855, + "grad_norm": 8.888398170471191, + "learning_rate": 2.3387889010398326e-05, + "loss": 0.1909, + "step": 34710 + }, + { + "epoch": 1.7375203302890028, + "grad_norm": 0.007723883260041475, + "learning_rate": 2.3378621341587738e-05, + "loss": 0.0171, + "step": 34720 + }, + { + "epoch": 1.7380207681721507, + "grad_norm": 14.052718162536621, + "learning_rate": 2.336935367277715e-05, + "loss": 0.2596, + "step": 34730 + }, + { + "epoch": 1.7385212060552984, + "grad_norm": 0.005513621494174004, + "learning_rate": 2.3360086003966562e-05, + "loss": 0.269, + "step": 34740 + }, + { + "epoch": 1.7390216439384463, + "grad_norm": 21.64529800415039, + "learning_rate": 2.3350818335155978e-05, + "loss": 0.9204, + "step": 34750 + }, + { + "epoch": 1.739522081821594, + "grad_norm": 0.058676473796367645, + "learning_rate": 2.3341550666345387e-05, + "loss": 0.1021, + "step": 34760 + }, + { + "epoch": 1.7400225197047416, + "grad_norm": 0.008240751922130585, + "learning_rate": 2.3332282997534802e-05, + "loss": 0.4438, + "step": 34770 + }, + { + "epoch": 1.7405229575878893, + "grad_norm": 4.327194690704346, + "learning_rate": 2.332301532872421e-05, + "loss": 0.1879, + "step": 34780 + }, + { + "epoch": 1.741023395471037, + "grad_norm": 0.004631643183529377, + "learning_rate": 2.3313747659913627e-05, + "loss": 0.2745, + "step": 34790 + }, + { + "epoch": 1.741523833354185, + "grad_norm": 0.04413685202598572, + "learning_rate": 2.330447999110304e-05, + "loss": 0.252, + "step": 34800 + }, + { + "epoch": 1.7420242712373326, + "grad_norm": 53.306678771972656, + "learning_rate": 2.329521232229245e-05, + "loss": 0.1485, + "step": 34810 + }, + { + "epoch": 1.7425247091204805, + "grad_norm": 0.03317927569150925, + "learning_rate": 2.3285944653481864e-05, + "loss": 0.0741, + "step": 34820 + }, + { + "epoch": 1.7430251470036282, + "grad_norm": 0.4268331825733185, + "learning_rate": 2.327667698467128e-05, + "loss": 0.23, + "step": 34830 + }, + { + "epoch": 1.743525584886776, + "grad_norm": 55.663841247558594, + "learning_rate": 2.3267409315860688e-05, + "loss": 0.5309, + "step": 34840 + }, + { + "epoch": 1.7440260227699236, + "grad_norm": 0.08571074157953262, + "learning_rate": 2.3258141647050104e-05, + "loss": 0.2613, + "step": 34850 + }, + { + "epoch": 1.7445264606530715, + "grad_norm": 0.001362709328532219, + "learning_rate": 2.3248873978239516e-05, + "loss": 0.0488, + "step": 34860 + }, + { + "epoch": 1.7450268985362192, + "grad_norm": 44.22722244262695, + "learning_rate": 2.323960630942893e-05, + "loss": 1.0911, + "step": 34870 + }, + { + "epoch": 1.745527336419367, + "grad_norm": 44.83208084106445, + "learning_rate": 2.323033864061834e-05, + "loss": 0.4632, + "step": 34880 + }, + { + "epoch": 1.7460277743025148, + "grad_norm": 24.044113159179688, + "learning_rate": 2.3221070971807753e-05, + "loss": 0.4784, + "step": 34890 + }, + { + "epoch": 1.7465282121856625, + "grad_norm": 22.513559341430664, + "learning_rate": 2.3211803302997165e-05, + "loss": 0.5457, + "step": 34900 + }, + { + "epoch": 1.7470286500688101, + "grad_norm": 30.285934448242188, + "learning_rate": 2.3202535634186577e-05, + "loss": 0.2643, + "step": 34910 + }, + { + "epoch": 1.7475290879519578, + "grad_norm": 0.5265253782272339, + "learning_rate": 2.319326796537599e-05, + "loss": 0.0015, + "step": 34920 + }, + { + "epoch": 1.7480295258351057, + "grad_norm": 65.88263702392578, + "learning_rate": 2.3184000296565402e-05, + "loss": 0.3438, + "step": 34930 + }, + { + "epoch": 1.7485299637182534, + "grad_norm": 5.153350830078125, + "learning_rate": 2.3174732627754817e-05, + "loss": 0.5004, + "step": 34940 + }, + { + "epoch": 1.7490304016014013, + "grad_norm": 2.93371319770813, + "learning_rate": 2.3165464958944226e-05, + "loss": 0.129, + "step": 34950 + }, + { + "epoch": 1.749530839484549, + "grad_norm": 0.011018266901373863, + "learning_rate": 2.3156197290133642e-05, + "loss": 0.3329, + "step": 34960 + }, + { + "epoch": 1.7500312773676967, + "grad_norm": 9.460107803344727, + "learning_rate": 2.3146929621323054e-05, + "loss": 0.6292, + "step": 34970 + }, + { + "epoch": 1.7505317152508444, + "grad_norm": 77.95601654052734, + "learning_rate": 2.3137661952512466e-05, + "loss": 0.4047, + "step": 34980 + }, + { + "epoch": 1.7510321531339923, + "grad_norm": 97.17278289794922, + "learning_rate": 2.312839428370188e-05, + "loss": 0.2286, + "step": 34990 + }, + { + "epoch": 1.75153259101714, + "grad_norm": 1.957470178604126, + "learning_rate": 2.3119126614891294e-05, + "loss": 0.1604, + "step": 35000 + }, + { + "epoch": 1.752033028900288, + "grad_norm": 42.893306732177734, + "learning_rate": 2.3109858946080703e-05, + "loss": 0.3204, + "step": 35010 + }, + { + "epoch": 1.7525334667834356, + "grad_norm": 8.978238105773926, + "learning_rate": 2.310059127727012e-05, + "loss": 0.3558, + "step": 35020 + }, + { + "epoch": 1.7530339046665833, + "grad_norm": 4.1642937660217285, + "learning_rate": 2.3091323608459528e-05, + "loss": 0.3792, + "step": 35030 + }, + { + "epoch": 1.753534342549731, + "grad_norm": 51.77234649658203, + "learning_rate": 2.3082055939648943e-05, + "loss": 0.5923, + "step": 35040 + }, + { + "epoch": 1.7540347804328786, + "grad_norm": 0.035350602120161057, + "learning_rate": 2.3072788270838356e-05, + "loss": 0.2369, + "step": 35050 + }, + { + "epoch": 1.7545352183160265, + "grad_norm": 16.438264846801758, + "learning_rate": 2.3063520602027768e-05, + "loss": 0.6796, + "step": 35060 + }, + { + "epoch": 1.7550356561991742, + "grad_norm": 15.922223091125488, + "learning_rate": 2.305425293321718e-05, + "loss": 0.1741, + "step": 35070 + }, + { + "epoch": 1.7555360940823221, + "grad_norm": 6.620477676391602, + "learning_rate": 2.3044985264406592e-05, + "loss": 0.0672, + "step": 35080 + }, + { + "epoch": 1.7560365319654698, + "grad_norm": 9.114898681640625, + "learning_rate": 2.3035717595596004e-05, + "loss": 0.3757, + "step": 35090 + }, + { + "epoch": 1.7565369698486175, + "grad_norm": 65.69085693359375, + "learning_rate": 2.3026449926785417e-05, + "loss": 0.3266, + "step": 35100 + }, + { + "epoch": 1.7570374077317652, + "grad_norm": 0.24977457523345947, + "learning_rate": 2.3017182257974832e-05, + "loss": 0.4493, + "step": 35110 + }, + { + "epoch": 1.7575378456149129, + "grad_norm": 25.75190544128418, + "learning_rate": 2.300791458916424e-05, + "loss": 0.4318, + "step": 35120 + }, + { + "epoch": 1.7580382834980608, + "grad_norm": 8.361320495605469, + "learning_rate": 2.2998646920353657e-05, + "loss": 0.1382, + "step": 35130 + }, + { + "epoch": 1.7585387213812087, + "grad_norm": 0.0015874310629442334, + "learning_rate": 2.2989379251543066e-05, + "loss": 0.2437, + "step": 35140 + }, + { + "epoch": 1.7590391592643564, + "grad_norm": 10.759787559509277, + "learning_rate": 2.298011158273248e-05, + "loss": 0.3078, + "step": 35150 + }, + { + "epoch": 1.759539597147504, + "grad_norm": 77.232666015625, + "learning_rate": 2.2970843913921894e-05, + "loss": 0.2849, + "step": 35160 + }, + { + "epoch": 1.7600400350306518, + "grad_norm": 51.74163055419922, + "learning_rate": 2.2961576245111306e-05, + "loss": 0.4742, + "step": 35170 + }, + { + "epoch": 1.7605404729137994, + "grad_norm": 78.97505187988281, + "learning_rate": 2.2952308576300718e-05, + "loss": 0.3558, + "step": 35180 + }, + { + "epoch": 1.7610409107969474, + "grad_norm": 66.3499526977539, + "learning_rate": 2.294304090749013e-05, + "loss": 0.3498, + "step": 35190 + }, + { + "epoch": 1.761541348680095, + "grad_norm": 91.49759674072266, + "learning_rate": 2.2933773238679543e-05, + "loss": 0.3098, + "step": 35200 + }, + { + "epoch": 1.762041786563243, + "grad_norm": 73.0001220703125, + "learning_rate": 2.2924505569868955e-05, + "loss": 0.4142, + "step": 35210 + }, + { + "epoch": 1.7625422244463906, + "grad_norm": 0.2705388367176056, + "learning_rate": 2.291523790105837e-05, + "loss": 0.2568, + "step": 35220 + }, + { + "epoch": 1.7630426623295383, + "grad_norm": 48.13319396972656, + "learning_rate": 2.290597023224778e-05, + "loss": 0.2573, + "step": 35230 + }, + { + "epoch": 1.763543100212686, + "grad_norm": 0.02865959331393242, + "learning_rate": 2.2896702563437195e-05, + "loss": 0.1511, + "step": 35240 + }, + { + "epoch": 1.7640435380958337, + "grad_norm": 87.63880920410156, + "learning_rate": 2.2887434894626604e-05, + "loss": 0.8985, + "step": 35250 + }, + { + "epoch": 1.7645439759789816, + "grad_norm": 0.9751858711242676, + "learning_rate": 2.287816722581602e-05, + "loss": 0.1872, + "step": 35260 + }, + { + "epoch": 1.7650444138621295, + "grad_norm": 14.628933906555176, + "learning_rate": 2.2868899557005432e-05, + "loss": 0.4084, + "step": 35270 + }, + { + "epoch": 1.7655448517452772, + "grad_norm": 146.8189697265625, + "learning_rate": 2.2859631888194844e-05, + "loss": 0.2726, + "step": 35280 + }, + { + "epoch": 1.7660452896284249, + "grad_norm": 0.10121644288301468, + "learning_rate": 2.2850364219384256e-05, + "loss": 0.0959, + "step": 35290 + }, + { + "epoch": 1.7665457275115726, + "grad_norm": 146.45860290527344, + "learning_rate": 2.2841096550573672e-05, + "loss": 0.6061, + "step": 35300 + }, + { + "epoch": 1.7670461653947203, + "grad_norm": 48.16352081298828, + "learning_rate": 2.283182888176308e-05, + "loss": 0.5184, + "step": 35310 + }, + { + "epoch": 1.7675466032778682, + "grad_norm": 69.53352355957031, + "learning_rate": 2.2822561212952496e-05, + "loss": 0.5884, + "step": 35320 + }, + { + "epoch": 1.7680470411610159, + "grad_norm": 0.17212708294391632, + "learning_rate": 2.281329354414191e-05, + "loss": 0.1386, + "step": 35330 + }, + { + "epoch": 1.7685474790441638, + "grad_norm": 44.4642333984375, + "learning_rate": 2.280402587533132e-05, + "loss": 0.3602, + "step": 35340 + }, + { + "epoch": 1.7690479169273114, + "grad_norm": 137.22406005859375, + "learning_rate": 2.2794758206520733e-05, + "loss": 0.4518, + "step": 35350 + }, + { + "epoch": 1.7695483548104591, + "grad_norm": 1.3393574953079224, + "learning_rate": 2.2785490537710145e-05, + "loss": 0.6505, + "step": 35360 + }, + { + "epoch": 1.7700487926936068, + "grad_norm": 0.062323592603206635, + "learning_rate": 2.2776222868899558e-05, + "loss": 0.1021, + "step": 35370 + }, + { + "epoch": 1.7705492305767545, + "grad_norm": 26.0643253326416, + "learning_rate": 2.276695520008897e-05, + "loss": 0.176, + "step": 35380 + }, + { + "epoch": 1.7710496684599024, + "grad_norm": 0.0666591078042984, + "learning_rate": 2.2757687531278382e-05, + "loss": 0.1918, + "step": 35390 + }, + { + "epoch": 1.7715501063430503, + "grad_norm": 101.77700805664062, + "learning_rate": 2.2748419862467794e-05, + "loss": 0.2844, + "step": 35400 + }, + { + "epoch": 1.772050544226198, + "grad_norm": 46.617801666259766, + "learning_rate": 2.273915219365721e-05, + "loss": 0.4497, + "step": 35410 + }, + { + "epoch": 1.7725509821093457, + "grad_norm": 0.8690857887268066, + "learning_rate": 2.272988452484662e-05, + "loss": 0.219, + "step": 35420 + }, + { + "epoch": 1.7730514199924934, + "grad_norm": 0.8222888708114624, + "learning_rate": 2.2720616856036034e-05, + "loss": 0.4633, + "step": 35430 + }, + { + "epoch": 1.773551857875641, + "grad_norm": 0.015981977805495262, + "learning_rate": 2.2711349187225447e-05, + "loss": 0.2939, + "step": 35440 + }, + { + "epoch": 1.774052295758789, + "grad_norm": 2.205671787261963, + "learning_rate": 2.270208151841486e-05, + "loss": 0.2454, + "step": 35450 + }, + { + "epoch": 1.7745527336419367, + "grad_norm": 45.32583999633789, + "learning_rate": 2.269281384960427e-05, + "loss": 0.2148, + "step": 35460 + }, + { + "epoch": 1.7750531715250846, + "grad_norm": 10.348443031311035, + "learning_rate": 2.2683546180793687e-05, + "loss": 0.0407, + "step": 35470 + }, + { + "epoch": 1.7755536094082323, + "grad_norm": 29.195524215698242, + "learning_rate": 2.2674278511983096e-05, + "loss": 0.5292, + "step": 35480 + }, + { + "epoch": 1.77605404729138, + "grad_norm": 0.3256097733974457, + "learning_rate": 2.266501084317251e-05, + "loss": 0.0198, + "step": 35490 + }, + { + "epoch": 1.7765544851745276, + "grad_norm": 0.02663148194551468, + "learning_rate": 2.265574317436192e-05, + "loss": 0.0649, + "step": 35500 + }, + { + "epoch": 1.7770549230576753, + "grad_norm": 0.18534736335277557, + "learning_rate": 2.2646475505551336e-05, + "loss": 0.318, + "step": 35510 + }, + { + "epoch": 1.7775553609408232, + "grad_norm": 14.132757186889648, + "learning_rate": 2.2637207836740748e-05, + "loss": 0.0571, + "step": 35520 + }, + { + "epoch": 1.7780557988239711, + "grad_norm": 0.42602047324180603, + "learning_rate": 2.262794016793016e-05, + "loss": 0.3856, + "step": 35530 + }, + { + "epoch": 1.7785562367071188, + "grad_norm": 23.865280151367188, + "learning_rate": 2.2618672499119573e-05, + "loss": 0.2717, + "step": 35540 + }, + { + "epoch": 1.7790566745902665, + "grad_norm": 0.013623963110148907, + "learning_rate": 2.2609404830308985e-05, + "loss": 0.1902, + "step": 35550 + }, + { + "epoch": 1.7795571124734142, + "grad_norm": 0.5346294641494751, + "learning_rate": 2.2600137161498397e-05, + "loss": 0.1271, + "step": 35560 + }, + { + "epoch": 1.7800575503565619, + "grad_norm": 53.593505859375, + "learning_rate": 2.259086949268781e-05, + "loss": 0.2292, + "step": 35570 + }, + { + "epoch": 1.7805579882397098, + "grad_norm": 0.007268376648426056, + "learning_rate": 2.2581601823877225e-05, + "loss": 0.4088, + "step": 35580 + }, + { + "epoch": 1.7810584261228575, + "grad_norm": 0.0003192540898453444, + "learning_rate": 2.2572334155066634e-05, + "loss": 0.2056, + "step": 35590 + }, + { + "epoch": 1.7815588640060054, + "grad_norm": 0.0016145213739946485, + "learning_rate": 2.256306648625605e-05, + "loss": 0.3497, + "step": 35600 + }, + { + "epoch": 1.782059301889153, + "grad_norm": 0.006183893885463476, + "learning_rate": 2.2553798817445458e-05, + "loss": 0.0909, + "step": 35610 + }, + { + "epoch": 1.7825597397723008, + "grad_norm": 0.12463249266147614, + "learning_rate": 2.2544531148634874e-05, + "loss": 0.5518, + "step": 35620 + }, + { + "epoch": 1.7830601776554484, + "grad_norm": 0.005212000571191311, + "learning_rate": 2.2535263479824286e-05, + "loss": 0.3234, + "step": 35630 + }, + { + "epoch": 1.7835606155385961, + "grad_norm": 0.07716991007328033, + "learning_rate": 2.25259958110137e-05, + "loss": 0.4496, + "step": 35640 + }, + { + "epoch": 1.784061053421744, + "grad_norm": 0.018094679340720177, + "learning_rate": 2.251672814220311e-05, + "loss": 0.196, + "step": 35650 + }, + { + "epoch": 1.784561491304892, + "grad_norm": 0.04424557834863663, + "learning_rate": 2.2507460473392526e-05, + "loss": 0.3632, + "step": 35660 + }, + { + "epoch": 1.7850619291880396, + "grad_norm": 0.5917732119560242, + "learning_rate": 2.2498192804581935e-05, + "loss": 0.4087, + "step": 35670 + }, + { + "epoch": 1.7855623670711873, + "grad_norm": 91.62684631347656, + "learning_rate": 2.248892513577135e-05, + "loss": 0.2956, + "step": 35680 + }, + { + "epoch": 1.786062804954335, + "grad_norm": 0.22325468063354492, + "learning_rate": 2.2479657466960763e-05, + "loss": 0.4562, + "step": 35690 + }, + { + "epoch": 1.7865632428374827, + "grad_norm": 0.027008648961782455, + "learning_rate": 2.2470389798150175e-05, + "loss": 0.3513, + "step": 35700 + }, + { + "epoch": 1.7870636807206306, + "grad_norm": 6.799362659454346, + "learning_rate": 2.2461122129339587e-05, + "loss": 0.46, + "step": 35710 + }, + { + "epoch": 1.7875641186037783, + "grad_norm": 39.22074508666992, + "learning_rate": 2.2451854460529e-05, + "loss": 0.4447, + "step": 35720 + }, + { + "epoch": 1.7880645564869262, + "grad_norm": 0.01904110051691532, + "learning_rate": 2.2442586791718412e-05, + "loss": 0.1791, + "step": 35730 + }, + { + "epoch": 1.7885649943700739, + "grad_norm": 57.86640167236328, + "learning_rate": 2.2433319122907824e-05, + "loss": 0.251, + "step": 35740 + }, + { + "epoch": 1.7890654322532216, + "grad_norm": 29.7217960357666, + "learning_rate": 2.2424051454097236e-05, + "loss": 0.3778, + "step": 35750 + }, + { + "epoch": 1.7895658701363693, + "grad_norm": 0.2912646532058716, + "learning_rate": 2.241478378528665e-05, + "loss": 0.2339, + "step": 35760 + }, + { + "epoch": 1.790066308019517, + "grad_norm": 67.52717590332031, + "learning_rate": 2.2405516116476064e-05, + "loss": 0.2299, + "step": 35770 + }, + { + "epoch": 1.7905667459026648, + "grad_norm": 0.18077419698238373, + "learning_rate": 2.2396248447665473e-05, + "loss": 0.1095, + "step": 35780 + }, + { + "epoch": 1.7910671837858125, + "grad_norm": 89.4803466796875, + "learning_rate": 2.238698077885489e-05, + "loss": 0.3971, + "step": 35790 + }, + { + "epoch": 1.7915676216689604, + "grad_norm": 59.1014518737793, + "learning_rate": 2.23777131100443e-05, + "loss": 0.4951, + "step": 35800 + }, + { + "epoch": 1.7920680595521081, + "grad_norm": 0.10562801361083984, + "learning_rate": 2.2368445441233713e-05, + "loss": 0.0629, + "step": 35810 + }, + { + "epoch": 1.7925684974352558, + "grad_norm": 0.3266245424747467, + "learning_rate": 2.2359177772423126e-05, + "loss": 0.0028, + "step": 35820 + }, + { + "epoch": 1.7930689353184035, + "grad_norm": 104.06383514404297, + "learning_rate": 2.234991010361254e-05, + "loss": 0.5054, + "step": 35830 + }, + { + "epoch": 1.7935693732015514, + "grad_norm": 79.35305786132812, + "learning_rate": 2.234064243480195e-05, + "loss": 0.2097, + "step": 35840 + }, + { + "epoch": 1.794069811084699, + "grad_norm": 93.92510223388672, + "learning_rate": 2.2331374765991366e-05, + "loss": 0.1954, + "step": 35850 + }, + { + "epoch": 1.794570248967847, + "grad_norm": 54.24437713623047, + "learning_rate": 2.2322107097180775e-05, + "loss": 0.4262, + "step": 35860 + }, + { + "epoch": 1.7950706868509947, + "grad_norm": 33.92999267578125, + "learning_rate": 2.231283942837019e-05, + "loss": 0.4556, + "step": 35870 + }, + { + "epoch": 1.7955711247341424, + "grad_norm": 24.68108367919922, + "learning_rate": 2.2303571759559602e-05, + "loss": 0.9264, + "step": 35880 + }, + { + "epoch": 1.79607156261729, + "grad_norm": 290.3980407714844, + "learning_rate": 2.2294304090749015e-05, + "loss": 0.7487, + "step": 35890 + }, + { + "epoch": 1.7965720005004377, + "grad_norm": 3.7732017040252686, + "learning_rate": 2.2285036421938427e-05, + "loss": 0.4082, + "step": 35900 + }, + { + "epoch": 1.7970724383835857, + "grad_norm": 128.47523498535156, + "learning_rate": 2.227576875312784e-05, + "loss": 0.2765, + "step": 35910 + }, + { + "epoch": 1.7975728762667333, + "grad_norm": 0.01737886667251587, + "learning_rate": 2.226650108431725e-05, + "loss": 0.124, + "step": 35920 + }, + { + "epoch": 1.7980733141498813, + "grad_norm": 0.01920807920396328, + "learning_rate": 2.2257233415506664e-05, + "loss": 0.2262, + "step": 35930 + }, + { + "epoch": 1.798573752033029, + "grad_norm": 121.92278289794922, + "learning_rate": 2.224796574669608e-05, + "loss": 0.3006, + "step": 35940 + }, + { + "epoch": 1.7990741899161766, + "grad_norm": 32.67146301269531, + "learning_rate": 2.2238698077885488e-05, + "loss": 0.5751, + "step": 35950 + }, + { + "epoch": 1.7995746277993243, + "grad_norm": 0.026594683527946472, + "learning_rate": 2.2229430409074904e-05, + "loss": 0.5084, + "step": 35960 + }, + { + "epoch": 1.800075065682472, + "grad_norm": 36.44878005981445, + "learning_rate": 2.2220162740264313e-05, + "loss": 0.8656, + "step": 35970 + }, + { + "epoch": 1.80057550356562, + "grad_norm": 58.922332763671875, + "learning_rate": 2.2210895071453728e-05, + "loss": 0.5674, + "step": 35980 + }, + { + "epoch": 1.8010759414487678, + "grad_norm": 1.8238654136657715, + "learning_rate": 2.220162740264314e-05, + "loss": 0.1266, + "step": 35990 + }, + { + "epoch": 1.8015763793319155, + "grad_norm": 49.93287658691406, + "learning_rate": 2.2192359733832553e-05, + "loss": 0.4535, + "step": 36000 + }, + { + "epoch": 1.8020768172150632, + "grad_norm": 0.445411741733551, + "learning_rate": 2.2183092065021965e-05, + "loss": 0.3265, + "step": 36010 + }, + { + "epoch": 1.8025772550982109, + "grad_norm": 4.6121506690979, + "learning_rate": 2.217382439621138e-05, + "loss": 0.2945, + "step": 36020 + }, + { + "epoch": 1.8030776929813586, + "grad_norm": 0.041555508971214294, + "learning_rate": 2.216455672740079e-05, + "loss": 0.7228, + "step": 36030 + }, + { + "epoch": 1.8035781308645065, + "grad_norm": 0.29446741938591003, + "learning_rate": 2.2155289058590202e-05, + "loss": 0.4424, + "step": 36040 + }, + { + "epoch": 1.8040785687476542, + "grad_norm": 89.879150390625, + "learning_rate": 2.2146021389779617e-05, + "loss": 0.1198, + "step": 36050 + }, + { + "epoch": 1.804579006630802, + "grad_norm": 48.31071472167969, + "learning_rate": 2.2136753720969026e-05, + "loss": 0.3127, + "step": 36060 + }, + { + "epoch": 1.8050794445139497, + "grad_norm": 0.011751716956496239, + "learning_rate": 2.2127486052158442e-05, + "loss": 0.4631, + "step": 36070 + }, + { + "epoch": 1.8055798823970974, + "grad_norm": 0.025256190448999405, + "learning_rate": 2.2118218383347854e-05, + "loss": 0.0724, + "step": 36080 + }, + { + "epoch": 1.8060803202802451, + "grad_norm": 0.015547518618404865, + "learning_rate": 2.2108950714537266e-05, + "loss": 0.2879, + "step": 36090 + }, + { + "epoch": 1.8065807581633928, + "grad_norm": 0.021095311269164085, + "learning_rate": 2.209968304572668e-05, + "loss": 0.5309, + "step": 36100 + }, + { + "epoch": 1.8070811960465407, + "grad_norm": 66.38638305664062, + "learning_rate": 2.209041537691609e-05, + "loss": 0.4198, + "step": 36110 + }, + { + "epoch": 1.8075816339296886, + "grad_norm": 19.6713809967041, + "learning_rate": 2.2081147708105503e-05, + "loss": 0.3057, + "step": 36120 + }, + { + "epoch": 1.8080820718128363, + "grad_norm": 0.4112485647201538, + "learning_rate": 2.207188003929492e-05, + "loss": 0.3186, + "step": 36130 + }, + { + "epoch": 1.808582509695984, + "grad_norm": 23.354997634887695, + "learning_rate": 2.2062612370484328e-05, + "loss": 0.1277, + "step": 36140 + }, + { + "epoch": 1.8090829475791317, + "grad_norm": 17.893112182617188, + "learning_rate": 2.2053344701673743e-05, + "loss": 0.6231, + "step": 36150 + }, + { + "epoch": 1.8095833854622794, + "grad_norm": 14.008752822875977, + "learning_rate": 2.2044077032863155e-05, + "loss": 0.5203, + "step": 36160 + }, + { + "epoch": 1.8100838233454273, + "grad_norm": 15.785538673400879, + "learning_rate": 2.2034809364052568e-05, + "loss": 0.6668, + "step": 36170 + }, + { + "epoch": 1.810584261228575, + "grad_norm": 0.16556011140346527, + "learning_rate": 2.202554169524198e-05, + "loss": 0.1358, + "step": 36180 + }, + { + "epoch": 1.8110846991117229, + "grad_norm": 0.011373938992619514, + "learning_rate": 2.2016274026431392e-05, + "loss": 0.2081, + "step": 36190 + }, + { + "epoch": 1.8115851369948706, + "grad_norm": 43.94941329956055, + "learning_rate": 2.2007006357620804e-05, + "loss": 0.2912, + "step": 36200 + }, + { + "epoch": 1.8120855748780182, + "grad_norm": 0.021698419004678726, + "learning_rate": 2.1997738688810217e-05, + "loss": 0.0613, + "step": 36210 + }, + { + "epoch": 1.812586012761166, + "grad_norm": 0.014907123520970345, + "learning_rate": 2.198847101999963e-05, + "loss": 0.2757, + "step": 36220 + }, + { + "epoch": 1.8130864506443136, + "grad_norm": 0.1800891011953354, + "learning_rate": 2.197920335118904e-05, + "loss": 0.2191, + "step": 36230 + }, + { + "epoch": 1.8135868885274615, + "grad_norm": 103.91694641113281, + "learning_rate": 2.1969935682378457e-05, + "loss": 0.1373, + "step": 36240 + }, + { + "epoch": 1.8140873264106094, + "grad_norm": 0.011495735496282578, + "learning_rate": 2.1960668013567866e-05, + "loss": 0.1227, + "step": 36250 + }, + { + "epoch": 1.8145877642937571, + "grad_norm": 10.910279273986816, + "learning_rate": 2.195140034475728e-05, + "loss": 0.4169, + "step": 36260 + }, + { + "epoch": 1.8150882021769048, + "grad_norm": 49.11826705932617, + "learning_rate": 2.1942132675946694e-05, + "loss": 0.1857, + "step": 36270 + }, + { + "epoch": 1.8155886400600525, + "grad_norm": 0.006984788458794355, + "learning_rate": 2.1932865007136106e-05, + "loss": 0.1363, + "step": 36280 + }, + { + "epoch": 1.8160890779432002, + "grad_norm": 0.06903201341629028, + "learning_rate": 2.1923597338325518e-05, + "loss": 0.166, + "step": 36290 + }, + { + "epoch": 1.816589515826348, + "grad_norm": 0.006401206366717815, + "learning_rate": 2.1914329669514934e-05, + "loss": 0.4362, + "step": 36300 + }, + { + "epoch": 1.8170899537094958, + "grad_norm": 0.39338043332099915, + "learning_rate": 2.1905062000704343e-05, + "loss": 0.1739, + "step": 36310 + }, + { + "epoch": 1.8175903915926437, + "grad_norm": 44.731101989746094, + "learning_rate": 2.1895794331893758e-05, + "loss": 0.5627, + "step": 36320 + }, + { + "epoch": 1.8180908294757914, + "grad_norm": 16.555343627929688, + "learning_rate": 2.1886526663083167e-05, + "loss": 0.4306, + "step": 36330 + }, + { + "epoch": 1.818591267358939, + "grad_norm": 28.271711349487305, + "learning_rate": 2.1877258994272583e-05, + "loss": 0.3503, + "step": 36340 + }, + { + "epoch": 1.8190917052420867, + "grad_norm": 0.08725877851247787, + "learning_rate": 2.1867991325461995e-05, + "loss": 0.2528, + "step": 36350 + }, + { + "epoch": 1.8195921431252344, + "grad_norm": 0.03677305951714516, + "learning_rate": 2.1858723656651407e-05, + "loss": 0.4761, + "step": 36360 + }, + { + "epoch": 1.8200925810083823, + "grad_norm": 47.35251235961914, + "learning_rate": 2.184945598784082e-05, + "loss": 0.4907, + "step": 36370 + }, + { + "epoch": 1.8205930188915302, + "grad_norm": 0.010965649969875813, + "learning_rate": 2.184018831903023e-05, + "loss": 0.2835, + "step": 36380 + }, + { + "epoch": 1.821093456774678, + "grad_norm": 41.9395637512207, + "learning_rate": 2.1830920650219644e-05, + "loss": 0.3103, + "step": 36390 + }, + { + "epoch": 1.8215938946578256, + "grad_norm": 0.2139580398797989, + "learning_rate": 2.1821652981409056e-05, + "loss": 0.344, + "step": 36400 + }, + { + "epoch": 1.8220943325409733, + "grad_norm": 0.12096220999956131, + "learning_rate": 2.1812385312598472e-05, + "loss": 0.1523, + "step": 36410 + }, + { + "epoch": 1.822594770424121, + "grad_norm": 0.7223994731903076, + "learning_rate": 2.180311764378788e-05, + "loss": 0.3344, + "step": 36420 + }, + { + "epoch": 1.823095208307269, + "grad_norm": 0.7662032246589661, + "learning_rate": 2.1793849974977296e-05, + "loss": 0.0098, + "step": 36430 + }, + { + "epoch": 1.8235956461904166, + "grad_norm": 0.004754338879138231, + "learning_rate": 2.178458230616671e-05, + "loss": 0.5085, + "step": 36440 + }, + { + "epoch": 1.8240960840735645, + "grad_norm": 50.44242858886719, + "learning_rate": 2.177531463735612e-05, + "loss": 0.4799, + "step": 36450 + }, + { + "epoch": 1.8245965219567122, + "grad_norm": 24.41664695739746, + "learning_rate": 2.1766046968545533e-05, + "loss": 0.267, + "step": 36460 + }, + { + "epoch": 1.8250969598398599, + "grad_norm": 0.06125974655151367, + "learning_rate": 2.1756779299734945e-05, + "loss": 0.0258, + "step": 36470 + }, + { + "epoch": 1.8255973977230076, + "grad_norm": 0.03127528727054596, + "learning_rate": 2.1747511630924358e-05, + "loss": 0.2823, + "step": 36480 + }, + { + "epoch": 1.8260978356061552, + "grad_norm": 0.16965565085411072, + "learning_rate": 2.1738243962113773e-05, + "loss": 0.3937, + "step": 36490 + }, + { + "epoch": 1.8265982734893031, + "grad_norm": 30.25301742553711, + "learning_rate": 2.1728976293303182e-05, + "loss": 0.4947, + "step": 36500 + }, + { + "epoch": 1.827098711372451, + "grad_norm": 0.018166866153478622, + "learning_rate": 2.1719708624492598e-05, + "loss": 0.2, + "step": 36510 + }, + { + "epoch": 1.8275991492555987, + "grad_norm": 0.3068659007549286, + "learning_rate": 2.171044095568201e-05, + "loss": 0.6746, + "step": 36520 + }, + { + "epoch": 1.8280995871387464, + "grad_norm": 56.1885986328125, + "learning_rate": 2.1701173286871422e-05, + "loss": 0.1142, + "step": 36530 + }, + { + "epoch": 1.8286000250218941, + "grad_norm": 43.63901138305664, + "learning_rate": 2.1691905618060834e-05, + "loss": 0.5185, + "step": 36540 + }, + { + "epoch": 1.8291004629050418, + "grad_norm": 0.9942418932914734, + "learning_rate": 2.1682637949250247e-05, + "loss": 0.4331, + "step": 36550 + }, + { + "epoch": 1.8296009007881897, + "grad_norm": 0.13348005712032318, + "learning_rate": 2.167337028043966e-05, + "loss": 0.333, + "step": 36560 + }, + { + "epoch": 1.8301013386713374, + "grad_norm": 0.031243445351719856, + "learning_rate": 2.166410261162907e-05, + "loss": 0.232, + "step": 36570 + }, + { + "epoch": 1.8306017765544853, + "grad_norm": 0.019990183413028717, + "learning_rate": 2.1654834942818483e-05, + "loss": 0.2694, + "step": 36580 + }, + { + "epoch": 1.831102214437633, + "grad_norm": 0.06679895520210266, + "learning_rate": 2.1645567274007896e-05, + "loss": 0.2122, + "step": 36590 + }, + { + "epoch": 1.8316026523207807, + "grad_norm": 10.911748886108398, + "learning_rate": 2.163629960519731e-05, + "loss": 0.4537, + "step": 36600 + }, + { + "epoch": 1.8321030902039284, + "grad_norm": 2.6960599422454834, + "learning_rate": 2.162703193638672e-05, + "loss": 0.1934, + "step": 36610 + }, + { + "epoch": 1.832603528087076, + "grad_norm": 0.17072725296020508, + "learning_rate": 2.1617764267576136e-05, + "loss": 0.6355, + "step": 36620 + }, + { + "epoch": 1.833103965970224, + "grad_norm": 0.6716978549957275, + "learning_rate": 2.1608496598765548e-05, + "loss": 0.2889, + "step": 36630 + }, + { + "epoch": 1.8336044038533716, + "grad_norm": 40.203914642333984, + "learning_rate": 2.159922892995496e-05, + "loss": 0.4625, + "step": 36640 + }, + { + "epoch": 1.8341048417365196, + "grad_norm": 0.01892678812146187, + "learning_rate": 2.1589961261144372e-05, + "loss": 0.3972, + "step": 36650 + }, + { + "epoch": 1.8346052796196672, + "grad_norm": 0.024436384439468384, + "learning_rate": 2.1580693592333788e-05, + "loss": 0.6017, + "step": 36660 + }, + { + "epoch": 1.835105717502815, + "grad_norm": 0.6054261922836304, + "learning_rate": 2.1571425923523197e-05, + "loss": 0.433, + "step": 36670 + }, + { + "epoch": 1.8356061553859626, + "grad_norm": 48.666690826416016, + "learning_rate": 2.1562158254712613e-05, + "loss": 0.4051, + "step": 36680 + }, + { + "epoch": 1.8361065932691105, + "grad_norm": 18.539897918701172, + "learning_rate": 2.155289058590202e-05, + "loss": 0.6238, + "step": 36690 + }, + { + "epoch": 1.8366070311522582, + "grad_norm": 0.19870835542678833, + "learning_rate": 2.1543622917091437e-05, + "loss": 0.7046, + "step": 36700 + }, + { + "epoch": 1.8371074690354061, + "grad_norm": 0.03675160929560661, + "learning_rate": 2.153435524828085e-05, + "loss": 0.2244, + "step": 36710 + }, + { + "epoch": 1.8376079069185538, + "grad_norm": 0.1298580765724182, + "learning_rate": 2.152508757947026e-05, + "loss": 0.2389, + "step": 36720 + }, + { + "epoch": 1.8381083448017015, + "grad_norm": 6.427433013916016, + "learning_rate": 2.1515819910659674e-05, + "loss": 0.1844, + "step": 36730 + }, + { + "epoch": 1.8386087826848492, + "grad_norm": 66.54005432128906, + "learning_rate": 2.1506552241849086e-05, + "loss": 0.6074, + "step": 36740 + }, + { + "epoch": 1.8391092205679969, + "grad_norm": 104.05426788330078, + "learning_rate": 2.14972845730385e-05, + "loss": 0.3865, + "step": 36750 + }, + { + "epoch": 1.8396096584511448, + "grad_norm": 0.09004459530115128, + "learning_rate": 2.148801690422791e-05, + "loss": 0.2494, + "step": 36760 + }, + { + "epoch": 1.8401100963342925, + "grad_norm": 2.426283597946167, + "learning_rate": 2.1478749235417326e-05, + "loss": 0.1349, + "step": 36770 + }, + { + "epoch": 1.8406105342174404, + "grad_norm": 15.57552719116211, + "learning_rate": 2.1469481566606735e-05, + "loss": 0.2136, + "step": 36780 + }, + { + "epoch": 1.841110972100588, + "grad_norm": 25.314517974853516, + "learning_rate": 2.146021389779615e-05, + "loss": 0.5049, + "step": 36790 + }, + { + "epoch": 1.8416114099837357, + "grad_norm": 4.397627830505371, + "learning_rate": 2.1450946228985563e-05, + "loss": 0.34, + "step": 36800 + }, + { + "epoch": 1.8421118478668834, + "grad_norm": 0.018764210864901543, + "learning_rate": 2.1441678560174975e-05, + "loss": 0.2889, + "step": 36810 + }, + { + "epoch": 1.8426122857500313, + "grad_norm": 0.26990124583244324, + "learning_rate": 2.1432410891364387e-05, + "loss": 0.2875, + "step": 36820 + }, + { + "epoch": 1.843112723633179, + "grad_norm": 0.12749311327934265, + "learning_rate": 2.14231432225538e-05, + "loss": 0.3077, + "step": 36830 + }, + { + "epoch": 1.843613161516327, + "grad_norm": 294.7828369140625, + "learning_rate": 2.1413875553743212e-05, + "loss": 0.2004, + "step": 36840 + }, + { + "epoch": 1.8441135993994746, + "grad_norm": 0.20693479478359222, + "learning_rate": 2.1404607884932628e-05, + "loss": 0.1047, + "step": 36850 + }, + { + "epoch": 1.8446140372826223, + "grad_norm": 5.780798435211182, + "learning_rate": 2.1395340216122036e-05, + "loss": 0.491, + "step": 36860 + }, + { + "epoch": 1.84511447516577, + "grad_norm": 0.049266599118709564, + "learning_rate": 2.1386072547311452e-05, + "loss": 0.089, + "step": 36870 + }, + { + "epoch": 1.8456149130489177, + "grad_norm": 0.5983725786209106, + "learning_rate": 2.1376804878500864e-05, + "loss": 0.6951, + "step": 36880 + }, + { + "epoch": 1.8461153509320656, + "grad_norm": 45.70246887207031, + "learning_rate": 2.1367537209690273e-05, + "loss": 0.7372, + "step": 36890 + }, + { + "epoch": 1.8466157888152133, + "grad_norm": 48.05126190185547, + "learning_rate": 2.135826954087969e-05, + "loss": 0.4989, + "step": 36900 + }, + { + "epoch": 1.8471162266983612, + "grad_norm": 14.787296295166016, + "learning_rate": 2.13490018720691e-05, + "loss": 0.276, + "step": 36910 + }, + { + "epoch": 1.8476166645815089, + "grad_norm": 54.241355895996094, + "learning_rate": 2.1339734203258513e-05, + "loss": 1.1148, + "step": 36920 + }, + { + "epoch": 1.8481171024646565, + "grad_norm": 0.17654715478420258, + "learning_rate": 2.1330466534447926e-05, + "loss": 0.1036, + "step": 36930 + }, + { + "epoch": 1.8486175403478042, + "grad_norm": 0.10604141652584076, + "learning_rate": 2.1321198865637338e-05, + "loss": 0.2584, + "step": 36940 + }, + { + "epoch": 1.849117978230952, + "grad_norm": 0.39898812770843506, + "learning_rate": 2.131193119682675e-05, + "loss": 0.367, + "step": 36950 + }, + { + "epoch": 1.8496184161140998, + "grad_norm": 0.043824415653944016, + "learning_rate": 2.1302663528016166e-05, + "loss": 0.2868, + "step": 36960 + }, + { + "epoch": 1.8501188539972477, + "grad_norm": 30.667285919189453, + "learning_rate": 2.1293395859205575e-05, + "loss": 0.2317, + "step": 36970 + }, + { + "epoch": 1.8506192918803954, + "grad_norm": 1.0943187475204468, + "learning_rate": 2.128412819039499e-05, + "loss": 0.1494, + "step": 36980 + }, + { + "epoch": 1.851119729763543, + "grad_norm": 20.45451545715332, + "learning_rate": 2.1274860521584402e-05, + "loss": 0.286, + "step": 36990 + }, + { + "epoch": 1.8516201676466908, + "grad_norm": 17.805410385131836, + "learning_rate": 2.1265592852773815e-05, + "loss": 0.3842, + "step": 37000 + }, + { + "epoch": 1.8521206055298385, + "grad_norm": 10.504836082458496, + "learning_rate": 2.1256325183963227e-05, + "loss": 0.3375, + "step": 37010 + }, + { + "epoch": 1.8526210434129864, + "grad_norm": 30.354015350341797, + "learning_rate": 2.124705751515264e-05, + "loss": 0.1099, + "step": 37020 + }, + { + "epoch": 1.853121481296134, + "grad_norm": 0.03985307365655899, + "learning_rate": 2.123778984634205e-05, + "loss": 0.1489, + "step": 37030 + }, + { + "epoch": 1.853621919179282, + "grad_norm": 0.007620376069098711, + "learning_rate": 2.1228522177531464e-05, + "loss": 0.2192, + "step": 37040 + }, + { + "epoch": 1.8541223570624297, + "grad_norm": 10.965964317321777, + "learning_rate": 2.1219254508720876e-05, + "loss": 0.3448, + "step": 37050 + }, + { + "epoch": 1.8546227949455774, + "grad_norm": 122.93539428710938, + "learning_rate": 2.1209986839910288e-05, + "loss": 0.4393, + "step": 37060 + }, + { + "epoch": 1.855123232828725, + "grad_norm": 0.1783834546804428, + "learning_rate": 2.1200719171099704e-05, + "loss": 0.2367, + "step": 37070 + }, + { + "epoch": 1.8556236707118727, + "grad_norm": 0.011519348248839378, + "learning_rate": 2.1191451502289113e-05, + "loss": 0.5087, + "step": 37080 + }, + { + "epoch": 1.8561241085950206, + "grad_norm": 111.02137756347656, + "learning_rate": 2.1182183833478528e-05, + "loss": 0.4737, + "step": 37090 + }, + { + "epoch": 1.8566245464781685, + "grad_norm": 0.09814456105232239, + "learning_rate": 2.117291616466794e-05, + "loss": 0.4133, + "step": 37100 + }, + { + "epoch": 1.8571249843613162, + "grad_norm": 0.25971850752830505, + "learning_rate": 2.1163648495857353e-05, + "loss": 0.4735, + "step": 37110 + }, + { + "epoch": 1.857625422244464, + "grad_norm": 0.1498715728521347, + "learning_rate": 2.1154380827046765e-05, + "loss": 0.0468, + "step": 37120 + }, + { + "epoch": 1.8581258601276116, + "grad_norm": 0.007408617064356804, + "learning_rate": 2.114511315823618e-05, + "loss": 0.371, + "step": 37130 + }, + { + "epoch": 1.8586262980107593, + "grad_norm": 99.17008209228516, + "learning_rate": 2.113584548942559e-05, + "loss": 0.6082, + "step": 37140 + }, + { + "epoch": 1.8591267358939072, + "grad_norm": 0.08124830573797226, + "learning_rate": 2.1126577820615005e-05, + "loss": 0.0293, + "step": 37150 + }, + { + "epoch": 1.8596271737770549, + "grad_norm": 2.598205327987671, + "learning_rate": 2.1117310151804414e-05, + "loss": 0.2392, + "step": 37160 + }, + { + "epoch": 1.8601276116602028, + "grad_norm": 0.00946068949997425, + "learning_rate": 2.110804248299383e-05, + "loss": 0.513, + "step": 37170 + }, + { + "epoch": 1.8606280495433505, + "grad_norm": 1.163857102394104, + "learning_rate": 2.1098774814183242e-05, + "loss": 0.0067, + "step": 37180 + }, + { + "epoch": 1.8611284874264982, + "grad_norm": 0.09203219413757324, + "learning_rate": 2.1089507145372654e-05, + "loss": 0.0533, + "step": 37190 + }, + { + "epoch": 1.8616289253096459, + "grad_norm": 44.388736724853516, + "learning_rate": 2.1080239476562066e-05, + "loss": 0.2571, + "step": 37200 + }, + { + "epoch": 1.8621293631927935, + "grad_norm": 0.06413394212722778, + "learning_rate": 2.107097180775148e-05, + "loss": 0.6785, + "step": 37210 + }, + { + "epoch": 1.8626298010759414, + "grad_norm": 0.3760691285133362, + "learning_rate": 2.106170413894089e-05, + "loss": 0.5935, + "step": 37220 + }, + { + "epoch": 1.8631302389590894, + "grad_norm": 0.007528044283390045, + "learning_rate": 2.1052436470130303e-05, + "loss": 0.117, + "step": 37230 + }, + { + "epoch": 1.863630676842237, + "grad_norm": 0.49975934624671936, + "learning_rate": 2.104316880131972e-05, + "loss": 0.637, + "step": 37240 + }, + { + "epoch": 1.8641311147253847, + "grad_norm": 0.1021186038851738, + "learning_rate": 2.1033901132509128e-05, + "loss": 0.3729, + "step": 37250 + }, + { + "epoch": 1.8646315526085324, + "grad_norm": 0.37049543857574463, + "learning_rate": 2.1024633463698543e-05, + "loss": 0.2795, + "step": 37260 + }, + { + "epoch": 1.86513199049168, + "grad_norm": 21.7613468170166, + "learning_rate": 2.1015365794887955e-05, + "loss": 0.5789, + "step": 37270 + }, + { + "epoch": 1.865632428374828, + "grad_norm": 2.3847222328186035, + "learning_rate": 2.1006098126077368e-05, + "loss": 0.187, + "step": 37280 + }, + { + "epoch": 1.8661328662579757, + "grad_norm": 0.2687782049179077, + "learning_rate": 2.099683045726678e-05, + "loss": 0.3523, + "step": 37290 + }, + { + "epoch": 1.8666333041411236, + "grad_norm": 0.10773049294948578, + "learning_rate": 2.0987562788456192e-05, + "loss": 0.1646, + "step": 37300 + }, + { + "epoch": 1.8671337420242713, + "grad_norm": 39.03302764892578, + "learning_rate": 2.0978295119645604e-05, + "loss": 0.5583, + "step": 37310 + }, + { + "epoch": 1.867634179907419, + "grad_norm": 17.857677459716797, + "learning_rate": 2.096902745083502e-05, + "loss": 0.422, + "step": 37320 + }, + { + "epoch": 1.8681346177905667, + "grad_norm": 33.34381866455078, + "learning_rate": 2.095975978202443e-05, + "loss": 0.5623, + "step": 37330 + }, + { + "epoch": 1.8686350556737144, + "grad_norm": 0.01266766618937254, + "learning_rate": 2.0950492113213845e-05, + "loss": 0.4701, + "step": 37340 + }, + { + "epoch": 1.8691354935568623, + "grad_norm": 34.289024353027344, + "learning_rate": 2.0941224444403257e-05, + "loss": 0.2499, + "step": 37350 + }, + { + "epoch": 1.8696359314400102, + "grad_norm": 0.04435868561267853, + "learning_rate": 2.093195677559267e-05, + "loss": 0.1273, + "step": 37360 + }, + { + "epoch": 1.8701363693231579, + "grad_norm": 0.5127589106559753, + "learning_rate": 2.092268910678208e-05, + "loss": 0.2888, + "step": 37370 + }, + { + "epoch": 1.8706368072063055, + "grad_norm": 0.8715260028839111, + "learning_rate": 2.0913421437971494e-05, + "loss": 0.1107, + "step": 37380 + }, + { + "epoch": 1.8711372450894532, + "grad_norm": 0.42933231592178345, + "learning_rate": 2.0904153769160906e-05, + "loss": 0.1777, + "step": 37390 + }, + { + "epoch": 1.871637682972601, + "grad_norm": 0.07614526152610779, + "learning_rate": 2.0894886100350318e-05, + "loss": 0.2791, + "step": 37400 + }, + { + "epoch": 1.8721381208557488, + "grad_norm": 0.0578426867723465, + "learning_rate": 2.088561843153973e-05, + "loss": 0.5403, + "step": 37410 + }, + { + "epoch": 1.8726385587388965, + "grad_norm": 63.44682693481445, + "learning_rate": 2.0876350762729143e-05, + "loss": 0.0905, + "step": 37420 + }, + { + "epoch": 1.8731389966220444, + "grad_norm": 1.2828755378723145, + "learning_rate": 2.0867083093918558e-05, + "loss": 0.1387, + "step": 37430 + }, + { + "epoch": 1.873639434505192, + "grad_norm": 32.66716384887695, + "learning_rate": 2.0857815425107967e-05, + "loss": 0.4102, + "step": 37440 + }, + { + "epoch": 1.8741398723883398, + "grad_norm": 79.36856842041016, + "learning_rate": 2.0848547756297383e-05, + "loss": 0.1003, + "step": 37450 + }, + { + "epoch": 1.8746403102714875, + "grad_norm": 76.4679946899414, + "learning_rate": 2.0839280087486795e-05, + "loss": 0.3565, + "step": 37460 + }, + { + "epoch": 1.8751407481546352, + "grad_norm": 0.12971659004688263, + "learning_rate": 2.0830012418676207e-05, + "loss": 0.0101, + "step": 37470 + }, + { + "epoch": 1.875641186037783, + "grad_norm": 0.6774689555168152, + "learning_rate": 2.082074474986562e-05, + "loss": 0.7674, + "step": 37480 + }, + { + "epoch": 1.876141623920931, + "grad_norm": 62.471923828125, + "learning_rate": 2.0811477081055035e-05, + "loss": 0.1523, + "step": 37490 + }, + { + "epoch": 1.8766420618040787, + "grad_norm": 0.15525639057159424, + "learning_rate": 2.0802209412244444e-05, + "loss": 0.0737, + "step": 37500 + }, + { + "epoch": 1.8771424996872264, + "grad_norm": 12.790932655334473, + "learning_rate": 2.079294174343386e-05, + "loss": 0.1853, + "step": 37510 + }, + { + "epoch": 1.877642937570374, + "grad_norm": 197.9672088623047, + "learning_rate": 2.078367407462327e-05, + "loss": 0.4632, + "step": 37520 + }, + { + "epoch": 1.8781433754535217, + "grad_norm": 0.11810034513473511, + "learning_rate": 2.0774406405812684e-05, + "loss": 0.3397, + "step": 37530 + }, + { + "epoch": 1.8786438133366696, + "grad_norm": 0.04318379983305931, + "learning_rate": 2.0765138737002096e-05, + "loss": 0.3368, + "step": 37540 + }, + { + "epoch": 1.8791442512198173, + "grad_norm": 105.40218353271484, + "learning_rate": 2.075587106819151e-05, + "loss": 0.1955, + "step": 37550 + }, + { + "epoch": 1.8796446891029652, + "grad_norm": 0.3675534129142761, + "learning_rate": 2.074660339938092e-05, + "loss": 0.2366, + "step": 37560 + }, + { + "epoch": 1.880145126986113, + "grad_norm": 57.306514739990234, + "learning_rate": 2.0737335730570333e-05, + "loss": 0.4528, + "step": 37570 + }, + { + "epoch": 1.8806455648692606, + "grad_norm": 0.012992950156331062, + "learning_rate": 2.0728068061759745e-05, + "loss": 0.2345, + "step": 37580 + }, + { + "epoch": 1.8811460027524083, + "grad_norm": 24.39497947692871, + "learning_rate": 2.0718800392949157e-05, + "loss": 0.3141, + "step": 37590 + }, + { + "epoch": 1.881646440635556, + "grad_norm": 220.0406494140625, + "learning_rate": 2.0709532724138573e-05, + "loss": 0.1814, + "step": 37600 + }, + { + "epoch": 1.8821468785187039, + "grad_norm": 0.07513434439897537, + "learning_rate": 2.0700265055327982e-05, + "loss": 0.7966, + "step": 37610 + }, + { + "epoch": 1.8826473164018516, + "grad_norm": 32.635162353515625, + "learning_rate": 2.0690997386517398e-05, + "loss": 0.339, + "step": 37620 + }, + { + "epoch": 1.8831477542849995, + "grad_norm": 7.458622455596924, + "learning_rate": 2.068172971770681e-05, + "loss": 0.3251, + "step": 37630 + }, + { + "epoch": 1.8836481921681472, + "grad_norm": 17.56751823425293, + "learning_rate": 2.0672462048896222e-05, + "loss": 0.1423, + "step": 37640 + }, + { + "epoch": 1.8841486300512948, + "grad_norm": 0.06591993570327759, + "learning_rate": 2.0663194380085634e-05, + "loss": 0.275, + "step": 37650 + }, + { + "epoch": 1.8846490679344425, + "grad_norm": 0.016062619164586067, + "learning_rate": 2.0653926711275047e-05, + "loss": 0.3795, + "step": 37660 + }, + { + "epoch": 1.8851495058175904, + "grad_norm": 0.0234084390103817, + "learning_rate": 2.064465904246446e-05, + "loss": 0.1408, + "step": 37670 + }, + { + "epoch": 1.8856499437007381, + "grad_norm": 0.05028308182954788, + "learning_rate": 2.0635391373653874e-05, + "loss": 0.7677, + "step": 37680 + }, + { + "epoch": 1.886150381583886, + "grad_norm": 84.11734008789062, + "learning_rate": 2.0626123704843283e-05, + "loss": 0.1825, + "step": 37690 + }, + { + "epoch": 1.8866508194670337, + "grad_norm": 0.005193856544792652, + "learning_rate": 2.06168560360327e-05, + "loss": 0.1099, + "step": 37700 + }, + { + "epoch": 1.8871512573501814, + "grad_norm": 91.68599700927734, + "learning_rate": 2.060758836722211e-05, + "loss": 0.1745, + "step": 37710 + }, + { + "epoch": 1.887651695233329, + "grad_norm": 0.06734602898359299, + "learning_rate": 2.0598320698411523e-05, + "loss": 0.2849, + "step": 37720 + }, + { + "epoch": 1.8881521331164768, + "grad_norm": 12.281442642211914, + "learning_rate": 2.0589053029600936e-05, + "loss": 0.2117, + "step": 37730 + }, + { + "epoch": 1.8886525709996247, + "grad_norm": 0.04940931126475334, + "learning_rate": 2.0579785360790348e-05, + "loss": 0.091, + "step": 37740 + }, + { + "epoch": 1.8891530088827724, + "grad_norm": 89.17977142333984, + "learning_rate": 2.057051769197976e-05, + "loss": 0.2922, + "step": 37750 + }, + { + "epoch": 1.8896534467659203, + "grad_norm": 0.016948577016592026, + "learning_rate": 2.0561250023169172e-05, + "loss": 0.3081, + "step": 37760 + }, + { + "epoch": 1.890153884649068, + "grad_norm": 32.07435989379883, + "learning_rate": 2.0551982354358585e-05, + "loss": 0.5511, + "step": 37770 + }, + { + "epoch": 1.8906543225322157, + "grad_norm": 66.67936706542969, + "learning_rate": 2.0542714685547997e-05, + "loss": 0.6165, + "step": 37780 + }, + { + "epoch": 1.8911547604153633, + "grad_norm": 156.69967651367188, + "learning_rate": 2.0533447016737413e-05, + "loss": 0.3891, + "step": 37790 + }, + { + "epoch": 1.891655198298511, + "grad_norm": 0.004467062186449766, + "learning_rate": 2.052417934792682e-05, + "loss": 0.2375, + "step": 37800 + }, + { + "epoch": 1.892155636181659, + "grad_norm": 0.38320139050483704, + "learning_rate": 2.0514911679116237e-05, + "loss": 0.6106, + "step": 37810 + }, + { + "epoch": 1.8926560740648068, + "grad_norm": 0.01657191850244999, + "learning_rate": 2.050564401030565e-05, + "loss": 0.3916, + "step": 37820 + }, + { + "epoch": 1.8931565119479545, + "grad_norm": 0.04788557440042496, + "learning_rate": 2.049637634149506e-05, + "loss": 0.0369, + "step": 37830 + }, + { + "epoch": 1.8936569498311022, + "grad_norm": 29.826955795288086, + "learning_rate": 2.0487108672684474e-05, + "loss": 0.4455, + "step": 37840 + }, + { + "epoch": 1.89415738771425, + "grad_norm": 5.009705543518066, + "learning_rate": 2.0477841003873886e-05, + "loss": 0.1145, + "step": 37850 + }, + { + "epoch": 1.8946578255973976, + "grad_norm": 0.2770562767982483, + "learning_rate": 2.0468573335063298e-05, + "loss": 0.1203, + "step": 37860 + }, + { + "epoch": 1.8951582634805455, + "grad_norm": 0.37686169147491455, + "learning_rate": 2.045930566625271e-05, + "loss": 0.2155, + "step": 37870 + }, + { + "epoch": 1.8956587013636932, + "grad_norm": 0.04225960001349449, + "learning_rate": 2.0450037997442123e-05, + "loss": 0.107, + "step": 37880 + }, + { + "epoch": 1.896159139246841, + "grad_norm": 0.9390984773635864, + "learning_rate": 2.0440770328631535e-05, + "loss": 0.284, + "step": 37890 + }, + { + "epoch": 1.8966595771299888, + "grad_norm": 30.980758666992188, + "learning_rate": 2.043150265982095e-05, + "loss": 0.679, + "step": 37900 + }, + { + "epoch": 1.8971600150131365, + "grad_norm": 0.48102912306785583, + "learning_rate": 2.042223499101036e-05, + "loss": 0.1449, + "step": 37910 + }, + { + "epoch": 1.8976604528962842, + "grad_norm": 1.111053705215454, + "learning_rate": 2.0412967322199775e-05, + "loss": 0.4942, + "step": 37920 + }, + { + "epoch": 1.8981608907794318, + "grad_norm": 0.3980999290943146, + "learning_rate": 2.0403699653389187e-05, + "loss": 0.189, + "step": 37930 + }, + { + "epoch": 1.8986613286625798, + "grad_norm": 11.73343276977539, + "learning_rate": 2.03944319845786e-05, + "loss": 0.4344, + "step": 37940 + }, + { + "epoch": 1.8991617665457277, + "grad_norm": 4.556815147399902, + "learning_rate": 2.0385164315768012e-05, + "loss": 0.4141, + "step": 37950 + }, + { + "epoch": 1.8996622044288753, + "grad_norm": 2.363901138305664, + "learning_rate": 2.0375896646957428e-05, + "loss": 0.1359, + "step": 37960 + }, + { + "epoch": 1.900162642312023, + "grad_norm": 0.009133118204772472, + "learning_rate": 2.0366628978146836e-05, + "loss": 0.2781, + "step": 37970 + }, + { + "epoch": 1.9006630801951707, + "grad_norm": 28.279172897338867, + "learning_rate": 2.0357361309336252e-05, + "loss": 0.3108, + "step": 37980 + }, + { + "epoch": 1.9011635180783184, + "grad_norm": 3.753739356994629, + "learning_rate": 2.0348093640525664e-05, + "loss": 0.1257, + "step": 37990 + }, + { + "epoch": 1.9016639559614663, + "grad_norm": 0.0606098398566246, + "learning_rate": 2.0338825971715077e-05, + "loss": 0.1737, + "step": 38000 + }, + { + "epoch": 1.902164393844614, + "grad_norm": 27.81208038330078, + "learning_rate": 2.032955830290449e-05, + "loss": 0.3096, + "step": 38010 + }, + { + "epoch": 1.902664831727762, + "grad_norm": 0.06868289411067963, + "learning_rate": 2.03202906340939e-05, + "loss": 0.0144, + "step": 38020 + }, + { + "epoch": 1.9031652696109096, + "grad_norm": 0.0072260224260389805, + "learning_rate": 2.0311022965283313e-05, + "loss": 0.411, + "step": 38030 + }, + { + "epoch": 1.9036657074940573, + "grad_norm": 2.708292245864868, + "learning_rate": 2.0301755296472725e-05, + "loss": 0.2609, + "step": 38040 + }, + { + "epoch": 1.904166145377205, + "grad_norm": 0.6285228729248047, + "learning_rate": 2.0292487627662138e-05, + "loss": 0.47, + "step": 38050 + }, + { + "epoch": 1.9046665832603527, + "grad_norm": 201.7383270263672, + "learning_rate": 2.028321995885155e-05, + "loss": 0.5306, + "step": 38060 + }, + { + "epoch": 1.9051670211435006, + "grad_norm": 0.02005821466445923, + "learning_rate": 2.0273952290040966e-05, + "loss": 0.2443, + "step": 38070 + }, + { + "epoch": 1.9056674590266485, + "grad_norm": 0.05378018319606781, + "learning_rate": 2.0264684621230374e-05, + "loss": 0.355, + "step": 38080 + }, + { + "epoch": 1.9061678969097962, + "grad_norm": 10.669534683227539, + "learning_rate": 2.025541695241979e-05, + "loss": 0.1992, + "step": 38090 + }, + { + "epoch": 1.9066683347929438, + "grad_norm": 0.14752936363220215, + "learning_rate": 2.0246149283609202e-05, + "loss": 0.7565, + "step": 38100 + }, + { + "epoch": 1.9071687726760915, + "grad_norm": 53.13887405395508, + "learning_rate": 2.0236881614798615e-05, + "loss": 0.5236, + "step": 38110 + }, + { + "epoch": 1.9076692105592392, + "grad_norm": 3.57295560836792, + "learning_rate": 2.0227613945988027e-05, + "loss": 0.2616, + "step": 38120 + }, + { + "epoch": 1.9081696484423871, + "grad_norm": 8.036920547485352, + "learning_rate": 2.021834627717744e-05, + "loss": 0.1247, + "step": 38130 + }, + { + "epoch": 1.9086700863255348, + "grad_norm": 1.4633432626724243, + "learning_rate": 2.020907860836685e-05, + "loss": 0.2525, + "step": 38140 + }, + { + "epoch": 1.9091705242086827, + "grad_norm": 70.59278869628906, + "learning_rate": 2.0199810939556267e-05, + "loss": 0.4322, + "step": 38150 + }, + { + "epoch": 1.9096709620918304, + "grad_norm": 0.18885450065135956, + "learning_rate": 2.0190543270745676e-05, + "loss": 0.1501, + "step": 38160 + }, + { + "epoch": 1.910171399974978, + "grad_norm": 3.3975775241851807, + "learning_rate": 2.018127560193509e-05, + "loss": 0.3504, + "step": 38170 + }, + { + "epoch": 1.9106718378581258, + "grad_norm": 0.10192982852458954, + "learning_rate": 2.0172007933124504e-05, + "loss": 0.5066, + "step": 38180 + }, + { + "epoch": 1.9111722757412735, + "grad_norm": 5.558974266052246, + "learning_rate": 2.0162740264313916e-05, + "loss": 0.1933, + "step": 38190 + }, + { + "epoch": 1.9116727136244214, + "grad_norm": 0.27394166588783264, + "learning_rate": 2.0153472595503328e-05, + "loss": 0.0265, + "step": 38200 + }, + { + "epoch": 1.9121731515075693, + "grad_norm": 0.11809088289737701, + "learning_rate": 2.014420492669274e-05, + "loss": 0.2456, + "step": 38210 + }, + { + "epoch": 1.912673589390717, + "grad_norm": 0.00890293437987566, + "learning_rate": 2.0134937257882153e-05, + "loss": 0.5438, + "step": 38220 + }, + { + "epoch": 1.9131740272738647, + "grad_norm": 3.0596389770507812, + "learning_rate": 2.0125669589071565e-05, + "loss": 0.2418, + "step": 38230 + }, + { + "epoch": 1.9136744651570123, + "grad_norm": 3.4457693099975586, + "learning_rate": 2.0116401920260977e-05, + "loss": 0.4921, + "step": 38240 + }, + { + "epoch": 1.91417490304016, + "grad_norm": 0.6279579401016235, + "learning_rate": 2.010713425145039e-05, + "loss": 0.3513, + "step": 38250 + }, + { + "epoch": 1.914675340923308, + "grad_norm": 45.83053207397461, + "learning_rate": 2.0097866582639805e-05, + "loss": 0.5769, + "step": 38260 + }, + { + "epoch": 1.9151757788064556, + "grad_norm": 6.307680130004883, + "learning_rate": 2.0088598913829214e-05, + "loss": 0.8254, + "step": 38270 + }, + { + "epoch": 1.9156762166896035, + "grad_norm": 0.06910669058561325, + "learning_rate": 2.007933124501863e-05, + "loss": 0.1113, + "step": 38280 + }, + { + "epoch": 1.9161766545727512, + "grad_norm": 0.026004932820796967, + "learning_rate": 2.0070063576208042e-05, + "loss": 0.3365, + "step": 38290 + }, + { + "epoch": 1.916677092455899, + "grad_norm": 0.01913640834391117, + "learning_rate": 2.0060795907397454e-05, + "loss": 0.3725, + "step": 38300 + }, + { + "epoch": 1.9171775303390466, + "grad_norm": 269.5062561035156, + "learning_rate": 2.0051528238586866e-05, + "loss": 0.0556, + "step": 38310 + }, + { + "epoch": 1.9176779682221943, + "grad_norm": 15.0125732421875, + "learning_rate": 2.0042260569776282e-05, + "loss": 0.1763, + "step": 38320 + }, + { + "epoch": 1.9181784061053422, + "grad_norm": 0.6028684377670288, + "learning_rate": 2.003299290096569e-05, + "loss": 0.5424, + "step": 38330 + }, + { + "epoch": 1.91867884398849, + "grad_norm": 0.06128727272152901, + "learning_rate": 2.0023725232155106e-05, + "loss": 0.2938, + "step": 38340 + }, + { + "epoch": 1.9191792818716378, + "grad_norm": 0.5892057418823242, + "learning_rate": 2.001445756334452e-05, + "loss": 0.3649, + "step": 38350 + }, + { + "epoch": 1.9196797197547855, + "grad_norm": 5.539007663726807, + "learning_rate": 2.000518989453393e-05, + "loss": 0.0791, + "step": 38360 + }, + { + "epoch": 1.9201801576379331, + "grad_norm": 40.56745147705078, + "learning_rate": 1.9995922225723343e-05, + "loss": 0.2469, + "step": 38370 + }, + { + "epoch": 1.9206805955210808, + "grad_norm": 0.039484478533267975, + "learning_rate": 1.9986654556912755e-05, + "loss": 0.4673, + "step": 38380 + }, + { + "epoch": 1.9211810334042287, + "grad_norm": 24.620361328125, + "learning_rate": 1.9977386888102168e-05, + "loss": 0.5188, + "step": 38390 + }, + { + "epoch": 1.9216814712873764, + "grad_norm": 2.6610538959503174, + "learning_rate": 1.996811921929158e-05, + "loss": 0.137, + "step": 38400 + }, + { + "epoch": 1.9221819091705243, + "grad_norm": 12.243138313293457, + "learning_rate": 1.9958851550480992e-05, + "loss": 0.3763, + "step": 38410 + }, + { + "epoch": 1.922682347053672, + "grad_norm": 1.1571323871612549, + "learning_rate": 1.9949583881670404e-05, + "loss": 0.3908, + "step": 38420 + }, + { + "epoch": 1.9231827849368197, + "grad_norm": 171.40101623535156, + "learning_rate": 1.994031621285982e-05, + "loss": 0.1568, + "step": 38430 + }, + { + "epoch": 1.9236832228199674, + "grad_norm": 0.05472568795084953, + "learning_rate": 1.993104854404923e-05, + "loss": 0.2378, + "step": 38440 + }, + { + "epoch": 1.924183660703115, + "grad_norm": 18.343496322631836, + "learning_rate": 1.9921780875238645e-05, + "loss": 0.2364, + "step": 38450 + }, + { + "epoch": 1.924684098586263, + "grad_norm": 0.011720729060471058, + "learning_rate": 1.9912513206428057e-05, + "loss": 0.5481, + "step": 38460 + }, + { + "epoch": 1.925184536469411, + "grad_norm": 0.03034079261124134, + "learning_rate": 1.990324553761747e-05, + "loss": 0.2344, + "step": 38470 + }, + { + "epoch": 1.9256849743525586, + "grad_norm": 0.1100611612200737, + "learning_rate": 1.989397786880688e-05, + "loss": 0.2187, + "step": 38480 + }, + { + "epoch": 1.9261854122357063, + "grad_norm": 0.0028235604986548424, + "learning_rate": 1.9884710199996294e-05, + "loss": 0.2562, + "step": 38490 + }, + { + "epoch": 1.926685850118854, + "grad_norm": 69.13214874267578, + "learning_rate": 1.9875442531185706e-05, + "loss": 0.571, + "step": 38500 + }, + { + "epoch": 1.9271862880020016, + "grad_norm": 0.01307985931634903, + "learning_rate": 1.986617486237512e-05, + "loss": 0.3741, + "step": 38510 + }, + { + "epoch": 1.9276867258851496, + "grad_norm": 0.07101128995418549, + "learning_rate": 1.985690719356453e-05, + "loss": 0.3192, + "step": 38520 + }, + { + "epoch": 1.9281871637682972, + "grad_norm": 0.031741708517074585, + "learning_rate": 1.9847639524753946e-05, + "loss": 0.4429, + "step": 38530 + }, + { + "epoch": 1.9286876016514451, + "grad_norm": 40.854793548583984, + "learning_rate": 1.9838371855943358e-05, + "loss": 0.1202, + "step": 38540 + }, + { + "epoch": 1.9291880395345928, + "grad_norm": 0.022560346871614456, + "learning_rate": 1.982910418713277e-05, + "loss": 0.2189, + "step": 38550 + }, + { + "epoch": 1.9296884774177405, + "grad_norm": 14.312662124633789, + "learning_rate": 1.9819836518322183e-05, + "loss": 0.1215, + "step": 38560 + }, + { + "epoch": 1.9301889153008882, + "grad_norm": 0.033199045807123184, + "learning_rate": 1.9810568849511595e-05, + "loss": 0.1576, + "step": 38570 + }, + { + "epoch": 1.930689353184036, + "grad_norm": 0.02994488552212715, + "learning_rate": 1.9801301180701007e-05, + "loss": 0.5478, + "step": 38580 + }, + { + "epoch": 1.9311897910671838, + "grad_norm": 0.021728241816163063, + "learning_rate": 1.979203351189042e-05, + "loss": 0.0402, + "step": 38590 + }, + { + "epoch": 1.9316902289503315, + "grad_norm": 102.66822052001953, + "learning_rate": 1.978276584307983e-05, + "loss": 0.7241, + "step": 38600 + }, + { + "epoch": 1.9321906668334794, + "grad_norm": 7.043759822845459, + "learning_rate": 1.9773498174269244e-05, + "loss": 0.1679, + "step": 38610 + }, + { + "epoch": 1.932691104716627, + "grad_norm": 6.715415000915527, + "learning_rate": 1.976423050545866e-05, + "loss": 0.2334, + "step": 38620 + }, + { + "epoch": 1.9331915425997748, + "grad_norm": 0.048299629241228104, + "learning_rate": 1.975496283664807e-05, + "loss": 0.081, + "step": 38630 + }, + { + "epoch": 1.9336919804829225, + "grad_norm": 6.945917129516602, + "learning_rate": 1.9745695167837484e-05, + "loss": 0.0675, + "step": 38640 + }, + { + "epoch": 1.9341924183660704, + "grad_norm": 0.003348106751218438, + "learning_rate": 1.9736427499026896e-05, + "loss": 0.1869, + "step": 38650 + }, + { + "epoch": 1.934692856249218, + "grad_norm": 33.25022888183594, + "learning_rate": 1.972715983021631e-05, + "loss": 0.3363, + "step": 38660 + }, + { + "epoch": 1.935193294132366, + "grad_norm": 71.23806762695312, + "learning_rate": 1.971789216140572e-05, + "loss": 0.5361, + "step": 38670 + }, + { + "epoch": 1.9356937320155136, + "grad_norm": 6.791610240936279, + "learning_rate": 1.9708624492595136e-05, + "loss": 0.2665, + "step": 38680 + }, + { + "epoch": 1.9361941698986613, + "grad_norm": 0.23762622475624084, + "learning_rate": 1.9699356823784545e-05, + "loss": 0.3839, + "step": 38690 + }, + { + "epoch": 1.936694607781809, + "grad_norm": 0.18446411192417145, + "learning_rate": 1.9690089154973957e-05, + "loss": 0.0271, + "step": 38700 + }, + { + "epoch": 1.9371950456649567, + "grad_norm": 0.057152871042490005, + "learning_rate": 1.9680821486163373e-05, + "loss": 0.2802, + "step": 38710 + }, + { + "epoch": 1.9376954835481046, + "grad_norm": 9.697153091430664, + "learning_rate": 1.9671553817352782e-05, + "loss": 0.0584, + "step": 38720 + }, + { + "epoch": 1.9381959214312523, + "grad_norm": 0.26081234216690063, + "learning_rate": 1.9662286148542198e-05, + "loss": 0.342, + "step": 38730 + }, + { + "epoch": 1.9386963593144002, + "grad_norm": 42.10904312133789, + "learning_rate": 1.9653018479731606e-05, + "loss": 0.3958, + "step": 38740 + }, + { + "epoch": 1.939196797197548, + "grad_norm": 0.2662922739982605, + "learning_rate": 1.9643750810921022e-05, + "loss": 0.4579, + "step": 38750 + }, + { + "epoch": 1.9396972350806956, + "grad_norm": 82.2423095703125, + "learning_rate": 1.9634483142110434e-05, + "loss": 0.5848, + "step": 38760 + }, + { + "epoch": 1.9401976729638433, + "grad_norm": 39.092464447021484, + "learning_rate": 1.9625215473299847e-05, + "loss": 0.6123, + "step": 38770 + }, + { + "epoch": 1.940698110846991, + "grad_norm": 33.71055603027344, + "learning_rate": 1.961594780448926e-05, + "loss": 0.1917, + "step": 38780 + }, + { + "epoch": 1.9411985487301389, + "grad_norm": 45.30207824707031, + "learning_rate": 1.9606680135678674e-05, + "loss": 0.5071, + "step": 38790 + }, + { + "epoch": 1.9416989866132868, + "grad_norm": 0.07529697567224503, + "learning_rate": 1.9597412466868083e-05, + "loss": 0.05, + "step": 38800 + }, + { + "epoch": 1.9421994244964345, + "grad_norm": 23.548137664794922, + "learning_rate": 1.95881447980575e-05, + "loss": 0.4317, + "step": 38810 + }, + { + "epoch": 1.9426998623795821, + "grad_norm": 0.28246650099754333, + "learning_rate": 1.957887712924691e-05, + "loss": 0.1419, + "step": 38820 + }, + { + "epoch": 1.9432003002627298, + "grad_norm": 77.95573425292969, + "learning_rate": 1.9569609460436323e-05, + "loss": 0.4006, + "step": 38830 + }, + { + "epoch": 1.9437007381458775, + "grad_norm": 0.030148694291710854, + "learning_rate": 1.9560341791625736e-05, + "loss": 0.2456, + "step": 38840 + }, + { + "epoch": 1.9442011760290254, + "grad_norm": 123.46881866455078, + "learning_rate": 1.9551074122815148e-05, + "loss": 0.4918, + "step": 38850 + }, + { + "epoch": 1.944701613912173, + "grad_norm": 13.518691062927246, + "learning_rate": 1.954180645400456e-05, + "loss": 0.5489, + "step": 38860 + }, + { + "epoch": 1.945202051795321, + "grad_norm": 0.10513529926538467, + "learning_rate": 1.9532538785193972e-05, + "loss": 0.3713, + "step": 38870 + }, + { + "epoch": 1.9457024896784687, + "grad_norm": 1.6589914560317993, + "learning_rate": 1.9523271116383385e-05, + "loss": 0.1135, + "step": 38880 + }, + { + "epoch": 1.9462029275616164, + "grad_norm": 0.016304541379213333, + "learning_rate": 1.9514003447572797e-05, + "loss": 0.431, + "step": 38890 + }, + { + "epoch": 1.946703365444764, + "grad_norm": 107.87229919433594, + "learning_rate": 1.9504735778762213e-05, + "loss": 0.6401, + "step": 38900 + }, + { + "epoch": 1.9472038033279118, + "grad_norm": 0.003196494420990348, + "learning_rate": 1.949546810995162e-05, + "loss": 0.0549, + "step": 38910 + }, + { + "epoch": 1.9477042412110597, + "grad_norm": 0.06463432312011719, + "learning_rate": 1.9486200441141037e-05, + "loss": 0.181, + "step": 38920 + }, + { + "epoch": 1.9482046790942076, + "grad_norm": 0.4441663324832916, + "learning_rate": 1.947693277233045e-05, + "loss": 0.2891, + "step": 38930 + }, + { + "epoch": 1.9487051169773553, + "grad_norm": 76.42790985107422, + "learning_rate": 1.946766510351986e-05, + "loss": 0.8265, + "step": 38940 + }, + { + "epoch": 1.949205554860503, + "grad_norm": 148.88485717773438, + "learning_rate": 1.9458397434709274e-05, + "loss": 0.4351, + "step": 38950 + }, + { + "epoch": 1.9497059927436506, + "grad_norm": 38.4464225769043, + "learning_rate": 1.9449129765898686e-05, + "loss": 0.6419, + "step": 38960 + }, + { + "epoch": 1.9502064306267983, + "grad_norm": 0.01637681946158409, + "learning_rate": 1.9439862097088098e-05, + "loss": 0.1851, + "step": 38970 + }, + { + "epoch": 1.9507068685099462, + "grad_norm": 18.69266700744629, + "learning_rate": 1.9430594428277514e-05, + "loss": 0.12, + "step": 38980 + }, + { + "epoch": 1.951207306393094, + "grad_norm": 0.032791733741760254, + "learning_rate": 1.9421326759466923e-05, + "loss": 0.1044, + "step": 38990 + }, + { + "epoch": 1.9517077442762418, + "grad_norm": 4.314128398895264, + "learning_rate": 1.941205909065634e-05, + "loss": 0.0435, + "step": 39000 + }, + { + "epoch": 1.9522081821593895, + "grad_norm": 28.199445724487305, + "learning_rate": 1.940279142184575e-05, + "loss": 0.1038, + "step": 39010 + }, + { + "epoch": 1.9527086200425372, + "grad_norm": 131.5802764892578, + "learning_rate": 1.9393523753035163e-05, + "loss": 0.4657, + "step": 39020 + }, + { + "epoch": 1.9532090579256849, + "grad_norm": 40.78411865234375, + "learning_rate": 1.9384256084224575e-05, + "loss": 0.631, + "step": 39030 + }, + { + "epoch": 1.9537094958088326, + "grad_norm": 17.69831657409668, + "learning_rate": 1.9374988415413987e-05, + "loss": 0.3653, + "step": 39040 + }, + { + "epoch": 1.9542099336919805, + "grad_norm": 0.007929614745080471, + "learning_rate": 1.93657207466034e-05, + "loss": 0.4712, + "step": 39050 + }, + { + "epoch": 1.9547103715751284, + "grad_norm": 133.8663330078125, + "learning_rate": 1.9356453077792812e-05, + "loss": 0.37, + "step": 39060 + }, + { + "epoch": 1.955210809458276, + "grad_norm": 0.049835480749607086, + "learning_rate": 1.9347185408982227e-05, + "loss": 0.4072, + "step": 39070 + }, + { + "epoch": 1.9557112473414238, + "grad_norm": 14.456281661987305, + "learning_rate": 1.9337917740171636e-05, + "loss": 0.4284, + "step": 39080 + }, + { + "epoch": 1.9562116852245715, + "grad_norm": 3.5736629962921143, + "learning_rate": 1.9328650071361052e-05, + "loss": 0.0975, + "step": 39090 + }, + { + "epoch": 1.9567121231077191, + "grad_norm": 0.1393398493528366, + "learning_rate": 1.931938240255046e-05, + "loss": 0.2134, + "step": 39100 + }, + { + "epoch": 1.957212560990867, + "grad_norm": 49.70804977416992, + "learning_rate": 1.9310114733739876e-05, + "loss": 0.2199, + "step": 39110 + }, + { + "epoch": 1.9577129988740147, + "grad_norm": 0.2116159051656723, + "learning_rate": 1.930084706492929e-05, + "loss": 0.0308, + "step": 39120 + }, + { + "epoch": 1.9582134367571626, + "grad_norm": 15.704398155212402, + "learning_rate": 1.92915793961187e-05, + "loss": 0.3336, + "step": 39130 + }, + { + "epoch": 1.9587138746403103, + "grad_norm": 0.5310855507850647, + "learning_rate": 1.9282311727308113e-05, + "loss": 0.279, + "step": 39140 + }, + { + "epoch": 1.959214312523458, + "grad_norm": 7.522899627685547, + "learning_rate": 1.927304405849753e-05, + "loss": 0.0129, + "step": 39150 + }, + { + "epoch": 1.9597147504066057, + "grad_norm": 65.16851806640625, + "learning_rate": 1.9263776389686938e-05, + "loss": 0.1165, + "step": 39160 + }, + { + "epoch": 1.9602151882897534, + "grad_norm": 0.0031264193821698427, + "learning_rate": 1.9254508720876353e-05, + "loss": 0.1863, + "step": 39170 + }, + { + "epoch": 1.9607156261729013, + "grad_norm": 9.622693061828613, + "learning_rate": 1.9245241052065766e-05, + "loss": 0.3588, + "step": 39180 + }, + { + "epoch": 1.9612160640560492, + "grad_norm": 3.1381494998931885, + "learning_rate": 1.9235973383255178e-05, + "loss": 0.3458, + "step": 39190 + }, + { + "epoch": 1.9617165019391969, + "grad_norm": 22.620786666870117, + "learning_rate": 1.922670571444459e-05, + "loss": 0.3601, + "step": 39200 + }, + { + "epoch": 1.9622169398223446, + "grad_norm": 100.99469757080078, + "learning_rate": 1.9217438045634002e-05, + "loss": 0.4682, + "step": 39210 + }, + { + "epoch": 1.9627173777054923, + "grad_norm": 78.294189453125, + "learning_rate": 1.9208170376823415e-05, + "loss": 0.4242, + "step": 39220 + }, + { + "epoch": 1.96321781558864, + "grad_norm": 24.582162857055664, + "learning_rate": 1.9198902708012827e-05, + "loss": 0.1758, + "step": 39230 + }, + { + "epoch": 1.9637182534717879, + "grad_norm": 111.17098999023438, + "learning_rate": 1.918963503920224e-05, + "loss": 0.4421, + "step": 39240 + }, + { + "epoch": 1.9642186913549355, + "grad_norm": 48.66032791137695, + "learning_rate": 1.918036737039165e-05, + "loss": 0.1993, + "step": 39250 + }, + { + "epoch": 1.9647191292380835, + "grad_norm": 47.789310455322266, + "learning_rate": 1.9171099701581067e-05, + "loss": 0.693, + "step": 39260 + }, + { + "epoch": 1.9652195671212311, + "grad_norm": 0.18001070618629456, + "learning_rate": 1.9161832032770476e-05, + "loss": 0.2761, + "step": 39270 + }, + { + "epoch": 1.9657200050043788, + "grad_norm": 0.06842699646949768, + "learning_rate": 1.915256436395989e-05, + "loss": 0.3865, + "step": 39280 + }, + { + "epoch": 1.9662204428875265, + "grad_norm": 133.1101837158203, + "learning_rate": 1.9143296695149304e-05, + "loss": 0.3797, + "step": 39290 + }, + { + "epoch": 1.9667208807706742, + "grad_norm": 0.001964255701750517, + "learning_rate": 1.9134029026338716e-05, + "loss": 0.1474, + "step": 39300 + }, + { + "epoch": 1.967221318653822, + "grad_norm": 16.45125389099121, + "learning_rate": 1.9124761357528128e-05, + "loss": 0.5324, + "step": 39310 + }, + { + "epoch": 1.96772175653697, + "grad_norm": 0.15499356389045715, + "learning_rate": 1.911549368871754e-05, + "loss": 0.1474, + "step": 39320 + }, + { + "epoch": 1.9682221944201177, + "grad_norm": 1.5570976734161377, + "learning_rate": 1.9106226019906953e-05, + "loss": 0.2248, + "step": 39330 + }, + { + "epoch": 1.9687226323032654, + "grad_norm": 17.196813583374023, + "learning_rate": 1.9096958351096368e-05, + "loss": 0.1063, + "step": 39340 + }, + { + "epoch": 1.969223070186413, + "grad_norm": 152.7604522705078, + "learning_rate": 1.9087690682285777e-05, + "loss": 0.1803, + "step": 39350 + }, + { + "epoch": 1.9697235080695608, + "grad_norm": 0.004978777840733528, + "learning_rate": 1.9078423013475193e-05, + "loss": 0.1799, + "step": 39360 + }, + { + "epoch": 1.9702239459527087, + "grad_norm": 37.901893615722656, + "learning_rate": 1.9069155344664605e-05, + "loss": 0.6565, + "step": 39370 + }, + { + "epoch": 1.9707243838358564, + "grad_norm": 0.25576865673065186, + "learning_rate": 1.9059887675854017e-05, + "loss": 0.5937, + "step": 39380 + }, + { + "epoch": 1.9712248217190043, + "grad_norm": 0.02231348119676113, + "learning_rate": 1.905062000704343e-05, + "loss": 0.4063, + "step": 39390 + }, + { + "epoch": 1.971725259602152, + "grad_norm": 13.795987129211426, + "learning_rate": 1.9041352338232842e-05, + "loss": 0.2079, + "step": 39400 + }, + { + "epoch": 1.9722256974852996, + "grad_norm": 0.23206686973571777, + "learning_rate": 1.9032084669422254e-05, + "loss": 0.219, + "step": 39410 + }, + { + "epoch": 1.9727261353684473, + "grad_norm": 50.55750274658203, + "learning_rate": 1.9022817000611666e-05, + "loss": 0.221, + "step": 39420 + }, + { + "epoch": 1.973226573251595, + "grad_norm": 54.07215118408203, + "learning_rate": 1.9013549331801082e-05, + "loss": 0.6284, + "step": 39430 + }, + { + "epoch": 1.973727011134743, + "grad_norm": 0.00926213152706623, + "learning_rate": 1.900428166299049e-05, + "loss": 0.0717, + "step": 39440 + }, + { + "epoch": 1.9742274490178906, + "grad_norm": 0.2374161332845688, + "learning_rate": 1.8995013994179906e-05, + "loss": 0.2238, + "step": 39450 + }, + { + "epoch": 1.9747278869010385, + "grad_norm": 0.42995256185531616, + "learning_rate": 1.8985746325369315e-05, + "loss": 0.0351, + "step": 39460 + }, + { + "epoch": 1.9752283247841862, + "grad_norm": 0.09314969927072525, + "learning_rate": 1.897647865655873e-05, + "loss": 0.1265, + "step": 39470 + }, + { + "epoch": 1.9757287626673339, + "grad_norm": 0.1101616770029068, + "learning_rate": 1.8967210987748143e-05, + "loss": 0.2749, + "step": 39480 + }, + { + "epoch": 1.9762292005504816, + "grad_norm": 77.97087860107422, + "learning_rate": 1.8957943318937555e-05, + "loss": 0.5518, + "step": 39490 + }, + { + "epoch": 1.9767296384336295, + "grad_norm": 0.3032289445400238, + "learning_rate": 1.8948675650126968e-05, + "loss": 0.1909, + "step": 39500 + }, + { + "epoch": 1.9772300763167772, + "grad_norm": 0.0001418236643075943, + "learning_rate": 1.8939407981316383e-05, + "loss": 0.1261, + "step": 39510 + }, + { + "epoch": 1.977730514199925, + "grad_norm": 0.2809264659881592, + "learning_rate": 1.8930140312505792e-05, + "loss": 0.1804, + "step": 39520 + }, + { + "epoch": 1.9782309520830728, + "grad_norm": 5.340816974639893, + "learning_rate": 1.8920872643695208e-05, + "loss": 0.1092, + "step": 39530 + }, + { + "epoch": 1.9787313899662204, + "grad_norm": 121.42462921142578, + "learning_rate": 1.891160497488462e-05, + "loss": 0.7047, + "step": 39540 + }, + { + "epoch": 1.9792318278493681, + "grad_norm": 0.29102838039398193, + "learning_rate": 1.890233730607403e-05, + "loss": 0.2219, + "step": 39550 + }, + { + "epoch": 1.9797322657325158, + "grad_norm": 0.9177095293998718, + "learning_rate": 1.8893069637263444e-05, + "loss": 0.3951, + "step": 39560 + }, + { + "epoch": 1.9802327036156637, + "grad_norm": 9.062356948852539, + "learning_rate": 1.8883801968452853e-05, + "loss": 0.3954, + "step": 39570 + }, + { + "epoch": 1.9807331414988114, + "grad_norm": 28.76822280883789, + "learning_rate": 1.887453429964227e-05, + "loss": 0.2041, + "step": 39580 + }, + { + "epoch": 1.9812335793819593, + "grad_norm": 86.29666900634766, + "learning_rate": 1.886526663083168e-05, + "loss": 0.5166, + "step": 39590 + }, + { + "epoch": 1.981734017265107, + "grad_norm": 0.03922783583402634, + "learning_rate": 1.8855998962021093e-05, + "loss": 0.4806, + "step": 39600 + }, + { + "epoch": 1.9822344551482547, + "grad_norm": 81.53506469726562, + "learning_rate": 1.8846731293210506e-05, + "loss": 1.1306, + "step": 39610 + }, + { + "epoch": 1.9827348930314024, + "grad_norm": 1.1444578170776367, + "learning_rate": 1.883746362439992e-05, + "loss": 0.2584, + "step": 39620 + }, + { + "epoch": 1.98323533091455, + "grad_norm": 1.7953673601150513, + "learning_rate": 1.882819595558933e-05, + "loss": 0.2887, + "step": 39630 + }, + { + "epoch": 1.983735768797698, + "grad_norm": 7.218924522399902, + "learning_rate": 1.8818928286778746e-05, + "loss": 0.2231, + "step": 39640 + }, + { + "epoch": 1.9842362066808459, + "grad_norm": 6.852556228637695, + "learning_rate": 1.8809660617968158e-05, + "loss": 0.4044, + "step": 39650 + }, + { + "epoch": 1.9847366445639936, + "grad_norm": 0.25927481055259705, + "learning_rate": 1.880039294915757e-05, + "loss": 0.177, + "step": 39660 + }, + { + "epoch": 1.9852370824471413, + "grad_norm": 16.382543563842773, + "learning_rate": 1.8791125280346983e-05, + "loss": 0.0945, + "step": 39670 + }, + { + "epoch": 1.985737520330289, + "grad_norm": 26.247037887573242, + "learning_rate": 1.8781857611536395e-05, + "loss": 0.4592, + "step": 39680 + }, + { + "epoch": 1.9862379582134366, + "grad_norm": 35.80073547363281, + "learning_rate": 1.8772589942725807e-05, + "loss": 0.1388, + "step": 39690 + }, + { + "epoch": 1.9867383960965845, + "grad_norm": 0.8223179578781128, + "learning_rate": 1.876332227391522e-05, + "loss": 0.1086, + "step": 39700 + }, + { + "epoch": 1.9872388339797322, + "grad_norm": 0.32429230213165283, + "learning_rate": 1.875405460510463e-05, + "loss": 0.4092, + "step": 39710 + }, + { + "epoch": 1.9877392718628801, + "grad_norm": 16.990327835083008, + "learning_rate": 1.8744786936294044e-05, + "loss": 0.2362, + "step": 39720 + }, + { + "epoch": 1.9882397097460278, + "grad_norm": 22.86969566345215, + "learning_rate": 1.873551926748346e-05, + "loss": 0.184, + "step": 39730 + }, + { + "epoch": 1.9887401476291755, + "grad_norm": 0.09800606966018677, + "learning_rate": 1.8726251598672868e-05, + "loss": 0.7, + "step": 39740 + }, + { + "epoch": 1.9892405855123232, + "grad_norm": 92.15171813964844, + "learning_rate": 1.8716983929862284e-05, + "loss": 0.1673, + "step": 39750 + }, + { + "epoch": 1.9897410233954709, + "grad_norm": 3.3917696475982666, + "learning_rate": 1.8707716261051696e-05, + "loss": 0.1873, + "step": 39760 + }, + { + "epoch": 1.9902414612786188, + "grad_norm": 6.319879055023193, + "learning_rate": 1.869844859224111e-05, + "loss": 0.3223, + "step": 39770 + }, + { + "epoch": 1.9907418991617667, + "grad_norm": 0.0378299206495285, + "learning_rate": 1.868918092343052e-05, + "loss": 0.2783, + "step": 39780 + }, + { + "epoch": 1.9912423370449144, + "grad_norm": 0.05343205854296684, + "learning_rate": 1.8679913254619936e-05, + "loss": 0.5748, + "step": 39790 + }, + { + "epoch": 1.991742774928062, + "grad_norm": 112.3886489868164, + "learning_rate": 1.8670645585809345e-05, + "loss": 0.2396, + "step": 39800 + }, + { + "epoch": 1.9922432128112098, + "grad_norm": 0.30969467759132385, + "learning_rate": 1.866137791699876e-05, + "loss": 0.3005, + "step": 39810 + }, + { + "epoch": 1.9927436506943574, + "grad_norm": 2.6278061866760254, + "learning_rate": 1.865211024818817e-05, + "loss": 0.2227, + "step": 39820 + }, + { + "epoch": 1.9932440885775053, + "grad_norm": 0.0008213632390834391, + "learning_rate": 1.8642842579377585e-05, + "loss": 0.8116, + "step": 39830 + }, + { + "epoch": 1.993744526460653, + "grad_norm": 5.412262439727783, + "learning_rate": 1.8633574910566998e-05, + "loss": 0.032, + "step": 39840 + }, + { + "epoch": 1.994244964343801, + "grad_norm": 0.04635142534971237, + "learning_rate": 1.862430724175641e-05, + "loss": 0.6638, + "step": 39850 + }, + { + "epoch": 1.9947454022269486, + "grad_norm": 9.537368774414062, + "learning_rate": 1.8615039572945822e-05, + "loss": 0.3742, + "step": 39860 + }, + { + "epoch": 1.9952458401100963, + "grad_norm": 0.018217923119664192, + "learning_rate": 1.8605771904135234e-05, + "loss": 0.3568, + "step": 39870 + }, + { + "epoch": 1.995746277993244, + "grad_norm": 90.65947723388672, + "learning_rate": 1.8596504235324647e-05, + "loss": 0.1278, + "step": 39880 + }, + { + "epoch": 1.9962467158763917, + "grad_norm": 36.24128341674805, + "learning_rate": 1.858723656651406e-05, + "loss": 0.8943, + "step": 39890 + }, + { + "epoch": 1.9967471537595396, + "grad_norm": 25.073434829711914, + "learning_rate": 1.8577968897703474e-05, + "loss": 0.2527, + "step": 39900 + }, + { + "epoch": 1.9972475916426875, + "grad_norm": 0.0025730018969625235, + "learning_rate": 1.8568701228892883e-05, + "loss": 0.496, + "step": 39910 + }, + { + "epoch": 1.9977480295258352, + "grad_norm": 0.026928216218948364, + "learning_rate": 1.85594335600823e-05, + "loss": 0.1918, + "step": 39920 + }, + { + "epoch": 1.9982484674089829, + "grad_norm": 55.383583068847656, + "learning_rate": 1.8550165891271708e-05, + "loss": 0.2728, + "step": 39930 + }, + { + "epoch": 1.9987489052921306, + "grad_norm": 83.79020690917969, + "learning_rate": 1.8540898222461123e-05, + "loss": 0.2642, + "step": 39940 + }, + { + "epoch": 1.9992493431752782, + "grad_norm": 0.629515528678894, + "learning_rate": 1.8531630553650536e-05, + "loss": 0.0246, + "step": 39950 + }, + { + "epoch": 1.9997497810584262, + "grad_norm": 0.03350881114602089, + "learning_rate": 1.8522362884839948e-05, + "loss": 0.3994, + "step": 39960 + }, + { + "epoch": 2.0, + "eval_accuracy": 0.9530511146138257, + "eval_loss": 0.1906302124261856, + "eval_runtime": 142.5982, + "eval_samples_per_second": 62.287, + "eval_steps_per_second": 62.287, + "step": 39965 + }, + { + "epoch": 2.000250218941574, + "grad_norm": 36.183475494384766, + "learning_rate": 1.851309521602936e-05, + "loss": 0.5012, + "step": 39970 + }, + { + "epoch": 2.0007506568247218, + "grad_norm": 74.43077850341797, + "learning_rate": 1.8503827547218776e-05, + "loss": 0.4733, + "step": 39980 + }, + { + "epoch": 2.0012510947078694, + "grad_norm": 17.011106491088867, + "learning_rate": 1.8494559878408185e-05, + "loss": 0.2518, + "step": 39990 + }, + { + "epoch": 2.001751532591017, + "grad_norm": 8.583924293518066, + "learning_rate": 1.84852922095976e-05, + "loss": 0.0577, + "step": 40000 + }, + { + "epoch": 2.002251970474165, + "grad_norm": 0.034581899642944336, + "learning_rate": 1.8476024540787012e-05, + "loss": 0.442, + "step": 40010 + }, + { + "epoch": 2.0027524083573125, + "grad_norm": 28.53826332092285, + "learning_rate": 1.8466756871976425e-05, + "loss": 0.1588, + "step": 40020 + }, + { + "epoch": 2.00325284624046, + "grad_norm": 53.56166076660156, + "learning_rate": 1.8457489203165837e-05, + "loss": 0.4005, + "step": 40030 + }, + { + "epoch": 2.0037532841236083, + "grad_norm": 2.86788272857666, + "learning_rate": 1.844822153435525e-05, + "loss": 0.2488, + "step": 40040 + }, + { + "epoch": 2.004253722006756, + "grad_norm": 0.16902998089790344, + "learning_rate": 1.843895386554466e-05, + "loss": 0.5776, + "step": 40050 + }, + { + "epoch": 2.0047541598899037, + "grad_norm": 0.018983514979481697, + "learning_rate": 1.8429686196734074e-05, + "loss": 0.49, + "step": 40060 + }, + { + "epoch": 2.0052545977730514, + "grad_norm": 80.67587280273438, + "learning_rate": 1.8420418527923486e-05, + "loss": 0.0472, + "step": 40070 + }, + { + "epoch": 2.005755035656199, + "grad_norm": 119.89305877685547, + "learning_rate": 1.8411150859112898e-05, + "loss": 0.3646, + "step": 40080 + }, + { + "epoch": 2.0062554735393467, + "grad_norm": 0.030567260459065437, + "learning_rate": 1.8401883190302314e-05, + "loss": 0.2772, + "step": 40090 + }, + { + "epoch": 2.006755911422495, + "grad_norm": 0.2291530817747116, + "learning_rate": 1.8392615521491723e-05, + "loss": 0.1388, + "step": 40100 + }, + { + "epoch": 2.0072563493056426, + "grad_norm": 25.05268096923828, + "learning_rate": 1.838334785268114e-05, + "loss": 0.3736, + "step": 40110 + }, + { + "epoch": 2.0077567871887902, + "grad_norm": 106.2945327758789, + "learning_rate": 1.837408018387055e-05, + "loss": 0.5956, + "step": 40120 + }, + { + "epoch": 2.008257225071938, + "grad_norm": 0.2427186369895935, + "learning_rate": 1.8364812515059963e-05, + "loss": 0.5374, + "step": 40130 + }, + { + "epoch": 2.0087576629550856, + "grad_norm": 0.15681257843971252, + "learning_rate": 1.8355544846249375e-05, + "loss": 0.1265, + "step": 40140 + }, + { + "epoch": 2.0092581008382333, + "grad_norm": 0.36968696117401123, + "learning_rate": 1.834627717743879e-05, + "loss": 0.0844, + "step": 40150 + }, + { + "epoch": 2.009758538721381, + "grad_norm": 0.03799310699105263, + "learning_rate": 1.83370095086282e-05, + "loss": 0.7535, + "step": 40160 + }, + { + "epoch": 2.010258976604529, + "grad_norm": 0.050830110907554626, + "learning_rate": 1.8327741839817615e-05, + "loss": 0.0983, + "step": 40170 + }, + { + "epoch": 2.010759414487677, + "grad_norm": 18.700965881347656, + "learning_rate": 1.8318474171007024e-05, + "loss": 0.4471, + "step": 40180 + }, + { + "epoch": 2.0112598523708245, + "grad_norm": 15.616438865661621, + "learning_rate": 1.830920650219644e-05, + "loss": 0.4519, + "step": 40190 + }, + { + "epoch": 2.011760290253972, + "grad_norm": 47.50605392456055, + "learning_rate": 1.8299938833385852e-05, + "loss": 0.423, + "step": 40200 + }, + { + "epoch": 2.01226072813712, + "grad_norm": 1.8525018692016602, + "learning_rate": 1.8290671164575264e-05, + "loss": 0.3398, + "step": 40210 + }, + { + "epoch": 2.0127611660202676, + "grad_norm": 0.21205556392669678, + "learning_rate": 1.8281403495764676e-05, + "loss": 0.048, + "step": 40220 + }, + { + "epoch": 2.0132616039034157, + "grad_norm": 22.824970245361328, + "learning_rate": 1.827213582695409e-05, + "loss": 0.3377, + "step": 40230 + }, + { + "epoch": 2.0137620417865634, + "grad_norm": 4.713993549346924, + "learning_rate": 1.82628681581435e-05, + "loss": 0.2176, + "step": 40240 + }, + { + "epoch": 2.014262479669711, + "grad_norm": 0.11285489797592163, + "learning_rate": 1.8253600489332913e-05, + "loss": 0.1977, + "step": 40250 + }, + { + "epoch": 2.0147629175528587, + "grad_norm": 4.33047342300415, + "learning_rate": 1.824433282052233e-05, + "loss": 0.3766, + "step": 40260 + }, + { + "epoch": 2.0152633554360064, + "grad_norm": 0.004346418660134077, + "learning_rate": 1.8235065151711738e-05, + "loss": 0.5763, + "step": 40270 + }, + { + "epoch": 2.015763793319154, + "grad_norm": 0.24993681907653809, + "learning_rate": 1.8225797482901153e-05, + "loss": 0.221, + "step": 40280 + }, + { + "epoch": 2.016264231202302, + "grad_norm": 0.06249694153666496, + "learning_rate": 1.8216529814090562e-05, + "loss": 0.1285, + "step": 40290 + }, + { + "epoch": 2.01676466908545, + "grad_norm": 1.6300166845321655, + "learning_rate": 1.8207262145279978e-05, + "loss": 0.1514, + "step": 40300 + }, + { + "epoch": 2.0172651069685976, + "grad_norm": 0.20850028097629547, + "learning_rate": 1.819799447646939e-05, + "loss": 0.4905, + "step": 40310 + }, + { + "epoch": 2.0177655448517453, + "grad_norm": 42.05246353149414, + "learning_rate": 1.8188726807658802e-05, + "loss": 0.3261, + "step": 40320 + }, + { + "epoch": 2.018265982734893, + "grad_norm": 0.018475279211997986, + "learning_rate": 1.8179459138848215e-05, + "loss": 0.5858, + "step": 40330 + }, + { + "epoch": 2.0187664206180407, + "grad_norm": 0.06797449290752411, + "learning_rate": 1.817019147003763e-05, + "loss": 0.5095, + "step": 40340 + }, + { + "epoch": 2.0192668585011884, + "grad_norm": 0.0018386917654424906, + "learning_rate": 1.816092380122704e-05, + "loss": 0.2144, + "step": 40350 + }, + { + "epoch": 2.0197672963843365, + "grad_norm": 72.31749725341797, + "learning_rate": 1.8151656132416455e-05, + "loss": 0.2882, + "step": 40360 + }, + { + "epoch": 2.020267734267484, + "grad_norm": 0.4082905650138855, + "learning_rate": 1.8142388463605867e-05, + "loss": 0.4986, + "step": 40370 + }, + { + "epoch": 2.020768172150632, + "grad_norm": 24.49420928955078, + "learning_rate": 1.813312079479528e-05, + "loss": 0.0943, + "step": 40380 + }, + { + "epoch": 2.0212686100337796, + "grad_norm": 0.1363367736339569, + "learning_rate": 1.812385312598469e-05, + "loss": 0.2112, + "step": 40390 + }, + { + "epoch": 2.0217690479169272, + "grad_norm": 24.658111572265625, + "learning_rate": 1.81145854571741e-05, + "loss": 0.4904, + "step": 40400 + }, + { + "epoch": 2.022269485800075, + "grad_norm": 30.858226776123047, + "learning_rate": 1.8105317788363516e-05, + "loss": 0.4938, + "step": 40410 + }, + { + "epoch": 2.0227699236832226, + "grad_norm": 30.39336395263672, + "learning_rate": 1.8096050119552928e-05, + "loss": 0.347, + "step": 40420 + }, + { + "epoch": 2.0232703615663707, + "grad_norm": 8.606295585632324, + "learning_rate": 1.808678245074234e-05, + "loss": 0.5334, + "step": 40430 + }, + { + "epoch": 2.0237707994495184, + "grad_norm": 0.04730116203427315, + "learning_rate": 1.8077514781931753e-05, + "loss": 0.4338, + "step": 40440 + }, + { + "epoch": 2.024271237332666, + "grad_norm": 26.66436195373535, + "learning_rate": 1.8068247113121168e-05, + "loss": 0.217, + "step": 40450 + }, + { + "epoch": 2.024771675215814, + "grad_norm": 84.3626480102539, + "learning_rate": 1.8058979444310577e-05, + "loss": 0.4095, + "step": 40460 + }, + { + "epoch": 2.0252721130989615, + "grad_norm": 13.096198081970215, + "learning_rate": 1.8049711775499993e-05, + "loss": 0.3339, + "step": 40470 + }, + { + "epoch": 2.025772550982109, + "grad_norm": 140.82504272460938, + "learning_rate": 1.8040444106689405e-05, + "loss": 0.7042, + "step": 40480 + }, + { + "epoch": 2.0262729888652573, + "grad_norm": 0.11263622343540192, + "learning_rate": 1.8031176437878817e-05, + "loss": 0.2196, + "step": 40490 + }, + { + "epoch": 2.026773426748405, + "grad_norm": 0.00596740934997797, + "learning_rate": 1.802190876906823e-05, + "loss": 0.0361, + "step": 40500 + }, + { + "epoch": 2.0272738646315527, + "grad_norm": 0.016670485958456993, + "learning_rate": 1.8012641100257642e-05, + "loss": 0.3699, + "step": 40510 + }, + { + "epoch": 2.0277743025147004, + "grad_norm": 9.679298400878906, + "learning_rate": 1.8003373431447054e-05, + "loss": 0.2621, + "step": 40520 + }, + { + "epoch": 2.028274740397848, + "grad_norm": 0.016727697104215622, + "learning_rate": 1.7994105762636466e-05, + "loss": 0.1422, + "step": 40530 + }, + { + "epoch": 2.0287751782809957, + "grad_norm": 3.3347113132476807, + "learning_rate": 1.798483809382588e-05, + "loss": 0.3457, + "step": 40540 + }, + { + "epoch": 2.0292756161641434, + "grad_norm": 0.1346331685781479, + "learning_rate": 1.797557042501529e-05, + "loss": 0.1885, + "step": 40550 + }, + { + "epoch": 2.0297760540472916, + "grad_norm": 21.049104690551758, + "learning_rate": 1.7966302756204706e-05, + "loss": 0.5957, + "step": 40560 + }, + { + "epoch": 2.0302764919304392, + "grad_norm": 8.052550315856934, + "learning_rate": 1.7957035087394115e-05, + "loss": 0.3279, + "step": 40570 + }, + { + "epoch": 2.030776929813587, + "grad_norm": 47.939815521240234, + "learning_rate": 1.794776741858353e-05, + "loss": 0.4315, + "step": 40580 + }, + { + "epoch": 2.0312773676967346, + "grad_norm": 0.011698291637003422, + "learning_rate": 1.7938499749772943e-05, + "loss": 0.0618, + "step": 40590 + }, + { + "epoch": 2.0317778055798823, + "grad_norm": 49.40618896484375, + "learning_rate": 1.7929232080962355e-05, + "loss": 0.2619, + "step": 40600 + }, + { + "epoch": 2.03227824346303, + "grad_norm": 66.57951354980469, + "learning_rate": 1.7919964412151768e-05, + "loss": 0.261, + "step": 40610 + }, + { + "epoch": 2.032778681346178, + "grad_norm": 2.0012264251708984, + "learning_rate": 1.7910696743341183e-05, + "loss": 0.2089, + "step": 40620 + }, + { + "epoch": 2.033279119229326, + "grad_norm": 38.68014907836914, + "learning_rate": 1.7901429074530592e-05, + "loss": 0.5501, + "step": 40630 + }, + { + "epoch": 2.0337795571124735, + "grad_norm": 36.08052444458008, + "learning_rate": 1.7892161405720008e-05, + "loss": 0.4128, + "step": 40640 + }, + { + "epoch": 2.034279994995621, + "grad_norm": 0.0028540717903524637, + "learning_rate": 1.7882893736909417e-05, + "loss": 0.2695, + "step": 40650 + }, + { + "epoch": 2.034780432878769, + "grad_norm": 0.19142566621303558, + "learning_rate": 1.7873626068098832e-05, + "loss": 0.1243, + "step": 40660 + }, + { + "epoch": 2.0352808707619165, + "grad_norm": 0.0079316021874547, + "learning_rate": 1.7864358399288244e-05, + "loss": 0.4231, + "step": 40670 + }, + { + "epoch": 2.0357813086450642, + "grad_norm": 67.60497283935547, + "learning_rate": 1.7855090730477657e-05, + "loss": 0.363, + "step": 40680 + }, + { + "epoch": 2.0362817465282124, + "grad_norm": 0.0099329287186265, + "learning_rate": 1.784582306166707e-05, + "loss": 0.2439, + "step": 40690 + }, + { + "epoch": 2.03678218441136, + "grad_norm": 1.6787437200546265, + "learning_rate": 1.783655539285648e-05, + "loss": 0.0845, + "step": 40700 + }, + { + "epoch": 2.0372826222945077, + "grad_norm": 0.007227650843560696, + "learning_rate": 1.7827287724045893e-05, + "loss": 0.2656, + "step": 40710 + }, + { + "epoch": 2.0377830601776554, + "grad_norm": 39.78553009033203, + "learning_rate": 1.7818020055235306e-05, + "loss": 0.3872, + "step": 40720 + }, + { + "epoch": 2.038283498060803, + "grad_norm": 0.314224511384964, + "learning_rate": 1.780875238642472e-05, + "loss": 0.1103, + "step": 40730 + }, + { + "epoch": 2.038783935943951, + "grad_norm": 0.10573425889015198, + "learning_rate": 1.779948471761413e-05, + "loss": 0.2695, + "step": 40740 + }, + { + "epoch": 2.039284373827099, + "grad_norm": 0.008868576027452946, + "learning_rate": 1.7790217048803546e-05, + "loss": 0.3601, + "step": 40750 + }, + { + "epoch": 2.0397848117102466, + "grad_norm": 0.0016389181837439537, + "learning_rate": 1.7780949379992955e-05, + "loss": 0.206, + "step": 40760 + }, + { + "epoch": 2.0402852495933943, + "grad_norm": 11.477750778198242, + "learning_rate": 1.777168171118237e-05, + "loss": 0.0399, + "step": 40770 + }, + { + "epoch": 2.040785687476542, + "grad_norm": 17.865495681762695, + "learning_rate": 1.7762414042371783e-05, + "loss": 0.2603, + "step": 40780 + }, + { + "epoch": 2.0412861253596897, + "grad_norm": 0.026078365743160248, + "learning_rate": 1.7753146373561195e-05, + "loss": 0.1058, + "step": 40790 + }, + { + "epoch": 2.0417865632428374, + "grad_norm": 0.05093768239021301, + "learning_rate": 1.7743878704750607e-05, + "loss": 0.1807, + "step": 40800 + }, + { + "epoch": 2.042287001125985, + "grad_norm": 69.58734893798828, + "learning_rate": 1.7734611035940023e-05, + "loss": 0.061, + "step": 40810 + }, + { + "epoch": 2.042787439009133, + "grad_norm": 0.77308189868927, + "learning_rate": 1.772534336712943e-05, + "loss": 0.2598, + "step": 40820 + }, + { + "epoch": 2.043287876892281, + "grad_norm": 3.8542232513427734, + "learning_rate": 1.7716075698318847e-05, + "loss": 0.3581, + "step": 40830 + }, + { + "epoch": 2.0437883147754285, + "grad_norm": 0.006630133371800184, + "learning_rate": 1.770680802950826e-05, + "loss": 0.3016, + "step": 40840 + }, + { + "epoch": 2.0442887526585762, + "grad_norm": 91.28841400146484, + "learning_rate": 1.769754036069767e-05, + "loss": 0.7471, + "step": 40850 + }, + { + "epoch": 2.044789190541724, + "grad_norm": 11.080266952514648, + "learning_rate": 1.7688272691887084e-05, + "loss": 0.1567, + "step": 40860 + }, + { + "epoch": 2.0452896284248716, + "grad_norm": 6.034518241882324, + "learning_rate": 1.7679005023076496e-05, + "loss": 0.1678, + "step": 40870 + }, + { + "epoch": 2.0457900663080197, + "grad_norm": 1.7420028448104858, + "learning_rate": 1.766973735426591e-05, + "loss": 0.3064, + "step": 40880 + }, + { + "epoch": 2.0462905041911674, + "grad_norm": 0.011813327670097351, + "learning_rate": 1.766046968545532e-05, + "loss": 0.3129, + "step": 40890 + }, + { + "epoch": 2.046790942074315, + "grad_norm": 34.51035690307617, + "learning_rate": 1.7651202016644733e-05, + "loss": 0.5289, + "step": 40900 + }, + { + "epoch": 2.047291379957463, + "grad_norm": 0.05686042085289955, + "learning_rate": 1.7641934347834145e-05, + "loss": 0.5642, + "step": 40910 + }, + { + "epoch": 2.0477918178406105, + "grad_norm": 0.004001266323029995, + "learning_rate": 1.763266667902356e-05, + "loss": 0.0799, + "step": 40920 + }, + { + "epoch": 2.048292255723758, + "grad_norm": 0.13982507586479187, + "learning_rate": 1.762339901021297e-05, + "loss": 0.2252, + "step": 40930 + }, + { + "epoch": 2.048792693606906, + "grad_norm": 120.44246673583984, + "learning_rate": 1.7614131341402385e-05, + "loss": 0.3626, + "step": 40940 + }, + { + "epoch": 2.049293131490054, + "grad_norm": 0.008860491216182709, + "learning_rate": 1.7604863672591797e-05, + "loss": 0.0849, + "step": 40950 + }, + { + "epoch": 2.0497935693732017, + "grad_norm": 0.3487808108329773, + "learning_rate": 1.759559600378121e-05, + "loss": 0.403, + "step": 40960 + }, + { + "epoch": 2.0502940072563494, + "grad_norm": 27.905179977416992, + "learning_rate": 1.7586328334970622e-05, + "loss": 0.2819, + "step": 40970 + }, + { + "epoch": 2.050794445139497, + "grad_norm": 59.00434494018555, + "learning_rate": 1.7577060666160038e-05, + "loss": 0.4935, + "step": 40980 + }, + { + "epoch": 2.0512948830226447, + "grad_norm": 2.3878912925720215, + "learning_rate": 1.7567792997349446e-05, + "loss": 0.285, + "step": 40990 + }, + { + "epoch": 2.0517953209057924, + "grad_norm": 90.27593994140625, + "learning_rate": 1.7558525328538862e-05, + "loss": 0.138, + "step": 41000 + }, + { + "epoch": 2.05229575878894, + "grad_norm": 8.914167404174805, + "learning_rate": 1.754925765972827e-05, + "loss": 0.0743, + "step": 41010 + }, + { + "epoch": 2.0527961966720882, + "grad_norm": 1.5360952615737915, + "learning_rate": 1.7539989990917687e-05, + "loss": 0.2433, + "step": 41020 + }, + { + "epoch": 2.053296634555236, + "grad_norm": 0.05739664286375046, + "learning_rate": 1.75307223221071e-05, + "loss": 0.1783, + "step": 41030 + }, + { + "epoch": 2.0537970724383836, + "grad_norm": 0.015166088938713074, + "learning_rate": 1.752145465329651e-05, + "loss": 0.7228, + "step": 41040 + }, + { + "epoch": 2.0542975103215313, + "grad_norm": 0.5518142580986023, + "learning_rate": 1.7512186984485923e-05, + "loss": 0.367, + "step": 41050 + }, + { + "epoch": 2.054797948204679, + "grad_norm": 0.04032757878303528, + "learning_rate": 1.7502919315675336e-05, + "loss": 0.4091, + "step": 41060 + }, + { + "epoch": 2.0552983860878267, + "grad_norm": 0.061078332364559174, + "learning_rate": 1.7493651646864748e-05, + "loss": 0.2246, + "step": 41070 + }, + { + "epoch": 2.055798823970975, + "grad_norm": 1.374756097793579, + "learning_rate": 1.748438397805416e-05, + "loss": 0.139, + "step": 41080 + }, + { + "epoch": 2.0562992618541225, + "grad_norm": 24.408458709716797, + "learning_rate": 1.7475116309243576e-05, + "loss": 0.2312, + "step": 41090 + }, + { + "epoch": 2.05679969973727, + "grad_norm": 5.169396877288818, + "learning_rate": 1.7465848640432985e-05, + "loss": 0.3377, + "step": 41100 + }, + { + "epoch": 2.057300137620418, + "grad_norm": 0.02238277718424797, + "learning_rate": 1.74565809716224e-05, + "loss": 0.1908, + "step": 41110 + }, + { + "epoch": 2.0578005755035655, + "grad_norm": 2.857543468475342, + "learning_rate": 1.744731330281181e-05, + "loss": 1.1043, + "step": 41120 + }, + { + "epoch": 2.0583010133867132, + "grad_norm": 0.006896019913256168, + "learning_rate": 1.7438045634001225e-05, + "loss": 0.112, + "step": 41130 + }, + { + "epoch": 2.058801451269861, + "grad_norm": 0.0179675854742527, + "learning_rate": 1.7428777965190637e-05, + "loss": 0.2813, + "step": 41140 + }, + { + "epoch": 2.059301889153009, + "grad_norm": 12.664907455444336, + "learning_rate": 1.741951029638005e-05, + "loss": 0.6567, + "step": 41150 + }, + { + "epoch": 2.0598023270361567, + "grad_norm": 0.003944510594010353, + "learning_rate": 1.741024262756946e-05, + "loss": 0.3663, + "step": 41160 + }, + { + "epoch": 2.0603027649193044, + "grad_norm": 4.230157852172852, + "learning_rate": 1.7400974958758877e-05, + "loss": 0.2224, + "step": 41170 + }, + { + "epoch": 2.060803202802452, + "grad_norm": 0.023941220715641975, + "learning_rate": 1.7391707289948286e-05, + "loss": 0.057, + "step": 41180 + }, + { + "epoch": 2.0613036406856, + "grad_norm": 0.15843987464904785, + "learning_rate": 1.73824396211377e-05, + "loss": 0.3322, + "step": 41190 + }, + { + "epoch": 2.0618040785687475, + "grad_norm": 26.0657901763916, + "learning_rate": 1.7373171952327114e-05, + "loss": 0.2586, + "step": 41200 + }, + { + "epoch": 2.0623045164518956, + "grad_norm": 88.1386947631836, + "learning_rate": 1.7363904283516526e-05, + "loss": 0.3918, + "step": 41210 + }, + { + "epoch": 2.0628049543350433, + "grad_norm": 2.5758790969848633, + "learning_rate": 1.7354636614705938e-05, + "loss": 0.3144, + "step": 41220 + }, + { + "epoch": 2.063305392218191, + "grad_norm": 0.21283535659313202, + "learning_rate": 1.734536894589535e-05, + "loss": 0.0951, + "step": 41230 + }, + { + "epoch": 2.0638058301013387, + "grad_norm": 0.19976390898227692, + "learning_rate": 1.7336101277084763e-05, + "loss": 0.1408, + "step": 41240 + }, + { + "epoch": 2.0643062679844864, + "grad_norm": 32.5446891784668, + "learning_rate": 1.7326833608274175e-05, + "loss": 0.4538, + "step": 41250 + }, + { + "epoch": 2.064806705867634, + "grad_norm": 12.253084182739258, + "learning_rate": 1.7317565939463587e-05, + "loss": 0.031, + "step": 41260 + }, + { + "epoch": 2.0653071437507817, + "grad_norm": 22.277263641357422, + "learning_rate": 1.7308298270653e-05, + "loss": 0.2714, + "step": 41270 + }, + { + "epoch": 2.06580758163393, + "grad_norm": 74.0071029663086, + "learning_rate": 1.7299030601842415e-05, + "loss": 0.1844, + "step": 41280 + }, + { + "epoch": 2.0663080195170775, + "grad_norm": 0.14201827347278595, + "learning_rate": 1.7289762933031824e-05, + "loss": 0.3728, + "step": 41290 + }, + { + "epoch": 2.0668084574002252, + "grad_norm": 0.2623313069343567, + "learning_rate": 1.728049526422124e-05, + "loss": 0.1296, + "step": 41300 + }, + { + "epoch": 2.067308895283373, + "grad_norm": 56.204837799072266, + "learning_rate": 1.7271227595410652e-05, + "loss": 0.5732, + "step": 41310 + }, + { + "epoch": 2.0678093331665206, + "grad_norm": 0.004416189156472683, + "learning_rate": 1.7261959926600064e-05, + "loss": 0.4985, + "step": 41320 + }, + { + "epoch": 2.0683097710496683, + "grad_norm": 0.06369306147098541, + "learning_rate": 1.7252692257789476e-05, + "loss": 0.1674, + "step": 41330 + }, + { + "epoch": 2.0688102089328164, + "grad_norm": 3.736593723297119, + "learning_rate": 1.7243424588978892e-05, + "loss": 0.0656, + "step": 41340 + }, + { + "epoch": 2.069310646815964, + "grad_norm": 131.93299865722656, + "learning_rate": 1.72341569201683e-05, + "loss": 0.1665, + "step": 41350 + }, + { + "epoch": 2.069811084699112, + "grad_norm": 31.923274993896484, + "learning_rate": 1.7224889251357713e-05, + "loss": 0.6633, + "step": 41360 + }, + { + "epoch": 2.0703115225822595, + "grad_norm": 151.2386932373047, + "learning_rate": 1.7215621582547125e-05, + "loss": 0.4985, + "step": 41370 + }, + { + "epoch": 2.070811960465407, + "grad_norm": 38.63837814331055, + "learning_rate": 1.7206353913736538e-05, + "loss": 0.735, + "step": 41380 + }, + { + "epoch": 2.071312398348555, + "grad_norm": 51.38584899902344, + "learning_rate": 1.7197086244925953e-05, + "loss": 0.0691, + "step": 41390 + }, + { + "epoch": 2.0718128362317025, + "grad_norm": 9.726037979125977, + "learning_rate": 1.7187818576115362e-05, + "loss": 0.2181, + "step": 41400 + }, + { + "epoch": 2.0723132741148507, + "grad_norm": 211.75538635253906, + "learning_rate": 1.7178550907304778e-05, + "loss": 0.562, + "step": 41410 + }, + { + "epoch": 2.0728137119979984, + "grad_norm": 106.3810806274414, + "learning_rate": 1.716928323849419e-05, + "loss": 0.1694, + "step": 41420 + }, + { + "epoch": 2.073314149881146, + "grad_norm": 76.61376953125, + "learning_rate": 1.7160015569683602e-05, + "loss": 0.1639, + "step": 41430 + }, + { + "epoch": 2.0738145877642937, + "grad_norm": 54.02947235107422, + "learning_rate": 1.7150747900873014e-05, + "loss": 0.586, + "step": 41440 + }, + { + "epoch": 2.0743150256474414, + "grad_norm": 0.7014523148536682, + "learning_rate": 1.714148023206243e-05, + "loss": 0.0296, + "step": 41450 + }, + { + "epoch": 2.074815463530589, + "grad_norm": 16.31682777404785, + "learning_rate": 1.713221256325184e-05, + "loss": 0.5282, + "step": 41460 + }, + { + "epoch": 2.0753159014137372, + "grad_norm": 2.4572389125823975, + "learning_rate": 1.7122944894441255e-05, + "loss": 0.1243, + "step": 41470 + }, + { + "epoch": 2.075816339296885, + "grad_norm": 10.312820434570312, + "learning_rate": 1.7113677225630663e-05, + "loss": 0.2557, + "step": 41480 + }, + { + "epoch": 2.0763167771800326, + "grad_norm": 0.043354202061891556, + "learning_rate": 1.710440955682008e-05, + "loss": 0.2972, + "step": 41490 + }, + { + "epoch": 2.0768172150631803, + "grad_norm": 30.08477020263672, + "learning_rate": 1.709514188800949e-05, + "loss": 0.5563, + "step": 41500 + }, + { + "epoch": 2.077317652946328, + "grad_norm": 23.584028244018555, + "learning_rate": 1.7085874219198904e-05, + "loss": 0.1951, + "step": 41510 + }, + { + "epoch": 2.0778180908294757, + "grad_norm": 0.38288307189941406, + "learning_rate": 1.7076606550388316e-05, + "loss": 0.1459, + "step": 41520 + }, + { + "epoch": 2.0783185287126233, + "grad_norm": 0.035065438598394394, + "learning_rate": 1.7067338881577728e-05, + "loss": 0.3192, + "step": 41530 + }, + { + "epoch": 2.0788189665957715, + "grad_norm": 32.40966033935547, + "learning_rate": 1.705807121276714e-05, + "loss": 0.4769, + "step": 41540 + }, + { + "epoch": 2.079319404478919, + "grad_norm": 77.58768463134766, + "learning_rate": 1.7048803543956553e-05, + "loss": 0.3629, + "step": 41550 + }, + { + "epoch": 2.079819842362067, + "grad_norm": 0.054218851029872894, + "learning_rate": 1.7039535875145968e-05, + "loss": 0.0249, + "step": 41560 + }, + { + "epoch": 2.0803202802452145, + "grad_norm": 0.2511400282382965, + "learning_rate": 1.7030268206335377e-05, + "loss": 0.3964, + "step": 41570 + }, + { + "epoch": 2.0808207181283622, + "grad_norm": 0.004853494465351105, + "learning_rate": 1.7021000537524793e-05, + "loss": 0.0146, + "step": 41580 + }, + { + "epoch": 2.08132115601151, + "grad_norm": 0.6762800812721252, + "learning_rate": 1.7011732868714205e-05, + "loss": 0.0481, + "step": 41590 + }, + { + "epoch": 2.081821593894658, + "grad_norm": 2.833867311477661, + "learning_rate": 1.7002465199903617e-05, + "loss": 0.0369, + "step": 41600 + }, + { + "epoch": 2.0823220317778057, + "grad_norm": 0.44816434383392334, + "learning_rate": 1.699319753109303e-05, + "loss": 0.1936, + "step": 41610 + }, + { + "epoch": 2.0828224696609534, + "grad_norm": 0.0207679383456707, + "learning_rate": 1.6983929862282442e-05, + "loss": 0.0453, + "step": 41620 + }, + { + "epoch": 2.083322907544101, + "grad_norm": 46.70305633544922, + "learning_rate": 1.6974662193471854e-05, + "loss": 0.6432, + "step": 41630 + }, + { + "epoch": 2.083823345427249, + "grad_norm": 5.063952922821045, + "learning_rate": 1.696539452466127e-05, + "loss": 0.1599, + "step": 41640 + }, + { + "epoch": 2.0843237833103965, + "grad_norm": 0.01050836406648159, + "learning_rate": 1.695612685585068e-05, + "loss": 0.5397, + "step": 41650 + }, + { + "epoch": 2.084824221193544, + "grad_norm": 0.08070944994688034, + "learning_rate": 1.6946859187040094e-05, + "loss": 0.1211, + "step": 41660 + }, + { + "epoch": 2.0853246590766923, + "grad_norm": 18.15068817138672, + "learning_rate": 1.6937591518229506e-05, + "loss": 0.3737, + "step": 41670 + }, + { + "epoch": 2.08582509695984, + "grad_norm": 32.95668029785156, + "learning_rate": 1.692832384941892e-05, + "loss": 0.2249, + "step": 41680 + }, + { + "epoch": 2.0863255348429877, + "grad_norm": 0.9353208541870117, + "learning_rate": 1.691905618060833e-05, + "loss": 0.1776, + "step": 41690 + }, + { + "epoch": 2.0868259727261353, + "grad_norm": 0.008900889195501804, + "learning_rate": 1.6909788511797743e-05, + "loss": 0.1741, + "step": 41700 + }, + { + "epoch": 2.087326410609283, + "grad_norm": 0.08964082598686218, + "learning_rate": 1.6900520842987155e-05, + "loss": 0.0835, + "step": 41710 + }, + { + "epoch": 2.0878268484924307, + "grad_norm": 0.03510541841387749, + "learning_rate": 1.6891253174176568e-05, + "loss": 0.3007, + "step": 41720 + }, + { + "epoch": 2.088327286375579, + "grad_norm": 0.02439655363559723, + "learning_rate": 1.688198550536598e-05, + "loss": 0.3839, + "step": 41730 + }, + { + "epoch": 2.0888277242587265, + "grad_norm": 0.6251195073127747, + "learning_rate": 1.6872717836555392e-05, + "loss": 0.3984, + "step": 41740 + }, + { + "epoch": 2.0893281621418742, + "grad_norm": 5.515605926513672, + "learning_rate": 1.6863450167744808e-05, + "loss": 0.3274, + "step": 41750 + }, + { + "epoch": 2.089828600025022, + "grad_norm": 0.04065871611237526, + "learning_rate": 1.6854182498934217e-05, + "loss": 0.3528, + "step": 41760 + }, + { + "epoch": 2.0903290379081696, + "grad_norm": 0.0017845227848738432, + "learning_rate": 1.6844914830123632e-05, + "loss": 0.3367, + "step": 41770 + }, + { + "epoch": 2.0908294757913173, + "grad_norm": 0.004426297266036272, + "learning_rate": 1.6835647161313044e-05, + "loss": 0.0599, + "step": 41780 + }, + { + "epoch": 2.091329913674465, + "grad_norm": 74.76866912841797, + "learning_rate": 1.6826379492502457e-05, + "loss": 0.3573, + "step": 41790 + }, + { + "epoch": 2.091830351557613, + "grad_norm": 12.55582332611084, + "learning_rate": 1.681711182369187e-05, + "loss": 0.3452, + "step": 41800 + }, + { + "epoch": 2.092330789440761, + "grad_norm": 0.055933818221092224, + "learning_rate": 1.6807844154881285e-05, + "loss": 0.2501, + "step": 41810 + }, + { + "epoch": 2.0928312273239085, + "grad_norm": 6.481377601623535, + "learning_rate": 1.6798576486070693e-05, + "loss": 0.3085, + "step": 41820 + }, + { + "epoch": 2.093331665207056, + "grad_norm": 0.01721583493053913, + "learning_rate": 1.678930881726011e-05, + "loss": 0.0719, + "step": 41830 + }, + { + "epoch": 2.093832103090204, + "grad_norm": 164.92823791503906, + "learning_rate": 1.6780041148449518e-05, + "loss": 0.1579, + "step": 41840 + }, + { + "epoch": 2.0943325409733515, + "grad_norm": 2.5871856212615967, + "learning_rate": 1.6770773479638934e-05, + "loss": 0.3734, + "step": 41850 + }, + { + "epoch": 2.094832978856499, + "grad_norm": 0.14298318326473236, + "learning_rate": 1.6761505810828346e-05, + "loss": 0.1828, + "step": 41860 + }, + { + "epoch": 2.0953334167396473, + "grad_norm": 0.008063302375376225, + "learning_rate": 1.6752238142017758e-05, + "loss": 0.1164, + "step": 41870 + }, + { + "epoch": 2.095833854622795, + "grad_norm": 99.90043640136719, + "learning_rate": 1.674297047320717e-05, + "loss": 0.2, + "step": 41880 + }, + { + "epoch": 2.0963342925059427, + "grad_norm": 0.0845799595117569, + "learning_rate": 1.6733702804396583e-05, + "loss": 0.311, + "step": 41890 + }, + { + "epoch": 2.0968347303890904, + "grad_norm": 0.6392174363136292, + "learning_rate": 1.6724435135585995e-05, + "loss": 0.1093, + "step": 41900 + }, + { + "epoch": 2.097335168272238, + "grad_norm": 0.008121958002448082, + "learning_rate": 1.6715167466775407e-05, + "loss": 0.1561, + "step": 41910 + }, + { + "epoch": 2.097835606155386, + "grad_norm": 0.0992736965417862, + "learning_rate": 1.6705899797964823e-05, + "loss": 0.0011, + "step": 41920 + }, + { + "epoch": 2.098336044038534, + "grad_norm": 53.072792053222656, + "learning_rate": 1.669663212915423e-05, + "loss": 0.318, + "step": 41930 + }, + { + "epoch": 2.0988364819216816, + "grad_norm": 101.37235260009766, + "learning_rate": 1.6687364460343647e-05, + "loss": 0.7573, + "step": 41940 + }, + { + "epoch": 2.0993369198048293, + "grad_norm": 0.011879120022058487, + "learning_rate": 1.667809679153306e-05, + "loss": 0.4385, + "step": 41950 + }, + { + "epoch": 2.099837357687977, + "grad_norm": 0.0541607066988945, + "learning_rate": 1.666882912272247e-05, + "loss": 0.5336, + "step": 41960 + }, + { + "epoch": 2.1003377955711247, + "grad_norm": 0.04441867023706436, + "learning_rate": 1.6659561453911884e-05, + "loss": 0.1949, + "step": 41970 + }, + { + "epoch": 2.1008382334542723, + "grad_norm": 0.05135912820696831, + "learning_rate": 1.6650293785101296e-05, + "loss": 0.3422, + "step": 41980 + }, + { + "epoch": 2.10133867133742, + "grad_norm": 0.21128815412521362, + "learning_rate": 1.664102611629071e-05, + "loss": 0.4421, + "step": 41990 + }, + { + "epoch": 2.101839109220568, + "grad_norm": 0.8196788430213928, + "learning_rate": 1.6631758447480124e-05, + "loss": 0.5453, + "step": 42000 + }, + { + "epoch": 2.102339547103716, + "grad_norm": 0.005452276207506657, + "learning_rate": 1.6622490778669533e-05, + "loss": 0.5112, + "step": 42010 + }, + { + "epoch": 2.1028399849868635, + "grad_norm": 0.07556823641061783, + "learning_rate": 1.661322310985895e-05, + "loss": 0.2275, + "step": 42020 + }, + { + "epoch": 2.103340422870011, + "grad_norm": 0.02606341801583767, + "learning_rate": 1.660395544104836e-05, + "loss": 0.2802, + "step": 42030 + }, + { + "epoch": 2.103840860753159, + "grad_norm": 53.23112106323242, + "learning_rate": 1.6594687772237773e-05, + "loss": 0.4616, + "step": 42040 + }, + { + "epoch": 2.1043412986363066, + "grad_norm": 33.94154739379883, + "learning_rate": 1.6585420103427185e-05, + "loss": 0.3759, + "step": 42050 + }, + { + "epoch": 2.1048417365194547, + "grad_norm": 0.0003220072539988905, + "learning_rate": 1.6576152434616597e-05, + "loss": 0.2496, + "step": 42060 + }, + { + "epoch": 2.1053421744026024, + "grad_norm": 0.0024805283173918724, + "learning_rate": 1.656688476580601e-05, + "loss": 0.231, + "step": 42070 + }, + { + "epoch": 2.10584261228575, + "grad_norm": 0.06745032221078873, + "learning_rate": 1.6557617096995422e-05, + "loss": 0.3867, + "step": 42080 + }, + { + "epoch": 2.106343050168898, + "grad_norm": 0.05683799460530281, + "learning_rate": 1.6548349428184834e-05, + "loss": 0.4002, + "step": 42090 + }, + { + "epoch": 2.1068434880520455, + "grad_norm": 0.2626964747905731, + "learning_rate": 1.6539081759374246e-05, + "loss": 0.156, + "step": 42100 + }, + { + "epoch": 2.107343925935193, + "grad_norm": 1.073984146118164, + "learning_rate": 1.6529814090563662e-05, + "loss": 0.0377, + "step": 42110 + }, + { + "epoch": 2.107844363818341, + "grad_norm": 29.017860412597656, + "learning_rate": 1.652054642175307e-05, + "loss": 0.4219, + "step": 42120 + }, + { + "epoch": 2.108344801701489, + "grad_norm": 101.26155090332031, + "learning_rate": 1.6511278752942487e-05, + "loss": 0.3997, + "step": 42130 + }, + { + "epoch": 2.1088452395846367, + "grad_norm": 0.0005412331665866077, + "learning_rate": 1.65020110841319e-05, + "loss": 0.3282, + "step": 42140 + }, + { + "epoch": 2.1093456774677843, + "grad_norm": 0.015363292768597603, + "learning_rate": 1.649274341532131e-05, + "loss": 0.2539, + "step": 42150 + }, + { + "epoch": 2.109846115350932, + "grad_norm": 36.90705490112305, + "learning_rate": 1.6483475746510723e-05, + "loss": 0.5786, + "step": 42160 + }, + { + "epoch": 2.1103465532340797, + "grad_norm": 51.79396057128906, + "learning_rate": 1.647420807770014e-05, + "loss": 0.4477, + "step": 42170 + }, + { + "epoch": 2.1108469911172274, + "grad_norm": 157.76930236816406, + "learning_rate": 1.6464940408889548e-05, + "loss": 0.643, + "step": 42180 + }, + { + "epoch": 2.1113474290003755, + "grad_norm": 92.49345397949219, + "learning_rate": 1.6455672740078963e-05, + "loss": 0.1753, + "step": 42190 + }, + { + "epoch": 2.111847866883523, + "grad_norm": 17.29288101196289, + "learning_rate": 1.6446405071268372e-05, + "loss": 0.4695, + "step": 42200 + }, + { + "epoch": 2.112348304766671, + "grad_norm": 6.6435747146606445, + "learning_rate": 1.6437137402457785e-05, + "loss": 0.3731, + "step": 42210 + }, + { + "epoch": 2.1128487426498186, + "grad_norm": 12.384690284729004, + "learning_rate": 1.64278697336472e-05, + "loss": 0.5606, + "step": 42220 + }, + { + "epoch": 2.1133491805329663, + "grad_norm": 0.21255692839622498, + "learning_rate": 1.641860206483661e-05, + "loss": 0.4191, + "step": 42230 + }, + { + "epoch": 2.113849618416114, + "grad_norm": 25.09225082397461, + "learning_rate": 1.6409334396026025e-05, + "loss": 0.4386, + "step": 42240 + }, + { + "epoch": 2.1143500562992616, + "grad_norm": 0.36174046993255615, + "learning_rate": 1.6400066727215437e-05, + "loss": 0.288, + "step": 42250 + }, + { + "epoch": 2.11485049418241, + "grad_norm": 29.42274284362793, + "learning_rate": 1.639079905840485e-05, + "loss": 0.5338, + "step": 42260 + }, + { + "epoch": 2.1153509320655575, + "grad_norm": 0.006259406451135874, + "learning_rate": 1.638153138959426e-05, + "loss": 0.1149, + "step": 42270 + }, + { + "epoch": 2.115851369948705, + "grad_norm": 23.08921241760254, + "learning_rate": 1.6372263720783677e-05, + "loss": 0.1511, + "step": 42280 + }, + { + "epoch": 2.116351807831853, + "grad_norm": 0.12182018160820007, + "learning_rate": 1.6362996051973086e-05, + "loss": 0.2842, + "step": 42290 + }, + { + "epoch": 2.1168522457150005, + "grad_norm": 0.10659777373075485, + "learning_rate": 1.63537283831625e-05, + "loss": 0.1545, + "step": 42300 + }, + { + "epoch": 2.117352683598148, + "grad_norm": 0.742087721824646, + "learning_rate": 1.6344460714351914e-05, + "loss": 0.0338, + "step": 42310 + }, + { + "epoch": 2.1178531214812963, + "grad_norm": 62.307403564453125, + "learning_rate": 1.6335193045541326e-05, + "loss": 0.5526, + "step": 42320 + }, + { + "epoch": 2.118353559364444, + "grad_norm": 18.4753360748291, + "learning_rate": 1.6325925376730738e-05, + "loss": 0.5186, + "step": 42330 + }, + { + "epoch": 2.1188539972475917, + "grad_norm": 47.794349670410156, + "learning_rate": 1.631665770792015e-05, + "loss": 0.2766, + "step": 42340 + }, + { + "epoch": 2.1193544351307394, + "grad_norm": 0.2705235183238983, + "learning_rate": 1.6307390039109563e-05, + "loss": 0.3207, + "step": 42350 + }, + { + "epoch": 2.119854873013887, + "grad_norm": 0.13273899257183075, + "learning_rate": 1.6298122370298975e-05, + "loss": 0.2408, + "step": 42360 + }, + { + "epoch": 2.1203553108970348, + "grad_norm": 19.757078170776367, + "learning_rate": 1.6288854701488387e-05, + "loss": 0.3571, + "step": 42370 + }, + { + "epoch": 2.1208557487801825, + "grad_norm": 7.831366539001465, + "learning_rate": 1.62795870326778e-05, + "loss": 0.2595, + "step": 42380 + }, + { + "epoch": 2.1213561866633306, + "grad_norm": 0.225729838013649, + "learning_rate": 1.6270319363867215e-05, + "loss": 0.3939, + "step": 42390 + }, + { + "epoch": 2.1218566245464783, + "grad_norm": 0.03716771677136421, + "learning_rate": 1.6261051695056624e-05, + "loss": 0.5303, + "step": 42400 + }, + { + "epoch": 2.122357062429626, + "grad_norm": 0.010199188254773617, + "learning_rate": 1.625178402624604e-05, + "loss": 0.5171, + "step": 42410 + }, + { + "epoch": 2.1228575003127736, + "grad_norm": 12.689889907836914, + "learning_rate": 1.6242516357435452e-05, + "loss": 0.398, + "step": 42420 + }, + { + "epoch": 2.1233579381959213, + "grad_norm": 0.12840276956558228, + "learning_rate": 1.6233248688624864e-05, + "loss": 0.1461, + "step": 42430 + }, + { + "epoch": 2.123858376079069, + "grad_norm": 0.035494960844516754, + "learning_rate": 1.6223981019814276e-05, + "loss": 0.3535, + "step": 42440 + }, + { + "epoch": 2.124358813962217, + "grad_norm": 0.007985730655491352, + "learning_rate": 1.621471335100369e-05, + "loss": 0.383, + "step": 42450 + }, + { + "epoch": 2.124859251845365, + "grad_norm": 3.560713052749634, + "learning_rate": 1.62054456821931e-05, + "loss": 0.3844, + "step": 42460 + }, + { + "epoch": 2.1253596897285125, + "grad_norm": 0.0971754714846611, + "learning_rate": 1.6196178013382516e-05, + "loss": 0.1856, + "step": 42470 + }, + { + "epoch": 2.12586012761166, + "grad_norm": 0.05653767287731171, + "learning_rate": 1.6186910344571925e-05, + "loss": 0.1947, + "step": 42480 + }, + { + "epoch": 2.126360565494808, + "grad_norm": 0.18255826830863953, + "learning_rate": 1.617764267576134e-05, + "loss": 0.2805, + "step": 42490 + }, + { + "epoch": 2.1268610033779556, + "grad_norm": 0.09389536082744598, + "learning_rate": 1.6168375006950753e-05, + "loss": 0.0068, + "step": 42500 + }, + { + "epoch": 2.1273614412611033, + "grad_norm": 55.73891067504883, + "learning_rate": 1.6159107338140165e-05, + "loss": 0.4257, + "step": 42510 + }, + { + "epoch": 2.1278618791442514, + "grad_norm": 40.96942138671875, + "learning_rate": 1.6149839669329578e-05, + "loss": 0.4026, + "step": 42520 + }, + { + "epoch": 2.128362317027399, + "grad_norm": 0.049225449562072754, + "learning_rate": 1.614057200051899e-05, + "loss": 0.2563, + "step": 42530 + }, + { + "epoch": 2.1288627549105468, + "grad_norm": 0.04504162818193436, + "learning_rate": 1.6131304331708402e-05, + "loss": 0.1068, + "step": 42540 + }, + { + "epoch": 2.1293631927936945, + "grad_norm": 3.5239040851593018, + "learning_rate": 1.6122036662897814e-05, + "loss": 0.076, + "step": 42550 + }, + { + "epoch": 2.129863630676842, + "grad_norm": 0.06804395467042923, + "learning_rate": 1.6112768994087227e-05, + "loss": 0.2688, + "step": 42560 + }, + { + "epoch": 2.13036406855999, + "grad_norm": 1.0224480628967285, + "learning_rate": 1.610350132527664e-05, + "loss": 0.5545, + "step": 42570 + }, + { + "epoch": 2.130864506443138, + "grad_norm": 0.24175766110420227, + "learning_rate": 1.6094233656466055e-05, + "loss": 0.1371, + "step": 42580 + }, + { + "epoch": 2.1313649443262856, + "grad_norm": 7.249225616455078, + "learning_rate": 1.6084965987655463e-05, + "loss": 0.0674, + "step": 42590 + }, + { + "epoch": 2.1318653822094333, + "grad_norm": 0.031010715290904045, + "learning_rate": 1.607569831884488e-05, + "loss": 0.1953, + "step": 42600 + }, + { + "epoch": 2.132365820092581, + "grad_norm": 0.624393105506897, + "learning_rate": 1.606643065003429e-05, + "loss": 0.2487, + "step": 42610 + }, + { + "epoch": 2.1328662579757287, + "grad_norm": 206.1861572265625, + "learning_rate": 1.6057162981223704e-05, + "loss": 0.3961, + "step": 42620 + }, + { + "epoch": 2.1333666958588764, + "grad_norm": 18.0457706451416, + "learning_rate": 1.6047895312413116e-05, + "loss": 0.2615, + "step": 42630 + }, + { + "epoch": 2.133867133742024, + "grad_norm": 0.020481456071138382, + "learning_rate": 1.603862764360253e-05, + "loss": 0.149, + "step": 42640 + }, + { + "epoch": 2.134367571625172, + "grad_norm": 0.812816858291626, + "learning_rate": 1.602935997479194e-05, + "loss": 0.0522, + "step": 42650 + }, + { + "epoch": 2.13486800950832, + "grad_norm": 0.05264083296060562, + "learning_rate": 1.6020092305981356e-05, + "loss": 0.1006, + "step": 42660 + }, + { + "epoch": 2.1353684473914676, + "grad_norm": 30.325477600097656, + "learning_rate": 1.6010824637170768e-05, + "loss": 0.63, + "step": 42670 + }, + { + "epoch": 2.1358688852746153, + "grad_norm": 0.06954088807106018, + "learning_rate": 1.600155696836018e-05, + "loss": 0.4189, + "step": 42680 + }, + { + "epoch": 2.136369323157763, + "grad_norm": 0.06339949369430542, + "learning_rate": 1.5992289299549593e-05, + "loss": 0.0491, + "step": 42690 + }, + { + "epoch": 2.1368697610409106, + "grad_norm": 0.7542130947113037, + "learning_rate": 1.5983021630739005e-05, + "loss": 0.0018, + "step": 42700 + }, + { + "epoch": 2.1373701989240583, + "grad_norm": 41.0650520324707, + "learning_rate": 1.5973753961928417e-05, + "loss": 0.4361, + "step": 42710 + }, + { + "epoch": 2.1378706368072065, + "grad_norm": 0.0015472650993615389, + "learning_rate": 1.596448629311783e-05, + "loss": 0.2581, + "step": 42720 + }, + { + "epoch": 2.138371074690354, + "grad_norm": 0.09102800488471985, + "learning_rate": 1.595521862430724e-05, + "loss": 0.2093, + "step": 42730 + }, + { + "epoch": 2.138871512573502, + "grad_norm": 53.93872833251953, + "learning_rate": 1.5945950955496654e-05, + "loss": 0.652, + "step": 42740 + }, + { + "epoch": 2.1393719504566495, + "grad_norm": 9.166760444641113, + "learning_rate": 1.593668328668607e-05, + "loss": 0.2107, + "step": 42750 + }, + { + "epoch": 2.139872388339797, + "grad_norm": 1.2169941663742065, + "learning_rate": 1.592741561787548e-05, + "loss": 0.1766, + "step": 42760 + }, + { + "epoch": 2.140372826222945, + "grad_norm": 0.30396318435668945, + "learning_rate": 1.5918147949064894e-05, + "loss": 0.515, + "step": 42770 + }, + { + "epoch": 2.140873264106093, + "grad_norm": 0.10006586462259293, + "learning_rate": 1.5908880280254306e-05, + "loss": 0.3261, + "step": 42780 + }, + { + "epoch": 2.1413737019892407, + "grad_norm": 0.015381998382508755, + "learning_rate": 1.589961261144372e-05, + "loss": 0.1284, + "step": 42790 + }, + { + "epoch": 2.1418741398723884, + "grad_norm": 0.009794034995138645, + "learning_rate": 1.589034494263313e-05, + "loss": 0.527, + "step": 42800 + }, + { + "epoch": 2.142374577755536, + "grad_norm": 7.208525657653809, + "learning_rate": 1.5881077273822543e-05, + "loss": 0.3455, + "step": 42810 + }, + { + "epoch": 2.1428750156386838, + "grad_norm": 11.619349479675293, + "learning_rate": 1.5871809605011955e-05, + "loss": 0.1449, + "step": 42820 + }, + { + "epoch": 2.1433754535218315, + "grad_norm": 0.33725085854530334, + "learning_rate": 1.586254193620137e-05, + "loss": 0.2987, + "step": 42830 + }, + { + "epoch": 2.1438758914049796, + "grad_norm": 41.4730224609375, + "learning_rate": 1.585327426739078e-05, + "loss": 0.5628, + "step": 42840 + }, + { + "epoch": 2.1443763292881273, + "grad_norm": 0.23227569460868835, + "learning_rate": 1.5844006598580195e-05, + "loss": 0.0182, + "step": 42850 + }, + { + "epoch": 2.144876767171275, + "grad_norm": 48.076534271240234, + "learning_rate": 1.5834738929769608e-05, + "loss": 0.3784, + "step": 42860 + }, + { + "epoch": 2.1453772050544226, + "grad_norm": 0.003831930225715041, + "learning_rate": 1.582547126095902e-05, + "loss": 0.2963, + "step": 42870 + }, + { + "epoch": 2.1458776429375703, + "grad_norm": 72.7916030883789, + "learning_rate": 1.5816203592148432e-05, + "loss": 0.7017, + "step": 42880 + }, + { + "epoch": 2.146378080820718, + "grad_norm": 0.49470022320747375, + "learning_rate": 1.5806935923337844e-05, + "loss": 0.0175, + "step": 42890 + }, + { + "epoch": 2.1468785187038657, + "grad_norm": 0.9450995922088623, + "learning_rate": 1.5797668254527257e-05, + "loss": 0.2102, + "step": 42900 + }, + { + "epoch": 2.147378956587014, + "grad_norm": 4.511043071746826, + "learning_rate": 1.578840058571667e-05, + "loss": 0.2864, + "step": 42910 + }, + { + "epoch": 2.1478793944701615, + "grad_norm": 0.022307192906737328, + "learning_rate": 1.577913291690608e-05, + "loss": 0.5044, + "step": 42920 + }, + { + "epoch": 2.148379832353309, + "grad_norm": 0.04127044230699539, + "learning_rate": 1.5769865248095493e-05, + "loss": 0.3739, + "step": 42930 + }, + { + "epoch": 2.148880270236457, + "grad_norm": 39.63435745239258, + "learning_rate": 1.576059757928491e-05, + "loss": 0.7126, + "step": 42940 + }, + { + "epoch": 2.1493807081196046, + "grad_norm": 0.03624890744686127, + "learning_rate": 1.5751329910474318e-05, + "loss": 0.132, + "step": 42950 + }, + { + "epoch": 2.1498811460027523, + "grad_norm": 0.03466876223683357, + "learning_rate": 1.5742062241663733e-05, + "loss": 0.3129, + "step": 42960 + }, + { + "epoch": 2.1503815838859, + "grad_norm": 40.60997772216797, + "learning_rate": 1.5732794572853146e-05, + "loss": 0.174, + "step": 42970 + }, + { + "epoch": 2.150882021769048, + "grad_norm": 129.5880126953125, + "learning_rate": 1.5723526904042558e-05, + "loss": 0.7251, + "step": 42980 + }, + { + "epoch": 2.1513824596521958, + "grad_norm": 26.291051864624023, + "learning_rate": 1.571425923523197e-05, + "loss": 0.1244, + "step": 42990 + }, + { + "epoch": 2.1518828975353435, + "grad_norm": 0.07951514422893524, + "learning_rate": 1.5704991566421386e-05, + "loss": 0.2185, + "step": 43000 + }, + { + "epoch": 2.152383335418491, + "grad_norm": 27.691545486450195, + "learning_rate": 1.5695723897610795e-05, + "loss": 0.5557, + "step": 43010 + }, + { + "epoch": 2.152883773301639, + "grad_norm": 0.0021667093969881535, + "learning_rate": 1.568645622880021e-05, + "loss": 0.1346, + "step": 43020 + }, + { + "epoch": 2.1533842111847865, + "grad_norm": 0.2929399907588959, + "learning_rate": 1.5677188559989623e-05, + "loss": 0.2344, + "step": 43030 + }, + { + "epoch": 2.1538846490679346, + "grad_norm": 0.02148512378334999, + "learning_rate": 1.5667920891179035e-05, + "loss": 0.2131, + "step": 43040 + }, + { + "epoch": 2.1543850869510823, + "grad_norm": 0.4256083071231842, + "learning_rate": 1.5658653222368447e-05, + "loss": 0.1271, + "step": 43050 + }, + { + "epoch": 2.15488552483423, + "grad_norm": 59.98033142089844, + "learning_rate": 1.5649385553557856e-05, + "loss": 0.1853, + "step": 43060 + }, + { + "epoch": 2.1553859627173777, + "grad_norm": 55.75909423828125, + "learning_rate": 1.564011788474727e-05, + "loss": 0.2129, + "step": 43070 + }, + { + "epoch": 2.1558864006005254, + "grad_norm": 66.91263580322266, + "learning_rate": 1.5630850215936684e-05, + "loss": 0.4451, + "step": 43080 + }, + { + "epoch": 2.156386838483673, + "grad_norm": 30.562902450561523, + "learning_rate": 1.5621582547126096e-05, + "loss": 0.653, + "step": 43090 + }, + { + "epoch": 2.1568872763668208, + "grad_norm": 162.46737670898438, + "learning_rate": 1.561231487831551e-05, + "loss": 0.4996, + "step": 43100 + }, + { + "epoch": 2.157387714249969, + "grad_norm": 39.42335891723633, + "learning_rate": 1.5603047209504924e-05, + "loss": 0.413, + "step": 43110 + }, + { + "epoch": 2.1578881521331166, + "grad_norm": 17.953468322753906, + "learning_rate": 1.5593779540694333e-05, + "loss": 0.4629, + "step": 43120 + }, + { + "epoch": 2.1583885900162643, + "grad_norm": 35.219566345214844, + "learning_rate": 1.558451187188375e-05, + "loss": 0.8171, + "step": 43130 + }, + { + "epoch": 2.158889027899412, + "grad_norm": 21.1070499420166, + "learning_rate": 1.557524420307316e-05, + "loss": 0.3878, + "step": 43140 + }, + { + "epoch": 2.1593894657825596, + "grad_norm": 0.7152191996574402, + "learning_rate": 1.5565976534262573e-05, + "loss": 0.3208, + "step": 43150 + }, + { + "epoch": 2.1598899036657073, + "grad_norm": 80.37777709960938, + "learning_rate": 1.5556708865451985e-05, + "loss": 0.2868, + "step": 43160 + }, + { + "epoch": 2.1603903415488555, + "grad_norm": 0.07893281430006027, + "learning_rate": 1.5547441196641397e-05, + "loss": 0.1246, + "step": 43170 + }, + { + "epoch": 2.160890779432003, + "grad_norm": 0.3749285936355591, + "learning_rate": 1.553817352783081e-05, + "loss": 0.5191, + "step": 43180 + }, + { + "epoch": 2.161391217315151, + "grad_norm": 0.15430530905723572, + "learning_rate": 1.5528905859020222e-05, + "loss": 0.1889, + "step": 43190 + }, + { + "epoch": 2.1618916551982985, + "grad_norm": 0.35895660519599915, + "learning_rate": 1.5519638190209634e-05, + "loss": 0.2369, + "step": 43200 + }, + { + "epoch": 2.162392093081446, + "grad_norm": 25.055025100708008, + "learning_rate": 1.5510370521399046e-05, + "loss": 0.2022, + "step": 43210 + }, + { + "epoch": 2.162892530964594, + "grad_norm": 166.9564971923828, + "learning_rate": 1.5501102852588462e-05, + "loss": 0.2987, + "step": 43220 + }, + { + "epoch": 2.1633929688477416, + "grad_norm": 28.642276763916016, + "learning_rate": 1.549183518377787e-05, + "loss": 0.3415, + "step": 43230 + }, + { + "epoch": 2.1638934067308897, + "grad_norm": 0.9292766451835632, + "learning_rate": 1.5482567514967287e-05, + "loss": 0.4494, + "step": 43240 + }, + { + "epoch": 2.1643938446140374, + "grad_norm": 0.19025924801826477, + "learning_rate": 1.54732998461567e-05, + "loss": 0.3231, + "step": 43250 + }, + { + "epoch": 2.164894282497185, + "grad_norm": 0.020142368972301483, + "learning_rate": 1.546403217734611e-05, + "loss": 0.1175, + "step": 43260 + }, + { + "epoch": 2.1653947203803328, + "grad_norm": 21.317852020263672, + "learning_rate": 1.5454764508535523e-05, + "loss": 0.2947, + "step": 43270 + }, + { + "epoch": 2.1658951582634804, + "grad_norm": 28.41099739074707, + "learning_rate": 1.5445496839724936e-05, + "loss": 0.057, + "step": 43280 + }, + { + "epoch": 2.166395596146628, + "grad_norm": 84.4126968383789, + "learning_rate": 1.5436229170914348e-05, + "loss": 0.6504, + "step": 43290 + }, + { + "epoch": 2.166896034029776, + "grad_norm": 0.05469006299972534, + "learning_rate": 1.5426961502103763e-05, + "loss": 0.1482, + "step": 43300 + }, + { + "epoch": 2.167396471912924, + "grad_norm": 0.8217666745185852, + "learning_rate": 1.5417693833293172e-05, + "loss": 0.26, + "step": 43310 + }, + { + "epoch": 2.1678969097960716, + "grad_norm": 0.015195644460618496, + "learning_rate": 1.5408426164482588e-05, + "loss": 0.1734, + "step": 43320 + }, + { + "epoch": 2.1683973476792193, + "grad_norm": 1.062322974205017, + "learning_rate": 1.5399158495672e-05, + "loss": 0.2539, + "step": 43330 + }, + { + "epoch": 2.168897785562367, + "grad_norm": 0.09187749773263931, + "learning_rate": 1.5389890826861412e-05, + "loss": 0.158, + "step": 43340 + }, + { + "epoch": 2.1693982234455147, + "grad_norm": 180.4571990966797, + "learning_rate": 1.5380623158050825e-05, + "loss": 0.0583, + "step": 43350 + }, + { + "epoch": 2.1698986613286624, + "grad_norm": 18.045860290527344, + "learning_rate": 1.5371355489240237e-05, + "loss": 0.0555, + "step": 43360 + }, + { + "epoch": 2.1703990992118105, + "grad_norm": 0.15957686305046082, + "learning_rate": 1.536208782042965e-05, + "loss": 0.0079, + "step": 43370 + }, + { + "epoch": 2.170899537094958, + "grad_norm": 28.831607818603516, + "learning_rate": 1.535282015161906e-05, + "loss": 0.5296, + "step": 43380 + }, + { + "epoch": 2.171399974978106, + "grad_norm": 25.021251678466797, + "learning_rate": 1.5343552482808477e-05, + "loss": 0.3388, + "step": 43390 + }, + { + "epoch": 2.1719004128612536, + "grad_norm": 2.169506072998047, + "learning_rate": 1.5334284813997886e-05, + "loss": 0.2356, + "step": 43400 + }, + { + "epoch": 2.1724008507444013, + "grad_norm": 0.00706471037119627, + "learning_rate": 1.53250171451873e-05, + "loss": 0.156, + "step": 43410 + }, + { + "epoch": 2.172901288627549, + "grad_norm": 11.976855278015137, + "learning_rate": 1.531574947637671e-05, + "loss": 0.3087, + "step": 43420 + }, + { + "epoch": 2.173401726510697, + "grad_norm": 0.01343289390206337, + "learning_rate": 1.5306481807566126e-05, + "loss": 0.4164, + "step": 43430 + }, + { + "epoch": 2.1739021643938448, + "grad_norm": 0.19636638462543488, + "learning_rate": 1.5297214138755538e-05, + "loss": 0.1443, + "step": 43440 + }, + { + "epoch": 2.1744026022769924, + "grad_norm": 0.005055873189121485, + "learning_rate": 1.528794646994495e-05, + "loss": 0.2237, + "step": 43450 + }, + { + "epoch": 2.17490304016014, + "grad_norm": 86.44271850585938, + "learning_rate": 1.5278678801134363e-05, + "loss": 0.1743, + "step": 43460 + }, + { + "epoch": 2.175403478043288, + "grad_norm": 48.36317443847656, + "learning_rate": 1.526941113232378e-05, + "loss": 0.4513, + "step": 43470 + }, + { + "epoch": 2.1759039159264355, + "grad_norm": 0.068246029317379, + "learning_rate": 1.5260143463513187e-05, + "loss": 0.168, + "step": 43480 + }, + { + "epoch": 2.176404353809583, + "grad_norm": 0.007941657677292824, + "learning_rate": 1.5250875794702601e-05, + "loss": 0.3827, + "step": 43490 + }, + { + "epoch": 2.1769047916927313, + "grad_norm": 0.8153774738311768, + "learning_rate": 1.5241608125892015e-05, + "loss": 0.4312, + "step": 43500 + }, + { + "epoch": 2.177405229575879, + "grad_norm": 0.0066399965435266495, + "learning_rate": 1.5232340457081426e-05, + "loss": 0.1896, + "step": 43510 + }, + { + "epoch": 2.1779056674590267, + "grad_norm": 31.326732635498047, + "learning_rate": 1.522307278827084e-05, + "loss": 0.451, + "step": 43520 + }, + { + "epoch": 2.1784061053421744, + "grad_norm": 0.23563729226589203, + "learning_rate": 1.521380511946025e-05, + "loss": 0.4659, + "step": 43530 + }, + { + "epoch": 2.178906543225322, + "grad_norm": 0.003946302458643913, + "learning_rate": 1.5204537450649664e-05, + "loss": 0.379, + "step": 43540 + }, + { + "epoch": 2.1794069811084698, + "grad_norm": 0.1072729080915451, + "learning_rate": 1.5195269781839078e-05, + "loss": 0.134, + "step": 43550 + }, + { + "epoch": 2.1799074189916174, + "grad_norm": 35.350223541259766, + "learning_rate": 1.5186002113028489e-05, + "loss": 0.3694, + "step": 43560 + }, + { + "epoch": 2.1804078568747656, + "grad_norm": 36.63623046875, + "learning_rate": 1.5176734444217903e-05, + "loss": 0.2256, + "step": 43570 + }, + { + "epoch": 2.1809082947579133, + "grad_norm": 0.7552362680435181, + "learning_rate": 1.5167466775407316e-05, + "loss": 0.1162, + "step": 43580 + }, + { + "epoch": 2.181408732641061, + "grad_norm": 45.04484939575195, + "learning_rate": 1.5158199106596727e-05, + "loss": 0.0526, + "step": 43590 + }, + { + "epoch": 2.1819091705242086, + "grad_norm": 53.492671966552734, + "learning_rate": 1.5148931437786141e-05, + "loss": 0.412, + "step": 43600 + }, + { + "epoch": 2.1824096084073563, + "grad_norm": 0.0010419083992019296, + "learning_rate": 1.5139663768975553e-05, + "loss": 0.3787, + "step": 43610 + }, + { + "epoch": 2.182910046290504, + "grad_norm": 39.30768966674805, + "learning_rate": 1.5130396100164965e-05, + "loss": 0.0894, + "step": 43620 + }, + { + "epoch": 2.183410484173652, + "grad_norm": 0.2000689059495926, + "learning_rate": 1.5121128431354378e-05, + "loss": 0.4556, + "step": 43630 + }, + { + "epoch": 2.1839109220568, + "grad_norm": 33.33097839355469, + "learning_rate": 1.511186076254379e-05, + "loss": 0.5457, + "step": 43640 + }, + { + "epoch": 2.1844113599399475, + "grad_norm": 0.00347686605527997, + "learning_rate": 1.5102593093733202e-05, + "loss": 0.0921, + "step": 43650 + }, + { + "epoch": 2.184911797823095, + "grad_norm": 38.55891418457031, + "learning_rate": 1.5093325424922616e-05, + "loss": 0.3899, + "step": 43660 + }, + { + "epoch": 2.185412235706243, + "grad_norm": 52.246646881103516, + "learning_rate": 1.5084057756112027e-05, + "loss": 0.1945, + "step": 43670 + }, + { + "epoch": 2.1859126735893906, + "grad_norm": 0.008213207125663757, + "learning_rate": 1.507479008730144e-05, + "loss": 0.0998, + "step": 43680 + }, + { + "epoch": 2.1864131114725387, + "grad_norm": 24.81956672668457, + "learning_rate": 1.5065522418490855e-05, + "loss": 0.4007, + "step": 43690 + }, + { + "epoch": 2.1869135493556864, + "grad_norm": 0.012843282893300056, + "learning_rate": 1.5056254749680265e-05, + "loss": 0.476, + "step": 43700 + }, + { + "epoch": 2.187413987238834, + "grad_norm": 30.12188720703125, + "learning_rate": 1.5046987080869679e-05, + "loss": 0.4623, + "step": 43710 + }, + { + "epoch": 2.1879144251219818, + "grad_norm": 0.07068731635808945, + "learning_rate": 1.5037719412059093e-05, + "loss": 0.1143, + "step": 43720 + }, + { + "epoch": 2.1884148630051294, + "grad_norm": 2.70628023147583, + "learning_rate": 1.5028451743248504e-05, + "loss": 0.6372, + "step": 43730 + }, + { + "epoch": 2.188915300888277, + "grad_norm": 16.989315032958984, + "learning_rate": 1.5019184074437917e-05, + "loss": 0.2752, + "step": 43740 + }, + { + "epoch": 2.189415738771425, + "grad_norm": 0.2855566740036011, + "learning_rate": 1.5009916405627331e-05, + "loss": 0.1575, + "step": 43750 + }, + { + "epoch": 2.189916176654573, + "grad_norm": 6.749683380126953, + "learning_rate": 1.5000648736816742e-05, + "loss": 0.1868, + "step": 43760 + }, + { + "epoch": 2.1904166145377206, + "grad_norm": 162.66078186035156, + "learning_rate": 1.4991381068006154e-05, + "loss": 0.2462, + "step": 43770 + }, + { + "epoch": 2.1909170524208683, + "grad_norm": 0.009674504399299622, + "learning_rate": 1.4982113399195566e-05, + "loss": 0.4766, + "step": 43780 + }, + { + "epoch": 2.191417490304016, + "grad_norm": 21.856779098510742, + "learning_rate": 1.4972845730384979e-05, + "loss": 0.0785, + "step": 43790 + }, + { + "epoch": 2.1919179281871637, + "grad_norm": 42.50467300415039, + "learning_rate": 1.4963578061574393e-05, + "loss": 0.2892, + "step": 43800 + }, + { + "epoch": 2.1924183660703114, + "grad_norm": 0.0004513936291914433, + "learning_rate": 1.4954310392763803e-05, + "loss": 0.0177, + "step": 43810 + }, + { + "epoch": 2.192918803953459, + "grad_norm": 0.20202696323394775, + "learning_rate": 1.4945042723953217e-05, + "loss": 0.211, + "step": 43820 + }, + { + "epoch": 2.193419241836607, + "grad_norm": 0.07267870754003525, + "learning_rate": 1.4935775055142631e-05, + "loss": 0.4623, + "step": 43830 + }, + { + "epoch": 2.193919679719755, + "grad_norm": 88.64801025390625, + "learning_rate": 1.4926507386332042e-05, + "loss": 0.2373, + "step": 43840 + }, + { + "epoch": 2.1944201176029026, + "grad_norm": 67.66937255859375, + "learning_rate": 1.4917239717521456e-05, + "loss": 0.2983, + "step": 43850 + }, + { + "epoch": 2.1949205554860503, + "grad_norm": 6.578775405883789, + "learning_rate": 1.490797204871087e-05, + "loss": 0.1944, + "step": 43860 + }, + { + "epoch": 2.195420993369198, + "grad_norm": 0.07934346795082092, + "learning_rate": 1.489870437990028e-05, + "loss": 0.1938, + "step": 43870 + }, + { + "epoch": 2.1959214312523456, + "grad_norm": 0.0010100409854203463, + "learning_rate": 1.4889436711089694e-05, + "loss": 0.5897, + "step": 43880 + }, + { + "epoch": 2.1964218691354938, + "grad_norm": 0.3934081196784973, + "learning_rate": 1.4880169042279105e-05, + "loss": 0.069, + "step": 43890 + }, + { + "epoch": 2.1969223070186414, + "grad_norm": 0.004565193317830563, + "learning_rate": 1.4870901373468518e-05, + "loss": 0.2859, + "step": 43900 + }, + { + "epoch": 2.197422744901789, + "grad_norm": 126.90725708007812, + "learning_rate": 1.4861633704657932e-05, + "loss": 0.1659, + "step": 43910 + }, + { + "epoch": 2.197923182784937, + "grad_norm": 148.3427734375, + "learning_rate": 1.4852366035847343e-05, + "loss": 0.3172, + "step": 43920 + }, + { + "epoch": 2.1984236206680845, + "grad_norm": 0.022296426817774773, + "learning_rate": 1.4843098367036757e-05, + "loss": 0.0019, + "step": 43930 + }, + { + "epoch": 2.198924058551232, + "grad_norm": 0.2782004773616791, + "learning_rate": 1.483383069822617e-05, + "loss": 0.2731, + "step": 43940 + }, + { + "epoch": 2.1994244964343803, + "grad_norm": 0.144475057721138, + "learning_rate": 1.4824563029415581e-05, + "loss": 0.0583, + "step": 43950 + }, + { + "epoch": 2.199924934317528, + "grad_norm": 0.10052917897701263, + "learning_rate": 1.4815295360604994e-05, + "loss": 0.1289, + "step": 43960 + }, + { + "epoch": 2.2004253722006757, + "grad_norm": 39.7257080078125, + "learning_rate": 1.4806027691794408e-05, + "loss": 0.3596, + "step": 43970 + }, + { + "epoch": 2.2009258100838234, + "grad_norm": 0.0021053075324743986, + "learning_rate": 1.4796760022983818e-05, + "loss": 0.0575, + "step": 43980 + }, + { + "epoch": 2.201426247966971, + "grad_norm": 45.098358154296875, + "learning_rate": 1.4787492354173232e-05, + "loss": 0.2716, + "step": 43990 + }, + { + "epoch": 2.2019266858501187, + "grad_norm": 0.0022995786275714636, + "learning_rate": 1.4778224685362643e-05, + "loss": 0.2381, + "step": 44000 + }, + { + "epoch": 2.2024271237332664, + "grad_norm": 17.514659881591797, + "learning_rate": 1.4768957016552057e-05, + "loss": 0.1217, + "step": 44010 + }, + { + "epoch": 2.2029275616164146, + "grad_norm": 100.1605453491211, + "learning_rate": 1.475968934774147e-05, + "loss": 0.2388, + "step": 44020 + }, + { + "epoch": 2.2034279994995623, + "grad_norm": 26.516361236572266, + "learning_rate": 1.4750421678930881e-05, + "loss": 0.4374, + "step": 44030 + }, + { + "epoch": 2.20392843738271, + "grad_norm": 4.329495429992676, + "learning_rate": 1.4741154010120295e-05, + "loss": 0.0497, + "step": 44040 + }, + { + "epoch": 2.2044288752658576, + "grad_norm": 39.086212158203125, + "learning_rate": 1.4731886341309709e-05, + "loss": 0.5246, + "step": 44050 + }, + { + "epoch": 2.2049293131490053, + "grad_norm": 0.04828022047877312, + "learning_rate": 1.472261867249912e-05, + "loss": 0.5221, + "step": 44060 + }, + { + "epoch": 2.205429751032153, + "grad_norm": 0.049591951072216034, + "learning_rate": 1.4713351003688533e-05, + "loss": 0.0965, + "step": 44070 + }, + { + "epoch": 2.2059301889153007, + "grad_norm": 0.06370294839143753, + "learning_rate": 1.4704083334877947e-05, + "loss": 0.405, + "step": 44080 + }, + { + "epoch": 2.206430626798449, + "grad_norm": 212.03875732421875, + "learning_rate": 1.4694815666067358e-05, + "loss": 0.1906, + "step": 44090 + }, + { + "epoch": 2.2069310646815965, + "grad_norm": 0.009467987343668938, + "learning_rate": 1.4685547997256772e-05, + "loss": 0.0897, + "step": 44100 + }, + { + "epoch": 2.207431502564744, + "grad_norm": 0.07870233803987503, + "learning_rate": 1.4676280328446184e-05, + "loss": 0.0222, + "step": 44110 + }, + { + "epoch": 2.207931940447892, + "grad_norm": 33.65857696533203, + "learning_rate": 1.4667012659635596e-05, + "loss": 0.2283, + "step": 44120 + }, + { + "epoch": 2.2084323783310396, + "grad_norm": 117.7280044555664, + "learning_rate": 1.4657744990825009e-05, + "loss": 0.2373, + "step": 44130 + }, + { + "epoch": 2.2089328162141872, + "grad_norm": 50.495540618896484, + "learning_rate": 1.4648477322014421e-05, + "loss": 0.6066, + "step": 44140 + }, + { + "epoch": 2.2094332540973354, + "grad_norm": 0.43756726384162903, + "learning_rate": 1.4639209653203833e-05, + "loss": 0.2244, + "step": 44150 + }, + { + "epoch": 2.209933691980483, + "grad_norm": 0.007698375731706619, + "learning_rate": 1.4629941984393247e-05, + "loss": 0.3085, + "step": 44160 + }, + { + "epoch": 2.2104341298636307, + "grad_norm": 16.984188079833984, + "learning_rate": 1.4620674315582658e-05, + "loss": 0.1914, + "step": 44170 + }, + { + "epoch": 2.2109345677467784, + "grad_norm": 41.35210037231445, + "learning_rate": 1.4611406646772072e-05, + "loss": 0.3351, + "step": 44180 + }, + { + "epoch": 2.211435005629926, + "grad_norm": 18.0213565826416, + "learning_rate": 1.4602138977961485e-05, + "loss": 0.1969, + "step": 44190 + }, + { + "epoch": 2.211935443513074, + "grad_norm": 0.01757592149078846, + "learning_rate": 1.4592871309150896e-05, + "loss": 0.0299, + "step": 44200 + }, + { + "epoch": 2.2124358813962215, + "grad_norm": 0.02950543351471424, + "learning_rate": 1.458360364034031e-05, + "loss": 0.5823, + "step": 44210 + }, + { + "epoch": 2.2129363192793696, + "grad_norm": 18.813953399658203, + "learning_rate": 1.4574335971529724e-05, + "loss": 0.3635, + "step": 44220 + }, + { + "epoch": 2.2134367571625173, + "grad_norm": 0.010398794896900654, + "learning_rate": 1.4565068302719134e-05, + "loss": 0.1269, + "step": 44230 + }, + { + "epoch": 2.213937195045665, + "grad_norm": 0.5973718762397766, + "learning_rate": 1.4555800633908548e-05, + "loss": 0.1788, + "step": 44240 + }, + { + "epoch": 2.2144376329288127, + "grad_norm": 94.56025695800781, + "learning_rate": 1.4546532965097959e-05, + "loss": 0.4713, + "step": 44250 + }, + { + "epoch": 2.2149380708119604, + "grad_norm": 0.006951294839382172, + "learning_rate": 1.4537265296287373e-05, + "loss": 0.0511, + "step": 44260 + }, + { + "epoch": 2.215438508695108, + "grad_norm": 0.01695949211716652, + "learning_rate": 1.4527997627476785e-05, + "loss": 0.1351, + "step": 44270 + }, + { + "epoch": 2.215938946578256, + "grad_norm": 16.852970123291016, + "learning_rate": 1.4518729958666197e-05, + "loss": 0.1609, + "step": 44280 + }, + { + "epoch": 2.216439384461404, + "grad_norm": 51.432533264160156, + "learning_rate": 1.450946228985561e-05, + "loss": 0.357, + "step": 44290 + }, + { + "epoch": 2.2169398223445516, + "grad_norm": 14.108671188354492, + "learning_rate": 1.4500194621045024e-05, + "loss": 0.1635, + "step": 44300 + }, + { + "epoch": 2.2174402602276992, + "grad_norm": 27.252653121948242, + "learning_rate": 1.4490926952234434e-05, + "loss": 0.4105, + "step": 44310 + }, + { + "epoch": 2.217940698110847, + "grad_norm": 0.043616414070129395, + "learning_rate": 1.4481659283423848e-05, + "loss": 0.4085, + "step": 44320 + }, + { + "epoch": 2.2184411359939946, + "grad_norm": 0.037259992212057114, + "learning_rate": 1.4472391614613262e-05, + "loss": 0.5539, + "step": 44330 + }, + { + "epoch": 2.2189415738771423, + "grad_norm": 0.08876165002584457, + "learning_rate": 1.4463123945802673e-05, + "loss": 0.0102, + "step": 44340 + }, + { + "epoch": 2.2194420117602904, + "grad_norm": 0.035416923463344574, + "learning_rate": 1.4453856276992087e-05, + "loss": 0.0527, + "step": 44350 + }, + { + "epoch": 2.219942449643438, + "grad_norm": 0.0008480912656523287, + "learning_rate": 1.4444588608181497e-05, + "loss": 0.4687, + "step": 44360 + }, + { + "epoch": 2.220442887526586, + "grad_norm": 0.005928483791649342, + "learning_rate": 1.4435320939370911e-05, + "loss": 0.4318, + "step": 44370 + }, + { + "epoch": 2.2209433254097335, + "grad_norm": 2.397148370742798, + "learning_rate": 1.4426053270560325e-05, + "loss": 0.2395, + "step": 44380 + }, + { + "epoch": 2.221443763292881, + "grad_norm": 0.009704225696623325, + "learning_rate": 1.4416785601749735e-05, + "loss": 0.047, + "step": 44390 + }, + { + "epoch": 2.221944201176029, + "grad_norm": 0.051907721906900406, + "learning_rate": 1.440751793293915e-05, + "loss": 0.0047, + "step": 44400 + }, + { + "epoch": 2.2224446390591766, + "grad_norm": 0.036177121102809906, + "learning_rate": 1.4398250264128563e-05, + "loss": 0.3677, + "step": 44410 + }, + { + "epoch": 2.2229450769423247, + "grad_norm": 0.0012542648473754525, + "learning_rate": 1.4388982595317974e-05, + "loss": 0.3033, + "step": 44420 + }, + { + "epoch": 2.2234455148254724, + "grad_norm": 0.0024243963416665792, + "learning_rate": 1.4379714926507388e-05, + "loss": 0.3047, + "step": 44430 + }, + { + "epoch": 2.22394595270862, + "grad_norm": 0.5606967806816101, + "learning_rate": 1.43704472576968e-05, + "loss": 0.5096, + "step": 44440 + }, + { + "epoch": 2.2244463905917677, + "grad_norm": 0.007200438994914293, + "learning_rate": 1.4361179588886212e-05, + "loss": 0.1184, + "step": 44450 + }, + { + "epoch": 2.2249468284749154, + "grad_norm": 18.30154037475586, + "learning_rate": 1.4351911920075625e-05, + "loss": 0.4841, + "step": 44460 + }, + { + "epoch": 2.225447266358063, + "grad_norm": 3.460865020751953, + "learning_rate": 1.4342644251265039e-05, + "loss": 1.012, + "step": 44470 + }, + { + "epoch": 2.2259477042412112, + "grad_norm": 0.018851377069950104, + "learning_rate": 1.4333376582454449e-05, + "loss": 0.1828, + "step": 44480 + }, + { + "epoch": 2.226448142124359, + "grad_norm": 116.6839828491211, + "learning_rate": 1.4324108913643863e-05, + "loss": 0.1664, + "step": 44490 + }, + { + "epoch": 2.2269485800075066, + "grad_norm": 47.037235260009766, + "learning_rate": 1.4314841244833274e-05, + "loss": 0.1222, + "step": 44500 + }, + { + "epoch": 2.2274490178906543, + "grad_norm": 0.20525795221328735, + "learning_rate": 1.4305573576022688e-05, + "loss": 0.1756, + "step": 44510 + }, + { + "epoch": 2.227949455773802, + "grad_norm": 83.3775863647461, + "learning_rate": 1.4296305907212101e-05, + "loss": 0.1111, + "step": 44520 + }, + { + "epoch": 2.2284498936569497, + "grad_norm": 0.4683445394039154, + "learning_rate": 1.4287038238401512e-05, + "loss": 0.0501, + "step": 44530 + }, + { + "epoch": 2.228950331540098, + "grad_norm": 45.115848541259766, + "learning_rate": 1.4277770569590926e-05, + "loss": 0.3561, + "step": 44540 + }, + { + "epoch": 2.2294507694232455, + "grad_norm": 0.09286727756261826, + "learning_rate": 1.426850290078034e-05, + "loss": 0.2915, + "step": 44550 + }, + { + "epoch": 2.229951207306393, + "grad_norm": 72.566650390625, + "learning_rate": 1.425923523196975e-05, + "loss": 0.3559, + "step": 44560 + }, + { + "epoch": 2.230451645189541, + "grad_norm": 1.1605892181396484, + "learning_rate": 1.4249967563159164e-05, + "loss": 0.1452, + "step": 44570 + }, + { + "epoch": 2.2309520830726886, + "grad_norm": 65.74857330322266, + "learning_rate": 1.4240699894348578e-05, + "loss": 0.5119, + "step": 44580 + }, + { + "epoch": 2.2314525209558362, + "grad_norm": 0.06521125882863998, + "learning_rate": 1.4231432225537989e-05, + "loss": 0.0093, + "step": 44590 + }, + { + "epoch": 2.231952958838984, + "grad_norm": 0.00927987415343523, + "learning_rate": 1.4222164556727403e-05, + "loss": 0.2812, + "step": 44600 + }, + { + "epoch": 2.232453396722132, + "grad_norm": 0.011272822506725788, + "learning_rate": 1.4212896887916813e-05, + "loss": 0.3221, + "step": 44610 + }, + { + "epoch": 2.2329538346052797, + "grad_norm": 0.679553210735321, + "learning_rate": 1.4203629219106226e-05, + "loss": 0.256, + "step": 44620 + }, + { + "epoch": 2.2334542724884274, + "grad_norm": 0.021406354382634163, + "learning_rate": 1.419436155029564e-05, + "loss": 0.257, + "step": 44630 + }, + { + "epoch": 2.233954710371575, + "grad_norm": 0.029126156121492386, + "learning_rate": 1.418509388148505e-05, + "loss": 0.3188, + "step": 44640 + }, + { + "epoch": 2.234455148254723, + "grad_norm": 0.0005510944174602628, + "learning_rate": 1.4175826212674464e-05, + "loss": 0.0066, + "step": 44650 + }, + { + "epoch": 2.2349555861378705, + "grad_norm": 1.1248005628585815, + "learning_rate": 1.4166558543863878e-05, + "loss": 0.7025, + "step": 44660 + }, + { + "epoch": 2.235456024021018, + "grad_norm": 0.007466370239853859, + "learning_rate": 1.4157290875053289e-05, + "loss": 0.0688, + "step": 44670 + }, + { + "epoch": 2.2359564619041663, + "grad_norm": 171.51119995117188, + "learning_rate": 1.4148023206242702e-05, + "loss": 0.1182, + "step": 44680 + }, + { + "epoch": 2.236456899787314, + "grad_norm": 0.9446596503257751, + "learning_rate": 1.4138755537432116e-05, + "loss": 0.4121, + "step": 44690 + }, + { + "epoch": 2.2369573376704617, + "grad_norm": 68.35132598876953, + "learning_rate": 1.4129487868621527e-05, + "loss": 0.1346, + "step": 44700 + }, + { + "epoch": 2.2374577755536094, + "grad_norm": 0.007891790941357613, + "learning_rate": 1.4120220199810941e-05, + "loss": 0.2062, + "step": 44710 + }, + { + "epoch": 2.237958213436757, + "grad_norm": 16.862375259399414, + "learning_rate": 1.4110952531000351e-05, + "loss": 0.3873, + "step": 44720 + }, + { + "epoch": 2.2384586513199047, + "grad_norm": 0.014194377698004246, + "learning_rate": 1.4101684862189765e-05, + "loss": 0.395, + "step": 44730 + }, + { + "epoch": 2.238959089203053, + "grad_norm": 24.388084411621094, + "learning_rate": 1.409241719337918e-05, + "loss": 0.1603, + "step": 44740 + }, + { + "epoch": 2.2394595270862006, + "grad_norm": 0.023043088614940643, + "learning_rate": 1.408314952456859e-05, + "loss": 0.1997, + "step": 44750 + }, + { + "epoch": 2.2399599649693482, + "grad_norm": 330.24737548828125, + "learning_rate": 1.4073881855758004e-05, + "loss": 0.8175, + "step": 44760 + }, + { + "epoch": 2.240460402852496, + "grad_norm": 0.021115800365805626, + "learning_rate": 1.4064614186947416e-05, + "loss": 0.4728, + "step": 44770 + }, + { + "epoch": 2.2409608407356436, + "grad_norm": 78.48406219482422, + "learning_rate": 1.4055346518136828e-05, + "loss": 0.2139, + "step": 44780 + }, + { + "epoch": 2.2414612786187913, + "grad_norm": 0.01454553846269846, + "learning_rate": 1.404607884932624e-05, + "loss": 0.0991, + "step": 44790 + }, + { + "epoch": 2.2419617165019394, + "grad_norm": 0.025081733241677284, + "learning_rate": 1.4036811180515655e-05, + "loss": 0.2275, + "step": 44800 + }, + { + "epoch": 2.242462154385087, + "grad_norm": 32.09824752807617, + "learning_rate": 1.4027543511705065e-05, + "loss": 0.2589, + "step": 44810 + }, + { + "epoch": 2.242962592268235, + "grad_norm": 22.3238582611084, + "learning_rate": 1.4018275842894479e-05, + "loss": 0.2309, + "step": 44820 + }, + { + "epoch": 2.2434630301513825, + "grad_norm": 0.6235008835792542, + "learning_rate": 1.4009008174083893e-05, + "loss": 0.1434, + "step": 44830 + }, + { + "epoch": 2.24396346803453, + "grad_norm": 1.8567147254943848, + "learning_rate": 1.3999740505273303e-05, + "loss": 0.1839, + "step": 44840 + }, + { + "epoch": 2.244463905917678, + "grad_norm": 0.09062235057353973, + "learning_rate": 1.3990472836462717e-05, + "loss": 0.0518, + "step": 44850 + }, + { + "epoch": 2.2449643438008255, + "grad_norm": 0.0030092066153883934, + "learning_rate": 1.3981205167652128e-05, + "loss": 0.1905, + "step": 44860 + }, + { + "epoch": 2.2454647816839737, + "grad_norm": 1.4141321182250977, + "learning_rate": 1.3971937498841542e-05, + "loss": 0.1308, + "step": 44870 + }, + { + "epoch": 2.2459652195671214, + "grad_norm": 1.3289310932159424, + "learning_rate": 1.3962669830030956e-05, + "loss": 0.2769, + "step": 44880 + }, + { + "epoch": 2.246465657450269, + "grad_norm": 24.6469783782959, + "learning_rate": 1.3953402161220366e-05, + "loss": 0.6161, + "step": 44890 + }, + { + "epoch": 2.2469660953334167, + "grad_norm": 0.2823737561702728, + "learning_rate": 1.394413449240978e-05, + "loss": 0.2726, + "step": 44900 + }, + { + "epoch": 2.2474665332165644, + "grad_norm": 0.11936521530151367, + "learning_rate": 1.3934866823599194e-05, + "loss": 0.2513, + "step": 44910 + }, + { + "epoch": 2.247966971099712, + "grad_norm": 19.864486694335938, + "learning_rate": 1.3925599154788605e-05, + "loss": 0.1404, + "step": 44920 + }, + { + "epoch": 2.24846740898286, + "grad_norm": 134.16697692871094, + "learning_rate": 1.3916331485978019e-05, + "loss": 0.4846, + "step": 44930 + }, + { + "epoch": 2.248967846866008, + "grad_norm": 30.209741592407227, + "learning_rate": 1.3907063817167431e-05, + "loss": 0.2094, + "step": 44940 + }, + { + "epoch": 2.2494682847491556, + "grad_norm": 50.358089447021484, + "learning_rate": 1.3897796148356843e-05, + "loss": 0.5026, + "step": 44950 + }, + { + "epoch": 2.2499687226323033, + "grad_norm": 0.011421519331634045, + "learning_rate": 1.3888528479546256e-05, + "loss": 0.3078, + "step": 44960 + }, + { + "epoch": 2.250469160515451, + "grad_norm": 0.0013304197927936912, + "learning_rate": 1.3879260810735668e-05, + "loss": 0.3208, + "step": 44970 + }, + { + "epoch": 2.2509695983985987, + "grad_norm": 65.94198608398438, + "learning_rate": 1.386999314192508e-05, + "loss": 0.3813, + "step": 44980 + }, + { + "epoch": 2.2514700362817464, + "grad_norm": 2.573643922805786, + "learning_rate": 1.3860725473114494e-05, + "loss": 0.1211, + "step": 44990 + }, + { + "epoch": 2.251970474164894, + "grad_norm": 7.698556423187256, + "learning_rate": 1.3851457804303905e-05, + "loss": 0.2533, + "step": 45000 + }, + { + "epoch": 2.252470912048042, + "grad_norm": 57.38174819946289, + "learning_rate": 1.3842190135493318e-05, + "loss": 0.6335, + "step": 45010 + }, + { + "epoch": 2.25297134993119, + "grad_norm": 35.37452697753906, + "learning_rate": 1.3832922466682732e-05, + "loss": 0.1572, + "step": 45020 + }, + { + "epoch": 2.2534717878143375, + "grad_norm": 0.2041410356760025, + "learning_rate": 1.3823654797872143e-05, + "loss": 0.3561, + "step": 45030 + }, + { + "epoch": 2.2539722256974852, + "grad_norm": 0.011455492116510868, + "learning_rate": 1.3814387129061557e-05, + "loss": 0.4566, + "step": 45040 + }, + { + "epoch": 2.254472663580633, + "grad_norm": 0.0013321515871211886, + "learning_rate": 1.380511946025097e-05, + "loss": 0.0744, + "step": 45050 + }, + { + "epoch": 2.254973101463781, + "grad_norm": 0.4941001534461975, + "learning_rate": 1.3795851791440381e-05, + "loss": 0.0375, + "step": 45060 + }, + { + "epoch": 2.2554735393469287, + "grad_norm": 41.816741943359375, + "learning_rate": 1.3786584122629795e-05, + "loss": 0.7687, + "step": 45070 + }, + { + "epoch": 2.2559739772300764, + "grad_norm": 0.007816538214683533, + "learning_rate": 1.3777316453819206e-05, + "loss": 0.2037, + "step": 45080 + }, + { + "epoch": 2.256474415113224, + "grad_norm": 29.924543380737305, + "learning_rate": 1.376804878500862e-05, + "loss": 0.5305, + "step": 45090 + }, + { + "epoch": 2.256974852996372, + "grad_norm": 0.5452877283096313, + "learning_rate": 1.3758781116198032e-05, + "loss": 0.0467, + "step": 45100 + }, + { + "epoch": 2.2574752908795195, + "grad_norm": 2.6906886100769043, + "learning_rate": 1.3749513447387444e-05, + "loss": 0.0992, + "step": 45110 + }, + { + "epoch": 2.257975728762667, + "grad_norm": 0.050248563289642334, + "learning_rate": 1.3740245778576857e-05, + "loss": 0.1952, + "step": 45120 + }, + { + "epoch": 2.2584761666458153, + "grad_norm": 16.287593841552734, + "learning_rate": 1.373097810976627e-05, + "loss": 0.1299, + "step": 45130 + }, + { + "epoch": 2.258976604528963, + "grad_norm": 1.385725498199463, + "learning_rate": 1.3721710440955681e-05, + "loss": 0.1207, + "step": 45140 + }, + { + "epoch": 2.2594770424121107, + "grad_norm": 111.57012939453125, + "learning_rate": 1.3712442772145095e-05, + "loss": 0.3983, + "step": 45150 + }, + { + "epoch": 2.2599774802952584, + "grad_norm": 0.004261493217200041, + "learning_rate": 1.3703175103334509e-05, + "loss": 0.1914, + "step": 45160 + }, + { + "epoch": 2.260477918178406, + "grad_norm": 0.030736548826098442, + "learning_rate": 1.369390743452392e-05, + "loss": 0.4423, + "step": 45170 + }, + { + "epoch": 2.2609783560615537, + "grad_norm": 7.221036911010742, + "learning_rate": 1.3684639765713333e-05, + "loss": 0.4523, + "step": 45180 + }, + { + "epoch": 2.2614787939447014, + "grad_norm": 0.1019936054944992, + "learning_rate": 1.3675372096902747e-05, + "loss": 0.2446, + "step": 45190 + }, + { + "epoch": 2.2619792318278495, + "grad_norm": 0.8982719779014587, + "learning_rate": 1.3666104428092158e-05, + "loss": 0.0341, + "step": 45200 + }, + { + "epoch": 2.2624796697109972, + "grad_norm": 92.34070587158203, + "learning_rate": 1.3656836759281572e-05, + "loss": 0.5568, + "step": 45210 + }, + { + "epoch": 2.262980107594145, + "grad_norm": 114.0129623413086, + "learning_rate": 1.3647569090470982e-05, + "loss": 0.5302, + "step": 45220 + }, + { + "epoch": 2.2634805454772926, + "grad_norm": 0.13583829998970032, + "learning_rate": 1.3638301421660396e-05, + "loss": 0.2221, + "step": 45230 + }, + { + "epoch": 2.2639809833604403, + "grad_norm": 0.06448232382535934, + "learning_rate": 1.362903375284981e-05, + "loss": 0.0331, + "step": 45240 + }, + { + "epoch": 2.264481421243588, + "grad_norm": 0.040147457271814346, + "learning_rate": 1.361976608403922e-05, + "loss": 0.1916, + "step": 45250 + }, + { + "epoch": 2.2649818591267357, + "grad_norm": 149.4015350341797, + "learning_rate": 1.3610498415228635e-05, + "loss": 0.0556, + "step": 45260 + }, + { + "epoch": 2.265482297009884, + "grad_norm": 0.026677453890442848, + "learning_rate": 1.3601230746418047e-05, + "loss": 0.2187, + "step": 45270 + }, + { + "epoch": 2.2659827348930315, + "grad_norm": 0.03679167851805687, + "learning_rate": 1.359196307760746e-05, + "loss": 0.3656, + "step": 45280 + }, + { + "epoch": 2.266483172776179, + "grad_norm": 53.67770767211914, + "learning_rate": 1.3582695408796872e-05, + "loss": 0.2997, + "step": 45290 + }, + { + "epoch": 2.266983610659327, + "grad_norm": 33.874996185302734, + "learning_rate": 1.3573427739986285e-05, + "loss": 0.5691, + "step": 45300 + }, + { + "epoch": 2.2674840485424745, + "grad_norm": 0.023645829409360886, + "learning_rate": 1.3564160071175696e-05, + "loss": 0.2441, + "step": 45310 + }, + { + "epoch": 2.2679844864256222, + "grad_norm": 169.5265350341797, + "learning_rate": 1.355489240236511e-05, + "loss": 0.3775, + "step": 45320 + }, + { + "epoch": 2.2684849243087704, + "grad_norm": 93.81895446777344, + "learning_rate": 1.354562473355452e-05, + "loss": 0.1932, + "step": 45330 + }, + { + "epoch": 2.268985362191918, + "grad_norm": 0.002436672570183873, + "learning_rate": 1.3536357064743934e-05, + "loss": 0.1035, + "step": 45340 + }, + { + "epoch": 2.2694858000750657, + "grad_norm": 122.69686126708984, + "learning_rate": 1.3527089395933348e-05, + "loss": 0.1385, + "step": 45350 + }, + { + "epoch": 2.2699862379582134, + "grad_norm": 0.05386127904057503, + "learning_rate": 1.3517821727122759e-05, + "loss": 0.6483, + "step": 45360 + }, + { + "epoch": 2.270486675841361, + "grad_norm": 19.582250595092773, + "learning_rate": 1.3508554058312173e-05, + "loss": 0.3689, + "step": 45370 + }, + { + "epoch": 2.270987113724509, + "grad_norm": 0.01484917476773262, + "learning_rate": 1.3499286389501587e-05, + "loss": 0.1837, + "step": 45380 + }, + { + "epoch": 2.271487551607657, + "grad_norm": 0.6557963490486145, + "learning_rate": 1.3490018720690997e-05, + "loss": 0.1333, + "step": 45390 + }, + { + "epoch": 2.2719879894908046, + "grad_norm": 0.1555948555469513, + "learning_rate": 1.3480751051880411e-05, + "loss": 0.3331, + "step": 45400 + }, + { + "epoch": 2.2724884273739523, + "grad_norm": 1.0058106184005737, + "learning_rate": 1.3471483383069825e-05, + "loss": 0.1276, + "step": 45410 + }, + { + "epoch": 2.2729888652571, + "grad_norm": 43.01150894165039, + "learning_rate": 1.3462215714259236e-05, + "loss": 0.078, + "step": 45420 + }, + { + "epoch": 2.2734893031402477, + "grad_norm": 0.05414634570479393, + "learning_rate": 1.345294804544865e-05, + "loss": 0.1443, + "step": 45430 + }, + { + "epoch": 2.2739897410233954, + "grad_norm": 71.2299575805664, + "learning_rate": 1.344368037663806e-05, + "loss": 0.1419, + "step": 45440 + }, + { + "epoch": 2.274490178906543, + "grad_norm": 1.5484466552734375, + "learning_rate": 1.3434412707827474e-05, + "loss": 0.4127, + "step": 45450 + }, + { + "epoch": 2.274990616789691, + "grad_norm": 96.4120864868164, + "learning_rate": 1.3425145039016886e-05, + "loss": 0.2431, + "step": 45460 + }, + { + "epoch": 2.275491054672839, + "grad_norm": 0.024097783491015434, + "learning_rate": 1.3415877370206299e-05, + "loss": 0.146, + "step": 45470 + }, + { + "epoch": 2.2759914925559865, + "grad_norm": 0.3115079998970032, + "learning_rate": 1.3406609701395711e-05, + "loss": 0.3412, + "step": 45480 + }, + { + "epoch": 2.2764919304391342, + "grad_norm": 8.267374038696289, + "learning_rate": 1.3397342032585125e-05, + "loss": 0.4033, + "step": 45490 + }, + { + "epoch": 2.276992368322282, + "grad_norm": 0.06923691183328629, + "learning_rate": 1.3388074363774535e-05, + "loss": 0.2594, + "step": 45500 + }, + { + "epoch": 2.2774928062054296, + "grad_norm": 8.234471321105957, + "learning_rate": 1.337880669496395e-05, + "loss": 0.3324, + "step": 45510 + }, + { + "epoch": 2.2779932440885773, + "grad_norm": 0.14031042158603668, + "learning_rate": 1.3369539026153363e-05, + "loss": 0.3218, + "step": 45520 + }, + { + "epoch": 2.2784936819717254, + "grad_norm": 0.0006781756528653204, + "learning_rate": 1.3360271357342774e-05, + "loss": 0.4438, + "step": 45530 + }, + { + "epoch": 2.278994119854873, + "grad_norm": 37.020721435546875, + "learning_rate": 1.3351003688532188e-05, + "loss": 0.4208, + "step": 45540 + }, + { + "epoch": 2.279494557738021, + "grad_norm": 15.332097053527832, + "learning_rate": 1.3341736019721602e-05, + "loss": 0.2464, + "step": 45550 + }, + { + "epoch": 2.2799949956211685, + "grad_norm": 0.5946381688117981, + "learning_rate": 1.3332468350911012e-05, + "loss": 0.2344, + "step": 45560 + }, + { + "epoch": 2.280495433504316, + "grad_norm": 0.1124095544219017, + "learning_rate": 1.3323200682100426e-05, + "loss": 0.1622, + "step": 45570 + }, + { + "epoch": 2.280995871387464, + "grad_norm": 0.04848816618323326, + "learning_rate": 1.3313933013289837e-05, + "loss": 0.0772, + "step": 45580 + }, + { + "epoch": 2.281496309270612, + "grad_norm": 148.11041259765625, + "learning_rate": 1.330466534447925e-05, + "loss": 0.1239, + "step": 45590 + }, + { + "epoch": 2.2819967471537597, + "grad_norm": 0.0025691238697618246, + "learning_rate": 1.3295397675668663e-05, + "loss": 0.2763, + "step": 45600 + }, + { + "epoch": 2.2824971850369074, + "grad_norm": 95.6934585571289, + "learning_rate": 1.3286130006858075e-05, + "loss": 0.5235, + "step": 45610 + }, + { + "epoch": 2.282997622920055, + "grad_norm": 0.6704521179199219, + "learning_rate": 1.3276862338047487e-05, + "loss": 0.2996, + "step": 45620 + }, + { + "epoch": 2.2834980608032027, + "grad_norm": 2.8785109519958496, + "learning_rate": 1.3267594669236901e-05, + "loss": 0.3948, + "step": 45630 + }, + { + "epoch": 2.2839984986863504, + "grad_norm": 1.1279758214950562, + "learning_rate": 1.3258327000426312e-05, + "loss": 0.2494, + "step": 45640 + }, + { + "epoch": 2.2844989365694985, + "grad_norm": 0.06598232686519623, + "learning_rate": 1.3249059331615726e-05, + "loss": 0.0617, + "step": 45650 + }, + { + "epoch": 2.2849993744526462, + "grad_norm": 0.05225912854075432, + "learning_rate": 1.323979166280514e-05, + "loss": 0.5303, + "step": 45660 + }, + { + "epoch": 2.285499812335794, + "grad_norm": 25.517169952392578, + "learning_rate": 1.323052399399455e-05, + "loss": 0.4941, + "step": 45670 + }, + { + "epoch": 2.2860002502189416, + "grad_norm": 0.1939866691827774, + "learning_rate": 1.3221256325183964e-05, + "loss": 0.4752, + "step": 45680 + }, + { + "epoch": 2.2865006881020893, + "grad_norm": 0.017244143411517143, + "learning_rate": 1.3211988656373375e-05, + "loss": 0.222, + "step": 45690 + }, + { + "epoch": 2.287001125985237, + "grad_norm": 66.26815032958984, + "learning_rate": 1.3202720987562789e-05, + "loss": 0.3502, + "step": 45700 + }, + { + "epoch": 2.2875015638683847, + "grad_norm": 0.07835496217012405, + "learning_rate": 1.3193453318752203e-05, + "loss": 0.32, + "step": 45710 + }, + { + "epoch": 2.288002001751533, + "grad_norm": 0.06845153123140335, + "learning_rate": 1.3184185649941613e-05, + "loss": 0.2033, + "step": 45720 + }, + { + "epoch": 2.2885024396346805, + "grad_norm": 36.82096481323242, + "learning_rate": 1.3174917981131027e-05, + "loss": 0.5353, + "step": 45730 + }, + { + "epoch": 2.289002877517828, + "grad_norm": 0.08082370460033417, + "learning_rate": 1.3165650312320441e-05, + "loss": 0.001, + "step": 45740 + }, + { + "epoch": 2.289503315400976, + "grad_norm": 0.5633445382118225, + "learning_rate": 1.3156382643509852e-05, + "loss": 0.3129, + "step": 45750 + }, + { + "epoch": 2.2900037532841235, + "grad_norm": 0.00664827274158597, + "learning_rate": 1.3147114974699266e-05, + "loss": 0.1555, + "step": 45760 + }, + { + "epoch": 2.290504191167271, + "grad_norm": 19.806303024291992, + "learning_rate": 1.3137847305888678e-05, + "loss": 0.1456, + "step": 45770 + }, + { + "epoch": 2.291004629050419, + "grad_norm": 38.117374420166016, + "learning_rate": 1.312857963707809e-05, + "loss": 0.467, + "step": 45780 + }, + { + "epoch": 2.291505066933567, + "grad_norm": 56.703163146972656, + "learning_rate": 1.3119311968267502e-05, + "loss": 0.4132, + "step": 45790 + }, + { + "epoch": 2.2920055048167147, + "grad_norm": 0.010894352570176125, + "learning_rate": 1.3110044299456915e-05, + "loss": 0.3599, + "step": 45800 + }, + { + "epoch": 2.2925059426998624, + "grad_norm": 0.20349276065826416, + "learning_rate": 1.3100776630646327e-05, + "loss": 0.131, + "step": 45810 + }, + { + "epoch": 2.29300638058301, + "grad_norm": 0.00853581354022026, + "learning_rate": 1.3091508961835741e-05, + "loss": 0.2031, + "step": 45820 + }, + { + "epoch": 2.293506818466158, + "grad_norm": 0.028037285432219505, + "learning_rate": 1.3082241293025151e-05, + "loss": 0.0036, + "step": 45830 + }, + { + "epoch": 2.2940072563493055, + "grad_norm": 0.031590431928634644, + "learning_rate": 1.3072973624214565e-05, + "loss": 0.3549, + "step": 45840 + }, + { + "epoch": 2.294507694232453, + "grad_norm": 3.197861909866333, + "learning_rate": 1.306370595540398e-05, + "loss": 0.2077, + "step": 45850 + }, + { + "epoch": 2.2950081321156013, + "grad_norm": 1.3115389347076416, + "learning_rate": 1.305443828659339e-05, + "loss": 0.1371, + "step": 45860 + }, + { + "epoch": 2.295508569998749, + "grad_norm": 0.005482860375195742, + "learning_rate": 1.3045170617782804e-05, + "loss": 0.4494, + "step": 45870 + }, + { + "epoch": 2.2960090078818967, + "grad_norm": 56.6987419128418, + "learning_rate": 1.3035902948972218e-05, + "loss": 0.5279, + "step": 45880 + }, + { + "epoch": 2.2965094457650443, + "grad_norm": 67.05369567871094, + "learning_rate": 1.3026635280161628e-05, + "loss": 0.0751, + "step": 45890 + }, + { + "epoch": 2.297009883648192, + "grad_norm": 0.04424736648797989, + "learning_rate": 1.3017367611351042e-05, + "loss": 0.4409, + "step": 45900 + }, + { + "epoch": 2.29751032153134, + "grad_norm": 0.12114602327346802, + "learning_rate": 1.3008099942540453e-05, + "loss": 0.2657, + "step": 45910 + }, + { + "epoch": 2.298010759414488, + "grad_norm": 0.02177262119948864, + "learning_rate": 1.2998832273729867e-05, + "loss": 0.0071, + "step": 45920 + }, + { + "epoch": 2.2985111972976355, + "grad_norm": 67.22406768798828, + "learning_rate": 1.298956460491928e-05, + "loss": 0.1934, + "step": 45930 + }, + { + "epoch": 2.299011635180783, + "grad_norm": 0.03393206000328064, + "learning_rate": 1.2980296936108691e-05, + "loss": 0.0315, + "step": 45940 + }, + { + "epoch": 2.299512073063931, + "grad_norm": 49.17835998535156, + "learning_rate": 1.2971029267298103e-05, + "loss": 0.4493, + "step": 45950 + }, + { + "epoch": 2.3000125109470786, + "grad_norm": 0.0008625603513792157, + "learning_rate": 1.2961761598487517e-05, + "loss": 0.4884, + "step": 45960 + }, + { + "epoch": 2.3005129488302263, + "grad_norm": 0.05268482863903046, + "learning_rate": 1.2952493929676928e-05, + "loss": 0.0315, + "step": 45970 + }, + { + "epoch": 2.3010133867133744, + "grad_norm": 0.6432932019233704, + "learning_rate": 1.2943226260866342e-05, + "loss": 0.546, + "step": 45980 + }, + { + "epoch": 2.301513824596522, + "grad_norm": 109.69371795654297, + "learning_rate": 1.2933958592055756e-05, + "loss": 0.2507, + "step": 45990 + }, + { + "epoch": 2.30201426247967, + "grad_norm": 33.37919235229492, + "learning_rate": 1.2924690923245166e-05, + "loss": 0.3109, + "step": 46000 + }, + { + "epoch": 2.3025147003628175, + "grad_norm": 0.03299299627542496, + "learning_rate": 1.291542325443458e-05, + "loss": 0.0998, + "step": 46010 + }, + { + "epoch": 2.303015138245965, + "grad_norm": 10.25894546508789, + "learning_rate": 1.2906155585623994e-05, + "loss": 0.1818, + "step": 46020 + }, + { + "epoch": 2.303515576129113, + "grad_norm": 3.0710012912750244, + "learning_rate": 1.2896887916813405e-05, + "loss": 0.1621, + "step": 46030 + }, + { + "epoch": 2.3040160140122605, + "grad_norm": 23.997282028198242, + "learning_rate": 1.2887620248002819e-05, + "loss": 0.5486, + "step": 46040 + }, + { + "epoch": 2.3045164518954087, + "grad_norm": 24.071516036987305, + "learning_rate": 1.287835257919223e-05, + "loss": 0.3723, + "step": 46050 + }, + { + "epoch": 2.3050168897785563, + "grad_norm": 38.31357955932617, + "learning_rate": 1.2869084910381643e-05, + "loss": 0.5623, + "step": 46060 + }, + { + "epoch": 2.305517327661704, + "grad_norm": 0.03500944748520851, + "learning_rate": 1.2859817241571057e-05, + "loss": 0.2855, + "step": 46070 + }, + { + "epoch": 2.3060177655448517, + "grad_norm": 0.26040738821029663, + "learning_rate": 1.2850549572760468e-05, + "loss": 0.401, + "step": 46080 + }, + { + "epoch": 2.3065182034279994, + "grad_norm": 0.01918499357998371, + "learning_rate": 1.2841281903949882e-05, + "loss": 0.3003, + "step": 46090 + }, + { + "epoch": 2.307018641311147, + "grad_norm": 60.9738655090332, + "learning_rate": 1.2832014235139294e-05, + "loss": 0.5717, + "step": 46100 + }, + { + "epoch": 2.3075190791942948, + "grad_norm": 205.7855987548828, + "learning_rate": 1.2822746566328706e-05, + "loss": 0.1673, + "step": 46110 + }, + { + "epoch": 2.308019517077443, + "grad_norm": 0.005664910655468702, + "learning_rate": 1.2813478897518118e-05, + "loss": 0.0736, + "step": 46120 + }, + { + "epoch": 2.3085199549605906, + "grad_norm": 0.03729207068681717, + "learning_rate": 1.2804211228707532e-05, + "loss": 0.2843, + "step": 46130 + }, + { + "epoch": 2.3090203928437383, + "grad_norm": 0.41630059480667114, + "learning_rate": 1.2794943559896943e-05, + "loss": 0.206, + "step": 46140 + }, + { + "epoch": 2.309520830726886, + "grad_norm": 6.176713943481445, + "learning_rate": 1.2785675891086357e-05, + "loss": 0.3756, + "step": 46150 + }, + { + "epoch": 2.3100212686100337, + "grad_norm": 185.4607696533203, + "learning_rate": 1.2776408222275767e-05, + "loss": 0.2418, + "step": 46160 + }, + { + "epoch": 2.310521706493182, + "grad_norm": 33.09423828125, + "learning_rate": 1.2767140553465181e-05, + "loss": 0.4437, + "step": 46170 + }, + { + "epoch": 2.3110221443763295, + "grad_norm": 18.801902770996094, + "learning_rate": 1.2757872884654595e-05, + "loss": 0.2062, + "step": 46180 + }, + { + "epoch": 2.311522582259477, + "grad_norm": 0.9967278242111206, + "learning_rate": 1.2748605215844006e-05, + "loss": 0.0164, + "step": 46190 + }, + { + "epoch": 2.312023020142625, + "grad_norm": 0.00993818324059248, + "learning_rate": 1.273933754703342e-05, + "loss": 0.2388, + "step": 46200 + }, + { + "epoch": 2.3125234580257725, + "grad_norm": 0.5151873230934143, + "learning_rate": 1.2730069878222834e-05, + "loss": 0.5016, + "step": 46210 + }, + { + "epoch": 2.31302389590892, + "grad_norm": 41.451934814453125, + "learning_rate": 1.2720802209412244e-05, + "loss": 0.505, + "step": 46220 + }, + { + "epoch": 2.313524333792068, + "grad_norm": 44.45586013793945, + "learning_rate": 1.2711534540601658e-05, + "loss": 0.3629, + "step": 46230 + }, + { + "epoch": 2.314024771675216, + "grad_norm": 55.653743743896484, + "learning_rate": 1.2702266871791072e-05, + "loss": 0.4957, + "step": 46240 + }, + { + "epoch": 2.3145252095583637, + "grad_norm": 0.10747212171554565, + "learning_rate": 1.2692999202980483e-05, + "loss": 0.1584, + "step": 46250 + }, + { + "epoch": 2.3150256474415114, + "grad_norm": 41.385555267333984, + "learning_rate": 1.2683731534169897e-05, + "loss": 0.2768, + "step": 46260 + }, + { + "epoch": 2.315526085324659, + "grad_norm": 5.493473052978516, + "learning_rate": 1.2674463865359307e-05, + "loss": 0.1158, + "step": 46270 + }, + { + "epoch": 2.3160265232078068, + "grad_norm": 2.0704243183135986, + "learning_rate": 1.2665196196548721e-05, + "loss": 0.2392, + "step": 46280 + }, + { + "epoch": 2.3165269610909545, + "grad_norm": 0.0683489590883255, + "learning_rate": 1.2655928527738133e-05, + "loss": 0.5935, + "step": 46290 + }, + { + "epoch": 2.317027398974102, + "grad_norm": 16.950410842895508, + "learning_rate": 1.2646660858927546e-05, + "loss": 0.2259, + "step": 46300 + }, + { + "epoch": 2.3175278368572503, + "grad_norm": 81.3865737915039, + "learning_rate": 1.2637393190116958e-05, + "loss": 0.2582, + "step": 46310 + }, + { + "epoch": 2.318028274740398, + "grad_norm": 0.24521417915821075, + "learning_rate": 1.2628125521306372e-05, + "loss": 0.3975, + "step": 46320 + }, + { + "epoch": 2.3185287126235457, + "grad_norm": 19.542295455932617, + "learning_rate": 1.2618857852495782e-05, + "loss": 0.483, + "step": 46330 + }, + { + "epoch": 2.3190291505066933, + "grad_norm": 32.216453552246094, + "learning_rate": 1.2609590183685196e-05, + "loss": 0.3658, + "step": 46340 + }, + { + "epoch": 2.319529588389841, + "grad_norm": 25.3648624420166, + "learning_rate": 1.260032251487461e-05, + "loss": 0.1879, + "step": 46350 + }, + { + "epoch": 2.3200300262729887, + "grad_norm": 45.45576858520508, + "learning_rate": 1.259105484606402e-05, + "loss": 0.3602, + "step": 46360 + }, + { + "epoch": 2.3205304641561364, + "grad_norm": 0.007335583213716745, + "learning_rate": 1.2581787177253435e-05, + "loss": 0.1673, + "step": 46370 + }, + { + "epoch": 2.3210309020392845, + "grad_norm": 0.14113162457942963, + "learning_rate": 1.2572519508442849e-05, + "loss": 0.0064, + "step": 46380 + }, + { + "epoch": 2.321531339922432, + "grad_norm": 0.02949310839176178, + "learning_rate": 1.256325183963226e-05, + "loss": 0.287, + "step": 46390 + }, + { + "epoch": 2.32203177780558, + "grad_norm": 0.0028248061425983906, + "learning_rate": 1.2553984170821673e-05, + "loss": 0.5847, + "step": 46400 + }, + { + "epoch": 2.3225322156887276, + "grad_norm": 64.15609741210938, + "learning_rate": 1.2544716502011084e-05, + "loss": 0.2892, + "step": 46410 + }, + { + "epoch": 2.3230326535718753, + "grad_norm": 0.04518118500709534, + "learning_rate": 1.2535448833200498e-05, + "loss": 0.2615, + "step": 46420 + }, + { + "epoch": 2.323533091455023, + "grad_norm": 37.429649353027344, + "learning_rate": 1.252618116438991e-05, + "loss": 0.1931, + "step": 46430 + }, + { + "epoch": 2.324033529338171, + "grad_norm": 0.03660832345485687, + "learning_rate": 1.2516913495579322e-05, + "loss": 0.0021, + "step": 46440 + }, + { + "epoch": 2.3245339672213188, + "grad_norm": 0.4264742136001587, + "learning_rate": 1.2507645826768734e-05, + "loss": 0.1868, + "step": 46450 + }, + { + "epoch": 2.3250344051044665, + "grad_norm": 0.0057344199158251286, + "learning_rate": 1.2498378157958147e-05, + "loss": 0.4873, + "step": 46460 + }, + { + "epoch": 2.325534842987614, + "grad_norm": 0.03797117993235588, + "learning_rate": 1.248911048914756e-05, + "loss": 0.0246, + "step": 46470 + }, + { + "epoch": 2.326035280870762, + "grad_norm": 0.04525899514555931, + "learning_rate": 1.2479842820336973e-05, + "loss": 0.4722, + "step": 46480 + }, + { + "epoch": 2.3265357187539095, + "grad_norm": 9.618682861328125, + "learning_rate": 1.2470575151526385e-05, + "loss": 0.1661, + "step": 46490 + }, + { + "epoch": 2.3270361566370577, + "grad_norm": 0.0054136840626597404, + "learning_rate": 1.2461307482715797e-05, + "loss": 0.0537, + "step": 46500 + }, + { + "epoch": 2.3275365945202053, + "grad_norm": 1.9780242443084717, + "learning_rate": 1.2452039813905211e-05, + "loss": 0.1363, + "step": 46510 + }, + { + "epoch": 2.328037032403353, + "grad_norm": 1.2078827619552612, + "learning_rate": 1.2442772145094624e-05, + "loss": 0.0828, + "step": 46520 + }, + { + "epoch": 2.3285374702865007, + "grad_norm": 0.017201382666826248, + "learning_rate": 1.2433504476284036e-05, + "loss": 0.0028, + "step": 46530 + }, + { + "epoch": 2.3290379081696484, + "grad_norm": 0.021844644099473953, + "learning_rate": 1.2424236807473448e-05, + "loss": 0.1449, + "step": 46540 + }, + { + "epoch": 2.329538346052796, + "grad_norm": 10.16336441040039, + "learning_rate": 1.2414969138662862e-05, + "loss": 0.3362, + "step": 46550 + }, + { + "epoch": 2.3300387839359438, + "grad_norm": 0.04892250895500183, + "learning_rate": 1.2405701469852274e-05, + "loss": 0.2961, + "step": 46560 + }, + { + "epoch": 2.330539221819092, + "grad_norm": 0.011948185972869396, + "learning_rate": 1.2396433801041686e-05, + "loss": 0.3764, + "step": 46570 + }, + { + "epoch": 2.3310396597022396, + "grad_norm": 4.510316848754883, + "learning_rate": 1.23871661322311e-05, + "loss": 0.3919, + "step": 46580 + }, + { + "epoch": 2.3315400975853873, + "grad_norm": 73.58092498779297, + "learning_rate": 1.2377898463420513e-05, + "loss": 0.299, + "step": 46590 + }, + { + "epoch": 2.332040535468535, + "grad_norm": 169.75997924804688, + "learning_rate": 1.2368630794609925e-05, + "loss": 0.3288, + "step": 46600 + }, + { + "epoch": 2.3325409733516826, + "grad_norm": 6.212255954742432, + "learning_rate": 1.2359363125799337e-05, + "loss": 0.2145, + "step": 46610 + }, + { + "epoch": 2.3330414112348303, + "grad_norm": 0.039561353623867035, + "learning_rate": 1.235009545698875e-05, + "loss": 0.102, + "step": 46620 + }, + { + "epoch": 2.333541849117978, + "grad_norm": 0.021576454862952232, + "learning_rate": 1.2340827788178162e-05, + "loss": 0.0826, + "step": 46630 + }, + { + "epoch": 2.334042287001126, + "grad_norm": 38.91325759887695, + "learning_rate": 1.2331560119367574e-05, + "loss": 0.5421, + "step": 46640 + }, + { + "epoch": 2.334542724884274, + "grad_norm": 0.22220951318740845, + "learning_rate": 1.2322292450556988e-05, + "loss": 0.2034, + "step": 46650 + }, + { + "epoch": 2.3350431627674215, + "grad_norm": 0.003946919459849596, + "learning_rate": 1.23130247817464e-05, + "loss": 0.1394, + "step": 46660 + }, + { + "epoch": 2.335543600650569, + "grad_norm": 0.059458814561367035, + "learning_rate": 1.2303757112935812e-05, + "loss": 0.2453, + "step": 46670 + }, + { + "epoch": 2.336044038533717, + "grad_norm": 0.5169283151626587, + "learning_rate": 1.2294489444125225e-05, + "loss": 0.3334, + "step": 46680 + }, + { + "epoch": 2.3365444764168646, + "grad_norm": 121.68727111816406, + "learning_rate": 1.2285221775314638e-05, + "loss": 0.3559, + "step": 46690 + }, + { + "epoch": 2.3370449143000123, + "grad_norm": 0.13285580277442932, + "learning_rate": 1.227595410650405e-05, + "loss": 0.0587, + "step": 46700 + }, + { + "epoch": 2.3375453521831604, + "grad_norm": 0.01884286478161812, + "learning_rate": 1.2266686437693463e-05, + "loss": 0.1478, + "step": 46710 + }, + { + "epoch": 2.338045790066308, + "grad_norm": 5.866711139678955, + "learning_rate": 1.2257418768882875e-05, + "loss": 0.7433, + "step": 46720 + }, + { + "epoch": 2.3385462279494558, + "grad_norm": 0.0013614718336611986, + "learning_rate": 1.2248151100072289e-05, + "loss": 0.0612, + "step": 46730 + }, + { + "epoch": 2.3390466658326035, + "grad_norm": 38.7133903503418, + "learning_rate": 1.2238883431261701e-05, + "loss": 0.2679, + "step": 46740 + }, + { + "epoch": 2.339547103715751, + "grad_norm": 0.019143469631671906, + "learning_rate": 1.2229615762451114e-05, + "loss": 0.2564, + "step": 46750 + }, + { + "epoch": 2.3400475415988993, + "grad_norm": 80.99140930175781, + "learning_rate": 1.2220348093640528e-05, + "loss": 0.4825, + "step": 46760 + }, + { + "epoch": 2.340547979482047, + "grad_norm": 0.012204098515212536, + "learning_rate": 1.221108042482994e-05, + "loss": 0.3109, + "step": 46770 + }, + { + "epoch": 2.3410484173651946, + "grad_norm": 0.0030309937428683043, + "learning_rate": 1.2201812756019352e-05, + "loss": 0.3108, + "step": 46780 + }, + { + "epoch": 2.3415488552483423, + "grad_norm": 0.1437019407749176, + "learning_rate": 1.2192545087208764e-05, + "loss": 0.3222, + "step": 46790 + }, + { + "epoch": 2.34204929313149, + "grad_norm": 69.80148315429688, + "learning_rate": 1.2183277418398177e-05, + "loss": 0.6198, + "step": 46800 + }, + { + "epoch": 2.3425497310146377, + "grad_norm": 0.0017795098247006536, + "learning_rate": 1.2174009749587589e-05, + "loss": 0.2506, + "step": 46810 + }, + { + "epoch": 2.3430501688977854, + "grad_norm": 0.006020568776875734, + "learning_rate": 1.2164742080777001e-05, + "loss": 0.004, + "step": 46820 + }, + { + "epoch": 2.3435506067809335, + "grad_norm": 140.6492156982422, + "learning_rate": 1.2155474411966415e-05, + "loss": 0.1415, + "step": 46830 + }, + { + "epoch": 2.344051044664081, + "grad_norm": 0.20103447139263153, + "learning_rate": 1.2146206743155827e-05, + "loss": 0.1632, + "step": 46840 + }, + { + "epoch": 2.344551482547229, + "grad_norm": 0.08536573499441147, + "learning_rate": 1.213693907434524e-05, + "loss": 0.7103, + "step": 46850 + }, + { + "epoch": 2.3450519204303766, + "grad_norm": 0.004113979637622833, + "learning_rate": 1.2127671405534652e-05, + "loss": 0.2509, + "step": 46860 + }, + { + "epoch": 2.3455523583135243, + "grad_norm": 0.0011104370933026075, + "learning_rate": 1.2118403736724066e-05, + "loss": 0.6485, + "step": 46870 + }, + { + "epoch": 2.346052796196672, + "grad_norm": 48.59470748901367, + "learning_rate": 1.2109136067913478e-05, + "loss": 0.4668, + "step": 46880 + }, + { + "epoch": 2.3465532340798196, + "grad_norm": 0.11525572836399078, + "learning_rate": 1.209986839910289e-05, + "loss": 0.0541, + "step": 46890 + }, + { + "epoch": 2.3470536719629678, + "grad_norm": 28.2167911529541, + "learning_rate": 1.2090600730292302e-05, + "loss": 0.1532, + "step": 46900 + }, + { + "epoch": 2.3475541098461155, + "grad_norm": 0.6706646680831909, + "learning_rate": 1.2081333061481716e-05, + "loss": 0.1592, + "step": 46910 + }, + { + "epoch": 2.348054547729263, + "grad_norm": 10.4028902053833, + "learning_rate": 1.2072065392671129e-05, + "loss": 0.6292, + "step": 46920 + }, + { + "epoch": 2.348554985612411, + "grad_norm": 0.005692063365131617, + "learning_rate": 1.206279772386054e-05, + "loss": 0.2709, + "step": 46930 + }, + { + "epoch": 2.3490554234955585, + "grad_norm": 0.04581728205084801, + "learning_rate": 1.2053530055049953e-05, + "loss": 0.0657, + "step": 46940 + }, + { + "epoch": 2.349555861378706, + "grad_norm": 82.3868179321289, + "learning_rate": 1.2044262386239365e-05, + "loss": 0.5925, + "step": 46950 + }, + { + "epoch": 2.350056299261854, + "grad_norm": 0.004800929222255945, + "learning_rate": 1.2034994717428778e-05, + "loss": 0.0199, + "step": 46960 + }, + { + "epoch": 2.350556737145002, + "grad_norm": 0.6279736161231995, + "learning_rate": 1.202572704861819e-05, + "loss": 0.6429, + "step": 46970 + }, + { + "epoch": 2.3510571750281497, + "grad_norm": 0.01637946255505085, + "learning_rate": 1.2016459379807604e-05, + "loss": 0.2948, + "step": 46980 + }, + { + "epoch": 2.3515576129112974, + "grad_norm": 25.047330856323242, + "learning_rate": 1.2007191710997016e-05, + "loss": 0.0767, + "step": 46990 + }, + { + "epoch": 2.352058050794445, + "grad_norm": 0.001451613730750978, + "learning_rate": 1.1997924042186428e-05, + "loss": 0.4218, + "step": 47000 + }, + { + "epoch": 2.3525584886775928, + "grad_norm": 109.4462661743164, + "learning_rate": 1.1988656373375842e-05, + "loss": 0.3336, + "step": 47010 + }, + { + "epoch": 2.353058926560741, + "grad_norm": 26.592052459716797, + "learning_rate": 1.1979388704565254e-05, + "loss": 0.2021, + "step": 47020 + }, + { + "epoch": 2.3535593644438886, + "grad_norm": 6.640713691711426, + "learning_rate": 1.1970121035754667e-05, + "loss": 0.2655, + "step": 47030 + }, + { + "epoch": 2.3540598023270363, + "grad_norm": 0.007408859208226204, + "learning_rate": 1.1960853366944079e-05, + "loss": 0.4388, + "step": 47040 + }, + { + "epoch": 2.354560240210184, + "grad_norm": 0.7837459444999695, + "learning_rate": 1.1951585698133493e-05, + "loss": 0.2683, + "step": 47050 + }, + { + "epoch": 2.3550606780933316, + "grad_norm": 37.47501754760742, + "learning_rate": 1.1942318029322905e-05, + "loss": 0.2243, + "step": 47060 + }, + { + "epoch": 2.3555611159764793, + "grad_norm": 0.18042755126953125, + "learning_rate": 1.1933050360512317e-05, + "loss": 0.0054, + "step": 47070 + }, + { + "epoch": 2.356061553859627, + "grad_norm": 0.5445796847343445, + "learning_rate": 1.192378269170173e-05, + "loss": 0.1613, + "step": 47080 + }, + { + "epoch": 2.356561991742775, + "grad_norm": 78.89828491210938, + "learning_rate": 1.1914515022891144e-05, + "loss": 0.2596, + "step": 47090 + }, + { + "epoch": 2.357062429625923, + "grad_norm": 0.6146403551101685, + "learning_rate": 1.1905247354080556e-05, + "loss": 0.1104, + "step": 47100 + }, + { + "epoch": 2.3575628675090705, + "grad_norm": 15.97066879272461, + "learning_rate": 1.1895979685269968e-05, + "loss": 0.3611, + "step": 47110 + }, + { + "epoch": 2.358063305392218, + "grad_norm": 0.018615275621414185, + "learning_rate": 1.188671201645938e-05, + "loss": 0.1893, + "step": 47120 + }, + { + "epoch": 2.358563743275366, + "grad_norm": 0.32413363456726074, + "learning_rate": 1.1877444347648793e-05, + "loss": 0.4177, + "step": 47130 + }, + { + "epoch": 2.3590641811585136, + "grad_norm": 0.03185757249593735, + "learning_rate": 1.1868176678838205e-05, + "loss": 0.0047, + "step": 47140 + }, + { + "epoch": 2.3595646190416613, + "grad_norm": 15.123984336853027, + "learning_rate": 1.1858909010027617e-05, + "loss": 0.296, + "step": 47150 + }, + { + "epoch": 2.3600650569248094, + "grad_norm": 33.875003814697266, + "learning_rate": 1.1849641341217031e-05, + "loss": 0.2211, + "step": 47160 + }, + { + "epoch": 2.360565494807957, + "grad_norm": 3.8303089141845703, + "learning_rate": 1.1840373672406443e-05, + "loss": 0.4028, + "step": 47170 + }, + { + "epoch": 2.3610659326911048, + "grad_norm": 21.458223342895508, + "learning_rate": 1.1831106003595855e-05, + "loss": 0.3215, + "step": 47180 + }, + { + "epoch": 2.3615663705742524, + "grad_norm": 2.552403211593628, + "learning_rate": 1.182183833478527e-05, + "loss": 0.1635, + "step": 47190 + }, + { + "epoch": 2.3620668084574, + "grad_norm": 26.714061737060547, + "learning_rate": 1.1812570665974682e-05, + "loss": 0.4568, + "step": 47200 + }, + { + "epoch": 2.362567246340548, + "grad_norm": 4.5844573974609375, + "learning_rate": 1.1803302997164094e-05, + "loss": 0.374, + "step": 47210 + }, + { + "epoch": 2.3630676842236955, + "grad_norm": 41.820430755615234, + "learning_rate": 1.1794035328353506e-05, + "loss": 0.1254, + "step": 47220 + }, + { + "epoch": 2.3635681221068436, + "grad_norm": 0.3956891596317291, + "learning_rate": 1.178476765954292e-05, + "loss": 0.0187, + "step": 47230 + }, + { + "epoch": 2.3640685599899913, + "grad_norm": 0.030479950830340385, + "learning_rate": 1.1775499990732332e-05, + "loss": 0.374, + "step": 47240 + }, + { + "epoch": 2.364568997873139, + "grad_norm": 30.373125076293945, + "learning_rate": 1.1766232321921745e-05, + "loss": 0.1416, + "step": 47250 + }, + { + "epoch": 2.3650694357562867, + "grad_norm": 48.6369743347168, + "learning_rate": 1.1756964653111157e-05, + "loss": 0.1272, + "step": 47260 + }, + { + "epoch": 2.3655698736394344, + "grad_norm": 0.00237075611948967, + "learning_rate": 1.174769698430057e-05, + "loss": 0.2377, + "step": 47270 + }, + { + "epoch": 2.366070311522582, + "grad_norm": 0.1153944805264473, + "learning_rate": 1.1738429315489981e-05, + "loss": 0.1698, + "step": 47280 + }, + { + "epoch": 2.36657074940573, + "grad_norm": 48.849456787109375, + "learning_rate": 1.1729161646679394e-05, + "loss": 0.3104, + "step": 47290 + }, + { + "epoch": 2.367071187288878, + "grad_norm": 1.0032455921173096, + "learning_rate": 1.1719893977868807e-05, + "loss": 0.1033, + "step": 47300 + }, + { + "epoch": 2.3675716251720256, + "grad_norm": 0.004756656475365162, + "learning_rate": 1.171062630905822e-05, + "loss": 0.2006, + "step": 47310 + }, + { + "epoch": 2.3680720630551733, + "grad_norm": 0.09549839049577713, + "learning_rate": 1.1701358640247632e-05, + "loss": 0.1251, + "step": 47320 + }, + { + "epoch": 2.368572500938321, + "grad_norm": 0.17827975749969482, + "learning_rate": 1.1692090971437044e-05, + "loss": 0.1071, + "step": 47330 + }, + { + "epoch": 2.3690729388214686, + "grad_norm": 0.1803271770477295, + "learning_rate": 1.1682823302626458e-05, + "loss": 0.1027, + "step": 47340 + }, + { + "epoch": 2.3695733767046168, + "grad_norm": 64.3800277709961, + "learning_rate": 1.167355563381587e-05, + "loss": 0.4235, + "step": 47350 + }, + { + "epoch": 2.3700738145877644, + "grad_norm": 0.0027432418428361416, + "learning_rate": 1.1664287965005283e-05, + "loss": 0.2376, + "step": 47360 + }, + { + "epoch": 2.370574252470912, + "grad_norm": 32.05485153198242, + "learning_rate": 1.1655020296194697e-05, + "loss": 0.0127, + "step": 47370 + }, + { + "epoch": 2.37107469035406, + "grad_norm": 0.11178874969482422, + "learning_rate": 1.1645752627384109e-05, + "loss": 0.0877, + "step": 47380 + }, + { + "epoch": 2.3715751282372075, + "grad_norm": 0.00759357213973999, + "learning_rate": 1.1636484958573521e-05, + "loss": 0.5538, + "step": 47390 + }, + { + "epoch": 2.372075566120355, + "grad_norm": 3.2188656330108643, + "learning_rate": 1.1627217289762933e-05, + "loss": 0.1137, + "step": 47400 + }, + { + "epoch": 2.372576004003503, + "grad_norm": 27.473119735717773, + "learning_rate": 1.1617949620952347e-05, + "loss": 0.0928, + "step": 47410 + }, + { + "epoch": 2.373076441886651, + "grad_norm": 87.31598663330078, + "learning_rate": 1.160868195214176e-05, + "loss": 0.0881, + "step": 47420 + }, + { + "epoch": 2.3735768797697987, + "grad_norm": 1.6711170673370361, + "learning_rate": 1.1599414283331172e-05, + "loss": 0.2514, + "step": 47430 + }, + { + "epoch": 2.3740773176529464, + "grad_norm": 0.002213681349530816, + "learning_rate": 1.1590146614520584e-05, + "loss": 0.0987, + "step": 47440 + }, + { + "epoch": 2.374577755536094, + "grad_norm": 125.16455841064453, + "learning_rate": 1.1580878945709996e-05, + "loss": 0.4432, + "step": 47450 + }, + { + "epoch": 2.3750781934192418, + "grad_norm": 14.105941772460938, + "learning_rate": 1.1571611276899409e-05, + "loss": 0.3295, + "step": 47460 + }, + { + "epoch": 2.3755786313023894, + "grad_norm": 12.410789489746094, + "learning_rate": 1.156234360808882e-05, + "loss": 0.3003, + "step": 47470 + }, + { + "epoch": 2.376079069185537, + "grad_norm": 4.43260383605957, + "learning_rate": 1.1553075939278235e-05, + "loss": 0.488, + "step": 47480 + }, + { + "epoch": 2.3765795070686853, + "grad_norm": 0.1424068957567215, + "learning_rate": 1.1543808270467647e-05, + "loss": 0.1057, + "step": 47490 + }, + { + "epoch": 2.377079944951833, + "grad_norm": 26.572214126586914, + "learning_rate": 1.153454060165706e-05, + "loss": 0.5736, + "step": 47500 + }, + { + "epoch": 2.3775803828349806, + "grad_norm": 0.012273930944502354, + "learning_rate": 1.1525272932846471e-05, + "loss": 0.2543, + "step": 47510 + }, + { + "epoch": 2.3780808207181283, + "grad_norm": 9.608736038208008, + "learning_rate": 1.1516005264035885e-05, + "loss": 0.263, + "step": 47520 + }, + { + "epoch": 2.378581258601276, + "grad_norm": 1.8663791418075562, + "learning_rate": 1.1506737595225298e-05, + "loss": 0.149, + "step": 47530 + }, + { + "epoch": 2.3790816964844237, + "grad_norm": 22.303871154785156, + "learning_rate": 1.149746992641471e-05, + "loss": 0.0306, + "step": 47540 + }, + { + "epoch": 2.3795821343675714, + "grad_norm": 0.30934542417526245, + "learning_rate": 1.1488202257604124e-05, + "loss": 0.1964, + "step": 47550 + }, + { + "epoch": 2.3800825722507195, + "grad_norm": 0.008517284877598286, + "learning_rate": 1.1478934588793536e-05, + "loss": 0.2059, + "step": 47560 + }, + { + "epoch": 2.380583010133867, + "grad_norm": 125.03191375732422, + "learning_rate": 1.1469666919982948e-05, + "loss": 0.3626, + "step": 47570 + }, + { + "epoch": 2.381083448017015, + "grad_norm": 0.033277831971645355, + "learning_rate": 1.146039925117236e-05, + "loss": 0.2766, + "step": 47580 + }, + { + "epoch": 2.3815838859001626, + "grad_norm": 24.197683334350586, + "learning_rate": 1.1451131582361774e-05, + "loss": 0.4807, + "step": 47590 + }, + { + "epoch": 2.3820843237833103, + "grad_norm": 77.28809356689453, + "learning_rate": 1.1441863913551187e-05, + "loss": 0.7029, + "step": 47600 + }, + { + "epoch": 2.3825847616664584, + "grad_norm": 1.7335530519485474, + "learning_rate": 1.1432596244740599e-05, + "loss": 0.4115, + "step": 47610 + }, + { + "epoch": 2.383085199549606, + "grad_norm": 68.68611145019531, + "learning_rate": 1.1423328575930011e-05, + "loss": 0.5497, + "step": 47620 + }, + { + "epoch": 2.3835856374327538, + "grad_norm": 9.567708969116211, + "learning_rate": 1.1414060907119423e-05, + "loss": 0.197, + "step": 47630 + }, + { + "epoch": 2.3840860753159014, + "grad_norm": 0.0034640019293874502, + "learning_rate": 1.1404793238308836e-05, + "loss": 0.4381, + "step": 47640 + }, + { + "epoch": 2.384586513199049, + "grad_norm": 0.5843839645385742, + "learning_rate": 1.1395525569498248e-05, + "loss": 0.257, + "step": 47650 + }, + { + "epoch": 2.385086951082197, + "grad_norm": 0.013266727328300476, + "learning_rate": 1.1386257900687662e-05, + "loss": 0.6066, + "step": 47660 + }, + { + "epoch": 2.3855873889653445, + "grad_norm": 0.0006409394554793835, + "learning_rate": 1.1376990231877074e-05, + "loss": 0.2755, + "step": 47670 + }, + { + "epoch": 2.3860878268484926, + "grad_norm": 0.26140084862709045, + "learning_rate": 1.1367722563066486e-05, + "loss": 0.0577, + "step": 47680 + }, + { + "epoch": 2.3865882647316403, + "grad_norm": 0.03888906538486481, + "learning_rate": 1.1358454894255899e-05, + "loss": 0.2702, + "step": 47690 + }, + { + "epoch": 2.387088702614788, + "grad_norm": 7.52365255355835, + "learning_rate": 1.1349187225445313e-05, + "loss": 0.014, + "step": 47700 + }, + { + "epoch": 2.3875891404979357, + "grad_norm": 0.4118958115577698, + "learning_rate": 1.1339919556634725e-05, + "loss": 0.2663, + "step": 47710 + }, + { + "epoch": 2.3880895783810834, + "grad_norm": 0.06595700234174728, + "learning_rate": 1.1330651887824137e-05, + "loss": 0.5184, + "step": 47720 + }, + { + "epoch": 2.388590016264231, + "grad_norm": 2.141462802886963, + "learning_rate": 1.1321384219013551e-05, + "loss": 0.6253, + "step": 47730 + }, + { + "epoch": 2.3890904541473788, + "grad_norm": 4.298454284667969, + "learning_rate": 1.1312116550202963e-05, + "loss": 0.1853, + "step": 47740 + }, + { + "epoch": 2.389590892030527, + "grad_norm": 1.1751389503479004, + "learning_rate": 1.1302848881392376e-05, + "loss": 0.4044, + "step": 47750 + }, + { + "epoch": 2.3900913299136746, + "grad_norm": 0.03360377252101898, + "learning_rate": 1.1293581212581788e-05, + "loss": 0.3421, + "step": 47760 + }, + { + "epoch": 2.3905917677968223, + "grad_norm": 0.07085932046175003, + "learning_rate": 1.12843135437712e-05, + "loss": 0.1825, + "step": 47770 + }, + { + "epoch": 2.39109220567997, + "grad_norm": 1.3266910314559937, + "learning_rate": 1.1275045874960612e-05, + "loss": 0.3259, + "step": 47780 + }, + { + "epoch": 2.3915926435631176, + "grad_norm": 3.7342753410339355, + "learning_rate": 1.1265778206150024e-05, + "loss": 0.0261, + "step": 47790 + }, + { + "epoch": 2.3920930814462653, + "grad_norm": 65.04846954345703, + "learning_rate": 1.1256510537339437e-05, + "loss": 0.0521, + "step": 47800 + }, + { + "epoch": 2.392593519329413, + "grad_norm": 77.73992919921875, + "learning_rate": 1.124724286852885e-05, + "loss": 0.1394, + "step": 47810 + }, + { + "epoch": 2.393093957212561, + "grad_norm": 0.34596872329711914, + "learning_rate": 1.1237975199718263e-05, + "loss": 0.6769, + "step": 47820 + }, + { + "epoch": 2.393594395095709, + "grad_norm": 71.05200958251953, + "learning_rate": 1.1228707530907675e-05, + "loss": 0.31, + "step": 47830 + }, + { + "epoch": 2.3940948329788565, + "grad_norm": 0.008082758635282516, + "learning_rate": 1.1219439862097089e-05, + "loss": 0.1462, + "step": 47840 + }, + { + "epoch": 2.394595270862004, + "grad_norm": 17.215991973876953, + "learning_rate": 1.1210172193286501e-05, + "loss": 0.2844, + "step": 47850 + }, + { + "epoch": 2.395095708745152, + "grad_norm": 24.90791893005371, + "learning_rate": 1.1200904524475914e-05, + "loss": 0.3435, + "step": 47860 + }, + { + "epoch": 2.3955961466283, + "grad_norm": 0.015266651287674904, + "learning_rate": 1.1191636855665326e-05, + "loss": 0.481, + "step": 47870 + }, + { + "epoch": 2.3960965845114477, + "grad_norm": 9.100876808166504, + "learning_rate": 1.118236918685474e-05, + "loss": 0.4798, + "step": 47880 + }, + { + "epoch": 2.3965970223945954, + "grad_norm": 0.09247326850891113, + "learning_rate": 1.1173101518044152e-05, + "loss": 0.4531, + "step": 47890 + }, + { + "epoch": 2.397097460277743, + "grad_norm": 9.965612411499023, + "learning_rate": 1.1163833849233564e-05, + "loss": 0.0951, + "step": 47900 + }, + { + "epoch": 2.3975978981608908, + "grad_norm": 0.046666111797094345, + "learning_rate": 1.1154566180422978e-05, + "loss": 0.2771, + "step": 47910 + }, + { + "epoch": 2.3980983360440384, + "grad_norm": 48.22270202636719, + "learning_rate": 1.114529851161239e-05, + "loss": 0.1313, + "step": 47920 + }, + { + "epoch": 2.398598773927186, + "grad_norm": 10.600062370300293, + "learning_rate": 1.1136030842801803e-05, + "loss": 0.2376, + "step": 47930 + }, + { + "epoch": 2.3990992118103343, + "grad_norm": 0.03481752425432205, + "learning_rate": 1.1126763173991215e-05, + "loss": 0.1345, + "step": 47940 + }, + { + "epoch": 2.399599649693482, + "grad_norm": 47.1296272277832, + "learning_rate": 1.1117495505180627e-05, + "loss": 0.6296, + "step": 47950 + }, + { + "epoch": 2.4001000875766296, + "grad_norm": 0.041479505598545074, + "learning_rate": 1.110822783637004e-05, + "loss": 0.4876, + "step": 47960 + }, + { + "epoch": 2.4006005254597773, + "grad_norm": 0.003074728650972247, + "learning_rate": 1.1098960167559452e-05, + "loss": 0.2961, + "step": 47970 + }, + { + "epoch": 2.401100963342925, + "grad_norm": 0.024408940225839615, + "learning_rate": 1.1089692498748864e-05, + "loss": 0.0552, + "step": 47980 + }, + { + "epoch": 2.4016014012260727, + "grad_norm": 0.010115785524249077, + "learning_rate": 1.1080424829938278e-05, + "loss": 0.1033, + "step": 47990 + }, + { + "epoch": 2.4021018391092204, + "grad_norm": 0.04313551262021065, + "learning_rate": 1.107115716112769e-05, + "loss": 0.2227, + "step": 48000 + }, + { + "epoch": 2.4026022769923685, + "grad_norm": 0.03095555491745472, + "learning_rate": 1.1061889492317102e-05, + "loss": 0.2255, + "step": 48010 + }, + { + "epoch": 2.403102714875516, + "grad_norm": 0.03192944824695587, + "learning_rate": 1.1052621823506516e-05, + "loss": 0.0304, + "step": 48020 + }, + { + "epoch": 2.403603152758664, + "grad_norm": 0.012537537142634392, + "learning_rate": 1.1043354154695929e-05, + "loss": 0.1727, + "step": 48030 + }, + { + "epoch": 2.4041035906418116, + "grad_norm": 1.3280067443847656, + "learning_rate": 1.103408648588534e-05, + "loss": 0.146, + "step": 48040 + }, + { + "epoch": 2.4046040285249592, + "grad_norm": 1.9084749221801758, + "learning_rate": 1.1024818817074753e-05, + "loss": 0.4275, + "step": 48050 + }, + { + "epoch": 2.405104466408107, + "grad_norm": 19.8612060546875, + "learning_rate": 1.1015551148264167e-05, + "loss": 0.45, + "step": 48060 + }, + { + "epoch": 2.4056049042912546, + "grad_norm": 0.03382042050361633, + "learning_rate": 1.100628347945358e-05, + "loss": 0.2519, + "step": 48070 + }, + { + "epoch": 2.4061053421744028, + "grad_norm": 69.65113067626953, + "learning_rate": 1.0997015810642991e-05, + "loss": 0.3031, + "step": 48080 + }, + { + "epoch": 2.4066057800575504, + "grad_norm": 23.008220672607422, + "learning_rate": 1.0987748141832405e-05, + "loss": 0.4471, + "step": 48090 + }, + { + "epoch": 2.407106217940698, + "grad_norm": 0.17284777760505676, + "learning_rate": 1.0978480473021818e-05, + "loss": 0.324, + "step": 48100 + }, + { + "epoch": 2.407606655823846, + "grad_norm": 3.9484729766845703, + "learning_rate": 1.096921280421123e-05, + "loss": 0.0565, + "step": 48110 + }, + { + "epoch": 2.4081070937069935, + "grad_norm": 31.74967384338379, + "learning_rate": 1.0959945135400642e-05, + "loss": 0.4654, + "step": 48120 + }, + { + "epoch": 2.408607531590141, + "grad_norm": 0.0009183812653645873, + "learning_rate": 1.0950677466590054e-05, + "loss": 0.2353, + "step": 48130 + }, + { + "epoch": 2.4091079694732893, + "grad_norm": 1.2431083917617798, + "learning_rate": 1.0941409797779467e-05, + "loss": 0.1727, + "step": 48140 + }, + { + "epoch": 2.409608407356437, + "grad_norm": 0.009539946913719177, + "learning_rate": 1.0932142128968879e-05, + "loss": 0.2597, + "step": 48150 + }, + { + "epoch": 2.4101088452395847, + "grad_norm": 0.8119586706161499, + "learning_rate": 1.0922874460158291e-05, + "loss": 0.2909, + "step": 48160 + }, + { + "epoch": 2.4106092831227324, + "grad_norm": 43.614768981933594, + "learning_rate": 1.0913606791347705e-05, + "loss": 0.3498, + "step": 48170 + }, + { + "epoch": 2.41110972100588, + "grad_norm": 34.95840835571289, + "learning_rate": 1.0904339122537117e-05, + "loss": 0.4431, + "step": 48180 + }, + { + "epoch": 2.4116101588890277, + "grad_norm": 1.2103159427642822, + "learning_rate": 1.089507145372653e-05, + "loss": 0.2534, + "step": 48190 + }, + { + "epoch": 2.412110596772176, + "grad_norm": 0.013783890753984451, + "learning_rate": 1.0885803784915944e-05, + "loss": 0.3225, + "step": 48200 + }, + { + "epoch": 2.4126110346553236, + "grad_norm": 71.94678497314453, + "learning_rate": 1.0876536116105356e-05, + "loss": 0.1225, + "step": 48210 + }, + { + "epoch": 2.4131114725384712, + "grad_norm": 16.59233856201172, + "learning_rate": 1.0867268447294768e-05, + "loss": 0.402, + "step": 48220 + }, + { + "epoch": 2.413611910421619, + "grad_norm": 11.487245559692383, + "learning_rate": 1.085800077848418e-05, + "loss": 0.0577, + "step": 48230 + }, + { + "epoch": 2.4141123483047666, + "grad_norm": 0.04708520695567131, + "learning_rate": 1.0848733109673594e-05, + "loss": 0.1701, + "step": 48240 + }, + { + "epoch": 2.4146127861879143, + "grad_norm": 0.030394993722438812, + "learning_rate": 1.0839465440863006e-05, + "loss": 0.2869, + "step": 48250 + }, + { + "epoch": 2.415113224071062, + "grad_norm": 0.019609294831752777, + "learning_rate": 1.0830197772052419e-05, + "loss": 0.1064, + "step": 48260 + }, + { + "epoch": 2.41561366195421, + "grad_norm": 79.69232940673828, + "learning_rate": 1.0820930103241831e-05, + "loss": 0.4649, + "step": 48270 + }, + { + "epoch": 2.416114099837358, + "grad_norm": 0.4760054349899292, + "learning_rate": 1.0811662434431243e-05, + "loss": 0.4058, + "step": 48280 + }, + { + "epoch": 2.4166145377205055, + "grad_norm": 0.007044341415166855, + "learning_rate": 1.0802394765620655e-05, + "loss": 0.3647, + "step": 48290 + }, + { + "epoch": 2.417114975603653, + "grad_norm": 112.56875610351562, + "learning_rate": 1.0793127096810068e-05, + "loss": 0.3068, + "step": 48300 + }, + { + "epoch": 2.417615413486801, + "grad_norm": 0.1573704183101654, + "learning_rate": 1.0783859427999482e-05, + "loss": 0.1558, + "step": 48310 + }, + { + "epoch": 2.4181158513699486, + "grad_norm": 0.004548450466245413, + "learning_rate": 1.0774591759188894e-05, + "loss": 0.2208, + "step": 48320 + }, + { + "epoch": 2.4186162892530962, + "grad_norm": 0.004708079155534506, + "learning_rate": 1.0765324090378306e-05, + "loss": 0.439, + "step": 48330 + }, + { + "epoch": 2.4191167271362444, + "grad_norm": 0.00685882056131959, + "learning_rate": 1.0756056421567718e-05, + "loss": 0.1251, + "step": 48340 + }, + { + "epoch": 2.419617165019392, + "grad_norm": 61.561553955078125, + "learning_rate": 1.0746788752757132e-05, + "loss": 0.3724, + "step": 48350 + }, + { + "epoch": 2.4201176029025397, + "grad_norm": 0.9751458168029785, + "learning_rate": 1.0737521083946545e-05, + "loss": 0.0114, + "step": 48360 + }, + { + "epoch": 2.4206180407856874, + "grad_norm": 0.024579040706157684, + "learning_rate": 1.0728253415135957e-05, + "loss": 0.1477, + "step": 48370 + }, + { + "epoch": 2.421118478668835, + "grad_norm": 0.4237092137336731, + "learning_rate": 1.071898574632537e-05, + "loss": 0.1453, + "step": 48380 + }, + { + "epoch": 2.421618916551983, + "grad_norm": 0.13820694386959076, + "learning_rate": 1.0709718077514783e-05, + "loss": 0.5966, + "step": 48390 + }, + { + "epoch": 2.422119354435131, + "grad_norm": 5.507770538330078, + "learning_rate": 1.0700450408704195e-05, + "loss": 0.2696, + "step": 48400 + }, + { + "epoch": 2.4226197923182786, + "grad_norm": 0.025263439863920212, + "learning_rate": 1.0691182739893607e-05, + "loss": 0.0383, + "step": 48410 + }, + { + "epoch": 2.4231202302014263, + "grad_norm": 0.0101917190477252, + "learning_rate": 1.0681915071083021e-05, + "loss": 0.0655, + "step": 48420 + }, + { + "epoch": 2.423620668084574, + "grad_norm": 0.025419721379876137, + "learning_rate": 1.0672647402272434e-05, + "loss": 0.7071, + "step": 48430 + }, + { + "epoch": 2.4241211059677217, + "grad_norm": 39.7987060546875, + "learning_rate": 1.0663379733461846e-05, + "loss": 0.4848, + "step": 48440 + }, + { + "epoch": 2.4246215438508694, + "grad_norm": 48.52802658081055, + "learning_rate": 1.0654112064651258e-05, + "loss": 0.1564, + "step": 48450 + }, + { + "epoch": 2.4251219817340175, + "grad_norm": 0.03447097912430763, + "learning_rate": 1.064484439584067e-05, + "loss": 0.1452, + "step": 48460 + }, + { + "epoch": 2.425622419617165, + "grad_norm": 4.396295547485352, + "learning_rate": 1.0635576727030083e-05, + "loss": 0.1827, + "step": 48470 + }, + { + "epoch": 2.426122857500313, + "grad_norm": 0.0013741620350629091, + "learning_rate": 1.0626309058219495e-05, + "loss": 0.1405, + "step": 48480 + }, + { + "epoch": 2.4266232953834606, + "grad_norm": 0.0424794964492321, + "learning_rate": 1.0617041389408909e-05, + "loss": 0.445, + "step": 48490 + }, + { + "epoch": 2.4271237332666082, + "grad_norm": 81.31024169921875, + "learning_rate": 1.0607773720598321e-05, + "loss": 0.043, + "step": 48500 + }, + { + "epoch": 2.427624171149756, + "grad_norm": 0.02487408183515072, + "learning_rate": 1.0598506051787733e-05, + "loss": 0.0496, + "step": 48510 + }, + { + "epoch": 2.4281246090329036, + "grad_norm": 0.013144459575414658, + "learning_rate": 1.0589238382977146e-05, + "loss": 0.1651, + "step": 48520 + }, + { + "epoch": 2.4286250469160517, + "grad_norm": 49.55827331542969, + "learning_rate": 1.057997071416656e-05, + "loss": 0.3415, + "step": 48530 + }, + { + "epoch": 2.4291254847991994, + "grad_norm": 0.1968480944633484, + "learning_rate": 1.0570703045355972e-05, + "loss": 0.2314, + "step": 48540 + }, + { + "epoch": 2.429625922682347, + "grad_norm": 10.632305145263672, + "learning_rate": 1.0561435376545384e-05, + "loss": 0.1854, + "step": 48550 + }, + { + "epoch": 2.430126360565495, + "grad_norm": 0.27595046162605286, + "learning_rate": 1.0552167707734798e-05, + "loss": 0.2666, + "step": 48560 + }, + { + "epoch": 2.4306267984486425, + "grad_norm": 0.34736040234565735, + "learning_rate": 1.054290003892421e-05, + "loss": 0.5922, + "step": 48570 + }, + { + "epoch": 2.43112723633179, + "grad_norm": 77.20977783203125, + "learning_rate": 1.0533632370113622e-05, + "loss": 0.1488, + "step": 48580 + }, + { + "epoch": 2.431627674214938, + "grad_norm": 0.7144792079925537, + "learning_rate": 1.0524364701303035e-05, + "loss": 0.4025, + "step": 48590 + }, + { + "epoch": 2.432128112098086, + "grad_norm": 0.3448299169540405, + "learning_rate": 1.0515097032492449e-05, + "loss": 0.3152, + "step": 48600 + }, + { + "epoch": 2.4326285499812337, + "grad_norm": 173.0098114013672, + "learning_rate": 1.050582936368186e-05, + "loss": 0.2315, + "step": 48610 + }, + { + "epoch": 2.4331289878643814, + "grad_norm": 138.0476531982422, + "learning_rate": 1.0496561694871271e-05, + "loss": 0.4403, + "step": 48620 + }, + { + "epoch": 2.433629425747529, + "grad_norm": 0.5109832286834717, + "learning_rate": 1.0487294026060685e-05, + "loss": 0.2061, + "step": 48630 + }, + { + "epoch": 2.4341298636306767, + "grad_norm": 0.020693346858024597, + "learning_rate": 1.0478026357250098e-05, + "loss": 0.2138, + "step": 48640 + }, + { + "epoch": 2.4346303015138244, + "grad_norm": 48.14086151123047, + "learning_rate": 1.046875868843951e-05, + "loss": 0.4071, + "step": 48650 + }, + { + "epoch": 2.435130739396972, + "grad_norm": 80.822509765625, + "learning_rate": 1.0459491019628922e-05, + "loss": 0.9627, + "step": 48660 + }, + { + "epoch": 2.4356311772801202, + "grad_norm": 29.59706687927246, + "learning_rate": 1.0450223350818336e-05, + "loss": 0.2993, + "step": 48670 + }, + { + "epoch": 2.436131615163268, + "grad_norm": 42.918243408203125, + "learning_rate": 1.0440955682007748e-05, + "loss": 0.4343, + "step": 48680 + }, + { + "epoch": 2.4366320530464156, + "grad_norm": 0.07886827737092972, + "learning_rate": 1.043168801319716e-05, + "loss": 0.2643, + "step": 48690 + }, + { + "epoch": 2.4371324909295633, + "grad_norm": 22.795669555664062, + "learning_rate": 1.0422420344386573e-05, + "loss": 0.5232, + "step": 48700 + }, + { + "epoch": 2.437632928812711, + "grad_norm": 11.127017974853516, + "learning_rate": 1.0413152675575987e-05, + "loss": 0.3193, + "step": 48710 + }, + { + "epoch": 2.438133366695859, + "grad_norm": 3.169090747833252, + "learning_rate": 1.0403885006765399e-05, + "loss": 0.2039, + "step": 48720 + }, + { + "epoch": 2.438633804579007, + "grad_norm": 73.1365966796875, + "learning_rate": 1.0394617337954811e-05, + "loss": 0.4812, + "step": 48730 + }, + { + "epoch": 2.4391342424621545, + "grad_norm": 11.758018493652344, + "learning_rate": 1.0385349669144225e-05, + "loss": 0.4502, + "step": 48740 + }, + { + "epoch": 2.439634680345302, + "grad_norm": 0.20626886188983917, + "learning_rate": 1.0376082000333637e-05, + "loss": 0.0426, + "step": 48750 + }, + { + "epoch": 2.44013511822845, + "grad_norm": 0.008836573921144009, + "learning_rate": 1.036681433152305e-05, + "loss": 0.2152, + "step": 48760 + }, + { + "epoch": 2.4406355561115975, + "grad_norm": 0.008960023522377014, + "learning_rate": 1.0357546662712462e-05, + "loss": 0.3464, + "step": 48770 + }, + { + "epoch": 2.4411359939947452, + "grad_norm": 25.21077537536621, + "learning_rate": 1.0348278993901874e-05, + "loss": 0.4454, + "step": 48780 + }, + { + "epoch": 2.4416364318778934, + "grad_norm": 0.20991413295269012, + "learning_rate": 1.0339011325091286e-05, + "loss": 0.2361, + "step": 48790 + }, + { + "epoch": 2.442136869761041, + "grad_norm": 0.007792830467224121, + "learning_rate": 1.0329743656280699e-05, + "loss": 0.1298, + "step": 48800 + }, + { + "epoch": 2.4426373076441887, + "grad_norm": 0.00042209969251416624, + "learning_rate": 1.0320475987470113e-05, + "loss": 0.2377, + "step": 48810 + }, + { + "epoch": 2.4431377455273364, + "grad_norm": 0.023788822814822197, + "learning_rate": 1.0311208318659525e-05, + "loss": 0.5366, + "step": 48820 + }, + { + "epoch": 2.443638183410484, + "grad_norm": 0.009356300346553326, + "learning_rate": 1.0301940649848937e-05, + "loss": 0.3622, + "step": 48830 + }, + { + "epoch": 2.444138621293632, + "grad_norm": 152.10458374023438, + "learning_rate": 1.029267298103835e-05, + "loss": 0.5359, + "step": 48840 + }, + { + "epoch": 2.4446390591767795, + "grad_norm": 0.3362799286842346, + "learning_rate": 1.0283405312227763e-05, + "loss": 0.0839, + "step": 48850 + }, + { + "epoch": 2.4451394970599276, + "grad_norm": 88.59455108642578, + "learning_rate": 1.0274137643417175e-05, + "loss": 0.1176, + "step": 48860 + }, + { + "epoch": 2.4456399349430753, + "grad_norm": 0.09900861233472824, + "learning_rate": 1.0264869974606588e-05, + "loss": 0.1163, + "step": 48870 + }, + { + "epoch": 2.446140372826223, + "grad_norm": 24.618967056274414, + "learning_rate": 1.0255602305796e-05, + "loss": 0.5526, + "step": 48880 + }, + { + "epoch": 2.4466408107093707, + "grad_norm": 0.13168767094612122, + "learning_rate": 1.0246334636985414e-05, + "loss": 0.1869, + "step": 48890 + }, + { + "epoch": 2.4471412485925184, + "grad_norm": 23.53903579711914, + "learning_rate": 1.0237066968174826e-05, + "loss": 0.4577, + "step": 48900 + }, + { + "epoch": 2.447641686475666, + "grad_norm": 0.6077121496200562, + "learning_rate": 1.0227799299364238e-05, + "loss": 0.4621, + "step": 48910 + }, + { + "epoch": 2.4481421243588137, + "grad_norm": 28.029115676879883, + "learning_rate": 1.0218531630553652e-05, + "loss": 0.2235, + "step": 48920 + }, + { + "epoch": 2.448642562241962, + "grad_norm": 40.39207077026367, + "learning_rate": 1.0209263961743065e-05, + "loss": 0.0933, + "step": 48930 + }, + { + "epoch": 2.4491430001251095, + "grad_norm": 0.013742350041866302, + "learning_rate": 1.0199996292932477e-05, + "loss": 0.6086, + "step": 48940 + }, + { + "epoch": 2.4496434380082572, + "grad_norm": 0.011610819958150387, + "learning_rate": 1.0190728624121889e-05, + "loss": 0.3236, + "step": 48950 + }, + { + "epoch": 2.450143875891405, + "grad_norm": 0.014272461645305157, + "learning_rate": 1.0181460955311301e-05, + "loss": 0.2731, + "step": 48960 + }, + { + "epoch": 2.4506443137745526, + "grad_norm": 0.0377456434071064, + "learning_rate": 1.0172193286500714e-05, + "loss": 0.3111, + "step": 48970 + }, + { + "epoch": 2.4511447516577007, + "grad_norm": 0.05823015421628952, + "learning_rate": 1.0162925617690126e-05, + "loss": 0.1585, + "step": 48980 + }, + { + "epoch": 2.4516451895408484, + "grad_norm": 0.022252066060900688, + "learning_rate": 1.015365794887954e-05, + "loss": 0.0379, + "step": 48990 + }, + { + "epoch": 2.452145627423996, + "grad_norm": 0.07585409283638, + "learning_rate": 1.0144390280068952e-05, + "loss": 0.2831, + "step": 49000 + }, + { + "epoch": 2.452646065307144, + "grad_norm": 81.115234375, + "learning_rate": 1.0135122611258364e-05, + "loss": 0.2924, + "step": 49010 + }, + { + "epoch": 2.4531465031902915, + "grad_norm": 0.6059202551841736, + "learning_rate": 1.0125854942447776e-05, + "loss": 0.3247, + "step": 49020 + }, + { + "epoch": 2.453646941073439, + "grad_norm": 51.88539123535156, + "learning_rate": 1.011658727363719e-05, + "loss": 0.395, + "step": 49030 + }, + { + "epoch": 2.454147378956587, + "grad_norm": 0.04751909524202347, + "learning_rate": 1.0107319604826603e-05, + "loss": 0.3137, + "step": 49040 + }, + { + "epoch": 2.454647816839735, + "grad_norm": 201.37269592285156, + "learning_rate": 1.0098051936016015e-05, + "loss": 0.2525, + "step": 49050 + }, + { + "epoch": 2.4551482547228827, + "grad_norm": 0.09677033871412277, + "learning_rate": 1.0088784267205427e-05, + "loss": 0.1384, + "step": 49060 + }, + { + "epoch": 2.4556486926060304, + "grad_norm": 26.68253517150879, + "learning_rate": 1.0079516598394841e-05, + "loss": 0.2292, + "step": 49070 + }, + { + "epoch": 2.456149130489178, + "grad_norm": 0.0020346343517303467, + "learning_rate": 1.0070248929584253e-05, + "loss": 0.4346, + "step": 49080 + }, + { + "epoch": 2.4566495683723257, + "grad_norm": 28.52943229675293, + "learning_rate": 1.0060981260773666e-05, + "loss": 0.3677, + "step": 49090 + }, + { + "epoch": 2.4571500062554734, + "grad_norm": 26.349788665771484, + "learning_rate": 1.0051713591963078e-05, + "loss": 0.3383, + "step": 49100 + }, + { + "epoch": 2.457650444138621, + "grad_norm": 0.007204275578260422, + "learning_rate": 1.004244592315249e-05, + "loss": 0.0188, + "step": 49110 + }, + { + "epoch": 2.4581508820217692, + "grad_norm": 0.014081922359764576, + "learning_rate": 1.0033178254341902e-05, + "loss": 0.1454, + "step": 49120 + }, + { + "epoch": 2.458651319904917, + "grad_norm": 99.62650299072266, + "learning_rate": 1.0023910585531315e-05, + "loss": 0.2127, + "step": 49130 + }, + { + "epoch": 2.4591517577880646, + "grad_norm": 1.6374752521514893, + "learning_rate": 1.0014642916720729e-05, + "loss": 0.0506, + "step": 49140 + }, + { + "epoch": 2.4596521956712123, + "grad_norm": 1.8240569829940796, + "learning_rate": 1.000537524791014e-05, + "loss": 0.6676, + "step": 49150 + }, + { + "epoch": 2.46015263355436, + "grad_norm": 0.00939291063696146, + "learning_rate": 9.996107579099553e-06, + "loss": 0.082, + "step": 49160 + }, + { + "epoch": 2.4606530714375077, + "grad_norm": 122.71465301513672, + "learning_rate": 9.986839910288967e-06, + "loss": 0.0856, + "step": 49170 + }, + { + "epoch": 2.4611535093206554, + "grad_norm": 0.08938247710466385, + "learning_rate": 9.97757224147838e-06, + "loss": 0.162, + "step": 49180 + }, + { + "epoch": 2.4616539472038035, + "grad_norm": 0.07040990889072418, + "learning_rate": 9.968304572667791e-06, + "loss": 0.5214, + "step": 49190 + }, + { + "epoch": 2.462154385086951, + "grad_norm": 0.03873102366924286, + "learning_rate": 9.959036903857204e-06, + "loss": 0.3442, + "step": 49200 + }, + { + "epoch": 2.462654822970099, + "grad_norm": 0.028000107035040855, + "learning_rate": 9.949769235046618e-06, + "loss": 0.1849, + "step": 49210 + }, + { + "epoch": 2.4631552608532465, + "grad_norm": 0.13059289753437042, + "learning_rate": 9.94050156623603e-06, + "loss": 0.4346, + "step": 49220 + }, + { + "epoch": 2.4636556987363942, + "grad_norm": 0.7737820744514465, + "learning_rate": 9.931233897425442e-06, + "loss": 0.2493, + "step": 49230 + }, + { + "epoch": 2.464156136619542, + "grad_norm": 0.2095870077610016, + "learning_rate": 9.921966228614854e-06, + "loss": 0.2145, + "step": 49240 + }, + { + "epoch": 2.46465657450269, + "grad_norm": 0.028412871062755585, + "learning_rate": 9.912698559804268e-06, + "loss": 0.2681, + "step": 49250 + }, + { + "epoch": 2.4651570123858377, + "grad_norm": 295.01824951171875, + "learning_rate": 9.90343089099368e-06, + "loss": 0.1375, + "step": 49260 + }, + { + "epoch": 2.4656574502689854, + "grad_norm": 0.03362424671649933, + "learning_rate": 9.894163222183093e-06, + "loss": 0.3866, + "step": 49270 + }, + { + "epoch": 2.466157888152133, + "grad_norm": 141.84011840820312, + "learning_rate": 9.884895553372505e-06, + "loss": 0.5383, + "step": 49280 + }, + { + "epoch": 2.466658326035281, + "grad_norm": 83.5831069946289, + "learning_rate": 9.875627884561917e-06, + "loss": 0.312, + "step": 49290 + }, + { + "epoch": 2.4671587639184285, + "grad_norm": 82.18689727783203, + "learning_rate": 9.86636021575133e-06, + "loss": 0.2034, + "step": 49300 + }, + { + "epoch": 2.4676592018015766, + "grad_norm": 1.9926230907440186, + "learning_rate": 9.857092546940742e-06, + "loss": 0.4439, + "step": 49310 + }, + { + "epoch": 2.4681596396847243, + "grad_norm": 0.004026865120977163, + "learning_rate": 9.847824878130156e-06, + "loss": 0.2943, + "step": 49320 + }, + { + "epoch": 2.468660077567872, + "grad_norm": 0.0327068567276001, + "learning_rate": 9.838557209319568e-06, + "loss": 0.0016, + "step": 49330 + }, + { + "epoch": 2.4691605154510197, + "grad_norm": 0.012458225712180138, + "learning_rate": 9.82928954050898e-06, + "loss": 0.1615, + "step": 49340 + }, + { + "epoch": 2.4696609533341674, + "grad_norm": 14.024452209472656, + "learning_rate": 9.820021871698394e-06, + "loss": 0.0313, + "step": 49350 + }, + { + "epoch": 2.470161391217315, + "grad_norm": 0.21738657355308533, + "learning_rate": 9.810754202887806e-06, + "loss": 0.1292, + "step": 49360 + }, + { + "epoch": 2.4706618291004627, + "grad_norm": 3.9768261909484863, + "learning_rate": 9.801486534077219e-06, + "loss": 0.1772, + "step": 49370 + }, + { + "epoch": 2.471162266983611, + "grad_norm": 30.317337036132812, + "learning_rate": 9.792218865266631e-06, + "loss": 0.2908, + "step": 49380 + }, + { + "epoch": 2.4716627048667585, + "grad_norm": 0.008647806942462921, + "learning_rate": 9.782951196456045e-06, + "loss": 0.5796, + "step": 49390 + }, + { + "epoch": 2.4721631427499062, + "grad_norm": 0.43476805090904236, + "learning_rate": 9.773683527645457e-06, + "loss": 0.0047, + "step": 49400 + }, + { + "epoch": 2.472663580633054, + "grad_norm": 43.897308349609375, + "learning_rate": 9.76441585883487e-06, + "loss": 0.3299, + "step": 49410 + }, + { + "epoch": 2.4731640185162016, + "grad_norm": 46.84210968017578, + "learning_rate": 9.755148190024282e-06, + "loss": 0.2692, + "step": 49420 + }, + { + "epoch": 2.4736644563993493, + "grad_norm": 0.05201515927910805, + "learning_rate": 9.745880521213696e-06, + "loss": 0.0912, + "step": 49430 + }, + { + "epoch": 2.474164894282497, + "grad_norm": 32.817405700683594, + "learning_rate": 9.736612852403108e-06, + "loss": 0.0808, + "step": 49440 + }, + { + "epoch": 2.474665332165645, + "grad_norm": 5.660813808441162, + "learning_rate": 9.72734518359252e-06, + "loss": 0.5096, + "step": 49450 + }, + { + "epoch": 2.475165770048793, + "grad_norm": 0.01828770898282528, + "learning_rate": 9.718077514781932e-06, + "loss": 0.0083, + "step": 49460 + }, + { + "epoch": 2.4756662079319405, + "grad_norm": 0.001398564549162984, + "learning_rate": 9.708809845971344e-06, + "loss": 0.4475, + "step": 49470 + }, + { + "epoch": 2.476166645815088, + "grad_norm": 0.23428107798099518, + "learning_rate": 9.699542177160757e-06, + "loss": 0.1073, + "step": 49480 + }, + { + "epoch": 2.476667083698236, + "grad_norm": 0.015836024656891823, + "learning_rate": 9.690274508350169e-06, + "loss": 0.293, + "step": 49490 + }, + { + "epoch": 2.4771675215813835, + "grad_norm": 0.004091959446668625, + "learning_rate": 9.681006839539583e-06, + "loss": 0.1668, + "step": 49500 + }, + { + "epoch": 2.4776679594645312, + "grad_norm": 0.07035493850708008, + "learning_rate": 9.671739170728995e-06, + "loss": 0.1687, + "step": 49510 + }, + { + "epoch": 2.4781683973476794, + "grad_norm": 5.062799453735352, + "learning_rate": 9.662471501918407e-06, + "loss": 0.3452, + "step": 49520 + }, + { + "epoch": 2.478668835230827, + "grad_norm": 1.642241358757019, + "learning_rate": 9.653203833107821e-06, + "loss": 0.3766, + "step": 49530 + }, + { + "epoch": 2.4791692731139747, + "grad_norm": 0.0712418481707573, + "learning_rate": 9.643936164297234e-06, + "loss": 0.4742, + "step": 49540 + }, + { + "epoch": 2.4796697109971224, + "grad_norm": 1.2408097982406616, + "learning_rate": 9.634668495486646e-06, + "loss": 0.1887, + "step": 49550 + }, + { + "epoch": 2.48017014888027, + "grad_norm": 84.28704071044922, + "learning_rate": 9.625400826676058e-06, + "loss": 0.1191, + "step": 49560 + }, + { + "epoch": 2.4806705867634182, + "grad_norm": 0.28964439034461975, + "learning_rate": 9.616133157865472e-06, + "loss": 0.5803, + "step": 49570 + }, + { + "epoch": 2.481171024646566, + "grad_norm": 0.01627482660114765, + "learning_rate": 9.606865489054884e-06, + "loss": 0.3798, + "step": 49580 + }, + { + "epoch": 2.4816714625297136, + "grad_norm": 2.7507681846618652, + "learning_rate": 9.597597820244297e-06, + "loss": 0.5194, + "step": 49590 + }, + { + "epoch": 2.4821719004128613, + "grad_norm": 0.8184736967086792, + "learning_rate": 9.588330151433709e-06, + "loss": 0.4337, + "step": 49600 + }, + { + "epoch": 2.482672338296009, + "grad_norm": 0.10450690984725952, + "learning_rate": 9.579062482623121e-06, + "loss": 0.0615, + "step": 49610 + }, + { + "epoch": 2.4831727761791567, + "grad_norm": 0.01512967236340046, + "learning_rate": 9.569794813812533e-06, + "loss": 0.2423, + "step": 49620 + }, + { + "epoch": 2.4836732140623043, + "grad_norm": 0.09881641715765, + "learning_rate": 9.560527145001946e-06, + "loss": 0.0632, + "step": 49630 + }, + { + "epoch": 2.4841736519454525, + "grad_norm": 0.2085515409708023, + "learning_rate": 9.55125947619136e-06, + "loss": 0.2542, + "step": 49640 + }, + { + "epoch": 2.4846740898286, + "grad_norm": 0.00563848065212369, + "learning_rate": 9.541991807380772e-06, + "loss": 0.1572, + "step": 49650 + }, + { + "epoch": 2.485174527711748, + "grad_norm": 149.6267852783203, + "learning_rate": 9.532724138570184e-06, + "loss": 0.499, + "step": 49660 + }, + { + "epoch": 2.4856749655948955, + "grad_norm": 123.03583526611328, + "learning_rate": 9.523456469759596e-06, + "loss": 0.2835, + "step": 49670 + }, + { + "epoch": 2.4861754034780432, + "grad_norm": 0.5631343126296997, + "learning_rate": 9.51418880094901e-06, + "loss": 0.0676, + "step": 49680 + }, + { + "epoch": 2.486675841361191, + "grad_norm": 5.4742631912231445, + "learning_rate": 9.504921132138422e-06, + "loss": 0.3871, + "step": 49690 + }, + { + "epoch": 2.4871762792443386, + "grad_norm": 0.008395788259804249, + "learning_rate": 9.495653463327835e-06, + "loss": 0.4332, + "step": 49700 + }, + { + "epoch": 2.4876767171274867, + "grad_norm": 0.06143414229154587, + "learning_rate": 9.486385794517249e-06, + "loss": 0.2596, + "step": 49710 + }, + { + "epoch": 2.4881771550106344, + "grad_norm": 0.2640905976295471, + "learning_rate": 9.47711812570666e-06, + "loss": 0.0093, + "step": 49720 + }, + { + "epoch": 2.488677592893782, + "grad_norm": 0.008664185181260109, + "learning_rate": 9.467850456896073e-06, + "loss": 0.1807, + "step": 49730 + }, + { + "epoch": 2.48917803077693, + "grad_norm": 0.009926139377057552, + "learning_rate": 9.458582788085485e-06, + "loss": 0.6022, + "step": 49740 + }, + { + "epoch": 2.4896784686600775, + "grad_norm": 0.0007698190747760236, + "learning_rate": 9.4493151192749e-06, + "loss": 0.1698, + "step": 49750 + }, + { + "epoch": 2.490178906543225, + "grad_norm": 0.0013606982538476586, + "learning_rate": 9.440047450464311e-06, + "loss": 0.6523, + "step": 49760 + }, + { + "epoch": 2.490679344426373, + "grad_norm": 0.039825648069381714, + "learning_rate": 9.430779781653724e-06, + "loss": 0.2665, + "step": 49770 + }, + { + "epoch": 2.491179782309521, + "grad_norm": 0.059538088738918304, + "learning_rate": 9.421512112843136e-06, + "loss": 0.1226, + "step": 49780 + }, + { + "epoch": 2.4916802201926687, + "grad_norm": 0.02083596959710121, + "learning_rate": 9.412244444032548e-06, + "loss": 0.3687, + "step": 49790 + }, + { + "epoch": 2.4921806580758163, + "grad_norm": 0.0958959311246872, + "learning_rate": 9.40297677522196e-06, + "loss": 0.675, + "step": 49800 + }, + { + "epoch": 2.492681095958964, + "grad_norm": 0.006893066689372063, + "learning_rate": 9.393709106411373e-06, + "loss": 0.3451, + "step": 49810 + }, + { + "epoch": 2.4931815338421117, + "grad_norm": 24.041959762573242, + "learning_rate": 9.384441437600787e-06, + "loss": 0.4875, + "step": 49820 + }, + { + "epoch": 2.49368197172526, + "grad_norm": 0.01730867475271225, + "learning_rate": 9.375173768790199e-06, + "loss": 0.2068, + "step": 49830 + }, + { + "epoch": 2.4941824096084075, + "grad_norm": 145.47207641601562, + "learning_rate": 9.365906099979611e-06, + "loss": 0.2607, + "step": 49840 + }, + { + "epoch": 2.4946828474915552, + "grad_norm": 64.14879608154297, + "learning_rate": 9.356638431169023e-06, + "loss": 0.5877, + "step": 49850 + }, + { + "epoch": 2.495183285374703, + "grad_norm": 0.02663533203303814, + "learning_rate": 9.347370762358437e-06, + "loss": 0.2855, + "step": 49860 + }, + { + "epoch": 2.4956837232578506, + "grad_norm": 0.006227743811905384, + "learning_rate": 9.33810309354785e-06, + "loss": 0.1116, + "step": 49870 + }, + { + "epoch": 2.4961841611409983, + "grad_norm": 0.009073487482964993, + "learning_rate": 9.328835424737262e-06, + "loss": 0.2782, + "step": 49880 + }, + { + "epoch": 2.496684599024146, + "grad_norm": 5.371424198150635, + "learning_rate": 9.319567755926676e-06, + "loss": 0.0569, + "step": 49890 + }, + { + "epoch": 2.497185036907294, + "grad_norm": 27.30664825439453, + "learning_rate": 9.310300087116088e-06, + "loss": 0.4868, + "step": 49900 + }, + { + "epoch": 2.497685474790442, + "grad_norm": 32.9210319519043, + "learning_rate": 9.3010324183055e-06, + "loss": 0.2704, + "step": 49910 + }, + { + "epoch": 2.4981859126735895, + "grad_norm": 1.8154101371765137, + "learning_rate": 9.291764749494913e-06, + "loss": 0.0055, + "step": 49920 + }, + { + "epoch": 2.498686350556737, + "grad_norm": 35.16851043701172, + "learning_rate": 9.282497080684326e-06, + "loss": 0.0068, + "step": 49930 + }, + { + "epoch": 2.499186788439885, + "grad_norm": 0.008242834359407425, + "learning_rate": 9.273229411873737e-06, + "loss": 0.0846, + "step": 49940 + }, + { + "epoch": 2.4996872263230325, + "grad_norm": 61.930484771728516, + "learning_rate": 9.26396174306315e-06, + "loss": 0.3179, + "step": 49950 + }, + { + "epoch": 2.50018766420618, + "grad_norm": 0.009819923900067806, + "learning_rate": 9.254694074252561e-06, + "loss": 0.2541, + "step": 49960 + }, + { + "epoch": 2.5006881020893283, + "grad_norm": 0.03993348404765129, + "learning_rate": 9.245426405441975e-06, + "loss": 0.4124, + "step": 49970 + }, + { + "epoch": 2.501188539972476, + "grad_norm": 0.0034491706173866987, + "learning_rate": 9.236158736631388e-06, + "loss": 0.0315, + "step": 49980 + }, + { + "epoch": 2.5016889778556237, + "grad_norm": 0.010300219058990479, + "learning_rate": 9.2268910678208e-06, + "loss": 0.4, + "step": 49990 + }, + { + "epoch": 2.5021894157387714, + "grad_norm": 26.157005310058594, + "learning_rate": 9.217623399010214e-06, + "loss": 0.23, + "step": 50000 + }, + { + "epoch": 2.502689853621919, + "grad_norm": 0.15513546764850616, + "learning_rate": 9.208355730199626e-06, + "loss": 0.1952, + "step": 50010 + }, + { + "epoch": 2.5031902915050668, + "grad_norm": 48.296142578125, + "learning_rate": 9.199088061389038e-06, + "loss": 0.322, + "step": 50020 + }, + { + "epoch": 2.5036907293882145, + "grad_norm": 89.03115844726562, + "learning_rate": 9.18982039257845e-06, + "loss": 0.0679, + "step": 50030 + }, + { + "epoch": 2.5041911672713626, + "grad_norm": 1.7722687721252441, + "learning_rate": 9.180552723767865e-06, + "loss": 0.3131, + "step": 50040 + }, + { + "epoch": 2.5046916051545103, + "grad_norm": 0.007387508172541857, + "learning_rate": 9.171285054957277e-06, + "loss": 0.5936, + "step": 50050 + }, + { + "epoch": 2.505192043037658, + "grad_norm": 0.10927948355674744, + "learning_rate": 9.162017386146689e-06, + "loss": 0.0603, + "step": 50060 + }, + { + "epoch": 2.5056924809208057, + "grad_norm": 0.0003206546825822443, + "learning_rate": 9.152749717336103e-06, + "loss": 0.5517, + "step": 50070 + }, + { + "epoch": 2.5061929188039533, + "grad_norm": 0.0060086739249527454, + "learning_rate": 9.143482048525515e-06, + "loss": 0.1317, + "step": 50080 + }, + { + "epoch": 2.5066933566871015, + "grad_norm": 125.4556884765625, + "learning_rate": 9.134214379714927e-06, + "loss": 0.2449, + "step": 50090 + }, + { + "epoch": 2.5071937945702487, + "grad_norm": 0.008306341245770454, + "learning_rate": 9.12494671090434e-06, + "loss": 0.0919, + "step": 50100 + }, + { + "epoch": 2.507694232453397, + "grad_norm": 0.009083365090191364, + "learning_rate": 9.115679042093752e-06, + "loss": 0.3059, + "step": 50110 + }, + { + "epoch": 2.5081946703365445, + "grad_norm": 1.4304612874984741, + "learning_rate": 9.106411373283164e-06, + "loss": 0.0033, + "step": 50120 + }, + { + "epoch": 2.508695108219692, + "grad_norm": 0.040249962359666824, + "learning_rate": 9.097143704472576e-06, + "loss": 0.3283, + "step": 50130 + }, + { + "epoch": 2.50919554610284, + "grad_norm": 1.2305465936660767, + "learning_rate": 9.087876035661989e-06, + "loss": 0.6231, + "step": 50140 + }, + { + "epoch": 2.5096959839859876, + "grad_norm": 0.7901120781898499, + "learning_rate": 9.078608366851403e-06, + "loss": 0.3743, + "step": 50150 + }, + { + "epoch": 2.5101964218691357, + "grad_norm": 1.4926519393920898, + "learning_rate": 9.069340698040815e-06, + "loss": 0.2604, + "step": 50160 + }, + { + "epoch": 2.5106968597522834, + "grad_norm": 5.039576530456543, + "learning_rate": 9.060073029230227e-06, + "loss": 0.3556, + "step": 50170 + }, + { + "epoch": 2.511197297635431, + "grad_norm": 0.008420703001320362, + "learning_rate": 9.050805360419641e-06, + "loss": 0.429, + "step": 50180 + }, + { + "epoch": 2.5116977355185788, + "grad_norm": 1.056878924369812, + "learning_rate": 9.041537691609053e-06, + "loss": 0.2243, + "step": 50190 + }, + { + "epoch": 2.5121981734017265, + "grad_norm": 107.99441528320312, + "learning_rate": 9.032270022798466e-06, + "loss": 0.1502, + "step": 50200 + }, + { + "epoch": 2.512698611284874, + "grad_norm": 32.20003890991211, + "learning_rate": 9.023002353987878e-06, + "loss": 0.2314, + "step": 50210 + }, + { + "epoch": 2.513199049168022, + "grad_norm": 0.004684048239141703, + "learning_rate": 9.013734685177292e-06, + "loss": 0.1092, + "step": 50220 + }, + { + "epoch": 2.51369948705117, + "grad_norm": 22.15615463256836, + "learning_rate": 9.004467016366704e-06, + "loss": 0.2204, + "step": 50230 + }, + { + "epoch": 2.5141999249343177, + "grad_norm": 0.11279484629631042, + "learning_rate": 8.995199347556116e-06, + "loss": 0.2411, + "step": 50240 + }, + { + "epoch": 2.5147003628174653, + "grad_norm": 42.21040725708008, + "learning_rate": 8.98593167874553e-06, + "loss": 0.1344, + "step": 50250 + }, + { + "epoch": 2.515200800700613, + "grad_norm": 0.03478037193417549, + "learning_rate": 8.976664009934942e-06, + "loss": 0.3488, + "step": 50260 + }, + { + "epoch": 2.5157012385837607, + "grad_norm": 0.010213949717581272, + "learning_rate": 8.967396341124355e-06, + "loss": 0.6846, + "step": 50270 + }, + { + "epoch": 2.5162016764669084, + "grad_norm": 0.004609443712979555, + "learning_rate": 8.958128672313767e-06, + "loss": 0.0084, + "step": 50280 + }, + { + "epoch": 2.516702114350056, + "grad_norm": 0.12136012315750122, + "learning_rate": 8.94886100350318e-06, + "loss": 0.0924, + "step": 50290 + }, + { + "epoch": 2.517202552233204, + "grad_norm": 0.07209935784339905, + "learning_rate": 8.939593334692591e-06, + "loss": 0.2963, + "step": 50300 + }, + { + "epoch": 2.517702990116352, + "grad_norm": 64.69906616210938, + "learning_rate": 8.930325665882004e-06, + "loss": 0.4713, + "step": 50310 + }, + { + "epoch": 2.5182034279994996, + "grad_norm": 0.002834426937624812, + "learning_rate": 8.921057997071416e-06, + "loss": 0.3407, + "step": 50320 + }, + { + "epoch": 2.5187038658826473, + "grad_norm": 0.0048147025518119335, + "learning_rate": 8.91179032826083e-06, + "loss": 0.0831, + "step": 50330 + }, + { + "epoch": 2.519204303765795, + "grad_norm": 0.005600831937044859, + "learning_rate": 8.902522659450242e-06, + "loss": 0.2456, + "step": 50340 + }, + { + "epoch": 2.519704741648943, + "grad_norm": 1.1772459745407104, + "learning_rate": 8.893254990639654e-06, + "loss": 0.1562, + "step": 50350 + }, + { + "epoch": 2.5202051795320903, + "grad_norm": 21.685794830322266, + "learning_rate": 8.883987321829068e-06, + "loss": 0.3596, + "step": 50360 + }, + { + "epoch": 2.5207056174152385, + "grad_norm": 0.013437144458293915, + "learning_rate": 8.87471965301848e-06, + "loss": 0.5416, + "step": 50370 + }, + { + "epoch": 2.521206055298386, + "grad_norm": 0.0006854833336547017, + "learning_rate": 8.865451984207893e-06, + "loss": 0.1323, + "step": 50380 + }, + { + "epoch": 2.521706493181534, + "grad_norm": 0.1378629058599472, + "learning_rate": 8.856184315397305e-06, + "loss": 0.1355, + "step": 50390 + }, + { + "epoch": 2.5222069310646815, + "grad_norm": 0.023243749514222145, + "learning_rate": 8.846916646586719e-06, + "loss": 0.1945, + "step": 50400 + }, + { + "epoch": 2.522707368947829, + "grad_norm": 80.98289489746094, + "learning_rate": 8.837648977776131e-06, + "loss": 0.3571, + "step": 50410 + }, + { + "epoch": 2.5232078068309773, + "grad_norm": 6.317680835723877, + "learning_rate": 8.828381308965543e-06, + "loss": 0.2094, + "step": 50420 + }, + { + "epoch": 2.5237082447141246, + "grad_norm": 16.898298263549805, + "learning_rate": 8.819113640154956e-06, + "loss": 0.2865, + "step": 50430 + }, + { + "epoch": 2.5242086825972727, + "grad_norm": 0.025098592042922974, + "learning_rate": 8.809845971344368e-06, + "loss": 0.2386, + "step": 50440 + }, + { + "epoch": 2.5247091204804204, + "grad_norm": 0.017326464876532555, + "learning_rate": 8.80057830253378e-06, + "loss": 0.4138, + "step": 50450 + }, + { + "epoch": 2.525209558363568, + "grad_norm": 0.010899503715336323, + "learning_rate": 8.791310633723192e-06, + "loss": 0.3277, + "step": 50460 + }, + { + "epoch": 2.5257099962467158, + "grad_norm": 0.0020219022408127785, + "learning_rate": 8.782042964912606e-06, + "loss": 0.2023, + "step": 50470 + }, + { + "epoch": 2.5262104341298635, + "grad_norm": 122.02760314941406, + "learning_rate": 8.772775296102019e-06, + "loss": 0.2213, + "step": 50480 + }, + { + "epoch": 2.5267108720130116, + "grad_norm": 0.041648365557193756, + "learning_rate": 8.763507627291431e-06, + "loss": 0.3999, + "step": 50490 + }, + { + "epoch": 2.5272113098961593, + "grad_norm": 27.885128021240234, + "learning_rate": 8.754239958480843e-06, + "loss": 0.4148, + "step": 50500 + }, + { + "epoch": 2.527711747779307, + "grad_norm": 0.027384595945477486, + "learning_rate": 8.744972289670257e-06, + "loss": 0.1024, + "step": 50510 + }, + { + "epoch": 2.5282121856624546, + "grad_norm": 62.9351806640625, + "learning_rate": 8.73570462085967e-06, + "loss": 0.24, + "step": 50520 + }, + { + "epoch": 2.5287126235456023, + "grad_norm": 0.06739214807748795, + "learning_rate": 8.726436952049082e-06, + "loss": 0.1198, + "step": 50530 + }, + { + "epoch": 2.52921306142875, + "grad_norm": 0.009853660129010677, + "learning_rate": 8.717169283238495e-06, + "loss": 0.429, + "step": 50540 + }, + { + "epoch": 2.5297134993118977, + "grad_norm": 3.2411370277404785, + "learning_rate": 8.707901614427908e-06, + "loss": 0.1108, + "step": 50550 + }, + { + "epoch": 2.530213937195046, + "grad_norm": 0.09043240547180176, + "learning_rate": 8.69863394561732e-06, + "loss": 0.0465, + "step": 50560 + }, + { + "epoch": 2.5307143750781935, + "grad_norm": 23.706024169921875, + "learning_rate": 8.689366276806732e-06, + "loss": 0.3912, + "step": 50570 + }, + { + "epoch": 2.531214812961341, + "grad_norm": 50.148983001708984, + "learning_rate": 8.680098607996146e-06, + "loss": 0.447, + "step": 50580 + }, + { + "epoch": 2.531715250844489, + "grad_norm": 153.34732055664062, + "learning_rate": 8.670830939185558e-06, + "loss": 0.3426, + "step": 50590 + }, + { + "epoch": 2.5322156887276366, + "grad_norm": 0.02839621715247631, + "learning_rate": 8.66156327037497e-06, + "loss": 0.2586, + "step": 50600 + }, + { + "epoch": 2.5327161266107847, + "grad_norm": 0.07615655660629272, + "learning_rate": 8.652295601564383e-06, + "loss": 0.1123, + "step": 50610 + }, + { + "epoch": 2.533216564493932, + "grad_norm": 20.799381256103516, + "learning_rate": 8.643027932753795e-06, + "loss": 0.2709, + "step": 50620 + }, + { + "epoch": 2.53371700237708, + "grad_norm": 0.022275077179074287, + "learning_rate": 8.633760263943207e-06, + "loss": 0.0838, + "step": 50630 + }, + { + "epoch": 2.5342174402602278, + "grad_norm": 0.0691380500793457, + "learning_rate": 8.62449259513262e-06, + "loss": 0.0382, + "step": 50640 + }, + { + "epoch": 2.5347178781433755, + "grad_norm": 0.1672557145357132, + "learning_rate": 8.615224926322034e-06, + "loss": 0.5715, + "step": 50650 + }, + { + "epoch": 2.535218316026523, + "grad_norm": 77.3583984375, + "learning_rate": 8.605957257511446e-06, + "loss": 0.1384, + "step": 50660 + }, + { + "epoch": 2.535718753909671, + "grad_norm": 32.251373291015625, + "learning_rate": 8.596689588700858e-06, + "loss": 0.3706, + "step": 50670 + }, + { + "epoch": 2.536219191792819, + "grad_norm": 0.03633183240890503, + "learning_rate": 8.58742191989027e-06, + "loss": 0.0431, + "step": 50680 + }, + { + "epoch": 2.536719629675966, + "grad_norm": 3.39093279838562, + "learning_rate": 8.578154251079684e-06, + "loss": 0.0702, + "step": 50690 + }, + { + "epoch": 2.5372200675591143, + "grad_norm": 0.029101818799972534, + "learning_rate": 8.568886582269096e-06, + "loss": 0.311, + "step": 50700 + }, + { + "epoch": 2.537720505442262, + "grad_norm": 0.01208474114537239, + "learning_rate": 8.559618913458509e-06, + "loss": 0.283, + "step": 50710 + }, + { + "epoch": 2.5382209433254097, + "grad_norm": 0.004686221480369568, + "learning_rate": 8.550351244647923e-06, + "loss": 0.4135, + "step": 50720 + }, + { + "epoch": 2.5387213812085574, + "grad_norm": 0.059223540127277374, + "learning_rate": 8.541083575837335e-06, + "loss": 0.0141, + "step": 50730 + }, + { + "epoch": 2.539221819091705, + "grad_norm": 0.08812157064676285, + "learning_rate": 8.531815907026747e-06, + "loss": 0.132, + "step": 50740 + }, + { + "epoch": 2.539722256974853, + "grad_norm": 27.426334381103516, + "learning_rate": 8.52254823821616e-06, + "loss": 0.2929, + "step": 50750 + }, + { + "epoch": 2.540222694858001, + "grad_norm": 0.5335895419120789, + "learning_rate": 8.513280569405573e-06, + "loss": 0.0765, + "step": 50760 + }, + { + "epoch": 2.5407231327411486, + "grad_norm": 37.13701248168945, + "learning_rate": 8.504012900594986e-06, + "loss": 0.5186, + "step": 50770 + }, + { + "epoch": 2.5412235706242963, + "grad_norm": 0.007291588932275772, + "learning_rate": 8.494745231784398e-06, + "loss": 0.3578, + "step": 50780 + }, + { + "epoch": 2.541724008507444, + "grad_norm": 13.138517379760742, + "learning_rate": 8.48547756297381e-06, + "loss": 0.2028, + "step": 50790 + }, + { + "epoch": 2.5422244463905916, + "grad_norm": 0.04381929337978363, + "learning_rate": 8.476209894163222e-06, + "loss": 0.1546, + "step": 50800 + }, + { + "epoch": 2.5427248842737393, + "grad_norm": 42.237064361572266, + "learning_rate": 8.466942225352635e-06, + "loss": 0.102, + "step": 50810 + }, + { + "epoch": 2.5432253221568875, + "grad_norm": 0.007749517913907766, + "learning_rate": 8.457674556542047e-06, + "loss": 0.0509, + "step": 50820 + }, + { + "epoch": 2.543725760040035, + "grad_norm": 0.0033161137253046036, + "learning_rate": 8.44840688773146e-06, + "loss": 0.4528, + "step": 50830 + }, + { + "epoch": 2.544226197923183, + "grad_norm": 88.66057586669922, + "learning_rate": 8.439139218920873e-06, + "loss": 0.4716, + "step": 50840 + }, + { + "epoch": 2.5447266358063305, + "grad_norm": 0.023146288469433784, + "learning_rate": 8.429871550110285e-06, + "loss": 0.4237, + "step": 50850 + }, + { + "epoch": 2.545227073689478, + "grad_norm": 0.3623051345348358, + "learning_rate": 8.420603881299698e-06, + "loss": 0.4005, + "step": 50860 + }, + { + "epoch": 2.545727511572626, + "grad_norm": 1.0579394102096558, + "learning_rate": 8.411336212489111e-06, + "loss": 0.2135, + "step": 50870 + }, + { + "epoch": 2.5462279494557736, + "grad_norm": 209.40365600585938, + "learning_rate": 8.402068543678524e-06, + "loss": 0.6191, + "step": 50880 + }, + { + "epoch": 2.5467283873389217, + "grad_norm": 36.30597686767578, + "learning_rate": 8.392800874867936e-06, + "loss": 0.38, + "step": 50890 + }, + { + "epoch": 2.5472288252220694, + "grad_norm": 0.4074292480945587, + "learning_rate": 8.38353320605735e-06, + "loss": 0.0836, + "step": 50900 + }, + { + "epoch": 2.547729263105217, + "grad_norm": 7.507017135620117, + "learning_rate": 8.374265537246762e-06, + "loss": 0.2505, + "step": 50910 + }, + { + "epoch": 2.5482297009883648, + "grad_norm": 6.353564739227295, + "learning_rate": 8.364997868436174e-06, + "loss": 0.1719, + "step": 50920 + }, + { + "epoch": 2.5487301388715125, + "grad_norm": 0.07381042093038559, + "learning_rate": 8.355730199625587e-06, + "loss": 0.2643, + "step": 50930 + }, + { + "epoch": 2.5492305767546606, + "grad_norm": 0.0025331934448331594, + "learning_rate": 8.346462530814999e-06, + "loss": 0.5038, + "step": 50940 + }, + { + "epoch": 2.549731014637808, + "grad_norm": 0.07080531120300293, + "learning_rate": 8.337194862004411e-06, + "loss": 0.4126, + "step": 50950 + }, + { + "epoch": 2.550231452520956, + "grad_norm": 54.66993713378906, + "learning_rate": 8.327927193193823e-06, + "loss": 0.0662, + "step": 50960 + }, + { + "epoch": 2.5507318904041036, + "grad_norm": 18.107858657836914, + "learning_rate": 8.318659524383237e-06, + "loss": 0.3231, + "step": 50970 + }, + { + "epoch": 2.5512323282872513, + "grad_norm": 3.4094717502593994, + "learning_rate": 8.30939185557265e-06, + "loss": 0.2464, + "step": 50980 + }, + { + "epoch": 2.551732766170399, + "grad_norm": 0.012637845240533352, + "learning_rate": 8.300124186762062e-06, + "loss": 0.0042, + "step": 50990 + }, + { + "epoch": 2.5522332040535467, + "grad_norm": 0.07471317797899246, + "learning_rate": 8.290856517951474e-06, + "loss": 0.3467, + "step": 51000 + }, + { + "epoch": 2.552733641936695, + "grad_norm": 2.9882328510284424, + "learning_rate": 8.281588849140888e-06, + "loss": 0.3361, + "step": 51010 + }, + { + "epoch": 2.5532340798198425, + "grad_norm": 0.08038809895515442, + "learning_rate": 8.2723211803303e-06, + "loss": 0.1862, + "step": 51020 + }, + { + "epoch": 2.55373451770299, + "grad_norm": 0.008654486387968063, + "learning_rate": 8.263053511519712e-06, + "loss": 0.2676, + "step": 51030 + }, + { + "epoch": 2.554234955586138, + "grad_norm": 0.11457152664661407, + "learning_rate": 8.253785842709125e-06, + "loss": 0.2304, + "step": 51040 + }, + { + "epoch": 2.5547353934692856, + "grad_norm": 0.0173015296459198, + "learning_rate": 8.244518173898539e-06, + "loss": 0.8369, + "step": 51050 + }, + { + "epoch": 2.5552358313524333, + "grad_norm": 0.09907073527574539, + "learning_rate": 8.235250505087951e-06, + "loss": 0.3685, + "step": 51060 + }, + { + "epoch": 2.555736269235581, + "grad_norm": 0.004656667821109295, + "learning_rate": 8.225982836277363e-06, + "loss": 0.0237, + "step": 51070 + }, + { + "epoch": 2.556236707118729, + "grad_norm": 30.800613403320312, + "learning_rate": 8.216715167466777e-06, + "loss": 0.3897, + "step": 51080 + }, + { + "epoch": 2.5567371450018768, + "grad_norm": 0.01568758115172386, + "learning_rate": 8.20744749865619e-06, + "loss": 0.8031, + "step": 51090 + }, + { + "epoch": 2.5572375828850245, + "grad_norm": 9.750901222229004, + "learning_rate": 8.198179829845602e-06, + "loss": 0.1411, + "step": 51100 + }, + { + "epoch": 2.557738020768172, + "grad_norm": 0.06261958926916122, + "learning_rate": 8.188912161035014e-06, + "loss": 0.0508, + "step": 51110 + }, + { + "epoch": 2.55823845865132, + "grad_norm": 0.021754607558250427, + "learning_rate": 8.179644492224426e-06, + "loss": 0.6241, + "step": 51120 + }, + { + "epoch": 2.5587388965344675, + "grad_norm": 0.18426750600337982, + "learning_rate": 8.170376823413838e-06, + "loss": 0.0157, + "step": 51130 + }, + { + "epoch": 2.559239334417615, + "grad_norm": 2.462371826171875, + "learning_rate": 8.16110915460325e-06, + "loss": 0.5344, + "step": 51140 + }, + { + "epoch": 2.5597397723007633, + "grad_norm": 53.20009994506836, + "learning_rate": 8.151841485792665e-06, + "loss": 0.1473, + "step": 51150 + }, + { + "epoch": 2.560240210183911, + "grad_norm": 0.054167989641427994, + "learning_rate": 8.142573816982077e-06, + "loss": 0.156, + "step": 51160 + }, + { + "epoch": 2.5607406480670587, + "grad_norm": 72.88970947265625, + "learning_rate": 8.133306148171489e-06, + "loss": 0.2512, + "step": 51170 + }, + { + "epoch": 2.5612410859502064, + "grad_norm": 0.7215125560760498, + "learning_rate": 8.124038479360901e-06, + "loss": 0.3787, + "step": 51180 + }, + { + "epoch": 2.561741523833354, + "grad_norm": 0.05100720748305321, + "learning_rate": 8.114770810550315e-06, + "loss": 0.2166, + "step": 51190 + }, + { + "epoch": 2.562241961716502, + "grad_norm": 2.2231333255767822, + "learning_rate": 8.105503141739727e-06, + "loss": 0.06, + "step": 51200 + }, + { + "epoch": 2.5627423995996494, + "grad_norm": 0.0044437856413424015, + "learning_rate": 8.09623547292914e-06, + "loss": 0.0445, + "step": 51210 + }, + { + "epoch": 2.5632428374827976, + "grad_norm": 135.8170928955078, + "learning_rate": 8.086967804118552e-06, + "loss": 0.2297, + "step": 51220 + }, + { + "epoch": 2.5637432753659453, + "grad_norm": 0.014283808879554272, + "learning_rate": 8.077700135307966e-06, + "loss": 0.5996, + "step": 51230 + }, + { + "epoch": 2.564243713249093, + "grad_norm": 10.648377418518066, + "learning_rate": 8.068432466497378e-06, + "loss": 0.6977, + "step": 51240 + }, + { + "epoch": 2.5647441511322406, + "grad_norm": 1.8232805728912354, + "learning_rate": 8.05916479768679e-06, + "loss": 0.1393, + "step": 51250 + }, + { + "epoch": 2.5652445890153883, + "grad_norm": 0.0024745664559304714, + "learning_rate": 8.049897128876204e-06, + "loss": 0.1853, + "step": 51260 + }, + { + "epoch": 2.5657450268985365, + "grad_norm": 0.006224095355719328, + "learning_rate": 8.040629460065615e-06, + "loss": 0.0232, + "step": 51270 + }, + { + "epoch": 2.566245464781684, + "grad_norm": 0.6033616065979004, + "learning_rate": 8.031361791255027e-06, + "loss": 0.1802, + "step": 51280 + }, + { + "epoch": 2.566745902664832, + "grad_norm": 0.6958104372024536, + "learning_rate": 8.02209412244444e-06, + "loss": 0.382, + "step": 51290 + }, + { + "epoch": 2.5672463405479795, + "grad_norm": 17.954374313354492, + "learning_rate": 8.012826453633853e-06, + "loss": 0.0154, + "step": 51300 + }, + { + "epoch": 2.567746778431127, + "grad_norm": 14.034693717956543, + "learning_rate": 8.003558784823266e-06, + "loss": 0.1717, + "step": 51310 + }, + { + "epoch": 2.568247216314275, + "grad_norm": 0.025846462696790695, + "learning_rate": 7.994291116012678e-06, + "loss": 0.2363, + "step": 51320 + }, + { + "epoch": 2.5687476541974226, + "grad_norm": 0.019017312675714493, + "learning_rate": 7.985023447202092e-06, + "loss": 0.1127, + "step": 51330 + }, + { + "epoch": 2.5692480920805707, + "grad_norm": 0.007190005853772163, + "learning_rate": 7.975755778391504e-06, + "loss": 0.2077, + "step": 51340 + }, + { + "epoch": 2.5697485299637184, + "grad_norm": 0.4465359151363373, + "learning_rate": 7.966488109580916e-06, + "loss": 0.1424, + "step": 51350 + }, + { + "epoch": 2.570248967846866, + "grad_norm": 0.028042752295732498, + "learning_rate": 7.957220440770328e-06, + "loss": 0.0636, + "step": 51360 + }, + { + "epoch": 2.5707494057300138, + "grad_norm": 4.2051682472229, + "learning_rate": 7.947952771959742e-06, + "loss": 0.1584, + "step": 51370 + }, + { + "epoch": 2.5712498436131614, + "grad_norm": 0.010359902866184711, + "learning_rate": 7.938685103149155e-06, + "loss": 0.0392, + "step": 51380 + }, + { + "epoch": 2.571750281496309, + "grad_norm": 1.9568545818328857, + "learning_rate": 7.929417434338567e-06, + "loss": 0.2499, + "step": 51390 + }, + { + "epoch": 2.572250719379457, + "grad_norm": 96.29537963867188, + "learning_rate": 7.920149765527979e-06, + "loss": 0.4412, + "step": 51400 + }, + { + "epoch": 2.572751157262605, + "grad_norm": 0.002529341261833906, + "learning_rate": 7.910882096717393e-06, + "loss": 0.2979, + "step": 51410 + }, + { + "epoch": 2.5732515951457526, + "grad_norm": 48.381011962890625, + "learning_rate": 7.901614427906805e-06, + "loss": 0.287, + "step": 51420 + }, + { + "epoch": 2.5737520330289003, + "grad_norm": 0.006842020433396101, + "learning_rate": 7.892346759096218e-06, + "loss": 0.2359, + "step": 51430 + }, + { + "epoch": 2.574252470912048, + "grad_norm": 23.082618713378906, + "learning_rate": 7.88307909028563e-06, + "loss": 0.4374, + "step": 51440 + }, + { + "epoch": 2.5747529087951957, + "grad_norm": 0.030651826411485672, + "learning_rate": 7.873811421475042e-06, + "loss": 0.1633, + "step": 51450 + }, + { + "epoch": 2.575253346678344, + "grad_norm": 11.766926765441895, + "learning_rate": 7.864543752664454e-06, + "loss": 0.5065, + "step": 51460 + }, + { + "epoch": 2.575753784561491, + "grad_norm": 0.044369492679834366, + "learning_rate": 7.855276083853867e-06, + "loss": 0.3243, + "step": 51470 + }, + { + "epoch": 2.576254222444639, + "grad_norm": 61.72634506225586, + "learning_rate": 7.84600841504328e-06, + "loss": 0.1021, + "step": 51480 + }, + { + "epoch": 2.576754660327787, + "grad_norm": 9.931803703308105, + "learning_rate": 7.836740746232693e-06, + "loss": 0.5061, + "step": 51490 + }, + { + "epoch": 2.5772550982109346, + "grad_norm": 34.75297164916992, + "learning_rate": 7.827473077422105e-06, + "loss": 0.2717, + "step": 51500 + }, + { + "epoch": 2.5777555360940823, + "grad_norm": 0.0015720182564109564, + "learning_rate": 7.818205408611519e-06, + "loss": 0.0108, + "step": 51510 + }, + { + "epoch": 2.57825597397723, + "grad_norm": 121.95532989501953, + "learning_rate": 7.808937739800931e-06, + "loss": 0.1719, + "step": 51520 + }, + { + "epoch": 2.578756411860378, + "grad_norm": 1.1453118324279785, + "learning_rate": 7.799670070990343e-06, + "loss": 0.1273, + "step": 51530 + }, + { + "epoch": 2.5792568497435253, + "grad_norm": 0.03382551297545433, + "learning_rate": 7.790402402179756e-06, + "loss": 0.1706, + "step": 51540 + }, + { + "epoch": 2.5797572876266734, + "grad_norm": 0.05402049422264099, + "learning_rate": 7.78113473336917e-06, + "loss": 0.2832, + "step": 51550 + }, + { + "epoch": 2.580257725509821, + "grad_norm": 0.3803606629371643, + "learning_rate": 7.771867064558582e-06, + "loss": 0.1052, + "step": 51560 + }, + { + "epoch": 2.580758163392969, + "grad_norm": 0.003631806932389736, + "learning_rate": 7.762599395747994e-06, + "loss": 0.4909, + "step": 51570 + }, + { + "epoch": 2.5812586012761165, + "grad_norm": 0.01680920273065567, + "learning_rate": 7.753331726937406e-06, + "loss": 0.0029, + "step": 51580 + }, + { + "epoch": 2.581759039159264, + "grad_norm": 0.5403025150299072, + "learning_rate": 7.74406405812682e-06, + "loss": 0.242, + "step": 51590 + }, + { + "epoch": 2.5822594770424123, + "grad_norm": 22.687776565551758, + "learning_rate": 7.734796389316233e-06, + "loss": 0.3041, + "step": 51600 + }, + { + "epoch": 2.58275991492556, + "grad_norm": 26.03846549987793, + "learning_rate": 7.725528720505645e-06, + "loss": 0.1807, + "step": 51610 + }, + { + "epoch": 2.5832603528087077, + "grad_norm": 32.02401351928711, + "learning_rate": 7.716261051695057e-06, + "loss": 0.0341, + "step": 51620 + }, + { + "epoch": 2.5837607906918554, + "grad_norm": 0.9253588914871216, + "learning_rate": 7.70699338288447e-06, + "loss": 0.0055, + "step": 51630 + }, + { + "epoch": 2.584261228575003, + "grad_norm": 0.01875016652047634, + "learning_rate": 7.697725714073882e-06, + "loss": 0.6152, + "step": 51640 + }, + { + "epoch": 2.5847616664581508, + "grad_norm": 80.61717224121094, + "learning_rate": 7.688458045263294e-06, + "loss": 0.6299, + "step": 51650 + }, + { + "epoch": 2.5852621043412984, + "grad_norm": 124.93114471435547, + "learning_rate": 7.679190376452708e-06, + "loss": 0.157, + "step": 51660 + }, + { + "epoch": 2.5857625422244466, + "grad_norm": 66.57306671142578, + "learning_rate": 7.66992270764212e-06, + "loss": 0.2773, + "step": 51670 + }, + { + "epoch": 2.5862629801075943, + "grad_norm": 0.17086796462535858, + "learning_rate": 7.660655038831532e-06, + "loss": 0.2018, + "step": 51680 + }, + { + "epoch": 2.586763417990742, + "grad_norm": 0.16188478469848633, + "learning_rate": 7.651387370020946e-06, + "loss": 0.174, + "step": 51690 + }, + { + "epoch": 2.5872638558738896, + "grad_norm": 0.8770257234573364, + "learning_rate": 7.642119701210358e-06, + "loss": 0.2388, + "step": 51700 + }, + { + "epoch": 2.5877642937570373, + "grad_norm": 0.7238094210624695, + "learning_rate": 7.63285203239977e-06, + "loss": 0.4644, + "step": 51710 + }, + { + "epoch": 2.588264731640185, + "grad_norm": 0.009960212744772434, + "learning_rate": 7.623584363589183e-06, + "loss": 0.2655, + "step": 51720 + }, + { + "epoch": 2.5887651695233327, + "grad_norm": 106.34603881835938, + "learning_rate": 7.614316694778596e-06, + "loss": 0.0397, + "step": 51730 + }, + { + "epoch": 2.589265607406481, + "grad_norm": 0.2729547321796417, + "learning_rate": 7.605049025968008e-06, + "loss": 0.021, + "step": 51740 + }, + { + "epoch": 2.5897660452896285, + "grad_norm": 0.12138912826776505, + "learning_rate": 7.5957813571574204e-06, + "loss": 0.4881, + "step": 51750 + }, + { + "epoch": 2.590266483172776, + "grad_norm": 0.07597094029188156, + "learning_rate": 7.586513688346833e-06, + "loss": 0.4294, + "step": 51760 + }, + { + "epoch": 2.590766921055924, + "grad_norm": 0.03761398419737816, + "learning_rate": 7.577246019536247e-06, + "loss": 0.3197, + "step": 51770 + }, + { + "epoch": 2.5912673589390716, + "grad_norm": 3.108227014541626, + "learning_rate": 7.567978350725659e-06, + "loss": 0.569, + "step": 51780 + }, + { + "epoch": 2.5917677968222197, + "grad_norm": 0.029373658820986748, + "learning_rate": 7.558710681915071e-06, + "loss": 0.3294, + "step": 51790 + }, + { + "epoch": 2.592268234705367, + "grad_norm": 2.84481143951416, + "learning_rate": 7.549443013104485e-06, + "loss": 0.4157, + "step": 51800 + }, + { + "epoch": 2.592768672588515, + "grad_norm": 18.17710304260254, + "learning_rate": 7.5401753442938965e-06, + "loss": 0.326, + "step": 51810 + }, + { + "epoch": 2.5932691104716628, + "grad_norm": 0.06834849715232849, + "learning_rate": 7.530907675483309e-06, + "loss": 0.1566, + "step": 51820 + }, + { + "epoch": 2.5937695483548104, + "grad_norm": 0.005538568831980228, + "learning_rate": 7.521640006672721e-06, + "loss": 0.0716, + "step": 51830 + }, + { + "epoch": 2.594269986237958, + "grad_norm": 0.24147456884384155, + "learning_rate": 7.512372337862135e-06, + "loss": 0.0628, + "step": 51840 + }, + { + "epoch": 2.594770424121106, + "grad_norm": 0.023202234879136086, + "learning_rate": 7.503104669051547e-06, + "loss": 0.0501, + "step": 51850 + }, + { + "epoch": 2.595270862004254, + "grad_norm": 0.011290473863482475, + "learning_rate": 7.493837000240959e-06, + "loss": 0.3112, + "step": 51860 + }, + { + "epoch": 2.5957712998874016, + "grad_norm": 0.030067047104239464, + "learning_rate": 7.484569331430373e-06, + "loss": 0.1693, + "step": 51870 + }, + { + "epoch": 2.5962717377705493, + "grad_norm": 0.029676998034119606, + "learning_rate": 7.4753016626197856e-06, + "loss": 0.4075, + "step": 51880 + }, + { + "epoch": 2.596772175653697, + "grad_norm": 0.09982117265462875, + "learning_rate": 7.466033993809198e-06, + "loss": 0.147, + "step": 51890 + }, + { + "epoch": 2.5972726135368447, + "grad_norm": 0.052596211433410645, + "learning_rate": 7.45676632499861e-06, + "loss": 0.1658, + "step": 51900 + }, + { + "epoch": 2.5977730514199924, + "grad_norm": 0.005605360493063927, + "learning_rate": 7.447498656188023e-06, + "loss": 0.3458, + "step": 51910 + }, + { + "epoch": 2.59827348930314, + "grad_norm": 0.0027085398323833942, + "learning_rate": 7.438230987377435e-06, + "loss": 0.0417, + "step": 51920 + }, + { + "epoch": 2.598773927186288, + "grad_norm": 37.90739822387695, + "learning_rate": 7.428963318566848e-06, + "loss": 0.071, + "step": 51930 + }, + { + "epoch": 2.599274365069436, + "grad_norm": 0.05480165779590607, + "learning_rate": 7.41969564975626e-06, + "loss": 0.1163, + "step": 51940 + }, + { + "epoch": 2.5997748029525836, + "grad_norm": 0.37215304374694824, + "learning_rate": 7.410427980945674e-06, + "loss": 0.4272, + "step": 51950 + }, + { + "epoch": 2.6002752408357312, + "grad_norm": 16.1798152923584, + "learning_rate": 7.401160312135086e-06, + "loss": 0.0895, + "step": 51960 + }, + { + "epoch": 2.600775678718879, + "grad_norm": 3.0715596675872803, + "learning_rate": 7.391892643324498e-06, + "loss": 0.2411, + "step": 51970 + }, + { + "epoch": 2.6012761166020266, + "grad_norm": 1.4701327085494995, + "learning_rate": 7.382624974513911e-06, + "loss": 0.4378, + "step": 51980 + }, + { + "epoch": 2.6017765544851743, + "grad_norm": 60.56190490722656, + "learning_rate": 7.373357305703324e-06, + "loss": 0.4352, + "step": 51990 + }, + { + "epoch": 2.6022769923683224, + "grad_norm": 0.02230219729244709, + "learning_rate": 7.364089636892736e-06, + "loss": 0.2032, + "step": 52000 + }, + { + "epoch": 2.60277743025147, + "grad_norm": 0.48772621154785156, + "learning_rate": 7.354821968082148e-06, + "loss": 0.211, + "step": 52010 + }, + { + "epoch": 2.603277868134618, + "grad_norm": 0.011699642054736614, + "learning_rate": 7.345554299271562e-06, + "loss": 0.3396, + "step": 52020 + }, + { + "epoch": 2.6037783060177655, + "grad_norm": 0.09288761019706726, + "learning_rate": 7.336286630460974e-06, + "loss": 0.316, + "step": 52030 + }, + { + "epoch": 2.604278743900913, + "grad_norm": 0.005879935342818499, + "learning_rate": 7.327018961650387e-06, + "loss": 0.002, + "step": 52040 + }, + { + "epoch": 2.6047791817840613, + "grad_norm": 0.4829367697238922, + "learning_rate": 7.3177512928398e-06, + "loss": 0.1019, + "step": 52050 + }, + { + "epoch": 2.6052796196672086, + "grad_norm": 0.010386830195784569, + "learning_rate": 7.308483624029212e-06, + "loss": 0.1301, + "step": 52060 + }, + { + "epoch": 2.6057800575503567, + "grad_norm": 98.31361389160156, + "learning_rate": 7.299215955218624e-06, + "loss": 0.1582, + "step": 52070 + }, + { + "epoch": 2.6062804954335044, + "grad_norm": 0.17241540551185608, + "learning_rate": 7.289948286408036e-06, + "loss": 0.0058, + "step": 52080 + }, + { + "epoch": 2.606780933316652, + "grad_norm": 0.00866694562137127, + "learning_rate": 7.28068061759745e-06, + "loss": 0.1934, + "step": 52090 + }, + { + "epoch": 2.6072813711997997, + "grad_norm": 12.643781661987305, + "learning_rate": 7.271412948786863e-06, + "loss": 0.1379, + "step": 52100 + }, + { + "epoch": 2.6077818090829474, + "grad_norm": 0.10478253662586212, + "learning_rate": 7.262145279976275e-06, + "loss": 0.4349, + "step": 52110 + }, + { + "epoch": 2.6082822469660956, + "grad_norm": 32.53661346435547, + "learning_rate": 7.252877611165687e-06, + "loss": 0.269, + "step": 52120 + }, + { + "epoch": 2.6087826848492432, + "grad_norm": 48.44936752319336, + "learning_rate": 7.243609942355101e-06, + "loss": 0.0543, + "step": 52130 + }, + { + "epoch": 2.609283122732391, + "grad_norm": 80.08451080322266, + "learning_rate": 7.234342273544513e-06, + "loss": 0.1994, + "step": 52140 + }, + { + "epoch": 2.6097835606155386, + "grad_norm": 0.006303401198238134, + "learning_rate": 7.2250746047339255e-06, + "loss": 0.1864, + "step": 52150 + }, + { + "epoch": 2.6102839984986863, + "grad_norm": 1.3577243089675903, + "learning_rate": 7.215806935923339e-06, + "loss": 0.4234, + "step": 52160 + }, + { + "epoch": 2.610784436381834, + "grad_norm": 18.328031539916992, + "learning_rate": 7.206539267112751e-06, + "loss": 0.2429, + "step": 52170 + }, + { + "epoch": 2.6112848742649817, + "grad_norm": 0.004299520514905453, + "learning_rate": 7.197271598302163e-06, + "loss": 0.3409, + "step": 52180 + }, + { + "epoch": 2.61178531214813, + "grad_norm": 239.3016815185547, + "learning_rate": 7.188003929491575e-06, + "loss": 0.4086, + "step": 52190 + }, + { + "epoch": 2.6122857500312775, + "grad_norm": 0.09346451610326767, + "learning_rate": 7.178736260680989e-06, + "loss": 0.4892, + "step": 52200 + }, + { + "epoch": 2.612786187914425, + "grad_norm": 2.720489025115967, + "learning_rate": 7.1694685918704015e-06, + "loss": 0.0755, + "step": 52210 + }, + { + "epoch": 2.613286625797573, + "grad_norm": 0.0038148167077451944, + "learning_rate": 7.160200923059814e-06, + "loss": 0.0125, + "step": 52220 + }, + { + "epoch": 2.6137870636807206, + "grad_norm": 56.91534423828125, + "learning_rate": 7.150933254249227e-06, + "loss": 0.4082, + "step": 52230 + }, + { + "epoch": 2.6142875015638682, + "grad_norm": 0.01905098557472229, + "learning_rate": 7.141665585438639e-06, + "loss": 0.28, + "step": 52240 + }, + { + "epoch": 2.614787939447016, + "grad_norm": 0.010157524608075619, + "learning_rate": 7.132397916628051e-06, + "loss": 0.0012, + "step": 52250 + }, + { + "epoch": 2.615288377330164, + "grad_norm": 0.01358792744576931, + "learning_rate": 7.123130247817464e-06, + "loss": 0.2214, + "step": 52260 + }, + { + "epoch": 2.6157888152133117, + "grad_norm": 25.78679656982422, + "learning_rate": 7.1138625790068776e-06, + "loss": 0.1439, + "step": 52270 + }, + { + "epoch": 2.6162892530964594, + "grad_norm": 0.019232574850320816, + "learning_rate": 7.10459491019629e-06, + "loss": 0.1351, + "step": 52280 + }, + { + "epoch": 2.616789690979607, + "grad_norm": 0.011393689550459385, + "learning_rate": 7.095327241385702e-06, + "loss": 0.2868, + "step": 52290 + }, + { + "epoch": 2.617290128862755, + "grad_norm": 0.019819244742393494, + "learning_rate": 7.086059572575114e-06, + "loss": 0.0222, + "step": 52300 + }, + { + "epoch": 2.617790566745903, + "grad_norm": 0.04753129556775093, + "learning_rate": 7.076791903764527e-06, + "loss": 0.4168, + "step": 52310 + }, + { + "epoch": 2.61829100462905, + "grad_norm": 59.13774108886719, + "learning_rate": 7.06752423495394e-06, + "loss": 0.3155, + "step": 52320 + }, + { + "epoch": 2.6187914425121983, + "grad_norm": 0.001492045703344047, + "learning_rate": 7.058256566143352e-06, + "loss": 0.2571, + "step": 52330 + }, + { + "epoch": 2.619291880395346, + "grad_norm": 87.72908020019531, + "learning_rate": 7.048988897332766e-06, + "loss": 0.1903, + "step": 52340 + }, + { + "epoch": 2.6197923182784937, + "grad_norm": 0.001018738723360002, + "learning_rate": 7.039721228522178e-06, + "loss": 0.3125, + "step": 52350 + }, + { + "epoch": 2.6202927561616414, + "grad_norm": 0.004480537958443165, + "learning_rate": 7.03045355971159e-06, + "loss": 0.2773, + "step": 52360 + }, + { + "epoch": 2.620793194044789, + "grad_norm": 0.41112565994262695, + "learning_rate": 7.0211858909010026e-06, + "loss": 0.2088, + "step": 52370 + }, + { + "epoch": 2.621293631927937, + "grad_norm": 5.500521659851074, + "learning_rate": 7.0119182220904165e-06, + "loss": 0.1689, + "step": 52380 + }, + { + "epoch": 2.6217940698110844, + "grad_norm": 72.07404327392578, + "learning_rate": 7.002650553279829e-06, + "loss": 0.1055, + "step": 52390 + }, + { + "epoch": 2.6222945076942326, + "grad_norm": 0.02349252998828888, + "learning_rate": 6.993382884469241e-06, + "loss": 0.1347, + "step": 52400 + }, + { + "epoch": 2.6227949455773802, + "grad_norm": 0.005084056872874498, + "learning_rate": 6.984115215658654e-06, + "loss": 0.4382, + "step": 52410 + }, + { + "epoch": 2.623295383460528, + "grad_norm": 0.03814268484711647, + "learning_rate": 6.974847546848066e-06, + "loss": 0.2788, + "step": 52420 + }, + { + "epoch": 2.6237958213436756, + "grad_norm": 24.344757080078125, + "learning_rate": 6.965579878037479e-06, + "loss": 0.2486, + "step": 52430 + }, + { + "epoch": 2.6242962592268233, + "grad_norm": 0.16131441295146942, + "learning_rate": 6.956312209226891e-06, + "loss": 0.0898, + "step": 52440 + }, + { + "epoch": 2.6247966971099714, + "grad_norm": 0.0005487446906045079, + "learning_rate": 6.947044540416305e-06, + "loss": 0.423, + "step": 52450 + }, + { + "epoch": 2.625297134993119, + "grad_norm": 2.015648603439331, + "learning_rate": 6.937776871605717e-06, + "loss": 0.0204, + "step": 52460 + }, + { + "epoch": 2.625797572876267, + "grad_norm": 0.005237853154540062, + "learning_rate": 6.928509202795129e-06, + "loss": 0.189, + "step": 52470 + }, + { + "epoch": 2.6262980107594145, + "grad_norm": 0.006784806493669748, + "learning_rate": 6.9192415339845415e-06, + "loss": 0.0362, + "step": 52480 + }, + { + "epoch": 2.626798448642562, + "grad_norm": 37.24993133544922, + "learning_rate": 6.909973865173955e-06, + "loss": 0.4123, + "step": 52490 + }, + { + "epoch": 2.62729888652571, + "grad_norm": 0.006358660291880369, + "learning_rate": 6.900706196363367e-06, + "loss": 0.1775, + "step": 52500 + }, + { + "epoch": 2.6277993244088576, + "grad_norm": 62.688690185546875, + "learning_rate": 6.891438527552779e-06, + "loss": 0.1894, + "step": 52510 + }, + { + "epoch": 2.6282997622920057, + "grad_norm": 0.2833355963230133, + "learning_rate": 6.882170858742193e-06, + "loss": 0.2964, + "step": 52520 + }, + { + "epoch": 2.6288002001751534, + "grad_norm": 0.005282938480377197, + "learning_rate": 6.872903189931605e-06, + "loss": 0.0204, + "step": 52530 + }, + { + "epoch": 2.629300638058301, + "grad_norm": 13.374032974243164, + "learning_rate": 6.8636355211210175e-06, + "loss": 0.4898, + "step": 52540 + }, + { + "epoch": 2.6298010759414487, + "grad_norm": 16.994359970092773, + "learning_rate": 6.85436785231043e-06, + "loss": 0.0892, + "step": 52550 + }, + { + "epoch": 2.6303015138245964, + "grad_norm": 53.371864318847656, + "learning_rate": 6.845100183499843e-06, + "loss": 0.2222, + "step": 52560 + }, + { + "epoch": 2.630801951707744, + "grad_norm": 32.42469787597656, + "learning_rate": 6.835832514689255e-06, + "loss": 0.3959, + "step": 52570 + }, + { + "epoch": 2.631302389590892, + "grad_norm": 0.00275307591073215, + "learning_rate": 6.826564845878667e-06, + "loss": 0.3762, + "step": 52580 + }, + { + "epoch": 2.63180282747404, + "grad_norm": 0.1788710057735443, + "learning_rate": 6.817297177068081e-06, + "loss": 0.3147, + "step": 52590 + }, + { + "epoch": 2.6323032653571876, + "grad_norm": 12.935569763183594, + "learning_rate": 6.8080295082574935e-06, + "loss": 0.0109, + "step": 52600 + }, + { + "epoch": 2.6328037032403353, + "grad_norm": 0.0045880889520049095, + "learning_rate": 6.798761839446906e-06, + "loss": 0.1632, + "step": 52610 + }, + { + "epoch": 2.633304141123483, + "grad_norm": 0.06666282564401627, + "learning_rate": 6.789494170636318e-06, + "loss": 0.4125, + "step": 52620 + }, + { + "epoch": 2.6338045790066307, + "grad_norm": 23.187952041625977, + "learning_rate": 6.780226501825732e-06, + "loss": 0.5151, + "step": 52630 + }, + { + "epoch": 2.634305016889779, + "grad_norm": 0.012001127935945988, + "learning_rate": 6.770958833015144e-06, + "loss": 0.1894, + "step": 52640 + }, + { + "epoch": 2.634805454772926, + "grad_norm": 40.68695831298828, + "learning_rate": 6.7616911642045565e-06, + "loss": 0.4854, + "step": 52650 + }, + { + "epoch": 2.635305892656074, + "grad_norm": 0.028561124578118324, + "learning_rate": 6.752423495393969e-06, + "loss": 0.2927, + "step": 52660 + }, + { + "epoch": 2.635806330539222, + "grad_norm": 20.536754608154297, + "learning_rate": 6.743155826583382e-06, + "loss": 0.21, + "step": 52670 + }, + { + "epoch": 2.6363067684223696, + "grad_norm": 0.02717406675219536, + "learning_rate": 6.733888157772794e-06, + "loss": 0.2228, + "step": 52680 + }, + { + "epoch": 2.6368072063055172, + "grad_norm": 0.345891535282135, + "learning_rate": 6.724620488962206e-06, + "loss": 0.1416, + "step": 52690 + }, + { + "epoch": 2.637307644188665, + "grad_norm": 1.9962353706359863, + "learning_rate": 6.71535282015162e-06, + "loss": 0.55, + "step": 52700 + }, + { + "epoch": 2.637808082071813, + "grad_norm": 0.04349164664745331, + "learning_rate": 6.7060851513410325e-06, + "loss": 0.2986, + "step": 52710 + }, + { + "epoch": 2.6383085199549607, + "grad_norm": 0.20431263744831085, + "learning_rate": 6.696817482530445e-06, + "loss": 0.1024, + "step": 52720 + }, + { + "epoch": 2.6388089578381084, + "grad_norm": 13.42419719696045, + "learning_rate": 6.687549813719857e-06, + "loss": 0.2463, + "step": 52730 + }, + { + "epoch": 2.639309395721256, + "grad_norm": 0.011551082134246826, + "learning_rate": 6.67828214490927e-06, + "loss": 0.0982, + "step": 52740 + }, + { + "epoch": 2.639809833604404, + "grad_norm": 0.018248075619339943, + "learning_rate": 6.669014476098682e-06, + "loss": 0.0972, + "step": 52750 + }, + { + "epoch": 2.6403102714875515, + "grad_norm": 63.713260650634766, + "learning_rate": 6.6597468072880946e-06, + "loss": 0.4374, + "step": 52760 + }, + { + "epoch": 2.640810709370699, + "grad_norm": 0.24352426826953888, + "learning_rate": 6.6504791384775085e-06, + "loss": 0.4735, + "step": 52770 + }, + { + "epoch": 2.6413111472538473, + "grad_norm": 0.04604744166135788, + "learning_rate": 6.641211469666921e-06, + "loss": 0.3255, + "step": 52780 + }, + { + "epoch": 2.641811585136995, + "grad_norm": 0.016437897458672523, + "learning_rate": 6.631943800856333e-06, + "loss": 0.1498, + "step": 52790 + }, + { + "epoch": 2.6423120230201427, + "grad_norm": 60.71615219116211, + "learning_rate": 6.622676132045745e-06, + "loss": 0.0891, + "step": 52800 + }, + { + "epoch": 2.6428124609032904, + "grad_norm": 2.2366764545440674, + "learning_rate": 6.613408463235158e-06, + "loss": 0.5004, + "step": 52810 + }, + { + "epoch": 2.643312898786438, + "grad_norm": 0.10194911062717438, + "learning_rate": 6.6041407944245706e-06, + "loss": 0.2329, + "step": 52820 + }, + { + "epoch": 2.6438133366695857, + "grad_norm": 0.014312250539660454, + "learning_rate": 6.594873125613983e-06, + "loss": 0.5647, + "step": 52830 + }, + { + "epoch": 2.6443137745527334, + "grad_norm": 0.05966499075293541, + "learning_rate": 6.585605456803395e-06, + "loss": 0.2765, + "step": 52840 + }, + { + "epoch": 2.6448142124358816, + "grad_norm": 0.009817738085985184, + "learning_rate": 6.576337787992809e-06, + "loss": 0.4144, + "step": 52850 + }, + { + "epoch": 2.6453146503190292, + "grad_norm": 48.640113830566406, + "learning_rate": 6.567070119182221e-06, + "loss": 0.1998, + "step": 52860 + }, + { + "epoch": 2.645815088202177, + "grad_norm": 0.007171668577939272, + "learning_rate": 6.5578024503716335e-06, + "loss": 0.0616, + "step": 52870 + }, + { + "epoch": 2.6463155260853246, + "grad_norm": 0.5569390654563904, + "learning_rate": 6.5485347815610474e-06, + "loss": 0.1094, + "step": 52880 + }, + { + "epoch": 2.6468159639684723, + "grad_norm": 11.962959289550781, + "learning_rate": 6.53926711275046e-06, + "loss": 0.2416, + "step": 52890 + }, + { + "epoch": 2.6473164018516204, + "grad_norm": 0.3956370949745178, + "learning_rate": 6.529999443939871e-06, + "loss": 0.1224, + "step": 52900 + }, + { + "epoch": 2.6478168397347677, + "grad_norm": 54.20707321166992, + "learning_rate": 6.520731775129283e-06, + "loss": 0.1175, + "step": 52910 + }, + { + "epoch": 2.648317277617916, + "grad_norm": 194.26190185546875, + "learning_rate": 6.511464106318697e-06, + "loss": 0.3181, + "step": 52920 + }, + { + "epoch": 2.6488177155010635, + "grad_norm": 54.66117477416992, + "learning_rate": 6.5021964375081095e-06, + "loss": 0.2358, + "step": 52930 + }, + { + "epoch": 2.649318153384211, + "grad_norm": 40.77069854736328, + "learning_rate": 6.492928768697522e-06, + "loss": 0.1398, + "step": 52940 + }, + { + "epoch": 2.649818591267359, + "grad_norm": 0.011637826450169086, + "learning_rate": 6.483661099886936e-06, + "loss": 0.6167, + "step": 52950 + }, + { + "epoch": 2.6503190291505065, + "grad_norm": 96.60619354248047, + "learning_rate": 6.474393431076348e-06, + "loss": 0.1553, + "step": 52960 + }, + { + "epoch": 2.6508194670336547, + "grad_norm": 0.028132647275924683, + "learning_rate": 6.46512576226576e-06, + "loss": 0.6265, + "step": 52970 + }, + { + "epoch": 2.6513199049168024, + "grad_norm": 0.03543921187520027, + "learning_rate": 6.4558580934551724e-06, + "loss": 0.3203, + "step": 52980 + }, + { + "epoch": 2.65182034279995, + "grad_norm": 70.86699676513672, + "learning_rate": 6.4465904246445855e-06, + "loss": 0.491, + "step": 52990 + }, + { + "epoch": 2.6523207806830977, + "grad_norm": 0.020841041579842567, + "learning_rate": 6.437322755833998e-06, + "loss": 0.2656, + "step": 53000 + }, + { + "epoch": 2.6528212185662454, + "grad_norm": 33.420440673828125, + "learning_rate": 6.42805508702341e-06, + "loss": 0.2128, + "step": 53010 + }, + { + "epoch": 2.653321656449393, + "grad_norm": 0.0024048995692282915, + "learning_rate": 6.418787418212822e-06, + "loss": 0.0735, + "step": 53020 + }, + { + "epoch": 2.653822094332541, + "grad_norm": 0.052168700844049454, + "learning_rate": 6.409519749402236e-06, + "loss": 0.1056, + "step": 53030 + }, + { + "epoch": 2.654322532215689, + "grad_norm": 0.26582249999046326, + "learning_rate": 6.4002520805916485e-06, + "loss": 0.2149, + "step": 53040 + }, + { + "epoch": 2.6548229700988366, + "grad_norm": 0.098860964179039, + "learning_rate": 6.390984411781061e-06, + "loss": 0.203, + "step": 53050 + }, + { + "epoch": 2.6553234079819843, + "grad_norm": 0.01828690804541111, + "learning_rate": 6.381716742970474e-06, + "loss": 0.3748, + "step": 53060 + }, + { + "epoch": 2.655823845865132, + "grad_norm": 0.39225882291793823, + "learning_rate": 6.372449074159886e-06, + "loss": 0.2396, + "step": 53070 + }, + { + "epoch": 2.6563242837482797, + "grad_norm": 0.16991892457008362, + "learning_rate": 6.363181405349298e-06, + "loss": 0.1513, + "step": 53080 + }, + { + "epoch": 2.6568247216314274, + "grad_norm": 0.008684629574418068, + "learning_rate": 6.3539137365387105e-06, + "loss": 0.2008, + "step": 53090 + }, + { + "epoch": 2.657325159514575, + "grad_norm": 0.12532608211040497, + "learning_rate": 6.3446460677281245e-06, + "loss": 0.0019, + "step": 53100 + }, + { + "epoch": 2.657825597397723, + "grad_norm": 37.6959114074707, + "learning_rate": 6.335378398917537e-06, + "loss": 0.2086, + "step": 53110 + }, + { + "epoch": 2.658326035280871, + "grad_norm": 1.5516233444213867, + "learning_rate": 6.326110730106949e-06, + "loss": 0.0497, + "step": 53120 + }, + { + "epoch": 2.6588264731640185, + "grad_norm": 3.731523036956787, + "learning_rate": 6.316843061296363e-06, + "loss": 0.1363, + "step": 53130 + }, + { + "epoch": 2.6593269110471662, + "grad_norm": 95.50447845458984, + "learning_rate": 6.307575392485774e-06, + "loss": 0.574, + "step": 53140 + }, + { + "epoch": 2.659827348930314, + "grad_norm": 127.68614196777344, + "learning_rate": 6.2983077236751866e-06, + "loss": 0.4309, + "step": 53150 + }, + { + "epoch": 2.660327786813462, + "grad_norm": 7.3641839027404785, + "learning_rate": 6.289040054864599e-06, + "loss": 0.005, + "step": 53160 + }, + { + "epoch": 2.6608282246966093, + "grad_norm": 0.005530231166630983, + "learning_rate": 6.279772386054013e-06, + "loss": 0.4135, + "step": 53170 + }, + { + "epoch": 2.6613286625797574, + "grad_norm": 0.001578154624439776, + "learning_rate": 6.270504717243425e-06, + "loss": 0.427, + "step": 53180 + }, + { + "epoch": 2.661829100462905, + "grad_norm": 0.21591638028621674, + "learning_rate": 6.261237048432837e-06, + "loss": 0.03, + "step": 53190 + }, + { + "epoch": 2.662329538346053, + "grad_norm": 115.83610534667969, + "learning_rate": 6.2519693796222495e-06, + "loss": 0.286, + "step": 53200 + }, + { + "epoch": 2.6628299762292005, + "grad_norm": 0.8543475270271301, + "learning_rate": 6.2427017108116626e-06, + "loss": 0.043, + "step": 53210 + }, + { + "epoch": 2.663330414112348, + "grad_norm": 0.13089485466480255, + "learning_rate": 6.233434042001076e-06, + "loss": 0.1288, + "step": 53220 + }, + { + "epoch": 2.6638308519954963, + "grad_norm": 0.15380859375, + "learning_rate": 6.224166373190488e-06, + "loss": 0.0876, + "step": 53230 + }, + { + "epoch": 2.6643312898786435, + "grad_norm": 2.310884952545166, + "learning_rate": 6.2148987043799e-06, + "loss": 0.2435, + "step": 53240 + }, + { + "epoch": 2.6648317277617917, + "grad_norm": 1.1977853775024414, + "learning_rate": 6.205631035569313e-06, + "loss": 0.6828, + "step": 53250 + }, + { + "epoch": 2.6653321656449394, + "grad_norm": 0.001773962751030922, + "learning_rate": 6.1963633667587255e-06, + "loss": 0.2003, + "step": 53260 + }, + { + "epoch": 2.665832603528087, + "grad_norm": 0.40930959582328796, + "learning_rate": 6.187095697948139e-06, + "loss": 0.3401, + "step": 53270 + }, + { + "epoch": 2.6663330414112347, + "grad_norm": 0.09972106665372849, + "learning_rate": 6.177828029137551e-06, + "loss": 0.1593, + "step": 53280 + }, + { + "epoch": 2.6668334792943824, + "grad_norm": 0.745758056640625, + "learning_rate": 6.168560360326964e-06, + "loss": 0.2078, + "step": 53290 + }, + { + "epoch": 2.6673339171775305, + "grad_norm": 33.90165710449219, + "learning_rate": 6.159292691516376e-06, + "loss": 0.1289, + "step": 53300 + }, + { + "epoch": 2.6678343550606782, + "grad_norm": 0.020202603191137314, + "learning_rate": 6.150025022705789e-06, + "loss": 0.1347, + "step": 53310 + }, + { + "epoch": 2.668334792943826, + "grad_norm": 0.01461669523268938, + "learning_rate": 6.1407573538952015e-06, + "loss": 0.1421, + "step": 53320 + }, + { + "epoch": 2.6688352308269736, + "grad_norm": 0.0074966540560126305, + "learning_rate": 6.131489685084614e-06, + "loss": 0.2021, + "step": 53330 + }, + { + "epoch": 2.6693356687101213, + "grad_norm": 0.0014433355536311865, + "learning_rate": 6.122222016274027e-06, + "loss": 0.2722, + "step": 53340 + }, + { + "epoch": 2.669836106593269, + "grad_norm": 0.0011452982435002923, + "learning_rate": 6.112954347463439e-06, + "loss": 0.3428, + "step": 53350 + }, + { + "epoch": 2.6703365444764167, + "grad_norm": 0.0071067665703594685, + "learning_rate": 6.103686678652852e-06, + "loss": 0.3101, + "step": 53360 + }, + { + "epoch": 2.670836982359565, + "grad_norm": 8.526885986328125, + "learning_rate": 6.0944190098422644e-06, + "loss": 0.2527, + "step": 53370 + }, + { + "epoch": 2.6713374202427125, + "grad_norm": 0.026250001043081284, + "learning_rate": 6.0851513410316775e-06, + "loss": 0.0643, + "step": 53380 + }, + { + "epoch": 2.67183785812586, + "grad_norm": 28.163471221923828, + "learning_rate": 6.07588367222109e-06, + "loss": 0.1984, + "step": 53390 + }, + { + "epoch": 2.672338296009008, + "grad_norm": 19.512283325195312, + "learning_rate": 6.066616003410502e-06, + "loss": 0.6004, + "step": 53400 + }, + { + "epoch": 2.6728387338921555, + "grad_norm": 28.954587936401367, + "learning_rate": 6.057348334599915e-06, + "loss": 0.2564, + "step": 53410 + }, + { + "epoch": 2.6733391717753032, + "grad_norm": 0.01020035520195961, + "learning_rate": 6.048080665789327e-06, + "loss": 0.1316, + "step": 53420 + }, + { + "epoch": 2.673839609658451, + "grad_norm": 41.50669479370117, + "learning_rate": 6.0388129969787405e-06, + "loss": 0.3139, + "step": 53430 + }, + { + "epoch": 2.674340047541599, + "grad_norm": 0.11358674615621567, + "learning_rate": 6.029545328168153e-06, + "loss": 0.3125, + "step": 53440 + }, + { + "epoch": 2.6748404854247467, + "grad_norm": 0.0016524195671081543, + "learning_rate": 6.020277659357566e-06, + "loss": 0.5317, + "step": 53450 + }, + { + "epoch": 2.6753409233078944, + "grad_norm": 5.80447244644165, + "learning_rate": 6.011009990546978e-06, + "loss": 0.2277, + "step": 53460 + }, + { + "epoch": 2.675841361191042, + "grad_norm": 0.022149043157696724, + "learning_rate": 6.001742321736391e-06, + "loss": 0.001, + "step": 53470 + }, + { + "epoch": 2.67634179907419, + "grad_norm": 0.24908429384231567, + "learning_rate": 5.992474652925803e-06, + "loss": 0.5037, + "step": 53480 + }, + { + "epoch": 2.676842236957338, + "grad_norm": 0.007026126142591238, + "learning_rate": 5.983206984115216e-06, + "loss": 0.4842, + "step": 53490 + }, + { + "epoch": 2.677342674840485, + "grad_norm": 57.95986557006836, + "learning_rate": 5.973939315304629e-06, + "loss": 0.0913, + "step": 53500 + }, + { + "epoch": 2.6778431127236333, + "grad_norm": 0.002470794366672635, + "learning_rate": 5.964671646494041e-06, + "loss": 0.3175, + "step": 53510 + }, + { + "epoch": 2.678343550606781, + "grad_norm": 0.49217313528060913, + "learning_rate": 5.955403977683454e-06, + "loss": 0.5046, + "step": 53520 + }, + { + "epoch": 2.6788439884899287, + "grad_norm": 0.010826364159584045, + "learning_rate": 5.946136308872866e-06, + "loss": 0.3838, + "step": 53530 + }, + { + "epoch": 2.6793444263730763, + "grad_norm": 199.83839416503906, + "learning_rate": 5.936868640062279e-06, + "loss": 0.5485, + "step": 53540 + }, + { + "epoch": 2.679844864256224, + "grad_norm": 0.6801645159721375, + "learning_rate": 5.927600971251692e-06, + "loss": 0.3033, + "step": 53550 + }, + { + "epoch": 2.680345302139372, + "grad_norm": 7.366738319396973, + "learning_rate": 5.918333302441105e-06, + "loss": 0.2931, + "step": 53560 + }, + { + "epoch": 2.68084574002252, + "grad_norm": 0.0387897789478302, + "learning_rate": 5.909065633630516e-06, + "loss": 0.0038, + "step": 53570 + }, + { + "epoch": 2.6813461779056675, + "grad_norm": 0.0021176764275878668, + "learning_rate": 5.899797964819929e-06, + "loss": 0.0515, + "step": 53580 + }, + { + "epoch": 2.6818466157888152, + "grad_norm": 0.15979740023612976, + "learning_rate": 5.890530296009342e-06, + "loss": 0.3091, + "step": 53590 + }, + { + "epoch": 2.682347053671963, + "grad_norm": 0.09495367109775543, + "learning_rate": 5.8812626271987546e-06, + "loss": 0.4744, + "step": 53600 + }, + { + "epoch": 2.6828474915551106, + "grad_norm": 0.00543246790766716, + "learning_rate": 5.871994958388168e-06, + "loss": 0.2857, + "step": 53610 + }, + { + "epoch": 2.6833479294382583, + "grad_norm": 0.11604252457618713, + "learning_rate": 5.86272728957758e-06, + "loss": 0.1527, + "step": 53620 + }, + { + "epoch": 2.6838483673214064, + "grad_norm": 15.720025062561035, + "learning_rate": 5.853459620766993e-06, + "loss": 0.157, + "step": 53630 + }, + { + "epoch": 2.684348805204554, + "grad_norm": 16.742429733276367, + "learning_rate": 5.844191951956405e-06, + "loss": 0.3116, + "step": 53640 + }, + { + "epoch": 2.684849243087702, + "grad_norm": 69.82891082763672, + "learning_rate": 5.8349242831458175e-06, + "loss": 0.1413, + "step": 53650 + }, + { + "epoch": 2.6853496809708495, + "grad_norm": 1.5985488891601562, + "learning_rate": 5.82565661433523e-06, + "loss": 0.1761, + "step": 53660 + }, + { + "epoch": 2.685850118853997, + "grad_norm": 7.359624862670898, + "learning_rate": 5.816388945524643e-06, + "loss": 0.0281, + "step": 53670 + }, + { + "epoch": 2.686350556737145, + "grad_norm": 1.950361967086792, + "learning_rate": 5.807121276714056e-06, + "loss": 0.2063, + "step": 53680 + }, + { + "epoch": 2.6868509946202925, + "grad_norm": 0.1081153005361557, + "learning_rate": 5.797853607903468e-06, + "loss": 0.3816, + "step": 53690 + }, + { + "epoch": 2.6873514325034407, + "grad_norm": 0.004088506568223238, + "learning_rate": 5.788585939092881e-06, + "loss": 0.0924, + "step": 53700 + }, + { + "epoch": 2.6878518703865883, + "grad_norm": 1.060620903968811, + "learning_rate": 5.7793182702822935e-06, + "loss": 0.0601, + "step": 53710 + }, + { + "epoch": 2.688352308269736, + "grad_norm": 0.014264395460486412, + "learning_rate": 5.770050601471707e-06, + "loss": 0.0065, + "step": 53720 + }, + { + "epoch": 2.6888527461528837, + "grad_norm": 16.35171890258789, + "learning_rate": 5.760782932661119e-06, + "loss": 0.0541, + "step": 53730 + }, + { + "epoch": 2.6893531840360314, + "grad_norm": 149.54856872558594, + "learning_rate": 5.751515263850531e-06, + "loss": 0.7565, + "step": 53740 + }, + { + "epoch": 2.6898536219191795, + "grad_norm": 81.81656646728516, + "learning_rate": 5.742247595039943e-06, + "loss": 0.345, + "step": 53750 + }, + { + "epoch": 2.690354059802327, + "grad_norm": 40.88715744018555, + "learning_rate": 5.7329799262293564e-06, + "loss": 0.4055, + "step": 53760 + }, + { + "epoch": 2.690854497685475, + "grad_norm": 0.03197389096021652, + "learning_rate": 5.7237122574187695e-06, + "loss": 0.1552, + "step": 53770 + }, + { + "epoch": 2.6913549355686226, + "grad_norm": 74.75322723388672, + "learning_rate": 5.714444588608182e-06, + "loss": 0.233, + "step": 53780 + }, + { + "epoch": 2.6918553734517703, + "grad_norm": 0.00996381975710392, + "learning_rate": 5.705176919797595e-06, + "loss": 0.1104, + "step": 53790 + }, + { + "epoch": 2.692355811334918, + "grad_norm": 47.85565948486328, + "learning_rate": 5.695909250987007e-06, + "loss": 0.4815, + "step": 53800 + }, + { + "epoch": 2.6928562492180657, + "grad_norm": 0.9291006326675415, + "learning_rate": 5.686641582176419e-06, + "loss": 0.2228, + "step": 53810 + }, + { + "epoch": 2.693356687101214, + "grad_norm": 0.039962075650691986, + "learning_rate": 5.677373913365832e-06, + "loss": 0.5983, + "step": 53820 + }, + { + "epoch": 2.6938571249843615, + "grad_norm": 22.67772102355957, + "learning_rate": 5.668106244555245e-06, + "loss": 0.4176, + "step": 53830 + }, + { + "epoch": 2.694357562867509, + "grad_norm": 0.12148747593164444, + "learning_rate": 5.658838575744657e-06, + "loss": 0.1488, + "step": 53840 + }, + { + "epoch": 2.694858000750657, + "grad_norm": 0.024220362305641174, + "learning_rate": 5.64957090693407e-06, + "loss": 0.4452, + "step": 53850 + }, + { + "epoch": 2.6953584386338045, + "grad_norm": 5.066788673400879, + "learning_rate": 5.640303238123483e-06, + "loss": 0.4318, + "step": 53860 + }, + { + "epoch": 2.695858876516952, + "grad_norm": 0.056153152137994766, + "learning_rate": 5.631035569312895e-06, + "loss": 0.039, + "step": 53870 + }, + { + "epoch": 2.6963593144001, + "grad_norm": 0.001981750363484025, + "learning_rate": 5.6217679005023085e-06, + "loss": 0.2162, + "step": 53880 + }, + { + "epoch": 2.696859752283248, + "grad_norm": 2.723752021789551, + "learning_rate": 5.612500231691721e-06, + "loss": 0.4477, + "step": 53890 + }, + { + "epoch": 2.6973601901663957, + "grad_norm": 0.45320022106170654, + "learning_rate": 5.603232562881133e-06, + "loss": 0.562, + "step": 53900 + }, + { + "epoch": 2.6978606280495434, + "grad_norm": 7.233700752258301, + "learning_rate": 5.593964894070545e-06, + "loss": 0.1526, + "step": 53910 + }, + { + "epoch": 2.698361065932691, + "grad_norm": 78.55522155761719, + "learning_rate": 5.584697225259958e-06, + "loss": 0.402, + "step": 53920 + }, + { + "epoch": 2.698861503815839, + "grad_norm": 0.010524014942348003, + "learning_rate": 5.5754295564493705e-06, + "loss": 0.008, + "step": 53930 + }, + { + "epoch": 2.6993619416989865, + "grad_norm": 0.0009086702484637499, + "learning_rate": 5.566161887638784e-06, + "loss": 0.5297, + "step": 53940 + }, + { + "epoch": 2.699862379582134, + "grad_norm": 0.03755944222211838, + "learning_rate": 5.556894218828197e-06, + "loss": 0.0555, + "step": 53950 + }, + { + "epoch": 2.7003628174652823, + "grad_norm": 0.010634690523147583, + "learning_rate": 5.547626550017609e-06, + "loss": 0.2907, + "step": 53960 + }, + { + "epoch": 2.70086325534843, + "grad_norm": 132.2368621826172, + "learning_rate": 5.538358881207022e-06, + "loss": 0.1447, + "step": 53970 + }, + { + "epoch": 2.7013636932315777, + "grad_norm": 4.581533432006836, + "learning_rate": 5.529091212396434e-06, + "loss": 0.135, + "step": 53980 + }, + { + "epoch": 2.7018641311147253, + "grad_norm": 0.002767818747088313, + "learning_rate": 5.5198235435858466e-06, + "loss": 0.2315, + "step": 53990 + }, + { + "epoch": 2.702364568997873, + "grad_norm": 0.005677152890712023, + "learning_rate": 5.510555874775259e-06, + "loss": 0.2221, + "step": 54000 + }, + { + "epoch": 2.702865006881021, + "grad_norm": 0.0049445428885519505, + "learning_rate": 5.501288205964672e-06, + "loss": 0.2746, + "step": 54010 + }, + { + "epoch": 2.7033654447641684, + "grad_norm": 0.016451798379421234, + "learning_rate": 5.492020537154084e-06, + "loss": 0.2177, + "step": 54020 + }, + { + "epoch": 2.7038658826473165, + "grad_norm": 25.13134002685547, + "learning_rate": 5.482752868343497e-06, + "loss": 0.1784, + "step": 54030 + }, + { + "epoch": 2.704366320530464, + "grad_norm": 0.07741358131170273, + "learning_rate": 5.47348519953291e-06, + "loss": 0.2342, + "step": 54040 + }, + { + "epoch": 2.704866758413612, + "grad_norm": 0.022271795198321342, + "learning_rate": 5.4642175307223226e-06, + "loss": 0.2773, + "step": 54050 + }, + { + "epoch": 2.7053671962967596, + "grad_norm": 10.753711700439453, + "learning_rate": 5.454949861911735e-06, + "loss": 0.2449, + "step": 54060 + }, + { + "epoch": 2.7058676341799073, + "grad_norm": 15.1447172164917, + "learning_rate": 5.445682193101147e-06, + "loss": 0.0713, + "step": 54070 + }, + { + "epoch": 2.7063680720630554, + "grad_norm": 0.007725929841399193, + "learning_rate": 5.43641452429056e-06, + "loss": 0.1497, + "step": 54080 + }, + { + "epoch": 2.706868509946203, + "grad_norm": 16.557029724121094, + "learning_rate": 5.427146855479972e-06, + "loss": 0.1342, + "step": 54090 + }, + { + "epoch": 2.707368947829351, + "grad_norm": 0.054356448352336884, + "learning_rate": 5.4178791866693855e-06, + "loss": 0.365, + "step": 54100 + }, + { + "epoch": 2.7078693857124985, + "grad_norm": 0.48548439145088196, + "learning_rate": 5.408611517858798e-06, + "loss": 0.43, + "step": 54110 + }, + { + "epoch": 2.708369823595646, + "grad_norm": 13.83474349975586, + "learning_rate": 5.399343849048211e-06, + "loss": 0.0973, + "step": 54120 + }, + { + "epoch": 2.708870261478794, + "grad_norm": 12.604772567749023, + "learning_rate": 5.390076180237624e-06, + "loss": 0.0725, + "step": 54130 + }, + { + "epoch": 2.7093706993619415, + "grad_norm": 0.11724910140037537, + "learning_rate": 5.380808511427036e-06, + "loss": 0.0089, + "step": 54140 + }, + { + "epoch": 2.7098711372450897, + "grad_norm": 0.07806672155857086, + "learning_rate": 5.371540842616448e-06, + "loss": 0.0014, + "step": 54150 + }, + { + "epoch": 2.7103715751282373, + "grad_norm": 64.52555847167969, + "learning_rate": 5.362273173805861e-06, + "loss": 0.3566, + "step": 54160 + }, + { + "epoch": 2.710872013011385, + "grad_norm": 0.021519210189580917, + "learning_rate": 5.353005504995274e-06, + "loss": 0.0829, + "step": 54170 + }, + { + "epoch": 2.7113724508945327, + "grad_norm": 2.329911231994629, + "learning_rate": 5.343737836184686e-06, + "loss": 0.0206, + "step": 54180 + }, + { + "epoch": 2.7118728887776804, + "grad_norm": 59.641719818115234, + "learning_rate": 5.334470167374099e-06, + "loss": 0.4831, + "step": 54190 + }, + { + "epoch": 2.712373326660828, + "grad_norm": 16.386096954345703, + "learning_rate": 5.325202498563511e-06, + "loss": 0.0213, + "step": 54200 + }, + { + "epoch": 2.7128737645439758, + "grad_norm": 60.064697265625, + "learning_rate": 5.3159348297529244e-06, + "loss": 0.3636, + "step": 54210 + }, + { + "epoch": 2.713374202427124, + "grad_norm": 0.0025766179896891117, + "learning_rate": 5.3066671609423375e-06, + "loss": 0.0265, + "step": 54220 + }, + { + "epoch": 2.7138746403102716, + "grad_norm": 0.008639090694487095, + "learning_rate": 5.297399492131749e-06, + "loss": 0.1539, + "step": 54230 + }, + { + "epoch": 2.7143750781934193, + "grad_norm": 0.05753583088517189, + "learning_rate": 5.288131823321162e-06, + "loss": 0.2859, + "step": 54240 + }, + { + "epoch": 2.714875516076567, + "grad_norm": 31.729856491088867, + "learning_rate": 5.278864154510574e-06, + "loss": 0.2261, + "step": 54250 + }, + { + "epoch": 2.7153759539597146, + "grad_norm": 26.112958908081055, + "learning_rate": 5.269596485699987e-06, + "loss": 0.1563, + "step": 54260 + }, + { + "epoch": 2.715876391842863, + "grad_norm": 0.01234404556453228, + "learning_rate": 5.2603288168894e-06, + "loss": 0.4809, + "step": 54270 + }, + { + "epoch": 2.71637682972601, + "grad_norm": 0.004819217603653669, + "learning_rate": 5.251061148078813e-06, + "loss": 0.4295, + "step": 54280 + }, + { + "epoch": 2.716877267609158, + "grad_norm": 10.893156051635742, + "learning_rate": 5.241793479268225e-06, + "loss": 0.114, + "step": 54290 + }, + { + "epoch": 2.717377705492306, + "grad_norm": 0.022931111976504326, + "learning_rate": 5.232525810457638e-06, + "loss": 0.2697, + "step": 54300 + }, + { + "epoch": 2.7178781433754535, + "grad_norm": 0.09070712327957153, + "learning_rate": 5.22325814164705e-06, + "loss": 0.0809, + "step": 54310 + }, + { + "epoch": 2.718378581258601, + "grad_norm": 0.053124405443668365, + "learning_rate": 5.2139904728364625e-06, + "loss": 0.2701, + "step": 54320 + }, + { + "epoch": 2.718879019141749, + "grad_norm": 58.14130783081055, + "learning_rate": 5.204722804025876e-06, + "loss": 0.2981, + "step": 54330 + }, + { + "epoch": 2.719379457024897, + "grad_norm": 0.2619777023792267, + "learning_rate": 5.195455135215288e-06, + "loss": 0.1866, + "step": 54340 + }, + { + "epoch": 2.7198798949080443, + "grad_norm": 0.43264833092689514, + "learning_rate": 5.186187466404701e-06, + "loss": 0.1732, + "step": 54350 + }, + { + "epoch": 2.7203803327911924, + "grad_norm": 115.37907409667969, + "learning_rate": 5.176919797594113e-06, + "loss": 0.2074, + "step": 54360 + }, + { + "epoch": 2.72088077067434, + "grad_norm": 0.010086285881698132, + "learning_rate": 5.167652128783526e-06, + "loss": 0.2319, + "step": 54370 + }, + { + "epoch": 2.7213812085574878, + "grad_norm": 0.03775709494948387, + "learning_rate": 5.1583844599729386e-06, + "loss": 0.2573, + "step": 54380 + }, + { + "epoch": 2.7218816464406355, + "grad_norm": 0.040614452213048935, + "learning_rate": 5.149116791162352e-06, + "loss": 0.0269, + "step": 54390 + }, + { + "epoch": 2.722382084323783, + "grad_norm": 0.07007140666246414, + "learning_rate": 5.139849122351764e-06, + "loss": 0.3058, + "step": 54400 + }, + { + "epoch": 2.7228825222069313, + "grad_norm": 8.75013542175293, + "learning_rate": 5.130581453541176e-06, + "loss": 0.4, + "step": 54410 + }, + { + "epoch": 2.723382960090079, + "grad_norm": 0.01710324175655842, + "learning_rate": 5.121313784730589e-06, + "loss": 0.2499, + "step": 54420 + }, + { + "epoch": 2.7238833979732266, + "grad_norm": 0.8458145260810852, + "learning_rate": 5.1120461159200015e-06, + "loss": 0.0922, + "step": 54430 + }, + { + "epoch": 2.7243838358563743, + "grad_norm": 0.06281821429729462, + "learning_rate": 5.1027784471094146e-06, + "loss": 0.3479, + "step": 54440 + }, + { + "epoch": 2.724884273739522, + "grad_norm": 178.5389404296875, + "learning_rate": 5.093510778298827e-06, + "loss": 0.3514, + "step": 54450 + }, + { + "epoch": 2.7253847116226697, + "grad_norm": 117.08137512207031, + "learning_rate": 5.08424310948824e-06, + "loss": 0.1163, + "step": 54460 + }, + { + "epoch": 2.7258851495058174, + "grad_norm": 13.209118843078613, + "learning_rate": 5.074975440677652e-06, + "loss": 0.4059, + "step": 54470 + }, + { + "epoch": 2.7263855873889655, + "grad_norm": 20.50191879272461, + "learning_rate": 5.065707771867064e-06, + "loss": 0.3248, + "step": 54480 + }, + { + "epoch": 2.726886025272113, + "grad_norm": 32.87042999267578, + "learning_rate": 5.0564401030564775e-06, + "loss": 0.1841, + "step": 54490 + }, + { + "epoch": 2.727386463155261, + "grad_norm": 0.0683615431189537, + "learning_rate": 5.04717243424589e-06, + "loss": 0.2927, + "step": 54500 + }, + { + "epoch": 2.7278869010384086, + "grad_norm": 0.008054709993302822, + "learning_rate": 5.037904765435303e-06, + "loss": 0.3059, + "step": 54510 + }, + { + "epoch": 2.7283873389215563, + "grad_norm": 0.3621916174888611, + "learning_rate": 5.028637096624715e-06, + "loss": 0.2706, + "step": 54520 + }, + { + "epoch": 2.728887776804704, + "grad_norm": 31.407835006713867, + "learning_rate": 5.019369427814128e-06, + "loss": 0.3337, + "step": 54530 + }, + { + "epoch": 2.7293882146878516, + "grad_norm": 31.91086769104004, + "learning_rate": 5.01010175900354e-06, + "loss": 0.1912, + "step": 54540 + }, + { + "epoch": 2.7298886525709998, + "grad_norm": 20.481220245361328, + "learning_rate": 5.0008340901929535e-06, + "loss": 0.0871, + "step": 54550 + }, + { + "epoch": 2.7303890904541475, + "grad_norm": 0.03942602872848511, + "learning_rate": 4.991566421382366e-06, + "loss": 0.2138, + "step": 54560 + }, + { + "epoch": 2.730889528337295, + "grad_norm": 0.5924592018127441, + "learning_rate": 4.982298752571778e-06, + "loss": 0.2619, + "step": 54570 + }, + { + "epoch": 2.731389966220443, + "grad_norm": 0.016124799847602844, + "learning_rate": 4.973031083761191e-06, + "loss": 0.1932, + "step": 54580 + }, + { + "epoch": 2.7318904041035905, + "grad_norm": 88.90430450439453, + "learning_rate": 4.963763414950603e-06, + "loss": 0.4366, + "step": 54590 + }, + { + "epoch": 2.7323908419867386, + "grad_norm": 3.0473339557647705, + "learning_rate": 4.9544957461400164e-06, + "loss": 0.7273, + "step": 54600 + }, + { + "epoch": 2.732891279869886, + "grad_norm": 3.1749298572540283, + "learning_rate": 4.945228077329429e-06, + "loss": 0.3972, + "step": 54610 + }, + { + "epoch": 2.733391717753034, + "grad_norm": 0.22681120038032532, + "learning_rate": 4.935960408518842e-06, + "loss": 0.2188, + "step": 54620 + }, + { + "epoch": 2.7338921556361817, + "grad_norm": 98.77107238769531, + "learning_rate": 4.926692739708254e-06, + "loss": 0.156, + "step": 54630 + }, + { + "epoch": 2.7343925935193294, + "grad_norm": 0.026721619069576263, + "learning_rate": 4.917425070897667e-06, + "loss": 0.3269, + "step": 54640 + }, + { + "epoch": 2.734893031402477, + "grad_norm": 9.76560115814209, + "learning_rate": 4.908157402087079e-06, + "loss": 0.2472, + "step": 54650 + }, + { + "epoch": 2.7353934692856248, + "grad_norm": 9.29918384552002, + "learning_rate": 4.898889733276492e-06, + "loss": 0.054, + "step": 54660 + }, + { + "epoch": 2.735893907168773, + "grad_norm": 23.824901580810547, + "learning_rate": 4.889622064465905e-06, + "loss": 0.0953, + "step": 54670 + }, + { + "epoch": 2.7363943450519206, + "grad_norm": 54.855064392089844, + "learning_rate": 4.880354395655317e-06, + "loss": 0.2574, + "step": 54680 + }, + { + "epoch": 2.7368947829350683, + "grad_norm": 34.88551330566406, + "learning_rate": 4.87108672684473e-06, + "loss": 0.2943, + "step": 54690 + }, + { + "epoch": 2.737395220818216, + "grad_norm": 48.88869857788086, + "learning_rate": 4.861819058034142e-06, + "loss": 0.0433, + "step": 54700 + }, + { + "epoch": 2.7378956587013636, + "grad_norm": 5.149670124053955, + "learning_rate": 4.852551389223555e-06, + "loss": 0.4208, + "step": 54710 + }, + { + "epoch": 2.7383960965845113, + "grad_norm": 1.990154504776001, + "learning_rate": 4.843283720412968e-06, + "loss": 0.0848, + "step": 54720 + }, + { + "epoch": 2.738896534467659, + "grad_norm": 67.8294677734375, + "learning_rate": 4.83401605160238e-06, + "loss": 0.4352, + "step": 54730 + }, + { + "epoch": 2.739396972350807, + "grad_norm": 30.571155548095703, + "learning_rate": 4.824748382791792e-06, + "loss": 0.2019, + "step": 54740 + }, + { + "epoch": 2.739897410233955, + "grad_norm": 0.3573492765426636, + "learning_rate": 4.815480713981205e-06, + "loss": 0.6738, + "step": 54750 + }, + { + "epoch": 2.7403978481171025, + "grad_norm": 69.4555435180664, + "learning_rate": 4.806213045170618e-06, + "loss": 0.1492, + "step": 54760 + }, + { + "epoch": 2.74089828600025, + "grad_norm": 14.499482154846191, + "learning_rate": 4.7969453763600305e-06, + "loss": 0.1405, + "step": 54770 + }, + { + "epoch": 2.741398723883398, + "grad_norm": 0.004668924957513809, + "learning_rate": 4.787677707549444e-06, + "loss": 0.1382, + "step": 54780 + }, + { + "epoch": 2.7418991617665456, + "grad_norm": 6.569698810577393, + "learning_rate": 4.778410038738856e-06, + "loss": 0.2546, + "step": 54790 + }, + { + "epoch": 2.7423995996496933, + "grad_norm": 0.012458263896405697, + "learning_rate": 4.769142369928269e-06, + "loss": 0.2011, + "step": 54800 + }, + { + "epoch": 2.7429000375328414, + "grad_norm": 108.79511260986328, + "learning_rate": 4.759874701117681e-06, + "loss": 0.4008, + "step": 54810 + }, + { + "epoch": 2.743400475415989, + "grad_norm": 12.66564655303955, + "learning_rate": 4.7506070323070935e-06, + "loss": 0.2218, + "step": 54820 + }, + { + "epoch": 2.7439009132991368, + "grad_norm": 0.02300100028514862, + "learning_rate": 4.741339363496506e-06, + "loss": 0.4622, + "step": 54830 + }, + { + "epoch": 2.7444013511822845, + "grad_norm": 0.0024990024976432323, + "learning_rate": 4.732071694685919e-06, + "loss": 0.2194, + "step": 54840 + }, + { + "epoch": 2.744901789065432, + "grad_norm": 0.12500186264514923, + "learning_rate": 4.722804025875332e-06, + "loss": 0.2508, + "step": 54850 + }, + { + "epoch": 2.7454022269485803, + "grad_norm": 0.021183166652917862, + "learning_rate": 4.713536357064744e-06, + "loss": 0.3159, + "step": 54860 + }, + { + "epoch": 2.7459026648317275, + "grad_norm": 81.45769500732422, + "learning_rate": 4.704268688254157e-06, + "loss": 0.3053, + "step": 54870 + }, + { + "epoch": 2.7464031027148756, + "grad_norm": 0.8351395130157471, + "learning_rate": 4.6950010194435695e-06, + "loss": 0.1235, + "step": 54880 + }, + { + "epoch": 2.7469035405980233, + "grad_norm": 0.008513366803526878, + "learning_rate": 4.685733350632983e-06, + "loss": 0.0765, + "step": 54890 + }, + { + "epoch": 2.747403978481171, + "grad_norm": 196.8218994140625, + "learning_rate": 4.676465681822394e-06, + "loss": 0.4472, + "step": 54900 + }, + { + "epoch": 2.7479044163643187, + "grad_norm": 51.34075164794922, + "learning_rate": 4.667198013011807e-06, + "loss": 0.1948, + "step": 54910 + }, + { + "epoch": 2.7484048542474664, + "grad_norm": 0.005238950252532959, + "learning_rate": 4.657930344201219e-06, + "loss": 0.501, + "step": 54920 + }, + { + "epoch": 2.7489052921306145, + "grad_norm": 28.19068717956543, + "learning_rate": 4.648662675390632e-06, + "loss": 0.4922, + "step": 54930 + }, + { + "epoch": 2.749405730013762, + "grad_norm": 0.09769978374242783, + "learning_rate": 4.6393950065800455e-06, + "loss": 0.2079, + "step": 54940 + }, + { + "epoch": 2.74990616789691, + "grad_norm": 0.0021790864411741495, + "learning_rate": 4.630127337769458e-06, + "loss": 0.2382, + "step": 54950 + }, + { + "epoch": 2.7504066057800576, + "grad_norm": 0.03794931247830391, + "learning_rate": 4.620859668958871e-06, + "loss": 0.1534, + "step": 54960 + }, + { + "epoch": 2.7509070436632053, + "grad_norm": 0.00910375826060772, + "learning_rate": 4.611592000148283e-06, + "loss": 0.1032, + "step": 54970 + }, + { + "epoch": 2.751407481546353, + "grad_norm": 0.02086549811065197, + "learning_rate": 4.602324331337695e-06, + "loss": 0.174, + "step": 54980 + }, + { + "epoch": 2.7519079194295006, + "grad_norm": 0.008065648376941681, + "learning_rate": 4.593056662527108e-06, + "loss": 0.1376, + "step": 54990 + }, + { + "epoch": 2.7524083573126488, + "grad_norm": 0.021750060841441154, + "learning_rate": 4.583788993716521e-06, + "loss": 0.752, + "step": 55000 + }, + { + "epoch": 2.7529087951957965, + "grad_norm": 0.017080871388316154, + "learning_rate": 4.574521324905933e-06, + "loss": 0.247, + "step": 55010 + }, + { + "epoch": 2.753409233078944, + "grad_norm": 1.251462697982788, + "learning_rate": 4.565253656095346e-06, + "loss": 0.1736, + "step": 55020 + }, + { + "epoch": 2.753909670962092, + "grad_norm": 0.0696200355887413, + "learning_rate": 4.555985987284759e-06, + "loss": 0.1435, + "step": 55030 + }, + { + "epoch": 2.7544101088452395, + "grad_norm": 96.4897232055664, + "learning_rate": 4.546718318474171e-06, + "loss": 0.1992, + "step": 55040 + }, + { + "epoch": 2.754910546728387, + "grad_norm": 20.5328311920166, + "learning_rate": 4.5374506496635844e-06, + "loss": 0.1497, + "step": 55050 + }, + { + "epoch": 2.755410984611535, + "grad_norm": 0.0073603973723948, + "learning_rate": 4.528182980852997e-06, + "loss": 0.3317, + "step": 55060 + }, + { + "epoch": 2.755911422494683, + "grad_norm": 33.31328201293945, + "learning_rate": 4.518915312042409e-06, + "loss": 0.2546, + "step": 55070 + }, + { + "epoch": 2.7564118603778307, + "grad_norm": 0.0033437800593674183, + "learning_rate": 4.509647643231821e-06, + "loss": 0.1874, + "step": 55080 + }, + { + "epoch": 2.7569122982609784, + "grad_norm": 52.506404876708984, + "learning_rate": 4.500379974421234e-06, + "loss": 0.6004, + "step": 55090 + }, + { + "epoch": 2.757412736144126, + "grad_norm": 0.8078111410140991, + "learning_rate": 4.4911123056106465e-06, + "loss": 0.4805, + "step": 55100 + }, + { + "epoch": 2.7579131740272738, + "grad_norm": 0.04266117513179779, + "learning_rate": 4.48184463680006e-06, + "loss": 0.0235, + "step": 55110 + }, + { + "epoch": 2.758413611910422, + "grad_norm": 0.07403316348791122, + "learning_rate": 4.472576967989473e-06, + "loss": 0.336, + "step": 55120 + }, + { + "epoch": 2.758914049793569, + "grad_norm": 0.01151516754180193, + "learning_rate": 4.463309299178885e-06, + "loss": 0.1771, + "step": 55130 + }, + { + "epoch": 2.7594144876767173, + "grad_norm": 0.02986159548163414, + "learning_rate": 4.454041630368297e-06, + "loss": 0.1945, + "step": 55140 + }, + { + "epoch": 2.759914925559865, + "grad_norm": 0.010891247540712357, + "learning_rate": 4.4447739615577094e-06, + "loss": 0.2106, + "step": 55150 + }, + { + "epoch": 2.7604153634430126, + "grad_norm": 0.1054176539182663, + "learning_rate": 4.4355062927471225e-06, + "loss": 0.0088, + "step": 55160 + }, + { + "epoch": 2.7609158013261603, + "grad_norm": 4.321850776672363, + "learning_rate": 4.426238623936535e-06, + "loss": 0.4108, + "step": 55170 + }, + { + "epoch": 2.761416239209308, + "grad_norm": 15.393594741821289, + "learning_rate": 4.416970955125948e-06, + "loss": 0.1417, + "step": 55180 + }, + { + "epoch": 2.761916677092456, + "grad_norm": 0.004040413070470095, + "learning_rate": 4.40770328631536e-06, + "loss": 0.0463, + "step": 55190 + }, + { + "epoch": 2.7624171149756034, + "grad_norm": 80.8304214477539, + "learning_rate": 4.398435617504773e-06, + "loss": 0.6473, + "step": 55200 + }, + { + "epoch": 2.7629175528587515, + "grad_norm": 33.82711410522461, + "learning_rate": 4.389167948694186e-06, + "loss": 0.9795, + "step": 55210 + }, + { + "epoch": 2.763417990741899, + "grad_norm": 0.00571279413998127, + "learning_rate": 4.3799002798835986e-06, + "loss": 0.122, + "step": 55220 + }, + { + "epoch": 2.763918428625047, + "grad_norm": 80.60511779785156, + "learning_rate": 4.370632611073011e-06, + "loss": 0.3531, + "step": 55230 + }, + { + "epoch": 2.7644188665081946, + "grad_norm": 24.383344650268555, + "learning_rate": 4.361364942262423e-06, + "loss": 0.146, + "step": 55240 + }, + { + "epoch": 2.7649193043913423, + "grad_norm": 0.05626457929611206, + "learning_rate": 4.352097273451836e-06, + "loss": 0.3697, + "step": 55250 + }, + { + "epoch": 2.7654197422744904, + "grad_norm": 0.004959566984325647, + "learning_rate": 4.342829604641248e-06, + "loss": 0.0414, + "step": 55260 + }, + { + "epoch": 2.765920180157638, + "grad_norm": 85.69996643066406, + "learning_rate": 4.3335619358306615e-06, + "loss": 0.2114, + "step": 55270 + }, + { + "epoch": 2.7664206180407858, + "grad_norm": 71.29188537597656, + "learning_rate": 4.324294267020074e-06, + "loss": 0.3223, + "step": 55280 + }, + { + "epoch": 2.7669210559239334, + "grad_norm": 1.9082945585250854, + "learning_rate": 4.315026598209487e-06, + "loss": 0.1044, + "step": 55290 + }, + { + "epoch": 2.767421493807081, + "grad_norm": 9.799867257243022e-05, + "learning_rate": 4.3057589293989e-06, + "loss": 0.5691, + "step": 55300 + }, + { + "epoch": 2.767921931690229, + "grad_norm": 0.010832109488546848, + "learning_rate": 4.296491260588312e-06, + "loss": 0.2058, + "step": 55310 + }, + { + "epoch": 2.7684223695733765, + "grad_norm": 89.16049194335938, + "learning_rate": 4.287223591777724e-06, + "loss": 0.2742, + "step": 55320 + }, + { + "epoch": 2.7689228074565246, + "grad_norm": 0.002366851083934307, + "learning_rate": 4.277955922967137e-06, + "loss": 0.1672, + "step": 55330 + }, + { + "epoch": 2.7694232453396723, + "grad_norm": 0.003948215860873461, + "learning_rate": 4.26868825415655e-06, + "loss": 0.088, + "step": 55340 + }, + { + "epoch": 2.76992368322282, + "grad_norm": 2.323450803756714, + "learning_rate": 4.259420585345962e-06, + "loss": 0.1216, + "step": 55350 + }, + { + "epoch": 2.7704241211059677, + "grad_norm": 1.6563665866851807, + "learning_rate": 4.250152916535375e-06, + "loss": 0.3666, + "step": 55360 + }, + { + "epoch": 2.7709245589891154, + "grad_norm": 0.014657512307167053, + "learning_rate": 4.240885247724787e-06, + "loss": 0.4367, + "step": 55370 + }, + { + "epoch": 2.771424996872263, + "grad_norm": 61.11927032470703, + "learning_rate": 4.2316175789142e-06, + "loss": 0.1291, + "step": 55380 + }, + { + "epoch": 2.7719254347554108, + "grad_norm": 0.021631337702274323, + "learning_rate": 4.222349910103613e-06, + "loss": 0.0591, + "step": 55390 + }, + { + "epoch": 2.772425872638559, + "grad_norm": 0.00997080747038126, + "learning_rate": 4.213082241293025e-06, + "loss": 0.082, + "step": 55400 + }, + { + "epoch": 2.7729263105217066, + "grad_norm": 0.46329039335250854, + "learning_rate": 4.203814572482438e-06, + "loss": 0.654, + "step": 55410 + }, + { + "epoch": 2.7734267484048543, + "grad_norm": 0.005985800176858902, + "learning_rate": 4.19454690367185e-06, + "loss": 0.4848, + "step": 55420 + }, + { + "epoch": 2.773927186288002, + "grad_norm": 0.02148817665874958, + "learning_rate": 4.185279234861263e-06, + "loss": 0.0643, + "step": 55430 + }, + { + "epoch": 2.7744276241711496, + "grad_norm": 0.004300635773688555, + "learning_rate": 4.176011566050676e-06, + "loss": 0.1773, + "step": 55440 + }, + { + "epoch": 2.7749280620542978, + "grad_norm": 0.26320725679397583, + "learning_rate": 4.166743897240089e-06, + "loss": 0.1167, + "step": 55450 + }, + { + "epoch": 2.775428499937445, + "grad_norm": 0.00832516048103571, + "learning_rate": 4.157476228429501e-06, + "loss": 0.2325, + "step": 55460 + }, + { + "epoch": 2.775928937820593, + "grad_norm": 0.8156567215919495, + "learning_rate": 4.148208559618914e-06, + "loss": 0.2683, + "step": 55470 + }, + { + "epoch": 2.776429375703741, + "grad_norm": 0.10897979140281677, + "learning_rate": 4.138940890808326e-06, + "loss": 0.0202, + "step": 55480 + }, + { + "epoch": 2.7769298135868885, + "grad_norm": 2.880709648132324, + "learning_rate": 4.1296732219977385e-06, + "loss": 0.4587, + "step": 55490 + }, + { + "epoch": 2.777430251470036, + "grad_norm": 0.0034044734202325344, + "learning_rate": 4.120405553187152e-06, + "loss": 0.0688, + "step": 55500 + }, + { + "epoch": 2.777930689353184, + "grad_norm": 64.66979217529297, + "learning_rate": 4.111137884376564e-06, + "loss": 0.1804, + "step": 55510 + }, + { + "epoch": 2.778431127236332, + "grad_norm": 0.03665924072265625, + "learning_rate": 4.101870215565977e-06, + "loss": 0.4007, + "step": 55520 + }, + { + "epoch": 2.7789315651194797, + "grad_norm": 90.83676147460938, + "learning_rate": 4.092602546755389e-06, + "loss": 0.2548, + "step": 55530 + }, + { + "epoch": 2.7794320030026274, + "grad_norm": 0.02998952753841877, + "learning_rate": 4.083334877944802e-06, + "loss": 0.2552, + "step": 55540 + }, + { + "epoch": 2.779932440885775, + "grad_norm": 50.251007080078125, + "learning_rate": 4.0740672091342145e-06, + "loss": 0.397, + "step": 55550 + }, + { + "epoch": 2.7804328787689228, + "grad_norm": 0.11932443082332611, + "learning_rate": 4.064799540323627e-06, + "loss": 0.43, + "step": 55560 + }, + { + "epoch": 2.7809333166520704, + "grad_norm": 41.43545913696289, + "learning_rate": 4.05553187151304e-06, + "loss": 0.0104, + "step": 55570 + }, + { + "epoch": 2.781433754535218, + "grad_norm": 0.012818706221878529, + "learning_rate": 4.046264202702452e-06, + "loss": 0.1018, + "step": 55580 + }, + { + "epoch": 2.7819341924183663, + "grad_norm": 1.7519643306732178, + "learning_rate": 4.036996533891865e-06, + "loss": 0.3483, + "step": 55590 + }, + { + "epoch": 2.782434630301514, + "grad_norm": 0.0046064602211117744, + "learning_rate": 4.0277288650812775e-06, + "loss": 0.0017, + "step": 55600 + }, + { + "epoch": 2.7829350681846616, + "grad_norm": 27.66427230834961, + "learning_rate": 4.0184611962706906e-06, + "loss": 0.3354, + "step": 55610 + }, + { + "epoch": 2.7834355060678093, + "grad_norm": 0.17287428677082062, + "learning_rate": 4.009193527460103e-06, + "loss": 0.1803, + "step": 55620 + }, + { + "epoch": 2.783935943950957, + "grad_norm": 8.709161758422852, + "learning_rate": 3.999925858649516e-06, + "loss": 0.4929, + "step": 55630 + }, + { + "epoch": 2.7844363818341047, + "grad_norm": 0.2020673155784607, + "learning_rate": 3.990658189838928e-06, + "loss": 0.363, + "step": 55640 + }, + { + "epoch": 2.7849368197172524, + "grad_norm": 1.0412259101867676, + "learning_rate": 3.98139052102834e-06, + "loss": 0.3213, + "step": 55650 + }, + { + "epoch": 2.7854372576004005, + "grad_norm": 0.005764560308307409, + "learning_rate": 3.9721228522177535e-06, + "loss": 0.0003, + "step": 55660 + }, + { + "epoch": 2.785937695483548, + "grad_norm": 20.361534118652344, + "learning_rate": 3.962855183407166e-06, + "loss": 0.404, + "step": 55670 + }, + { + "epoch": 2.786438133366696, + "grad_norm": 0.0026090245228260756, + "learning_rate": 3.953587514596579e-06, + "loss": 0.0011, + "step": 55680 + }, + { + "epoch": 2.7869385712498436, + "grad_norm": 0.009606342762708664, + "learning_rate": 3.944319845785991e-06, + "loss": 0.3467, + "step": 55690 + }, + { + "epoch": 2.7874390091329913, + "grad_norm": 115.53101348876953, + "learning_rate": 3.935052176975404e-06, + "loss": 0.2227, + "step": 55700 + }, + { + "epoch": 2.7879394470161394, + "grad_norm": 0.683100700378418, + "learning_rate": 3.925784508164816e-06, + "loss": 0.0022, + "step": 55710 + }, + { + "epoch": 2.7884398848992866, + "grad_norm": 29.48588752746582, + "learning_rate": 3.9165168393542295e-06, + "loss": 0.2784, + "step": 55720 + }, + { + "epoch": 2.7889403227824348, + "grad_norm": 0.03518044948577881, + "learning_rate": 3.907249170543642e-06, + "loss": 0.0896, + "step": 55730 + }, + { + "epoch": 2.7894407606655824, + "grad_norm": 0.005893390625715256, + "learning_rate": 3.897981501733054e-06, + "loss": 0.3281, + "step": 55740 + }, + { + "epoch": 2.78994119854873, + "grad_norm": 0.019368739798665047, + "learning_rate": 3.888713832922467e-06, + "loss": 0.0003, + "step": 55750 + }, + { + "epoch": 2.790441636431878, + "grad_norm": 0.009706887416541576, + "learning_rate": 3.879446164111879e-06, + "loss": 0.2317, + "step": 55760 + }, + { + "epoch": 2.7909420743150255, + "grad_norm": 0.002712317742407322, + "learning_rate": 3.870178495301292e-06, + "loss": 0.1009, + "step": 55770 + }, + { + "epoch": 2.7914425121981736, + "grad_norm": 0.002232688246294856, + "learning_rate": 3.860910826490705e-06, + "loss": 0.1877, + "step": 55780 + }, + { + "epoch": 2.7919429500813213, + "grad_norm": 0.025917494669556618, + "learning_rate": 3.851643157680118e-06, + "loss": 0.1673, + "step": 55790 + }, + { + "epoch": 2.792443387964469, + "grad_norm": 1.0872734785079956, + "learning_rate": 3.84237548886953e-06, + "loss": 0.2492, + "step": 55800 + }, + { + "epoch": 2.7929438258476167, + "grad_norm": 0.036920156329870224, + "learning_rate": 3.833107820058942e-06, + "loss": 0.1045, + "step": 55810 + }, + { + "epoch": 2.7934442637307644, + "grad_norm": 51.51325225830078, + "learning_rate": 3.8238401512483545e-06, + "loss": 0.1224, + "step": 55820 + }, + { + "epoch": 2.793944701613912, + "grad_norm": 0.007304209750145674, + "learning_rate": 3.8145724824377676e-06, + "loss": 0.5248, + "step": 55830 + }, + { + "epoch": 2.7944451394970597, + "grad_norm": 0.5639947056770325, + "learning_rate": 3.8053048136271807e-06, + "loss": 0.2979, + "step": 55840 + }, + { + "epoch": 2.794945577380208, + "grad_norm": 0.1505047231912613, + "learning_rate": 3.796037144816593e-06, + "loss": 0.3045, + "step": 55850 + }, + { + "epoch": 2.7954460152633556, + "grad_norm": 16.375810623168945, + "learning_rate": 3.786769476006006e-06, + "loss": 0.3403, + "step": 55860 + }, + { + "epoch": 2.7959464531465033, + "grad_norm": 0.8366689682006836, + "learning_rate": 3.7775018071954183e-06, + "loss": 0.2906, + "step": 55870 + }, + { + "epoch": 2.796446891029651, + "grad_norm": 39.30411911010742, + "learning_rate": 3.768234138384831e-06, + "loss": 0.2824, + "step": 55880 + }, + { + "epoch": 2.7969473289127986, + "grad_norm": 67.98787689208984, + "learning_rate": 3.758966469574243e-06, + "loss": 0.1908, + "step": 55890 + }, + { + "epoch": 2.7974477667959463, + "grad_norm": 65.95452880859375, + "learning_rate": 3.7496988007636563e-06, + "loss": 0.1817, + "step": 55900 + }, + { + "epoch": 2.797948204679094, + "grad_norm": 0.5367636680603027, + "learning_rate": 3.7404311319530685e-06, + "loss": 0.2535, + "step": 55910 + }, + { + "epoch": 2.798448642562242, + "grad_norm": 2.5635721683502197, + "learning_rate": 3.731163463142481e-06, + "loss": 0.2654, + "step": 55920 + }, + { + "epoch": 2.79894908044539, + "grad_norm": 0.00294335070066154, + "learning_rate": 3.7218957943318943e-06, + "loss": 0.3908, + "step": 55930 + }, + { + "epoch": 2.7994495183285375, + "grad_norm": 0.06766127794981003, + "learning_rate": 3.7126281255213065e-06, + "loss": 0.185, + "step": 55940 + }, + { + "epoch": 2.799949956211685, + "grad_norm": 0.0023460641968995333, + "learning_rate": 3.703360456710719e-06, + "loss": 0.4078, + "step": 55950 + }, + { + "epoch": 2.800450394094833, + "grad_norm": 0.004410709720104933, + "learning_rate": 3.6940927879001314e-06, + "loss": 0.005, + "step": 55960 + }, + { + "epoch": 2.800950831977981, + "grad_norm": 0.05864166468381882, + "learning_rate": 3.6848251190895445e-06, + "loss": 0.6145, + "step": 55970 + }, + { + "epoch": 2.8014512698611282, + "grad_norm": 0.006647009402513504, + "learning_rate": 3.6755574502789568e-06, + "loss": 0.4356, + "step": 55980 + }, + { + "epoch": 2.8019517077442764, + "grad_norm": 0.0024092216044664383, + "learning_rate": 3.66628978146837e-06, + "loss": 0.2969, + "step": 55990 + }, + { + "epoch": 2.802452145627424, + "grad_norm": 1.6439789533615112, + "learning_rate": 3.657022112657782e-06, + "loss": 0.0068, + "step": 56000 + }, + { + "epoch": 2.8029525835105717, + "grad_norm": 0.011989391408860683, + "learning_rate": 3.647754443847195e-06, + "loss": 0.2793, + "step": 56010 + }, + { + "epoch": 2.8034530213937194, + "grad_norm": 0.1002710685133934, + "learning_rate": 3.638486775036608e-06, + "loss": 0.253, + "step": 56020 + }, + { + "epoch": 2.803953459276867, + "grad_norm": 44.029293060302734, + "learning_rate": 3.62921910622602e-06, + "loss": 0.5553, + "step": 56030 + }, + { + "epoch": 2.8044538971600153, + "grad_norm": 0.04861004278063774, + "learning_rate": 3.619951437415433e-06, + "loss": 0.2997, + "step": 56040 + }, + { + "epoch": 2.8049543350431625, + "grad_norm": 0.14872054755687714, + "learning_rate": 3.610683768604845e-06, + "loss": 0.3464, + "step": 56050 + }, + { + "epoch": 2.8054547729263106, + "grad_norm": 0.0290598813444376, + "learning_rate": 3.601416099794258e-06, + "loss": 0.2237, + "step": 56060 + }, + { + "epoch": 2.8059552108094583, + "grad_norm": 17.53714370727539, + "learning_rate": 3.5921484309836704e-06, + "loss": 0.1621, + "step": 56070 + }, + { + "epoch": 2.806455648692606, + "grad_norm": 232.04490661621094, + "learning_rate": 3.582880762173083e-06, + "loss": 0.2182, + "step": 56080 + }, + { + "epoch": 2.8069560865757537, + "grad_norm": 0.37230491638183594, + "learning_rate": 3.5736130933624953e-06, + "loss": 0.2298, + "step": 56090 + }, + { + "epoch": 2.8074565244589014, + "grad_norm": 0.04505237191915512, + "learning_rate": 3.5643454245519084e-06, + "loss": 0.4138, + "step": 56100 + }, + { + "epoch": 2.8079569623420495, + "grad_norm": 0.12896820902824402, + "learning_rate": 3.5550777557413215e-06, + "loss": 0.0067, + "step": 56110 + }, + { + "epoch": 2.808457400225197, + "grad_norm": 15.294682502746582, + "learning_rate": 3.5458100869307337e-06, + "loss": 0.1836, + "step": 56120 + }, + { + "epoch": 2.808957838108345, + "grad_norm": 109.61554718017578, + "learning_rate": 3.5365424181201464e-06, + "loss": 0.326, + "step": 56130 + }, + { + "epoch": 2.8094582759914926, + "grad_norm": 53.49571990966797, + "learning_rate": 3.5272747493095586e-06, + "loss": 0.0712, + "step": 56140 + }, + { + "epoch": 2.8099587138746402, + "grad_norm": 8.788214683532715, + "learning_rate": 3.5180070804989717e-06, + "loss": 0.26, + "step": 56150 + }, + { + "epoch": 2.810459151757788, + "grad_norm": 54.61404037475586, + "learning_rate": 3.508739411688384e-06, + "loss": 0.2743, + "step": 56160 + }, + { + "epoch": 2.8109595896409356, + "grad_norm": 0.06914973258972168, + "learning_rate": 3.4994717428777967e-06, + "loss": 0.3135, + "step": 56170 + }, + { + "epoch": 2.8114600275240837, + "grad_norm": 189.26292419433594, + "learning_rate": 3.490204074067209e-06, + "loss": 0.6619, + "step": 56180 + }, + { + "epoch": 2.8119604654072314, + "grad_norm": 0.011231365613639355, + "learning_rate": 3.480936405256622e-06, + "loss": 0.1662, + "step": 56190 + }, + { + "epoch": 2.812460903290379, + "grad_norm": 1.1944348812103271, + "learning_rate": 3.4716687364460347e-06, + "loss": 0.273, + "step": 56200 + }, + { + "epoch": 2.812961341173527, + "grad_norm": 0.34605464339256287, + "learning_rate": 3.462401067635447e-06, + "loss": 0.2326, + "step": 56210 + }, + { + "epoch": 2.8134617790566745, + "grad_norm": 0.0017933743074536324, + "learning_rate": 3.45313339882486e-06, + "loss": 0.7578, + "step": 56220 + }, + { + "epoch": 2.813962216939822, + "grad_norm": 0.3642039895057678, + "learning_rate": 3.4438657300142722e-06, + "loss": 0.5515, + "step": 56230 + }, + { + "epoch": 2.81446265482297, + "grad_norm": 0.06854122877120972, + "learning_rate": 3.4345980612036853e-06, + "loss": 0.2399, + "step": 56240 + }, + { + "epoch": 2.814963092706118, + "grad_norm": 19.791412353515625, + "learning_rate": 3.4253303923930976e-06, + "loss": 0.1277, + "step": 56250 + }, + { + "epoch": 2.8154635305892657, + "grad_norm": 0.2917163670063019, + "learning_rate": 3.4160627235825103e-06, + "loss": 0.4003, + "step": 56260 + }, + { + "epoch": 2.8159639684724134, + "grad_norm": 0.05127083510160446, + "learning_rate": 3.4067950547719225e-06, + "loss": 0.4466, + "step": 56270 + }, + { + "epoch": 2.816464406355561, + "grad_norm": 65.57665252685547, + "learning_rate": 3.3975273859613356e-06, + "loss": 0.1116, + "step": 56280 + }, + { + "epoch": 2.8169648442387087, + "grad_norm": 0.00406739441677928, + "learning_rate": 3.3882597171507483e-06, + "loss": 0.0238, + "step": 56290 + }, + { + "epoch": 2.817465282121857, + "grad_norm": 106.0880126953125, + "learning_rate": 3.3789920483401605e-06, + "loss": 0.7374, + "step": 56300 + }, + { + "epoch": 2.817965720005004, + "grad_norm": 1.9034944772720337, + "learning_rate": 3.3697243795295736e-06, + "loss": 0.1402, + "step": 56310 + }, + { + "epoch": 2.8184661578881522, + "grad_norm": 1.3206653594970703, + "learning_rate": 3.360456710718986e-06, + "loss": 0.2444, + "step": 56320 + }, + { + "epoch": 2.8189665957713, + "grad_norm": 0.008442246355116367, + "learning_rate": 3.3511890419083985e-06, + "loss": 0.2255, + "step": 56330 + }, + { + "epoch": 2.8194670336544476, + "grad_norm": 0.04630701243877411, + "learning_rate": 3.3419213730978108e-06, + "loss": 0.0686, + "step": 56340 + }, + { + "epoch": 2.8199674715375953, + "grad_norm": 0.04034515097737312, + "learning_rate": 3.332653704287224e-06, + "loss": 0.0202, + "step": 56350 + }, + { + "epoch": 2.820467909420743, + "grad_norm": 0.005361557472497225, + "learning_rate": 3.323386035476636e-06, + "loss": 0.3274, + "step": 56360 + }, + { + "epoch": 2.820968347303891, + "grad_norm": 0.006438547279685736, + "learning_rate": 3.314118366666049e-06, + "loss": 0.3913, + "step": 56370 + }, + { + "epoch": 2.821468785187039, + "grad_norm": 0.41555559635162354, + "learning_rate": 3.304850697855462e-06, + "loss": 0.1769, + "step": 56380 + }, + { + "epoch": 2.8219692230701865, + "grad_norm": 0.004337988793849945, + "learning_rate": 3.295583029044874e-06, + "loss": 0.2532, + "step": 56390 + }, + { + "epoch": 2.822469660953334, + "grad_norm": 0.0038082327228039503, + "learning_rate": 3.286315360234287e-06, + "loss": 0.3106, + "step": 56400 + }, + { + "epoch": 2.822970098836482, + "grad_norm": 0.053402721881866455, + "learning_rate": 3.2770476914236995e-06, + "loss": 0.1731, + "step": 56410 + }, + { + "epoch": 2.8234705367196296, + "grad_norm": 0.05418894812464714, + "learning_rate": 3.267780022613112e-06, + "loss": 0.0021, + "step": 56420 + }, + { + "epoch": 2.8239709746027772, + "grad_norm": 0.027050824835896492, + "learning_rate": 3.2585123538025244e-06, + "loss": 0.0009, + "step": 56430 + }, + { + "epoch": 2.8244714124859254, + "grad_norm": 0.8067432045936584, + "learning_rate": 3.2492446849919375e-06, + "loss": 0.2397, + "step": 56440 + }, + { + "epoch": 2.824971850369073, + "grad_norm": 0.030323436483740807, + "learning_rate": 3.2399770161813497e-06, + "loss": 0.2167, + "step": 56450 + }, + { + "epoch": 2.8254722882522207, + "grad_norm": 83.72158813476562, + "learning_rate": 3.2307093473707624e-06, + "loss": 0.5268, + "step": 56460 + }, + { + "epoch": 2.8259727261353684, + "grad_norm": 0.07849568873643875, + "learning_rate": 3.2214416785601755e-06, + "loss": 0.1834, + "step": 56470 + }, + { + "epoch": 2.826473164018516, + "grad_norm": 0.005567236803472042, + "learning_rate": 3.2121740097495877e-06, + "loss": 0.1015, + "step": 56480 + }, + { + "epoch": 2.826973601901664, + "grad_norm": 65.05257415771484, + "learning_rate": 3.202906340939001e-06, + "loss": 0.1932, + "step": 56490 + }, + { + "epoch": 2.8274740397848115, + "grad_norm": 0.07315672934055328, + "learning_rate": 3.1936386721284126e-06, + "loss": 0.2394, + "step": 56500 + }, + { + "epoch": 2.8279744776679596, + "grad_norm": 4.73793363571167, + "learning_rate": 3.1843710033178257e-06, + "loss": 0.0657, + "step": 56510 + }, + { + "epoch": 2.8284749155511073, + "grad_norm": 18.7728271484375, + "learning_rate": 3.175103334507238e-06, + "loss": 0.5215, + "step": 56520 + }, + { + "epoch": 2.828975353434255, + "grad_norm": 0.0005170708172954619, + "learning_rate": 3.165835665696651e-06, + "loss": 0.1398, + "step": 56530 + }, + { + "epoch": 2.8294757913174027, + "grad_norm": 0.08288062363862991, + "learning_rate": 3.1565679968860633e-06, + "loss": 0.1153, + "step": 56540 + }, + { + "epoch": 2.8299762292005504, + "grad_norm": 0.04877622798085213, + "learning_rate": 3.147300328075476e-06, + "loss": 0.3149, + "step": 56550 + }, + { + "epoch": 2.8304766670836985, + "grad_norm": 88.89527130126953, + "learning_rate": 3.138032659264889e-06, + "loss": 0.2743, + "step": 56560 + }, + { + "epoch": 2.8309771049668457, + "grad_norm": 0.002313077449798584, + "learning_rate": 3.1287649904543013e-06, + "loss": 0.0911, + "step": 56570 + }, + { + "epoch": 2.831477542849994, + "grad_norm": 0.02744070254266262, + "learning_rate": 3.119497321643714e-06, + "loss": 0.0283, + "step": 56580 + }, + { + "epoch": 2.8319779807331416, + "grad_norm": 0.01000681146979332, + "learning_rate": 3.1102296528331262e-06, + "loss": 0.2016, + "step": 56590 + }, + { + "epoch": 2.8324784186162892, + "grad_norm": 0.003604018362239003, + "learning_rate": 3.1009619840225393e-06, + "loss": 0.2997, + "step": 56600 + }, + { + "epoch": 2.832978856499437, + "grad_norm": 0.0030105006881058216, + "learning_rate": 3.091694315211952e-06, + "loss": 0.0032, + "step": 56610 + }, + { + "epoch": 2.8334792943825846, + "grad_norm": 88.21735382080078, + "learning_rate": 3.0824266464013642e-06, + "loss": 0.0747, + "step": 56620 + }, + { + "epoch": 2.8339797322657327, + "grad_norm": 0.009555390104651451, + "learning_rate": 3.073158977590777e-06, + "loss": 0.1435, + "step": 56630 + }, + { + "epoch": 2.8344801701488804, + "grad_norm": 0.2209535837173462, + "learning_rate": 3.0638913087801896e-06, + "loss": 0.1065, + "step": 56640 + }, + { + "epoch": 2.834980608032028, + "grad_norm": 0.5542692542076111, + "learning_rate": 3.0546236399696023e-06, + "loss": 0.2646, + "step": 56650 + }, + { + "epoch": 2.835481045915176, + "grad_norm": 250.3510284423828, + "learning_rate": 3.045355971159015e-06, + "loss": 0.2205, + "step": 56660 + }, + { + "epoch": 2.8359814837983235, + "grad_norm": 0.07686138153076172, + "learning_rate": 3.036088302348427e-06, + "loss": 0.0464, + "step": 56670 + }, + { + "epoch": 2.836481921681471, + "grad_norm": 0.021961160004138947, + "learning_rate": 3.02682063353784e-06, + "loss": 0.477, + "step": 56680 + }, + { + "epoch": 2.836982359564619, + "grad_norm": 60.41271209716797, + "learning_rate": 3.017552964727253e-06, + "loss": 0.6285, + "step": 56690 + }, + { + "epoch": 2.837482797447767, + "grad_norm": 0.01938127540051937, + "learning_rate": 3.0082852959166656e-06, + "loss": 0.3276, + "step": 56700 + }, + { + "epoch": 2.8379832353309147, + "grad_norm": 27.98394775390625, + "learning_rate": 2.999017627106078e-06, + "loss": 0.4906, + "step": 56710 + }, + { + "epoch": 2.8384836732140624, + "grad_norm": 130.48989868164062, + "learning_rate": 2.9897499582954905e-06, + "loss": 0.2253, + "step": 56720 + }, + { + "epoch": 2.83898411109721, + "grad_norm": 185.4452362060547, + "learning_rate": 2.980482289484903e-06, + "loss": 0.5129, + "step": 56730 + }, + { + "epoch": 2.8394845489803577, + "grad_norm": 137.61302185058594, + "learning_rate": 2.971214620674316e-06, + "loss": 0.3195, + "step": 56740 + }, + { + "epoch": 2.8399849868635054, + "grad_norm": 51.37843704223633, + "learning_rate": 2.961946951863728e-06, + "loss": 0.2681, + "step": 56750 + }, + { + "epoch": 2.840485424746653, + "grad_norm": 72.2567138671875, + "learning_rate": 2.9526792830531408e-06, + "loss": 0.4307, + "step": 56760 + }, + { + "epoch": 2.8409858626298012, + "grad_norm": 0.023341728374361992, + "learning_rate": 2.9434116142425534e-06, + "loss": 0.1687, + "step": 56770 + }, + { + "epoch": 2.841486300512949, + "grad_norm": 83.3965072631836, + "learning_rate": 2.9341439454319665e-06, + "loss": 0.1411, + "step": 56780 + }, + { + "epoch": 2.8419867383960966, + "grad_norm": 0.062252357602119446, + "learning_rate": 2.9248762766213788e-06, + "loss": 0.0018, + "step": 56790 + }, + { + "epoch": 2.8424871762792443, + "grad_norm": 11.896427154541016, + "learning_rate": 2.9156086078107914e-06, + "loss": 0.3895, + "step": 56800 + }, + { + "epoch": 2.842987614162392, + "grad_norm": 0.03020811267197132, + "learning_rate": 2.906340939000204e-06, + "loss": 0.1709, + "step": 56810 + }, + { + "epoch": 2.84348805204554, + "grad_norm": 33.46907424926758, + "learning_rate": 2.8970732701896168e-06, + "loss": 0.3145, + "step": 56820 + }, + { + "epoch": 2.8439884899286874, + "grad_norm": 124.77019500732422, + "learning_rate": 2.887805601379029e-06, + "loss": 0.1258, + "step": 56830 + }, + { + "epoch": 2.8444889278118355, + "grad_norm": 0.002428194275125861, + "learning_rate": 2.8785379325684417e-06, + "loss": 0.1627, + "step": 56840 + }, + { + "epoch": 2.844989365694983, + "grad_norm": 0.06242990866303444, + "learning_rate": 2.8692702637578544e-06, + "loss": 0.0753, + "step": 56850 + }, + { + "epoch": 2.845489803578131, + "grad_norm": 37.001033782958984, + "learning_rate": 2.860002594947267e-06, + "loss": 0.185, + "step": 56860 + }, + { + "epoch": 2.8459902414612785, + "grad_norm": 120.3897476196289, + "learning_rate": 2.8507349261366797e-06, + "loss": 0.2211, + "step": 56870 + }, + { + "epoch": 2.8464906793444262, + "grad_norm": 84.41731262207031, + "learning_rate": 2.8414672573260924e-06, + "loss": 0.6197, + "step": 56880 + }, + { + "epoch": 2.8469911172275744, + "grad_norm": 0.01950308494269848, + "learning_rate": 2.832199588515505e-06, + "loss": 0.259, + "step": 56890 + }, + { + "epoch": 2.8474915551107216, + "grad_norm": 0.010406963527202606, + "learning_rate": 2.8229319197049177e-06, + "loss": 0.0972, + "step": 56900 + }, + { + "epoch": 2.8479919929938697, + "grad_norm": 0.022837677970528603, + "learning_rate": 2.8136642508943304e-06, + "loss": 0.1648, + "step": 56910 + }, + { + "epoch": 2.8484924308770174, + "grad_norm": 0.09012983739376068, + "learning_rate": 2.8043965820837426e-06, + "loss": 0.0051, + "step": 56920 + }, + { + "epoch": 2.848992868760165, + "grad_norm": 57.45946502685547, + "learning_rate": 2.7951289132731553e-06, + "loss": 0.1971, + "step": 56930 + }, + { + "epoch": 2.849493306643313, + "grad_norm": 0.02808561734855175, + "learning_rate": 2.785861244462568e-06, + "loss": 0.0532, + "step": 56940 + }, + { + "epoch": 2.8499937445264605, + "grad_norm": 0.01179296150803566, + "learning_rate": 2.7765935756519806e-06, + "loss": 0.0107, + "step": 56950 + }, + { + "epoch": 2.8504941824096086, + "grad_norm": 0.006684999912977219, + "learning_rate": 2.767325906841393e-06, + "loss": 0.3201, + "step": 56960 + }, + { + "epoch": 2.8509946202927563, + "grad_norm": 0.01490097213536501, + "learning_rate": 2.758058238030806e-06, + "loss": 0.1388, + "step": 56970 + }, + { + "epoch": 2.851495058175904, + "grad_norm": 0.005457353312522173, + "learning_rate": 2.7487905692202187e-06, + "loss": 0.3249, + "step": 56980 + }, + { + "epoch": 2.8519954960590517, + "grad_norm": 0.0017573570366948843, + "learning_rate": 2.7395229004096313e-06, + "loss": 0.1366, + "step": 56990 + }, + { + "epoch": 2.8524959339421994, + "grad_norm": 0.0021893077064305544, + "learning_rate": 2.7302552315990436e-06, + "loss": 0.1436, + "step": 57000 + }, + { + "epoch": 2.852996371825347, + "grad_norm": 0.011630503460764885, + "learning_rate": 2.7209875627884562e-06, + "loss": 0.0136, + "step": 57010 + }, + { + "epoch": 2.8534968097084947, + "grad_norm": 0.03460406884551048, + "learning_rate": 2.711719893977869e-06, + "loss": 0.0284, + "step": 57020 + }, + { + "epoch": 2.853997247591643, + "grad_norm": 32.94295120239258, + "learning_rate": 2.7024522251672816e-06, + "loss": 0.5535, + "step": 57030 + }, + { + "epoch": 2.8544976854747905, + "grad_norm": 44.114906311035156, + "learning_rate": 2.693184556356694e-06, + "loss": 0.1638, + "step": 57040 + }, + { + "epoch": 2.8549981233579382, + "grad_norm": 0.5732833743095398, + "learning_rate": 2.6839168875461065e-06, + "loss": 0.2057, + "step": 57050 + }, + { + "epoch": 2.855498561241086, + "grad_norm": 0.009473584592342377, + "learning_rate": 2.6746492187355196e-06, + "loss": 0.439, + "step": 57060 + }, + { + "epoch": 2.8559989991242336, + "grad_norm": 0.014999611303210258, + "learning_rate": 2.6653815499249323e-06, + "loss": 0.0505, + "step": 57070 + }, + { + "epoch": 2.8564994370073813, + "grad_norm": 8.126874923706055, + "learning_rate": 2.6561138811143445e-06, + "loss": 0.1972, + "step": 57080 + }, + { + "epoch": 2.856999874890529, + "grad_norm": 42.540374755859375, + "learning_rate": 2.646846212303757e-06, + "loss": 0.2928, + "step": 57090 + }, + { + "epoch": 2.857500312773677, + "grad_norm": 0.002807572251185775, + "learning_rate": 2.63757854349317e-06, + "loss": 0.274, + "step": 57100 + }, + { + "epoch": 2.858000750656825, + "grad_norm": 51.46083068847656, + "learning_rate": 2.6283108746825825e-06, + "loss": 0.0995, + "step": 57110 + }, + { + "epoch": 2.8585011885399725, + "grad_norm": 0.28614938259124756, + "learning_rate": 2.619043205871995e-06, + "loss": 0.1386, + "step": 57120 + }, + { + "epoch": 2.85900162642312, + "grad_norm": 0.06321782618761063, + "learning_rate": 2.6097755370614074e-06, + "loss": 0.2531, + "step": 57130 + }, + { + "epoch": 2.859502064306268, + "grad_norm": 0.002996381837874651, + "learning_rate": 2.60050786825082e-06, + "loss": 0.0218, + "step": 57140 + }, + { + "epoch": 2.860002502189416, + "grad_norm": 95.47913360595703, + "learning_rate": 2.591240199440233e-06, + "loss": 0.2694, + "step": 57150 + }, + { + "epoch": 2.8605029400725632, + "grad_norm": 0.020139893516898155, + "learning_rate": 2.5819725306296454e-06, + "loss": 0.0199, + "step": 57160 + }, + { + "epoch": 2.8610033779557114, + "grad_norm": 0.0895015150308609, + "learning_rate": 2.572704861819058e-06, + "loss": 0.2532, + "step": 57170 + }, + { + "epoch": 2.861503815838859, + "grad_norm": 0.9309127330780029, + "learning_rate": 2.5634371930084708e-06, + "loss": 0.2005, + "step": 57180 + }, + { + "epoch": 2.8620042537220067, + "grad_norm": 0.013479774817824364, + "learning_rate": 2.5541695241978834e-06, + "loss": 0.2128, + "step": 57190 + }, + { + "epoch": 2.8625046916051544, + "grad_norm": 28.290531158447266, + "learning_rate": 2.544901855387296e-06, + "loss": 0.2602, + "step": 57200 + }, + { + "epoch": 2.863005129488302, + "grad_norm": 0.014987978152930737, + "learning_rate": 2.5356341865767084e-06, + "loss": 0.3943, + "step": 57210 + }, + { + "epoch": 2.8635055673714502, + "grad_norm": 0.06580891460180283, + "learning_rate": 2.526366517766121e-06, + "loss": 0.2314, + "step": 57220 + }, + { + "epoch": 2.864006005254598, + "grad_norm": 0.3221062123775482, + "learning_rate": 2.5170988489555337e-06, + "loss": 0.5757, + "step": 57230 + }, + { + "epoch": 2.8645064431377456, + "grad_norm": 0.017948484048247337, + "learning_rate": 2.507831180144947e-06, + "loss": 0.1276, + "step": 57240 + }, + { + "epoch": 2.8650068810208933, + "grad_norm": 0.0067477538250386715, + "learning_rate": 2.498563511334359e-06, + "loss": 0.0065, + "step": 57250 + }, + { + "epoch": 2.865507318904041, + "grad_norm": 0.0005999404820613563, + "learning_rate": 2.4892958425237717e-06, + "loss": 0.2248, + "step": 57260 + }, + { + "epoch": 2.8660077567871887, + "grad_norm": 0.005863881204277277, + "learning_rate": 2.4800281737131844e-06, + "loss": 0.1663, + "step": 57270 + }, + { + "epoch": 2.8665081946703364, + "grad_norm": 0.18658415973186493, + "learning_rate": 2.470760504902597e-06, + "loss": 0.4067, + "step": 57280 + }, + { + "epoch": 2.8670086325534845, + "grad_norm": 116.64852142333984, + "learning_rate": 2.4614928360920093e-06, + "loss": 0.2936, + "step": 57290 + }, + { + "epoch": 2.867509070436632, + "grad_norm": 18.121173858642578, + "learning_rate": 2.452225167281422e-06, + "loss": 0.2097, + "step": 57300 + }, + { + "epoch": 2.86800950831978, + "grad_norm": 151.27587890625, + "learning_rate": 2.4429574984708346e-06, + "loss": 0.0668, + "step": 57310 + }, + { + "epoch": 2.8685099462029275, + "grad_norm": 0.007292787544429302, + "learning_rate": 2.4336898296602473e-06, + "loss": 0.1545, + "step": 57320 + }, + { + "epoch": 2.8690103840860752, + "grad_norm": 123.95307159423828, + "learning_rate": 2.42442216084966e-06, + "loss": 0.4289, + "step": 57330 + }, + { + "epoch": 2.869510821969223, + "grad_norm": 18.258882522583008, + "learning_rate": 2.4151544920390726e-06, + "loss": 0.6845, + "step": 57340 + }, + { + "epoch": 2.8700112598523706, + "grad_norm": 145.63687133789062, + "learning_rate": 2.4058868232284853e-06, + "loss": 0.2324, + "step": 57350 + }, + { + "epoch": 2.8705116977355187, + "grad_norm": 0.004687258042395115, + "learning_rate": 2.396619154417898e-06, + "loss": 0.1198, + "step": 57360 + }, + { + "epoch": 2.8710121356186664, + "grad_norm": 0.27505671977996826, + "learning_rate": 2.3873514856073106e-06, + "loss": 0.2051, + "step": 57370 + }, + { + "epoch": 2.871512573501814, + "grad_norm": 1.1144750118255615, + "learning_rate": 2.378083816796723e-06, + "loss": 0.1571, + "step": 57380 + }, + { + "epoch": 2.872013011384962, + "grad_norm": 0.003093757899478078, + "learning_rate": 2.3688161479861356e-06, + "loss": 0.0346, + "step": 57390 + }, + { + "epoch": 2.8725134492681095, + "grad_norm": 104.54627227783203, + "learning_rate": 2.3595484791755482e-06, + "loss": 0.3109, + "step": 57400 + }, + { + "epoch": 2.8730138871512576, + "grad_norm": 0.003922802861779928, + "learning_rate": 2.350280810364961e-06, + "loss": 0.1462, + "step": 57410 + }, + { + "epoch": 2.873514325034405, + "grad_norm": 22.985321044921875, + "learning_rate": 2.3410131415543736e-06, + "loss": 0.4676, + "step": 57420 + }, + { + "epoch": 2.874014762917553, + "grad_norm": 30.556291580200195, + "learning_rate": 2.3317454727437862e-06, + "loss": 0.2562, + "step": 57430 + }, + { + "epoch": 2.8745152008007007, + "grad_norm": 0.012784902937710285, + "learning_rate": 2.322477803933199e-06, + "loss": 0.2976, + "step": 57440 + }, + { + "epoch": 2.8750156386838484, + "grad_norm": 80.40872955322266, + "learning_rate": 2.3132101351226116e-06, + "loss": 0.1389, + "step": 57450 + }, + { + "epoch": 2.875516076566996, + "grad_norm": 0.030993985012173653, + "learning_rate": 2.303942466312024e-06, + "loss": 0.3759, + "step": 57460 + }, + { + "epoch": 2.8760165144501437, + "grad_norm": 70.9173583984375, + "learning_rate": 2.2946747975014365e-06, + "loss": 0.0734, + "step": 57470 + }, + { + "epoch": 2.876516952333292, + "grad_norm": 0.02930140309035778, + "learning_rate": 2.285407128690849e-06, + "loss": 0.126, + "step": 57480 + }, + { + "epoch": 2.8770173902164395, + "grad_norm": 0.03737739101052284, + "learning_rate": 2.276139459880262e-06, + "loss": 0.0511, + "step": 57490 + }, + { + "epoch": 2.8775178280995872, + "grad_norm": 0.8011598587036133, + "learning_rate": 2.266871791069674e-06, + "loss": 0.0578, + "step": 57500 + }, + { + "epoch": 2.878018265982735, + "grad_norm": 1.275637149810791, + "learning_rate": 2.257604122259087e-06, + "loss": 0.2952, + "step": 57510 + }, + { + "epoch": 2.8785187038658826, + "grad_norm": 0.01602550409734249, + "learning_rate": 2.2483364534485e-06, + "loss": 0.0029, + "step": 57520 + }, + { + "epoch": 2.8790191417490303, + "grad_norm": 0.44147783517837524, + "learning_rate": 2.2390687846379125e-06, + "loss": 0.1359, + "step": 57530 + }, + { + "epoch": 2.879519579632178, + "grad_norm": 0.004879973828792572, + "learning_rate": 2.2298011158273248e-06, + "loss": 0.3584, + "step": 57540 + }, + { + "epoch": 2.880020017515326, + "grad_norm": 0.5507230758666992, + "learning_rate": 2.2205334470167374e-06, + "loss": 0.0689, + "step": 57550 + }, + { + "epoch": 2.880520455398474, + "grad_norm": 0.03170169144868851, + "learning_rate": 2.21126577820615e-06, + "loss": 0.4541, + "step": 57560 + }, + { + "epoch": 2.8810208932816215, + "grad_norm": 20.67360496520996, + "learning_rate": 2.2019981093955628e-06, + "loss": 0.2426, + "step": 57570 + }, + { + "epoch": 2.881521331164769, + "grad_norm": 21.11448860168457, + "learning_rate": 2.1927304405849754e-06, + "loss": 0.391, + "step": 57580 + }, + { + "epoch": 2.882021769047917, + "grad_norm": 31.926895141601562, + "learning_rate": 2.1834627717743877e-06, + "loss": 0.3115, + "step": 57590 + }, + { + "epoch": 2.8825222069310645, + "grad_norm": 0.018563488498330116, + "learning_rate": 2.1741951029638008e-06, + "loss": 0.224, + "step": 57600 + }, + { + "epoch": 2.883022644814212, + "grad_norm": 0.7045378684997559, + "learning_rate": 2.1649274341532134e-06, + "loss": 0.0644, + "step": 57610 + }, + { + "epoch": 2.8835230826973604, + "grad_norm": 0.1984180361032486, + "learning_rate": 2.1556597653426257e-06, + "loss": 0.2356, + "step": 57620 + }, + { + "epoch": 2.884023520580508, + "grad_norm": 0.023076830431818962, + "learning_rate": 2.1463920965320384e-06, + "loss": 0.2756, + "step": 57630 + }, + { + "epoch": 2.8845239584636557, + "grad_norm": 45.02309799194336, + "learning_rate": 2.137124427721451e-06, + "loss": 0.5762, + "step": 57640 + }, + { + "epoch": 2.8850243963468034, + "grad_norm": 16.89664649963379, + "learning_rate": 2.1278567589108637e-06, + "loss": 0.4741, + "step": 57650 + }, + { + "epoch": 2.885524834229951, + "grad_norm": 0.00340336374938488, + "learning_rate": 2.1185890901002764e-06, + "loss": 0.4534, + "step": 57660 + }, + { + "epoch": 2.8860252721130992, + "grad_norm": 157.74478149414062, + "learning_rate": 2.1093214212896886e-06, + "loss": 0.1162, + "step": 57670 + }, + { + "epoch": 2.8865257099962465, + "grad_norm": 0.017850741744041443, + "learning_rate": 2.1000537524791013e-06, + "loss": 0.3583, + "step": 57680 + }, + { + "epoch": 2.8870261478793946, + "grad_norm": 88.53473663330078, + "learning_rate": 2.0907860836685144e-06, + "loss": 0.3802, + "step": 57690 + }, + { + "epoch": 2.8875265857625423, + "grad_norm": 0.050205446779727936, + "learning_rate": 2.081518414857927e-06, + "loss": 0.3723, + "step": 57700 + }, + { + "epoch": 2.88802702364569, + "grad_norm": 24.77780532836914, + "learning_rate": 2.0722507460473393e-06, + "loss": 0.1768, + "step": 57710 + }, + { + "epoch": 2.8885274615288377, + "grad_norm": 0.03377895429730415, + "learning_rate": 2.062983077236752e-06, + "loss": 0.2209, + "step": 57720 + }, + { + "epoch": 2.8890278994119853, + "grad_norm": 0.021542275324463844, + "learning_rate": 2.0537154084261646e-06, + "loss": 0.7557, + "step": 57730 + }, + { + "epoch": 2.8895283372951335, + "grad_norm": 0.11822956055402756, + "learning_rate": 2.0444477396155773e-06, + "loss": 0.0661, + "step": 57740 + }, + { + "epoch": 2.890028775178281, + "grad_norm": 50.23988723754883, + "learning_rate": 2.0351800708049895e-06, + "loss": 0.5575, + "step": 57750 + }, + { + "epoch": 2.890529213061429, + "grad_norm": 1.0382297039031982, + "learning_rate": 2.0259124019944022e-06, + "loss": 0.119, + "step": 57760 + }, + { + "epoch": 2.8910296509445765, + "grad_norm": 30.948564529418945, + "learning_rate": 2.016644733183815e-06, + "loss": 0.3242, + "step": 57770 + }, + { + "epoch": 2.891530088827724, + "grad_norm": 3.8033528327941895, + "learning_rate": 2.007377064373228e-06, + "loss": 0.208, + "step": 57780 + }, + { + "epoch": 2.892030526710872, + "grad_norm": 0.09451092779636383, + "learning_rate": 1.9981093955626402e-06, + "loss": 0.4567, + "step": 57790 + }, + { + "epoch": 2.8925309645940196, + "grad_norm": 1.9720643758773804, + "learning_rate": 1.988841726752053e-06, + "loss": 0.0202, + "step": 57800 + }, + { + "epoch": 2.8930314024771677, + "grad_norm": 139.41281127929688, + "learning_rate": 1.9795740579414656e-06, + "loss": 0.2376, + "step": 57810 + }, + { + "epoch": 2.8935318403603154, + "grad_norm": 0.0021596518345177174, + "learning_rate": 1.9703063891308782e-06, + "loss": 0.4882, + "step": 57820 + }, + { + "epoch": 2.894032278243463, + "grad_norm": 0.18559619784355164, + "learning_rate": 1.9610387203202905e-06, + "loss": 0.2075, + "step": 57830 + }, + { + "epoch": 2.894532716126611, + "grad_norm": 50.10270309448242, + "learning_rate": 1.951771051509703e-06, + "loss": 0.278, + "step": 57840 + }, + { + "epoch": 2.8950331540097585, + "grad_norm": 0.015525158494710922, + "learning_rate": 1.942503382699116e-06, + "loss": 0.1462, + "step": 57850 + }, + { + "epoch": 2.895533591892906, + "grad_norm": 0.0673438161611557, + "learning_rate": 1.9332357138885285e-06, + "loss": 0.012, + "step": 57860 + }, + { + "epoch": 2.896034029776054, + "grad_norm": 0.042604707181453705, + "learning_rate": 1.923968045077941e-06, + "loss": 0.1973, + "step": 57870 + }, + { + "epoch": 2.896534467659202, + "grad_norm": 62.38845443725586, + "learning_rate": 1.914700376267354e-06, + "loss": 0.4405, + "step": 57880 + }, + { + "epoch": 2.8970349055423497, + "grad_norm": 0.0029982724227011204, + "learning_rate": 1.9054327074567665e-06, + "loss": 0.1015, + "step": 57890 + }, + { + "epoch": 2.8975353434254973, + "grad_norm": 50.723175048828125, + "learning_rate": 1.896165038646179e-06, + "loss": 0.2293, + "step": 57900 + }, + { + "epoch": 2.898035781308645, + "grad_norm": 38.37306213378906, + "learning_rate": 1.8868973698355916e-06, + "loss": 0.207, + "step": 57910 + }, + { + "epoch": 2.8985362191917927, + "grad_norm": 19.938661575317383, + "learning_rate": 1.8776297010250043e-06, + "loss": 0.1622, + "step": 57920 + }, + { + "epoch": 2.899036657074941, + "grad_norm": 0.006251406855881214, + "learning_rate": 1.8683620322144168e-06, + "loss": 0.2969, + "step": 57930 + }, + { + "epoch": 2.899537094958088, + "grad_norm": 0.011950446292757988, + "learning_rate": 1.8590943634038294e-06, + "loss": 0.0714, + "step": 57940 + }, + { + "epoch": 2.900037532841236, + "grad_norm": 9.673850059509277, + "learning_rate": 1.8498266945932419e-06, + "loss": 0.0578, + "step": 57950 + }, + { + "epoch": 2.900537970724384, + "grad_norm": 0.3221280574798584, + "learning_rate": 1.8405590257826548e-06, + "loss": 0.1514, + "step": 57960 + }, + { + "epoch": 2.9010384086075316, + "grad_norm": 0.8872140645980835, + "learning_rate": 1.8312913569720674e-06, + "loss": 0.5601, + "step": 57970 + }, + { + "epoch": 2.9015388464906793, + "grad_norm": 0.0020398376509547234, + "learning_rate": 1.82202368816148e-06, + "loss": 0.2234, + "step": 57980 + }, + { + "epoch": 2.902039284373827, + "grad_norm": 12.743843078613281, + "learning_rate": 1.8127560193508926e-06, + "loss": 0.3077, + "step": 57990 + }, + { + "epoch": 2.902539722256975, + "grad_norm": 0.0018078135326504707, + "learning_rate": 1.8034883505403052e-06, + "loss": 0.2663, + "step": 58000 + }, + { + "epoch": 2.9030401601401223, + "grad_norm": 0.06716956198215485, + "learning_rate": 1.7942206817297177e-06, + "loss": 0.0281, + "step": 58010 + }, + { + "epoch": 2.9035405980232705, + "grad_norm": 0.1360960453748703, + "learning_rate": 1.7849530129191304e-06, + "loss": 0.3311, + "step": 58020 + }, + { + "epoch": 2.904041035906418, + "grad_norm": 16.065956115722656, + "learning_rate": 1.7756853441085428e-06, + "loss": 0.3143, + "step": 58030 + }, + { + "epoch": 2.904541473789566, + "grad_norm": 0.007522755302488804, + "learning_rate": 1.7664176752979555e-06, + "loss": 0.0825, + "step": 58040 + }, + { + "epoch": 2.9050419116727135, + "grad_norm": 0.3241857588291168, + "learning_rate": 1.7571500064873684e-06, + "loss": 0.1666, + "step": 58050 + }, + { + "epoch": 2.905542349555861, + "grad_norm": 17.277433395385742, + "learning_rate": 1.747882337676781e-06, + "loss": 0.2569, + "step": 58060 + }, + { + "epoch": 2.9060427874390093, + "grad_norm": 0.1808914989233017, + "learning_rate": 1.7386146688661935e-06, + "loss": 0.1136, + "step": 58070 + }, + { + "epoch": 2.906543225322157, + "grad_norm": 156.6179962158203, + "learning_rate": 1.7293470000556062e-06, + "loss": 0.0751, + "step": 58080 + }, + { + "epoch": 2.9070436632053047, + "grad_norm": 0.11545134335756302, + "learning_rate": 1.7200793312450186e-06, + "loss": 0.0623, + "step": 58090 + }, + { + "epoch": 2.9075441010884524, + "grad_norm": 0.014872241765260696, + "learning_rate": 1.7108116624344313e-06, + "loss": 0.1092, + "step": 58100 + }, + { + "epoch": 2.9080445389716, + "grad_norm": 62.74874496459961, + "learning_rate": 1.701543993623844e-06, + "loss": 0.1047, + "step": 58110 + }, + { + "epoch": 2.9085449768547478, + "grad_norm": 100.64775848388672, + "learning_rate": 1.6922763248132564e-06, + "loss": 0.0886, + "step": 58120 + }, + { + "epoch": 2.9090454147378955, + "grad_norm": 8.051138877868652, + "learning_rate": 1.683008656002669e-06, + "loss": 0.1498, + "step": 58130 + }, + { + "epoch": 2.9095458526210436, + "grad_norm": 71.98966979980469, + "learning_rate": 1.673740987192082e-06, + "loss": 0.2428, + "step": 58140 + }, + { + "epoch": 2.9100462905041913, + "grad_norm": 47.967262268066406, + "learning_rate": 1.6644733183814944e-06, + "loss": 0.3486, + "step": 58150 + }, + { + "epoch": 2.910546728387339, + "grad_norm": 0.0013771761441603303, + "learning_rate": 1.655205649570907e-06, + "loss": 0.1668, + "step": 58160 + }, + { + "epoch": 2.9110471662704867, + "grad_norm": 4.346522331237793, + "learning_rate": 1.6459379807603198e-06, + "loss": 0.1706, + "step": 58170 + }, + { + "epoch": 2.9115476041536343, + "grad_norm": 39.09217834472656, + "learning_rate": 1.6366703119497322e-06, + "loss": 0.2558, + "step": 58180 + }, + { + "epoch": 2.912048042036782, + "grad_norm": 0.005327288061380386, + "learning_rate": 1.6274026431391449e-06, + "loss": 0.177, + "step": 58190 + }, + { + "epoch": 2.9125484799199297, + "grad_norm": 0.08121754974126816, + "learning_rate": 1.6181349743285573e-06, + "loss": 0.0839, + "step": 58200 + }, + { + "epoch": 2.913048917803078, + "grad_norm": 0.027167102321982384, + "learning_rate": 1.60886730551797e-06, + "loss": 0.1995, + "step": 58210 + }, + { + "epoch": 2.9135493556862255, + "grad_norm": 9.671104431152344, + "learning_rate": 1.5995996367073825e-06, + "loss": 0.1514, + "step": 58220 + }, + { + "epoch": 2.914049793569373, + "grad_norm": 67.68254852294922, + "learning_rate": 1.5903319678967956e-06, + "loss": 0.3353, + "step": 58230 + }, + { + "epoch": 2.914550231452521, + "grad_norm": 22.217012405395508, + "learning_rate": 1.581064299086208e-06, + "loss": 0.7104, + "step": 58240 + }, + { + "epoch": 2.9150506693356686, + "grad_norm": 0.051831141114234924, + "learning_rate": 1.5717966302756207e-06, + "loss": 0.1556, + "step": 58250 + }, + { + "epoch": 2.9155511072188167, + "grad_norm": 0.06793921440839767, + "learning_rate": 1.5625289614650332e-06, + "loss": 0.316, + "step": 58260 + }, + { + "epoch": 2.916051545101964, + "grad_norm": 0.6066509485244751, + "learning_rate": 1.5532612926544458e-06, + "loss": 0.0942, + "step": 58270 + }, + { + "epoch": 2.916551982985112, + "grad_norm": 1.324234962463379, + "learning_rate": 1.5439936238438583e-06, + "loss": 0.0994, + "step": 58280 + }, + { + "epoch": 2.9170524208682598, + "grad_norm": 108.70734405517578, + "learning_rate": 1.534725955033271e-06, + "loss": 0.3785, + "step": 58290 + }, + { + "epoch": 2.9175528587514075, + "grad_norm": 0.05135975033044815, + "learning_rate": 1.5254582862226836e-06, + "loss": 0.1959, + "step": 58300 + }, + { + "epoch": 2.918053296634555, + "grad_norm": 0.025919415056705475, + "learning_rate": 1.5161906174120963e-06, + "loss": 0.2276, + "step": 58310 + }, + { + "epoch": 2.918553734517703, + "grad_norm": 0.3541892468929291, + "learning_rate": 1.5069229486015087e-06, + "loss": 0.1937, + "step": 58320 + }, + { + "epoch": 2.919054172400851, + "grad_norm": 0.45804429054260254, + "learning_rate": 1.4976552797909214e-06, + "loss": 0.1052, + "step": 58330 + }, + { + "epoch": 2.9195546102839987, + "grad_norm": 35.640892028808594, + "learning_rate": 1.488387610980334e-06, + "loss": 0.193, + "step": 58340 + }, + { + "epoch": 2.9200550481671463, + "grad_norm": 2.090437650680542, + "learning_rate": 1.4791199421697468e-06, + "loss": 0.2077, + "step": 58350 + }, + { + "epoch": 2.920555486050294, + "grad_norm": 1.5553261041641235, + "learning_rate": 1.4698522733591592e-06, + "loss": 0.2782, + "step": 58360 + }, + { + "epoch": 2.9210559239334417, + "grad_norm": 70.23728942871094, + "learning_rate": 1.4605846045485719e-06, + "loss": 0.0913, + "step": 58370 + }, + { + "epoch": 2.9215563618165894, + "grad_norm": 0.003356712870299816, + "learning_rate": 1.4513169357379846e-06, + "loss": 0.435, + "step": 58380 + }, + { + "epoch": 2.922056799699737, + "grad_norm": 0.5276179909706116, + "learning_rate": 1.4420492669273972e-06, + "loss": 0.2201, + "step": 58390 + }, + { + "epoch": 2.922557237582885, + "grad_norm": 82.59899139404297, + "learning_rate": 1.4327815981168097e-06, + "loss": 0.1618, + "step": 58400 + }, + { + "epoch": 2.923057675466033, + "grad_norm": 3.3380796909332275, + "learning_rate": 1.4235139293062223e-06, + "loss": 0.197, + "step": 58410 + }, + { + "epoch": 2.9235581133491806, + "grad_norm": 0.000848713971208781, + "learning_rate": 1.414246260495635e-06, + "loss": 0.1062, + "step": 58420 + }, + { + "epoch": 2.9240585512323283, + "grad_norm": 5.775787830352783, + "learning_rate": 1.4049785916850477e-06, + "loss": 0.0091, + "step": 58430 + }, + { + "epoch": 2.924558989115476, + "grad_norm": 26.41167640686035, + "learning_rate": 1.3957109228744604e-06, + "loss": 0.2386, + "step": 58440 + }, + { + "epoch": 2.9250594269986236, + "grad_norm": 121.17642974853516, + "learning_rate": 1.3864432540638728e-06, + "loss": 0.1029, + "step": 58450 + }, + { + "epoch": 2.9255598648817713, + "grad_norm": 0.1362910270690918, + "learning_rate": 1.3771755852532855e-06, + "loss": 0.0044, + "step": 58460 + }, + { + "epoch": 2.9260603027649195, + "grad_norm": 0.0032885263208299875, + "learning_rate": 1.367907916442698e-06, + "loss": 0.1526, + "step": 58470 + }, + { + "epoch": 2.926560740648067, + "grad_norm": 159.34970092773438, + "learning_rate": 1.3586402476321108e-06, + "loss": 0.0395, + "step": 58480 + }, + { + "epoch": 2.927061178531215, + "grad_norm": 0.317752480506897, + "learning_rate": 1.3493725788215233e-06, + "loss": 0.0423, + "step": 58490 + }, + { + "epoch": 2.9275616164143625, + "grad_norm": 0.0293453186750412, + "learning_rate": 1.340104910010936e-06, + "loss": 0.1596, + "step": 58500 + }, + { + "epoch": 2.92806205429751, + "grad_norm": 0.00717025063931942, + "learning_rate": 1.3308372412003484e-06, + "loss": 0.2075, + "step": 58510 + }, + { + "epoch": 2.9285624921806583, + "grad_norm": 0.003746991977095604, + "learning_rate": 1.3215695723897613e-06, + "loss": 0.3826, + "step": 58520 + }, + { + "epoch": 2.9290629300638056, + "grad_norm": 0.7668344974517822, + "learning_rate": 1.3123019035791737e-06, + "loss": 0.1634, + "step": 58530 + }, + { + "epoch": 2.9295633679469537, + "grad_norm": 0.00465210760012269, + "learning_rate": 1.3030342347685864e-06, + "loss": 0.1988, + "step": 58540 + }, + { + "epoch": 2.9300638058301014, + "grad_norm": 0.002582026645541191, + "learning_rate": 1.2937665659579989e-06, + "loss": 0.3547, + "step": 58550 + }, + { + "epoch": 2.930564243713249, + "grad_norm": 0.015318874269723892, + "learning_rate": 1.2844988971474115e-06, + "loss": 0.1064, + "step": 58560 + }, + { + "epoch": 2.9310646815963968, + "grad_norm": 1.5629428625106812, + "learning_rate": 1.2752312283368242e-06, + "loss": 0.3798, + "step": 58570 + }, + { + "epoch": 2.9315651194795445, + "grad_norm": 0.016781432554125786, + "learning_rate": 1.2659635595262369e-06, + "loss": 0.0618, + "step": 58580 + }, + { + "epoch": 2.9320655573626926, + "grad_norm": 0.05590541288256645, + "learning_rate": 1.2566958907156493e-06, + "loss": 0.1598, + "step": 58590 + }, + { + "epoch": 2.9325659952458403, + "grad_norm": 0.006095320917665958, + "learning_rate": 1.247428221905062e-06, + "loss": 0.4666, + "step": 58600 + }, + { + "epoch": 2.933066433128988, + "grad_norm": 0.018803561106324196, + "learning_rate": 1.2381605530944747e-06, + "loss": 0.4078, + "step": 58610 + }, + { + "epoch": 2.9335668710121356, + "grad_norm": 0.022977612912654877, + "learning_rate": 1.2288928842838873e-06, + "loss": 0.0545, + "step": 58620 + }, + { + "epoch": 2.9340673088952833, + "grad_norm": 0.05562311410903931, + "learning_rate": 1.2196252154732998e-06, + "loss": 0.6782, + "step": 58630 + }, + { + "epoch": 2.934567746778431, + "grad_norm": 0.001480288221500814, + "learning_rate": 1.2103575466627125e-06, + "loss": 0.2471, + "step": 58640 + }, + { + "epoch": 2.9350681846615787, + "grad_norm": 0.4216262698173523, + "learning_rate": 1.2010898778521251e-06, + "loss": 0.4904, + "step": 58650 + }, + { + "epoch": 2.935568622544727, + "grad_norm": 108.7698745727539, + "learning_rate": 1.1918222090415378e-06, + "loss": 0.2163, + "step": 58660 + }, + { + "epoch": 2.9360690604278745, + "grad_norm": 11.047207832336426, + "learning_rate": 1.1825545402309505e-06, + "loss": 0.1825, + "step": 58670 + }, + { + "epoch": 2.936569498311022, + "grad_norm": 99.78950500488281, + "learning_rate": 1.173286871420363e-06, + "loss": 0.4781, + "step": 58680 + }, + { + "epoch": 2.93706993619417, + "grad_norm": 75.3531494140625, + "learning_rate": 1.1640192026097756e-06, + "loss": 0.1407, + "step": 58690 + }, + { + "epoch": 2.9375703740773176, + "grad_norm": 0.027641665190458298, + "learning_rate": 1.1547515337991883e-06, + "loss": 0.3069, + "step": 58700 + }, + { + "epoch": 2.9380708119604653, + "grad_norm": 0.005255357827991247, + "learning_rate": 1.145483864988601e-06, + "loss": 0.1317, + "step": 58710 + }, + { + "epoch": 2.938571249843613, + "grad_norm": 35.853668212890625, + "learning_rate": 1.1362161961780134e-06, + "loss": 0.191, + "step": 58720 + }, + { + "epoch": 2.939071687726761, + "grad_norm": 0.40039631724357605, + "learning_rate": 1.126948527367426e-06, + "loss": 0.0562, + "step": 58730 + }, + { + "epoch": 2.9395721256099088, + "grad_norm": 0.0034841536544263363, + "learning_rate": 1.1176808585568385e-06, + "loss": 0.3539, + "step": 58740 + }, + { + "epoch": 2.9400725634930565, + "grad_norm": 0.23719123005867004, + "learning_rate": 1.1084131897462514e-06, + "loss": 0.1971, + "step": 58750 + }, + { + "epoch": 2.940573001376204, + "grad_norm": 0.0011781504144892097, + "learning_rate": 1.0991455209356639e-06, + "loss": 0.6654, + "step": 58760 + }, + { + "epoch": 2.941073439259352, + "grad_norm": 0.2979811728000641, + "learning_rate": 1.0898778521250765e-06, + "loss": 0.0817, + "step": 58770 + }, + { + "epoch": 2.9415738771425, + "grad_norm": 1.790711522102356, + "learning_rate": 1.080610183314489e-06, + "loss": 0.3577, + "step": 58780 + }, + { + "epoch": 2.942074315025647, + "grad_norm": 4.476901531219482, + "learning_rate": 1.0713425145039019e-06, + "loss": 0.1118, + "step": 58790 + }, + { + "epoch": 2.9425747529087953, + "grad_norm": 30.635068893432617, + "learning_rate": 1.0620748456933143e-06, + "loss": 0.0423, + "step": 58800 + }, + { + "epoch": 2.943075190791943, + "grad_norm": 62.91170883178711, + "learning_rate": 1.052807176882727e-06, + "loss": 0.2266, + "step": 58810 + }, + { + "epoch": 2.9435756286750907, + "grad_norm": 0.02068292535841465, + "learning_rate": 1.0435395080721395e-06, + "loss": 0.0928, + "step": 58820 + }, + { + "epoch": 2.9440760665582384, + "grad_norm": 0.002519358182325959, + "learning_rate": 1.0342718392615521e-06, + "loss": 0.2409, + "step": 58830 + }, + { + "epoch": 2.944576504441386, + "grad_norm": 0.005876005161553621, + "learning_rate": 1.0250041704509648e-06, + "loss": 0.5027, + "step": 58840 + }, + { + "epoch": 2.945076942324534, + "grad_norm": 0.004355174023658037, + "learning_rate": 1.0157365016403775e-06, + "loss": 0.2541, + "step": 58850 + }, + { + "epoch": 2.9455773802076815, + "grad_norm": 0.32604560256004333, + "learning_rate": 1.00646883282979e-06, + "loss": 0.2606, + "step": 58860 + }, + { + "epoch": 2.9460778180908296, + "grad_norm": 0.030545631423592567, + "learning_rate": 9.972011640192026e-07, + "loss": 0.1467, + "step": 58870 + }, + { + "epoch": 2.9465782559739773, + "grad_norm": 0.002536026295274496, + "learning_rate": 9.879334952086153e-07, + "loss": 0.0562, + "step": 58880 + }, + { + "epoch": 2.947078693857125, + "grad_norm": 53.60538101196289, + "learning_rate": 9.78665826398028e-07, + "loss": 0.0103, + "step": 58890 + }, + { + "epoch": 2.9475791317402726, + "grad_norm": 13.053601264953613, + "learning_rate": 9.693981575874404e-07, + "loss": 0.3115, + "step": 58900 + }, + { + "epoch": 2.9480795696234203, + "grad_norm": 0.006596755236387253, + "learning_rate": 9.60130488776853e-07, + "loss": 0.0879, + "step": 58910 + }, + { + "epoch": 2.9485800075065685, + "grad_norm": 0.4454554617404938, + "learning_rate": 9.508628199662656e-07, + "loss": 0.2372, + "step": 58920 + }, + { + "epoch": 2.949080445389716, + "grad_norm": 16.73801040649414, + "learning_rate": 9.415951511556784e-07, + "loss": 0.4041, + "step": 58930 + }, + { + "epoch": 2.949580883272864, + "grad_norm": 41.52074432373047, + "learning_rate": 9.32327482345091e-07, + "loss": 0.0101, + "step": 58940 + }, + { + "epoch": 2.9500813211560115, + "grad_norm": 0.008064993657171726, + "learning_rate": 9.230598135345035e-07, + "loss": 0.039, + "step": 58950 + }, + { + "epoch": 2.950581759039159, + "grad_norm": 123.44711303710938, + "learning_rate": 9.137921447239161e-07, + "loss": 0.2844, + "step": 58960 + }, + { + "epoch": 2.951082196922307, + "grad_norm": 0.003305513644590974, + "learning_rate": 9.045244759133289e-07, + "loss": 0.1827, + "step": 58970 + }, + { + "epoch": 2.9515826348054546, + "grad_norm": 0.0027956031262874603, + "learning_rate": 8.952568071027414e-07, + "loss": 0.5056, + "step": 58980 + }, + { + "epoch": 2.9520830726886027, + "grad_norm": 0.006068565417081118, + "learning_rate": 8.85989138292154e-07, + "loss": 0.4222, + "step": 58990 + }, + { + "epoch": 2.9525835105717504, + "grad_norm": 0.0025336602702736855, + "learning_rate": 8.767214694815666e-07, + "loss": 0.2098, + "step": 59000 + }, + { + "epoch": 2.953083948454898, + "grad_norm": 0.0016803753096610308, + "learning_rate": 8.674538006709792e-07, + "loss": 0.0014, + "step": 59010 + }, + { + "epoch": 2.9535843863380458, + "grad_norm": 0.005916744936257601, + "learning_rate": 8.581861318603919e-07, + "loss": 0.279, + "step": 59020 + }, + { + "epoch": 2.9540848242211935, + "grad_norm": 0.27994003891944885, + "learning_rate": 8.489184630498045e-07, + "loss": 0.1295, + "step": 59030 + }, + { + "epoch": 2.954585262104341, + "grad_norm": 0.0045267571695148945, + "learning_rate": 8.396507942392171e-07, + "loss": 0.3857, + "step": 59040 + }, + { + "epoch": 2.955085699987489, + "grad_norm": 179.03053283691406, + "learning_rate": 8.303831254286297e-07, + "loss": 0.224, + "step": 59050 + }, + { + "epoch": 2.955586137870637, + "grad_norm": 3.9205451011657715, + "learning_rate": 8.211154566180424e-07, + "loss": 0.013, + "step": 59060 + }, + { + "epoch": 2.9560865757537846, + "grad_norm": 0.010136112570762634, + "learning_rate": 8.11847787807455e-07, + "loss": 0.1031, + "step": 59070 + }, + { + "epoch": 2.9565870136369323, + "grad_norm": 0.0384041853249073, + "learning_rate": 8.025801189968676e-07, + "loss": 0.1855, + "step": 59080 + }, + { + "epoch": 2.95708745152008, + "grad_norm": 0.0821397602558136, + "learning_rate": 7.933124501862802e-07, + "loss": 0.3675, + "step": 59090 + }, + { + "epoch": 2.9575878894032277, + "grad_norm": 23.979175567626953, + "learning_rate": 7.840447813756927e-07, + "loss": 0.3205, + "step": 59100 + }, + { + "epoch": 2.958088327286376, + "grad_norm": 21.24807357788086, + "learning_rate": 7.747771125651054e-07, + "loss": 0.6426, + "step": 59110 + }, + { + "epoch": 2.958588765169523, + "grad_norm": 0.0045844255946576595, + "learning_rate": 7.655094437545181e-07, + "loss": 0.2783, + "step": 59120 + }, + { + "epoch": 2.959089203052671, + "grad_norm": 15.054691314697266, + "learning_rate": 7.562417749439306e-07, + "loss": 0.4768, + "step": 59130 + }, + { + "epoch": 2.959589640935819, + "grad_norm": 0.005498078651726246, + "learning_rate": 7.469741061333433e-07, + "loss": 0.1134, + "step": 59140 + }, + { + "epoch": 2.9600900788189666, + "grad_norm": 1.2120693922042847, + "learning_rate": 7.377064373227559e-07, + "loss": 0.3042, + "step": 59150 + }, + { + "epoch": 2.9605905167021143, + "grad_norm": 38.234397888183594, + "learning_rate": 7.284387685121685e-07, + "loss": 0.1737, + "step": 59160 + }, + { + "epoch": 2.961090954585262, + "grad_norm": 0.07520147413015366, + "learning_rate": 7.191710997015811e-07, + "loss": 0.6588, + "step": 59170 + }, + { + "epoch": 2.96159139246841, + "grad_norm": 392.56292724609375, + "learning_rate": 7.099034308909937e-07, + "loss": 0.5489, + "step": 59180 + }, + { + "epoch": 2.9620918303515578, + "grad_norm": 0.007796322461217642, + "learning_rate": 7.006357620804063e-07, + "loss": 0.1697, + "step": 59190 + }, + { + "epoch": 2.9625922682347055, + "grad_norm": 0.008702303282916546, + "learning_rate": 6.913680932698189e-07, + "loss": 0.4736, + "step": 59200 + }, + { + "epoch": 2.963092706117853, + "grad_norm": 81.98515319824219, + "learning_rate": 6.821004244592316e-07, + "loss": 0.201, + "step": 59210 + }, + { + "epoch": 2.963593144001001, + "grad_norm": 2.0835118293762207, + "learning_rate": 6.728327556486441e-07, + "loss": 0.0738, + "step": 59220 + }, + { + "epoch": 2.9640935818841485, + "grad_norm": 35.40370559692383, + "learning_rate": 6.635650868380568e-07, + "loss": 0.3172, + "step": 59230 + }, + { + "epoch": 2.964594019767296, + "grad_norm": 23.444080352783203, + "learning_rate": 6.542974180274694e-07, + "loss": 0.1134, + "step": 59240 + }, + { + "epoch": 2.9650944576504443, + "grad_norm": 0.002383539918810129, + "learning_rate": 6.45029749216882e-07, + "loss": 0.2857, + "step": 59250 + }, + { + "epoch": 2.965594895533592, + "grad_norm": 0.006513621192425489, + "learning_rate": 6.357620804062946e-07, + "loss": 0.2536, + "step": 59260 + }, + { + "epoch": 2.9660953334167397, + "grad_norm": 11.840142250061035, + "learning_rate": 6.264944115957072e-07, + "loss": 0.1163, + "step": 59270 + }, + { + "epoch": 2.9665957712998874, + "grad_norm": 0.029289498925209045, + "learning_rate": 6.172267427851198e-07, + "loss": 0.1546, + "step": 59280 + }, + { + "epoch": 2.967096209183035, + "grad_norm": 0.013420775532722473, + "learning_rate": 6.079590739745325e-07, + "loss": 0.2893, + "step": 59290 + }, + { + "epoch": 2.9675966470661828, + "grad_norm": 0.006180759519338608, + "learning_rate": 5.986914051639451e-07, + "loss": 0.0433, + "step": 59300 + }, + { + "epoch": 2.9680970849493304, + "grad_norm": 0.007466780487447977, + "learning_rate": 5.894237363533577e-07, + "loss": 0.2379, + "step": 59310 + }, + { + "epoch": 2.9685975228324786, + "grad_norm": 0.08655251562595367, + "learning_rate": 5.801560675427704e-07, + "loss": 0.2814, + "step": 59320 + }, + { + "epoch": 2.9690979607156263, + "grad_norm": 20.327959060668945, + "learning_rate": 5.70888398732183e-07, + "loss": 0.2562, + "step": 59330 + }, + { + "epoch": 2.969598398598774, + "grad_norm": 0.07779091596603394, + "learning_rate": 5.616207299215956e-07, + "loss": 0.3112, + "step": 59340 + }, + { + "epoch": 2.9700988364819216, + "grad_norm": 0.06243646889925003, + "learning_rate": 5.523530611110082e-07, + "loss": 0.0724, + "step": 59350 + }, + { + "epoch": 2.9705992743650693, + "grad_norm": 55.93368148803711, + "learning_rate": 5.430853923004208e-07, + "loss": 0.3917, + "step": 59360 + }, + { + "epoch": 2.9710997122482175, + "grad_norm": 0.005560718011111021, + "learning_rate": 5.338177234898334e-07, + "loss": 0.0599, + "step": 59370 + }, + { + "epoch": 2.9716001501313647, + "grad_norm": 0.0036988495849072933, + "learning_rate": 5.24550054679246e-07, + "loss": 0.0732, + "step": 59380 + }, + { + "epoch": 2.972100588014513, + "grad_norm": 0.0028758614789694548, + "learning_rate": 5.152823858686587e-07, + "loss": 0.1537, + "step": 59390 + }, + { + "epoch": 2.9726010258976605, + "grad_norm": 2.886552572250366, + "learning_rate": 5.060147170580712e-07, + "loss": 0.234, + "step": 59400 + }, + { + "epoch": 2.973101463780808, + "grad_norm": 17.29117774963379, + "learning_rate": 4.967470482474839e-07, + "loss": 0.2236, + "step": 59410 + }, + { + "epoch": 2.973601901663956, + "grad_norm": 0.9561861157417297, + "learning_rate": 4.874793794368965e-07, + "loss": 0.275, + "step": 59420 + }, + { + "epoch": 2.9741023395471036, + "grad_norm": 0.00983459036797285, + "learning_rate": 4.782117106263091e-07, + "loss": 0.2009, + "step": 59430 + }, + { + "epoch": 2.9746027774302517, + "grad_norm": 122.1671371459961, + "learning_rate": 4.689440418157217e-07, + "loss": 0.1751, + "step": 59440 + }, + { + "epoch": 2.9751032153133994, + "grad_norm": 43.181785583496094, + "learning_rate": 4.5967637300513426e-07, + "loss": 0.488, + "step": 59450 + }, + { + "epoch": 2.975603653196547, + "grad_norm": 119.95027923583984, + "learning_rate": 4.5040870419454693e-07, + "loss": 0.2681, + "step": 59460 + }, + { + "epoch": 2.9761040910796948, + "grad_norm": 0.024042019620537758, + "learning_rate": 4.411410353839595e-07, + "loss": 0.4886, + "step": 59470 + }, + { + "epoch": 2.9766045289628424, + "grad_norm": 0.14669445157051086, + "learning_rate": 4.3187336657337216e-07, + "loss": 0.3357, + "step": 59480 + }, + { + "epoch": 2.97710496684599, + "grad_norm": 64.50975036621094, + "learning_rate": 4.226056977627848e-07, + "loss": 0.2092, + "step": 59490 + }, + { + "epoch": 2.977605404729138, + "grad_norm": 0.01565660908818245, + "learning_rate": 4.133380289521974e-07, + "loss": 0.185, + "step": 59500 + }, + { + "epoch": 2.978105842612286, + "grad_norm": 0.005355158355087042, + "learning_rate": 4.0407036014161e-07, + "loss": 0.2534, + "step": 59510 + }, + { + "epoch": 2.9786062804954336, + "grad_norm": 0.0025346274487674236, + "learning_rate": 3.948026913310227e-07, + "loss": 0.5326, + "step": 59520 + }, + { + "epoch": 2.9791067183785813, + "grad_norm": 0.009735047817230225, + "learning_rate": 3.8553502252043525e-07, + "loss": 0.0154, + "step": 59530 + }, + { + "epoch": 2.979607156261729, + "grad_norm": 0.0035545136779546738, + "learning_rate": 3.7626735370984786e-07, + "loss": 0.6573, + "step": 59540 + }, + { + "epoch": 2.9801075941448767, + "grad_norm": 17.558156967163086, + "learning_rate": 3.669996848992605e-07, + "loss": 0.1264, + "step": 59550 + }, + { + "epoch": 2.9806080320280244, + "grad_norm": 0.0014485775027424097, + "learning_rate": 3.577320160886731e-07, + "loss": 0.0852, + "step": 59560 + }, + { + "epoch": 2.981108469911172, + "grad_norm": 43.37870788574219, + "learning_rate": 3.4846434727808566e-07, + "loss": 0.4489, + "step": 59570 + }, + { + "epoch": 2.98160890779432, + "grad_norm": 0.0016763253370299935, + "learning_rate": 3.391966784674983e-07, + "loss": 0.2821, + "step": 59580 + }, + { + "epoch": 2.982109345677468, + "grad_norm": 54.27371597290039, + "learning_rate": 3.299290096569109e-07, + "loss": 0.4316, + "step": 59590 + }, + { + "epoch": 2.9826097835606156, + "grad_norm": 0.015180137939751148, + "learning_rate": 3.206613408463235e-07, + "loss": 0.0767, + "step": 59600 + }, + { + "epoch": 2.9831102214437633, + "grad_norm": 98.58396911621094, + "learning_rate": 3.113936720357361e-07, + "loss": 0.6636, + "step": 59610 + }, + { + "epoch": 2.983610659326911, + "grad_norm": 0.009758652187883854, + "learning_rate": 3.0212600322514874e-07, + "loss": 0.2168, + "step": 59620 + }, + { + "epoch": 2.984111097210059, + "grad_norm": 0.001770585891790688, + "learning_rate": 2.928583344145614e-07, + "loss": 0.0067, + "step": 59630 + }, + { + "epoch": 2.9846115350932063, + "grad_norm": 0.3700677156448364, + "learning_rate": 2.8359066560397403e-07, + "loss": 0.1624, + "step": 59640 + }, + { + "epoch": 2.9851119729763544, + "grad_norm": 28.579029083251953, + "learning_rate": 2.7432299679338664e-07, + "loss": 0.3308, + "step": 59650 + }, + { + "epoch": 2.985612410859502, + "grad_norm": 58.15632629394531, + "learning_rate": 2.650553279827992e-07, + "loss": 0.1768, + "step": 59660 + }, + { + "epoch": 2.98611284874265, + "grad_norm": 94.88214111328125, + "learning_rate": 2.557876591722118e-07, + "loss": 0.311, + "step": 59670 + }, + { + "epoch": 2.9866132866257975, + "grad_norm": 12.443111419677734, + "learning_rate": 2.4651999036162444e-07, + "loss": 0.1023, + "step": 59680 + }, + { + "epoch": 2.987113724508945, + "grad_norm": 25.858402252197266, + "learning_rate": 2.3725232155103706e-07, + "loss": 0.2205, + "step": 59690 + }, + { + "epoch": 2.9876141623920933, + "grad_norm": 0.023004285991191864, + "learning_rate": 2.2798465274044967e-07, + "loss": 0.0845, + "step": 59700 + }, + { + "epoch": 2.9881146002752406, + "grad_norm": 0.032680694013834, + "learning_rate": 2.187169839298623e-07, + "loss": 0.3715, + "step": 59710 + }, + { + "epoch": 2.9886150381583887, + "grad_norm": 0.0007050809217616916, + "learning_rate": 2.094493151192749e-07, + "loss": 0.0511, + "step": 59720 + }, + { + "epoch": 2.9891154760415364, + "grad_norm": 0.010106697678565979, + "learning_rate": 2.0018164630868755e-07, + "loss": 0.0851, + "step": 59730 + }, + { + "epoch": 2.989615913924684, + "grad_norm": 1.3041093349456787, + "learning_rate": 1.9091397749810014e-07, + "loss": 0.2248, + "step": 59740 + }, + { + "epoch": 2.9901163518078318, + "grad_norm": 0.0020095950458198786, + "learning_rate": 1.8164630868751276e-07, + "loss": 0.0778, + "step": 59750 + }, + { + "epoch": 2.9906167896909794, + "grad_norm": 10.908820152282715, + "learning_rate": 1.7237863987692537e-07, + "loss": 0.2578, + "step": 59760 + }, + { + "epoch": 2.9911172275741276, + "grad_norm": 0.1526775062084198, + "learning_rate": 1.6311097106633796e-07, + "loss": 0.1862, + "step": 59770 + }, + { + "epoch": 2.9916176654572753, + "grad_norm": 0.10212847590446472, + "learning_rate": 1.5384330225575058e-07, + "loss": 0.0216, + "step": 59780 + }, + { + "epoch": 2.992118103340423, + "grad_norm": 7.3319315910339355, + "learning_rate": 1.4457563344516322e-07, + "loss": 0.5454, + "step": 59790 + }, + { + "epoch": 2.9926185412235706, + "grad_norm": 0.056509800255298615, + "learning_rate": 1.3530796463457584e-07, + "loss": 0.2681, + "step": 59800 + }, + { + "epoch": 2.9931189791067183, + "grad_norm": 0.0017340255435556173, + "learning_rate": 1.2604029582398843e-07, + "loss": 0.0044, + "step": 59810 + }, + { + "epoch": 2.993619416989866, + "grad_norm": 0.002209821017459035, + "learning_rate": 1.1677262701340105e-07, + "loss": 0.1034, + "step": 59820 + }, + { + "epoch": 2.9941198548730137, + "grad_norm": 0.007244156673550606, + "learning_rate": 1.0750495820281366e-07, + "loss": 0.0639, + "step": 59830 + }, + { + "epoch": 2.994620292756162, + "grad_norm": 0.051756639033555984, + "learning_rate": 9.823728939222629e-08, + "loss": 0.1901, + "step": 59840 + }, + { + "epoch": 2.9951207306393095, + "grad_norm": 166.4386749267578, + "learning_rate": 8.89696205816389e-08, + "loss": 0.3324, + "step": 59850 + }, + { + "epoch": 2.995621168522457, + "grad_norm": 0.015056375414133072, + "learning_rate": 7.970195177105151e-08, + "loss": 0.1145, + "step": 59860 + }, + { + "epoch": 2.996121606405605, + "grad_norm": 0.5218987464904785, + "learning_rate": 7.043428296046413e-08, + "loss": 0.2514, + "step": 59870 + }, + { + "epoch": 2.9966220442887526, + "grad_norm": 43.13042449951172, + "learning_rate": 6.116661414987675e-08, + "loss": 0.4488, + "step": 59880 + }, + { + "epoch": 2.9971224821719002, + "grad_norm": 0.5391528606414795, + "learning_rate": 5.1898945339289356e-08, + "loss": 0.2437, + "step": 59890 + }, + { + "epoch": 2.997622920055048, + "grad_norm": 0.010746831074357033, + "learning_rate": 4.263127652870197e-08, + "loss": 0.2943, + "step": 59900 + }, + { + "epoch": 2.998123357938196, + "grad_norm": 0.0012645687675103545, + "learning_rate": 3.336360771811459e-08, + "loss": 0.0989, + "step": 59910 + }, + { + "epoch": 2.9986237958213438, + "grad_norm": 0.005694156046956778, + "learning_rate": 2.4095938907527203e-08, + "loss": 0.0061, + "step": 59920 + }, + { + "epoch": 2.9991242337044914, + "grad_norm": 33.80691146850586, + "learning_rate": 1.4828270096939816e-08, + "loss": 0.154, + "step": 59930 + }, + { + "epoch": 2.999624671587639, + "grad_norm": 0.007596906740218401, + "learning_rate": 5.560601286352431e-09, + "loss": 0.4436, + "step": 59940 + }, + { + "epoch": 2.999924934317528, + "eval_accuracy": 0.9675748705246566, + "eval_loss": 0.13822685182094574, + "eval_runtime": 142.0495, + "eval_samples_per_second": 62.527, + "eval_steps_per_second": 62.527, + "step": 59946 + }, + { + "epoch": 2.999924934317528, + "step": 59946, + "total_flos": 5.96040154031702e+18, + "train_loss": 0.3818838064128899, + "train_runtime": 9137.9419, + "train_samples_per_second": 26.241, + "train_steps_per_second": 6.56 + } + ], + "logging_steps": 10, + "max_steps": 59946, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 500, + "total_flos": 5.96040154031702e+18, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +}