Qwen3-8B-ODA-Mixture-100k / trainer_state.json
GX-XinGao's picture
Initial release v1.0
cc8d18b verified
Raw History Blame Contribute Delete
65.5 kB
{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 3708, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.008090614886731391, "grad_norm": 22.52252141115457, "learning_rate": 1.2129380053908356e-06, "loss": 2.5582, "step": 10 }, { "epoch": 0.016181229773462782, "grad_norm": 11.441648898342923, "learning_rate": 2.560646900269542e-06, "loss": 2.2025, "step": 20 }, { "epoch": 0.024271844660194174, "grad_norm": 5.370882200215843, "learning_rate": 3.908355795148248e-06, "loss": 1.668, "step": 30 }, { "epoch": 0.032362459546925564, "grad_norm": 2.0115313434456827, "learning_rate": 5.256064690026954e-06, "loss": 1.1266, "step": 40 }, { "epoch": 0.040453074433656956, "grad_norm": 1.2036774860036918, "learning_rate": 6.60377358490566e-06, "loss": 0.7799, "step": 50 }, { "epoch": 0.04854368932038835, "grad_norm": 0.8106871294631399, "learning_rate": 7.951482479784367e-06, "loss": 0.5907, "step": 60 }, { "epoch": 0.05663430420711974, "grad_norm": 0.6572037767143685, "learning_rate": 9.299191374663074e-06, "loss": 0.4977, "step": 70 }, { "epoch": 0.06472491909385113, "grad_norm": 0.9743925361281945, "learning_rate": 1.0646900269541779e-05, "loss": 0.4728, "step": 80 }, { "epoch": 0.07281553398058252, "grad_norm": 0.5302813737105591, "learning_rate": 1.1994609164420486e-05, "loss": 0.446, "step": 90 }, { "epoch": 0.08090614886731391, "grad_norm": 0.5504892707748702, "learning_rate": 1.3342318059299191e-05, "loss": 0.4164, "step": 100 }, { "epoch": 0.0889967637540453, "grad_norm": 0.46842577857609835, "learning_rate": 1.4690026954177898e-05, "loss": 0.412, "step": 110 }, { "epoch": 0.0970873786407767, "grad_norm": 0.5228315101871648, "learning_rate": 1.6037735849056604e-05, "loss": 0.3975, "step": 120 }, { "epoch": 0.10517799352750809, "grad_norm": 0.5623706583567197, "learning_rate": 1.738544474393531e-05, "loss": 0.3857, "step": 130 }, { "epoch": 0.11326860841423948, "grad_norm": 0.3506071062876399, "learning_rate": 1.8733153638814018e-05, "loss": 0.3837, "step": 140 }, { "epoch": 0.12135922330097088, "grad_norm": 0.491171903870991, "learning_rate": 2.0080862533692725e-05, "loss": 0.3721, "step": 150 }, { "epoch": 0.12944983818770225, "grad_norm": 0.5505284472221431, "learning_rate": 2.1428571428571428e-05, "loss": 0.3734, "step": 160 }, { "epoch": 0.13754045307443366, "grad_norm": 0.8767975628519997, "learning_rate": 2.2776280323450135e-05, "loss": 0.365, "step": 170 }, { "epoch": 0.14563106796116504, "grad_norm": 0.5343891120951508, "learning_rate": 2.4123989218328842e-05, "loss": 0.3631, "step": 180 }, { "epoch": 0.15372168284789645, "grad_norm": 0.47266629407427385, "learning_rate": 2.547169811320755e-05, "loss": 0.3509, "step": 190 }, { "epoch": 0.16181229773462782, "grad_norm": 0.49039579917599835, "learning_rate": 2.6819407008086256e-05, "loss": 0.3474, "step": 200 }, { "epoch": 0.16990291262135923, "grad_norm": 0.5095673082261557, "learning_rate": 2.8167115902964963e-05, "loss": 0.3439, "step": 210 }, { "epoch": 0.1779935275080906, "grad_norm": 0.7559655324445135, "learning_rate": 2.9514824797843667e-05, "loss": 0.3477, "step": 220 }, { "epoch": 0.18608414239482202, "grad_norm": 0.717837164870727, "learning_rate": 3.086253369272237e-05, "loss": 0.3424, "step": 230 }, { "epoch": 0.1941747572815534, "grad_norm": 0.44154340671189124, "learning_rate": 3.221024258760108e-05, "loss": 0.3432, "step": 240 }, { "epoch": 0.2022653721682848, "grad_norm": 0.645302977385022, "learning_rate": 3.355795148247979e-05, "loss": 0.3378, "step": 250 }, { "epoch": 0.21035598705501618, "grad_norm": 0.45392402289172107, "learning_rate": 3.490566037735849e-05, "loss": 0.34, "step": 260 }, { "epoch": 0.21844660194174756, "grad_norm": 0.7064296610826926, "learning_rate": 3.62533692722372e-05, "loss": 0.339, "step": 270 }, { "epoch": 0.22653721682847897, "grad_norm": 0.5611292380408083, "learning_rate": 3.76010781671159e-05, "loss": 0.3284, "step": 280 }, { "epoch": 0.23462783171521034, "grad_norm": 0.5448155574311832, "learning_rate": 3.894878706199461e-05, "loss": 0.3291, "step": 290 }, { "epoch": 0.24271844660194175, "grad_norm": 0.43795871099809547, "learning_rate": 4.0296495956873316e-05, "loss": 0.3206, "step": 300 }, { "epoch": 0.25080906148867316, "grad_norm": 0.5584253534316864, "learning_rate": 4.164420485175202e-05, "loss": 0.3276, "step": 310 }, { "epoch": 0.2588996763754045, "grad_norm": 0.5617857932874714, "learning_rate": 4.299191374663073e-05, "loss": 0.3258, "step": 320 }, { "epoch": 0.2669902912621359, "grad_norm": 0.5441488728977509, "learning_rate": 4.433962264150944e-05, "loss": 0.3227, "step": 330 }, { "epoch": 0.2750809061488673, "grad_norm": 0.8386999082473845, "learning_rate": 4.5687331536388144e-05, "loss": 0.3196, "step": 340 }, { "epoch": 0.28317152103559873, "grad_norm": 0.43608786652562637, "learning_rate": 4.703504043126685e-05, "loss": 0.3191, "step": 350 }, { "epoch": 0.2912621359223301, "grad_norm": 0.4937683097487802, "learning_rate": 4.838274932614555e-05, "loss": 0.3255, "step": 360 }, { "epoch": 0.2993527508090615, "grad_norm": 0.5061197818365496, "learning_rate": 4.973045822102426e-05, "loss": 0.3119, "step": 370 }, { "epoch": 0.3074433656957929, "grad_norm": 0.4744335618999143, "learning_rate": 4.9999290952604396e-05, "loss": 0.3191, "step": 380 }, { "epoch": 0.3155339805825243, "grad_norm": 0.5037259460580787, "learning_rate": 4.9996410516491115e-05, "loss": 0.3098, "step": 390 }, { "epoch": 0.32362459546925565, "grad_norm": 0.5710713036434062, "learning_rate": 4.9991314631296585e-05, "loss": 0.3096, "step": 400 }, { "epoch": 0.33171521035598706, "grad_norm": 0.33451540591505297, "learning_rate": 4.9984003748672604e-05, "loss": 0.3095, "step": 410 }, { "epoch": 0.33980582524271846, "grad_norm": 0.4340252206831366, "learning_rate": 4.997447851658774e-05, "loss": 0.3076, "step": 420 }, { "epoch": 0.3478964401294498, "grad_norm": 0.3838413262491291, "learning_rate": 4.9962739779269887e-05, "loss": 0.3067, "step": 430 }, { "epoch": 0.3559870550161812, "grad_norm": 0.42248192127450584, "learning_rate": 4.9948788577131414e-05, "loss": 0.2989, "step": 440 }, { "epoch": 0.3640776699029126, "grad_norm": 0.4026141748836561, "learning_rate": 4.993262614667696e-05, "loss": 0.3018, "step": 450 }, { "epoch": 0.37216828478964403, "grad_norm": 0.4295230581487718, "learning_rate": 4.9914253920393884e-05, "loss": 0.3048, "step": 460 }, { "epoch": 0.3802588996763754, "grad_norm": 0.3384646839950234, "learning_rate": 4.9893673526625265e-05, "loss": 0.3025, "step": 470 }, { "epoch": 0.3883495145631068, "grad_norm": 0.41348113959511545, "learning_rate": 4.987088678942555e-05, "loss": 0.2997, "step": 480 }, { "epoch": 0.3964401294498382, "grad_norm": 0.3673430436712548, "learning_rate": 4.984589572839897e-05, "loss": 0.2981, "step": 490 }, { "epoch": 0.4045307443365696, "grad_norm": 0.3637541700718482, "learning_rate": 4.9818702558520485e-05, "loss": 0.2994, "step": 500 }, { "epoch": 0.41262135922330095, "grad_norm": 0.3667744606459147, "learning_rate": 4.978930968993946e-05, "loss": 0.3027, "step": 510 }, { "epoch": 0.42071197411003236, "grad_norm": 0.3783856834139639, "learning_rate": 4.9757719727766085e-05, "loss": 0.2995, "step": 520 }, { "epoch": 0.42880258899676377, "grad_norm": 0.39358597558975966, "learning_rate": 4.972393547184046e-05, "loss": 0.2968, "step": 530 }, { "epoch": 0.4368932038834951, "grad_norm": 0.3961355130202433, "learning_rate": 4.968795991648446e-05, "loss": 0.2951, "step": 540 }, { "epoch": 0.4449838187702265, "grad_norm": 0.37157203976493997, "learning_rate": 4.9649796250236344e-05, "loss": 0.2842, "step": 550 }, { "epoch": 0.45307443365695793, "grad_norm": 0.2840234936379272, "learning_rate": 4.960944785556814e-05, "loss": 0.2862, "step": 560 }, { "epoch": 0.46116504854368934, "grad_norm": 0.320746304504261, "learning_rate": 4.956691830858585e-05, "loss": 0.2925, "step": 570 }, { "epoch": 0.4692556634304207, "grad_norm": 0.3627941299734378, "learning_rate": 4.952221137871252e-05, "loss": 0.2884, "step": 580 }, { "epoch": 0.4773462783171521, "grad_norm": 0.44860484414052815, "learning_rate": 4.947533102835413e-05, "loss": 0.2937, "step": 590 }, { "epoch": 0.4854368932038835, "grad_norm": 0.3404390735893878, "learning_rate": 4.942628141254843e-05, "loss": 0.2888, "step": 600 }, { "epoch": 0.4935275080906149, "grad_norm": 0.3346590749564513, "learning_rate": 4.937506687859666e-05, "loss": 0.2839, "step": 610 }, { "epoch": 0.5016181229773463, "grad_norm": 0.3540605558733981, "learning_rate": 4.932169196567824e-05, "loss": 0.2926, "step": 620 }, { "epoch": 0.5097087378640777, "grad_norm": 0.3225115085834403, "learning_rate": 4.9266161404448454e-05, "loss": 0.2914, "step": 630 }, { "epoch": 0.517799352750809, "grad_norm": 0.333757773957392, "learning_rate": 4.920848011661919e-05, "loss": 0.2844, "step": 640 }, { "epoch": 0.5258899676375405, "grad_norm": 0.3691189283201263, "learning_rate": 4.914865321452274e-05, "loss": 0.2838, "step": 650 }, { "epoch": 0.5339805825242718, "grad_norm": 0.41635131408315823, "learning_rate": 4.908668600065862e-05, "loss": 0.2853, "step": 660 }, { "epoch": 0.5420711974110033, "grad_norm": 0.3161547689784877, "learning_rate": 4.90225839672237e-05, "loss": 0.2844, "step": 670 }, { "epoch": 0.5501618122977346, "grad_norm": 0.32264889950016085, "learning_rate": 4.8956352795625325e-05, "loss": 0.2912, "step": 680 }, { "epoch": 0.558252427184466, "grad_norm": 0.3354534030777558, "learning_rate": 4.8887998355977886e-05, "loss": 0.2826, "step": 690 }, { "epoch": 0.5663430420711975, "grad_norm": 0.3409096712904377, "learning_rate": 4.881752670658244e-05, "loss": 0.2799, "step": 700 }, { "epoch": 0.5744336569579288, "grad_norm": 0.3012793393888447, "learning_rate": 4.87449440933898e-05, "loss": 0.2713, "step": 710 }, { "epoch": 0.5825242718446602, "grad_norm": 0.3127062042429517, "learning_rate": 4.867025694944698e-05, "loss": 0.2813, "step": 720 }, { "epoch": 0.5906148867313916, "grad_norm": 0.3269471051340985, "learning_rate": 4.859347189432699e-05, "loss": 0.28, "step": 730 }, { "epoch": 0.598705501618123, "grad_norm": 0.32543025148595867, "learning_rate": 4.8514595733542144e-05, "loss": 0.2709, "step": 740 }, { "epoch": 0.6067961165048543, "grad_norm": 0.3147551999550132, "learning_rate": 4.8433635457940915e-05, "loss": 0.2814, "step": 750 }, { "epoch": 0.6148867313915858, "grad_norm": 0.26606594884988966, "learning_rate": 4.8350598243088283e-05, "loss": 0.2776, "step": 760 }, { "epoch": 0.6229773462783171, "grad_norm": 0.3007757906909445, "learning_rate": 4.8265491448629804e-05, "loss": 0.2767, "step": 770 }, { "epoch": 0.6310679611650486, "grad_norm": 0.3795450058305756, "learning_rate": 4.817832261763928e-05, "loss": 0.2772, "step": 780 }, { "epoch": 0.63915857605178, "grad_norm": 0.3106723870622729, "learning_rate": 4.8089099475950257e-05, "loss": 0.276, "step": 790 }, { "epoch": 0.6472491909385113, "grad_norm": 0.34373131058866446, "learning_rate": 4.7997829931471225e-05, "loss": 0.2765, "step": 800 }, { "epoch": 0.6553398058252428, "grad_norm": 0.2846997869981113, "learning_rate": 4.7904522073484786e-05, "loss": 0.2778, "step": 810 }, { "epoch": 0.6634304207119741, "grad_norm": 0.31767113534168995, "learning_rate": 4.780918417193065e-05, "loss": 0.2775, "step": 820 }, { "epoch": 0.6715210355987055, "grad_norm": 0.31169855988969003, "learning_rate": 4.7711824676672726e-05, "loss": 0.2733, "step": 830 }, { "epoch": 0.6796116504854369, "grad_norm": 0.3175387528791534, "learning_rate": 4.76124522167501e-05, "loss": 0.2746, "step": 840 }, { "epoch": 0.6877022653721683, "grad_norm": 0.2975263968714564, "learning_rate": 4.751107559961238e-05, "loss": 0.2745, "step": 850 }, { "epoch": 0.6957928802588996, "grad_norm": 0.29446496624170493, "learning_rate": 4.740770381033894e-05, "loss": 0.275, "step": 860 }, { "epoch": 0.7038834951456311, "grad_norm": 0.2973639432877628, "learning_rate": 4.730234601084268e-05, "loss": 0.2716, "step": 870 }, { "epoch": 0.7119741100323624, "grad_norm": 0.2497924694108393, "learning_rate": 4.719501153905793e-05, "loss": 0.2673, "step": 880 }, { "epoch": 0.7200647249190939, "grad_norm": 0.2948558193473281, "learning_rate": 4.7085709908112866e-05, "loss": 0.2731, "step": 890 }, { "epoch": 0.7281553398058253, "grad_norm": 0.2459558041018532, "learning_rate": 4.6974450805486305e-05, "loss": 0.2639, "step": 900 }, { "epoch": 0.7362459546925566, "grad_norm": 0.2559758327802977, "learning_rate": 4.686124409214917e-05, "loss": 0.2715, "step": 910 }, { "epoch": 0.7443365695792881, "grad_norm": 0.2811398346458149, "learning_rate": 4.674609980169042e-05, "loss": 0.2712, "step": 920 }, { "epoch": 0.7524271844660194, "grad_norm": 0.2686221572337129, "learning_rate": 4.662902813942784e-05, "loss": 0.272, "step": 930 }, { "epoch": 0.7605177993527508, "grad_norm": 0.2755997369035635, "learning_rate": 4.651003948150349e-05, "loss": 0.2689, "step": 940 }, { "epoch": 0.7686084142394822, "grad_norm": 0.2672412475643565, "learning_rate": 4.638914437396408e-05, "loss": 0.2693, "step": 950 }, { "epoch": 0.7766990291262136, "grad_norm": 0.23259525487399887, "learning_rate": 4.626635353182626e-05, "loss": 0.2659, "step": 960 }, { "epoch": 0.7847896440129449, "grad_norm": 0.26440852988259955, "learning_rate": 4.614167783812694e-05, "loss": 0.2753, "step": 970 }, { "epoch": 0.7928802588996764, "grad_norm": 0.2786648778182051, "learning_rate": 4.601512834295874e-05, "loss": 0.272, "step": 980 }, { "epoch": 0.8009708737864077, "grad_norm": 0.2475631464266887, "learning_rate": 4.588671626249057e-05, "loss": 0.2631, "step": 990 }, { "epoch": 0.8090614886731392, "grad_norm": 0.23380841398688618, "learning_rate": 4.5756452977973585e-05, "loss": 0.2609, "step": 1000 }, { "epoch": 0.8171521035598706, "grad_norm": 0.27565297705248776, "learning_rate": 4.56243500347324e-05, "loss": 0.2651, "step": 1010 }, { "epoch": 0.8252427184466019, "grad_norm": 0.29059051987658346, "learning_rate": 4.549041914114188e-05, "loss": 0.2687, "step": 1020 }, { "epoch": 0.8333333333333334, "grad_norm": 0.29594618957533503, "learning_rate": 4.535467216758936e-05, "loss": 0.2666, "step": 1030 }, { "epoch": 0.8414239482200647, "grad_norm": 0.3931170619188329, "learning_rate": 4.5217121145422616e-05, "loss": 0.2631, "step": 1040 }, { "epoch": 0.8495145631067961, "grad_norm": 0.2857742902635426, "learning_rate": 4.5077778265883477e-05, "loss": 0.2632, "step": 1050 }, { "epoch": 0.8576051779935275, "grad_norm": 0.24489808715354155, "learning_rate": 4.4936655879027336e-05, "loss": 0.2634, "step": 1060 }, { "epoch": 0.8656957928802589, "grad_norm": 0.26376201661023657, "learning_rate": 4.479376649262855e-05, "loss": 0.266, "step": 1070 }, { "epoch": 0.8737864077669902, "grad_norm": 0.25183442403581696, "learning_rate": 4.464912277107185e-05, "loss": 0.2658, "step": 1080 }, { "epoch": 0.8818770226537217, "grad_norm": 0.2613869002025676, "learning_rate": 4.450273753422992e-05, "loss": 0.2653, "step": 1090 }, { "epoch": 0.889967637540453, "grad_norm": 0.29304343706553704, "learning_rate": 4.435462375632711e-05, "loss": 0.2651, "step": 1100 }, { "epoch": 0.8980582524271845, "grad_norm": 0.263122059677539, "learning_rate": 4.420479456478957e-05, "loss": 0.2621, "step": 1110 }, { "epoch": 0.9061488673139159, "grad_norm": 0.24882087207880094, "learning_rate": 4.405326323908178e-05, "loss": 0.2635, "step": 1120 }, { "epoch": 0.9142394822006472, "grad_norm": 0.2744368878036377, "learning_rate": 4.390004320952947e-05, "loss": 0.2639, "step": 1130 }, { "epoch": 0.9223300970873787, "grad_norm": 0.25617953359336987, "learning_rate": 4.374514805612942e-05, "loss": 0.2667, "step": 1140 }, { "epoch": 0.93042071197411, "grad_norm": 0.26969276785611856, "learning_rate": 4.358859150734576e-05, "loss": 0.2592, "step": 1150 }, { "epoch": 0.9385113268608414, "grad_norm": 0.2703864111603607, "learning_rate": 4.343038743889324e-05, "loss": 0.2602, "step": 1160 }, { "epoch": 0.9466019417475728, "grad_norm": 0.3115866006041078, "learning_rate": 4.3270549872507415e-05, "loss": 0.26, "step": 1170 }, { "epoch": 0.9546925566343042, "grad_norm": 0.2611547498092126, "learning_rate": 4.3109092974701895e-05, "loss": 0.2591, "step": 1180 }, { "epoch": 0.9627831715210357, "grad_norm": 0.26931477864354814, "learning_rate": 4.2946031055512733e-05, "loss": 0.26, "step": 1190 }, { "epoch": 0.970873786407767, "grad_norm": 0.2578523252517678, "learning_rate": 4.2781378567230145e-05, "loss": 0.2633, "step": 1200 }, { "epoch": 0.9789644012944984, "grad_norm": 0.27451378253592124, "learning_rate": 4.2615150103117576e-05, "loss": 0.2598, "step": 1210 }, { "epoch": 0.9870550161812298, "grad_norm": 0.2609329028667746, "learning_rate": 4.24473603961183e-05, "loss": 0.2587, "step": 1220 }, { "epoch": 0.9951456310679612, "grad_norm": 0.2280958023284859, "learning_rate": 4.227802431754961e-05, "loss": 0.2589, "step": 1230 }, { "epoch": 1.0032362459546926, "grad_norm": 0.30070050651075947, "learning_rate": 4.210715687578481e-05, "loss": 0.2468, "step": 1240 }, { "epoch": 1.0113268608414239, "grad_norm": 0.26616355843357287, "learning_rate": 4.193477321492293e-05, "loss": 0.2219, "step": 1250 }, { "epoch": 1.0194174757281553, "grad_norm": 0.2713492665233822, "learning_rate": 4.176088861344657e-05, "loss": 0.2251, "step": 1260 }, { "epoch": 1.0275080906148868, "grad_norm": 0.21687794735432822, "learning_rate": 4.158551848286773e-05, "loss": 0.2178, "step": 1270 }, { "epoch": 1.035598705501618, "grad_norm": 0.2539615253588423, "learning_rate": 4.140867836636189e-05, "loss": 0.2289, "step": 1280 }, { "epoch": 1.0436893203883495, "grad_norm": 0.24657642734847438, "learning_rate": 4.1230383937390374e-05, "loss": 0.2258, "step": 1290 }, { "epoch": 1.051779935275081, "grad_norm": 0.2609125461360237, "learning_rate": 4.1050650998311215e-05, "loss": 0.2216, "step": 1300 }, { "epoch": 1.0598705501618122, "grad_norm": 0.3630380994135679, "learning_rate": 4.086949547897862e-05, "loss": 0.2225, "step": 1310 }, { "epoch": 1.0679611650485437, "grad_norm": 0.2504138070534478, "learning_rate": 4.068693343533103e-05, "loss": 0.2227, "step": 1320 }, { "epoch": 1.0760517799352751, "grad_norm": 0.3126812459563885, "learning_rate": 4.050298104796812e-05, "loss": 0.2268, "step": 1330 }, { "epoch": 1.0841423948220066, "grad_norm": 0.2257523743221347, "learning_rate": 4.0317654620716704e-05, "loss": 0.2306, "step": 1340 }, { "epoch": 1.0922330097087378, "grad_norm": 0.21960504136006673, "learning_rate": 4.013097057918566e-05, "loss": 0.2254, "step": 1350 }, { "epoch": 1.1003236245954693, "grad_norm": 0.250009270408735, "learning_rate": 3.9942945469310175e-05, "loss": 0.2279, "step": 1360 }, { "epoch": 1.1084142394822007, "grad_norm": 0.23480544241701537, "learning_rate": 3.9753595955885264e-05, "loss": 0.2256, "step": 1370 }, { "epoch": 1.116504854368932, "grad_norm": 0.23720387815460467, "learning_rate": 3.9562938821088705e-05, "loss": 0.2235, "step": 1380 }, { "epoch": 1.1245954692556634, "grad_norm": 0.23437981665878366, "learning_rate": 3.9370990962993695e-05, "loss": 0.2237, "step": 1390 }, { "epoch": 1.132686084142395, "grad_norm": 0.2286079680729574, "learning_rate": 3.9177769394071086e-05, "loss": 0.2215, "step": 1400 }, { "epoch": 1.1407766990291262, "grad_norm": 0.22687828650289346, "learning_rate": 3.898329123968163e-05, "loss": 0.2254, "step": 1410 }, { "epoch": 1.1488673139158576, "grad_norm": 0.2224235012119624, "learning_rate": 3.87875737365581e-05, "loss": 0.2184, "step": 1420 }, { "epoch": 1.156957928802589, "grad_norm": 0.20195576636612733, "learning_rate": 3.8590634231277626e-05, "loss": 0.2176, "step": 1430 }, { "epoch": 1.1650485436893203, "grad_norm": 0.20327918718818466, "learning_rate": 3.8392490178724184e-05, "loss": 0.2207, "step": 1440 }, { "epoch": 1.1731391585760518, "grad_norm": 0.21744881881616449, "learning_rate": 3.8193159140541645e-05, "loss": 0.223, "step": 1450 }, { "epoch": 1.1812297734627832, "grad_norm": 0.30658821143823883, "learning_rate": 3.7992658783577215e-05, "loss": 0.2264, "step": 1460 }, { "epoch": 1.1893203883495145, "grad_norm": 0.25174532799295857, "learning_rate": 3.779100687831563e-05, "loss": 0.2216, "step": 1470 }, { "epoch": 1.197411003236246, "grad_norm": 0.2561981869189091, "learning_rate": 3.758822129730415e-05, "loss": 0.2201, "step": 1480 }, { "epoch": 1.2055016181229774, "grad_norm": 0.22464952614110853, "learning_rate": 3.738432001356851e-05, "loss": 0.2239, "step": 1490 }, { "epoch": 1.2135922330097086, "grad_norm": 0.2100829197882277, "learning_rate": 3.7179321099019916e-05, "loss": 0.2281, "step": 1500 }, { "epoch": 1.22168284789644, "grad_norm": 0.2469541669228544, "learning_rate": 3.6973242722853365e-05, "loss": 0.225, "step": 1510 }, { "epoch": 1.2297734627831716, "grad_norm": 0.23312096116267617, "learning_rate": 3.6766103149937295e-05, "loss": 0.2246, "step": 1520 }, { "epoch": 1.237864077669903, "grad_norm": 0.20677868772581298, "learning_rate": 3.655792073919471e-05, "loss": 0.2249, "step": 1530 }, { "epoch": 1.2459546925566343, "grad_norm": 0.2563183131089248, "learning_rate": 3.634871394197607e-05, "loss": 0.2263, "step": 1540 }, { "epoch": 1.2540453074433657, "grad_norm": 0.22128405881374966, "learning_rate": 3.6138501300423934e-05, "loss": 0.2211, "step": 1550 }, { "epoch": 1.262135922330097, "grad_norm": 0.21006117612744632, "learning_rate": 3.592730144582948e-05, "loss": 0.2243, "step": 1560 }, { "epoch": 1.2702265372168284, "grad_norm": 0.1975344463355082, "learning_rate": 3.571513309698131e-05, "loss": 0.2202, "step": 1570 }, { "epoch": 1.27831715210356, "grad_norm": 0.20997495813930359, "learning_rate": 3.5502015058506335e-05, "loss": 0.2229, "step": 1580 }, { "epoch": 1.2864077669902914, "grad_norm": 0.2332593631600526, "learning_rate": 3.528796621920307e-05, "loss": 0.219, "step": 1590 }, { "epoch": 1.2944983818770226, "grad_norm": 0.22815002833576079, "learning_rate": 3.50730055503676e-05, "loss": 0.222, "step": 1600 }, { "epoch": 1.302588996763754, "grad_norm": 0.20437252443820447, "learning_rate": 3.485715210411204e-05, "loss": 0.2262, "step": 1610 }, { "epoch": 1.3106796116504853, "grad_norm": 0.2395647563898874, "learning_rate": 3.4640425011676034e-05, "loss": 0.2226, "step": 1620 }, { "epoch": 1.3187702265372168, "grad_norm": 0.22072548502378347, "learning_rate": 3.442284348173106e-05, "loss": 0.2254, "step": 1630 }, { "epoch": 1.3268608414239482, "grad_norm": 0.21606415951025024, "learning_rate": 3.420442679867796e-05, "loss": 0.2244, "step": 1640 }, { "epoch": 1.3349514563106797, "grad_norm": 0.27863651532858746, "learning_rate": 3.398519432093782e-05, "loss": 0.2227, "step": 1650 }, { "epoch": 1.343042071197411, "grad_norm": 0.2202632134109216, "learning_rate": 3.376516547923614e-05, "loss": 0.2195, "step": 1660 }, { "epoch": 1.3511326860841424, "grad_norm": 0.21524751059743225, "learning_rate": 3.3544359774880714e-05, "loss": 0.2203, "step": 1670 }, { "epoch": 1.3592233009708738, "grad_norm": 0.20473236269563994, "learning_rate": 3.3322796778033204e-05, "loss": 0.2151, "step": 1680 }, { "epoch": 1.367313915857605, "grad_norm": 0.21469153751522554, "learning_rate": 3.3100496125974624e-05, "loss": 0.216, "step": 1690 }, { "epoch": 1.3754045307443366, "grad_norm": 0.19954679234440933, "learning_rate": 3.2877477521364895e-05, "loss": 0.2162, "step": 1700 }, { "epoch": 1.383495145631068, "grad_norm": 0.18861748166722786, "learning_rate": 3.2653760730496555e-05, "loss": 0.2187, "step": 1710 }, { "epoch": 1.3915857605177995, "grad_norm": 0.20341053182705085, "learning_rate": 3.242936558154285e-05, "loss": 0.225, "step": 1720 }, { "epoch": 1.3996763754045307, "grad_norm": 0.19372792735024014, "learning_rate": 3.2204311962800426e-05, "loss": 0.2239, "step": 1730 }, { "epoch": 1.4077669902912622, "grad_norm": 0.27663774087598847, "learning_rate": 3.197861982092651e-05, "loss": 0.2201, "step": 1740 }, { "epoch": 1.4158576051779934, "grad_norm": 0.20101300631018146, "learning_rate": 3.175230915917108e-05, "loss": 0.2234, "step": 1750 }, { "epoch": 1.4239482200647249, "grad_norm": 0.20722987740988344, "learning_rate": 3.152540003560398e-05, "loss": 0.219, "step": 1760 }, { "epoch": 1.4320388349514563, "grad_norm": 0.23987446750061972, "learning_rate": 3.129791256133712e-05, "loss": 0.2158, "step": 1770 }, { "epoch": 1.4401294498381878, "grad_norm": 0.2072184752121984, "learning_rate": 3.106986689874204e-05, "loss": 0.2208, "step": 1780 }, { "epoch": 1.448220064724919, "grad_norm": 0.19280709770923388, "learning_rate": 3.0841283259662875e-05, "loss": 0.2227, "step": 1790 }, { "epoch": 1.4563106796116505, "grad_norm": 0.20536031119127787, "learning_rate": 3.0612181903625014e-05, "loss": 0.222, "step": 1800 }, { "epoch": 1.4644012944983817, "grad_norm": 0.22626011728625453, "learning_rate": 3.0382583136039444e-05, "loss": 0.2174, "step": 1810 }, { "epoch": 1.4724919093851132, "grad_norm": 0.2105569387760891, "learning_rate": 3.015250730640308e-05, "loss": 0.2167, "step": 1820 }, { "epoch": 1.4805825242718447, "grad_norm": 0.20632917153940236, "learning_rate": 2.9921974806495178e-05, "loss": 0.2141, "step": 1830 }, { "epoch": 1.4886731391585761, "grad_norm": 0.2021951348802143, "learning_rate": 2.969100606856998e-05, "loss": 0.2189, "step": 1840 }, { "epoch": 1.4967637540453074, "grad_norm": 0.21580708092456813, "learning_rate": 2.9459621563545825e-05, "loss": 0.2236, "step": 1850 }, { "epoch": 1.5048543689320388, "grad_norm": 0.19252241914699963, "learning_rate": 2.9227841799190775e-05, "loss": 0.2182, "step": 1860 }, { "epoch": 1.51294498381877, "grad_norm": 0.28670687006718837, "learning_rate": 2.8995687318304975e-05, "loss": 0.2224, "step": 1870 }, { "epoch": 1.5210355987055015, "grad_norm": 0.19972242558140968, "learning_rate": 2.8763178696899995e-05, "loss": 0.219, "step": 1880 }, { "epoch": 1.529126213592233, "grad_norm": 0.21361706883180026, "learning_rate": 2.853033654237507e-05, "loss": 0.2207, "step": 1890 }, { "epoch": 1.5372168284789645, "grad_norm": 0.22498931655736576, "learning_rate": 2.8297181491690756e-05, "loss": 0.2191, "step": 1900 }, { "epoch": 1.545307443365696, "grad_norm": 0.23189215817184283, "learning_rate": 2.8063734209539773e-05, "loss": 0.2192, "step": 1910 }, { "epoch": 1.5533980582524272, "grad_norm": 0.22214336417666555, "learning_rate": 2.783001538651554e-05, "loss": 0.2196, "step": 1920 }, { "epoch": 1.5614886731391586, "grad_norm": 0.22032252634525124, "learning_rate": 2.7596045737278336e-05, "loss": 0.2202, "step": 1930 }, { "epoch": 1.5695792880258899, "grad_norm": 0.19861461201811947, "learning_rate": 2.7361845998719315e-05, "loss": 0.2224, "step": 1940 }, { "epoch": 1.5776699029126213, "grad_norm": 0.20017500987570222, "learning_rate": 2.7127436928122612e-05, "loss": 0.2196, "step": 1950 }, { "epoch": 1.5857605177993528, "grad_norm": 0.2065059828986781, "learning_rate": 2.6892839301325623e-05, "loss": 0.2134, "step": 1960 }, { "epoch": 1.5938511326860842, "grad_norm": 0.20590200374163903, "learning_rate": 2.6658073910877603e-05, "loss": 0.2167, "step": 1970 }, { "epoch": 1.6019417475728155, "grad_norm": 0.1855510838919491, "learning_rate": 2.6423161564196803e-05, "loss": 0.2161, "step": 1980 }, { "epoch": 1.610032362459547, "grad_norm": 0.23296838329760938, "learning_rate": 2.6188123081726306e-05, "loss": 0.2174, "step": 1990 }, { "epoch": 1.6181229773462782, "grad_norm": 0.21045668674547632, "learning_rate": 2.5952979295088714e-05, "loss": 0.2149, "step": 2000 }, { "epoch": 1.6262135922330097, "grad_norm": 0.19447307206087575, "learning_rate": 2.57177510452398e-05, "loss": 0.2148, "step": 2010 }, { "epoch": 1.6343042071197411, "grad_norm": 0.23090130774168546, "learning_rate": 2.5482459180621377e-05, "loss": 0.2191, "step": 2020 }, { "epoch": 1.6423948220064726, "grad_norm": 0.18913378090947097, "learning_rate": 2.524712455531347e-05, "loss": 0.2136, "step": 2030 }, { "epoch": 1.650485436893204, "grad_norm": 0.20240602934546922, "learning_rate": 2.501176802718599e-05, "loss": 0.2153, "step": 2040 }, { "epoch": 1.6585760517799353, "grad_norm": 0.18200383138065016, "learning_rate": 2.4776410456050165e-05, "loss": 0.2149, "step": 2050 }, { "epoch": 1.6666666666666665, "grad_norm": 0.1919414220183083, "learning_rate": 2.4541072701809624e-05, "loss": 0.2187, "step": 2060 }, { "epoch": 1.674757281553398, "grad_norm": 0.194424204771118, "learning_rate": 2.4305775622611627e-05, "loss": 0.2156, "step": 2070 }, { "epoch": 1.6828478964401294, "grad_norm": 0.1986457800050313, "learning_rate": 2.4070540072998372e-05, "loss": 0.2186, "step": 2080 }, { "epoch": 1.690938511326861, "grad_norm": 0.20702148931842815, "learning_rate": 2.3835386902058637e-05, "loss": 0.2145, "step": 2090 }, { "epoch": 1.6990291262135924, "grad_norm": 0.1933456613050236, "learning_rate": 2.360033695157995e-05, "loss": 0.213, "step": 2100 }, { "epoch": 1.7071197411003236, "grad_norm": 0.1902051798469252, "learning_rate": 2.3365411054201315e-05, "loss": 0.2154, "step": 2110 }, { "epoch": 1.715210355987055, "grad_norm": 0.2128582736922773, "learning_rate": 2.3130630031566818e-05, "loss": 0.2209, "step": 2120 }, { "epoch": 1.7233009708737863, "grad_norm": 0.16887432451943601, "learning_rate": 2.2896014692480226e-05, "loss": 0.2156, "step": 2130 }, { "epoch": 1.7313915857605178, "grad_norm": 0.20654290160014185, "learning_rate": 2.266158583106063e-05, "loss": 0.2156, "step": 2140 }, { "epoch": 1.7394822006472492, "grad_norm": 0.1889181780602885, "learning_rate": 2.2427364224899502e-05, "loss": 0.2152, "step": 2150 }, { "epoch": 1.7475728155339807, "grad_norm": 0.20357991974988612, "learning_rate": 2.2193370633219115e-05, "loss": 0.2137, "step": 2160 }, { "epoch": 1.755663430420712, "grad_norm": 0.18958212912498537, "learning_rate": 2.1959625795032664e-05, "loss": 0.2113, "step": 2170 }, { "epoch": 1.7637540453074434, "grad_norm": 0.18671823093768689, "learning_rate": 2.1726150427306182e-05, "loss": 0.2146, "step": 2180 }, { "epoch": 1.7718446601941746, "grad_norm": 0.18915530672104663, "learning_rate": 2.1492965223122305e-05, "loss": 0.2132, "step": 2190 }, { "epoch": 1.779935275080906, "grad_norm": 0.23999090664815534, "learning_rate": 2.126009084984629e-05, "loss": 0.2124, "step": 2200 }, { "epoch": 1.7880258899676376, "grad_norm": 0.1958651936225041, "learning_rate": 2.102754794729426e-05, "loss": 0.2151, "step": 2210 }, { "epoch": 1.796116504854369, "grad_norm": 0.1977750621585522, "learning_rate": 2.079535712590382e-05, "loss": 0.2107, "step": 2220 }, { "epoch": 1.8042071197411005, "grad_norm": 0.17805905963116433, "learning_rate": 2.056353896490742e-05, "loss": 0.2091, "step": 2230 }, { "epoch": 1.8122977346278317, "grad_norm": 0.17973603360437174, "learning_rate": 2.0332114010508334e-05, "loss": 0.2114, "step": 2240 }, { "epoch": 1.820388349514563, "grad_norm": 0.18242429802007595, "learning_rate": 2.010110277405966e-05, "loss": 0.213, "step": 2250 }, { "epoch": 1.8284789644012944, "grad_norm": 0.18356014531964865, "learning_rate": 1.9870525730246424e-05, "loss": 0.2142, "step": 2260 }, { "epoch": 1.8365695792880259, "grad_norm": 0.18174264807445242, "learning_rate": 1.9640403315270824e-05, "loss": 0.212, "step": 2270 }, { "epoch": 1.8446601941747574, "grad_norm": 0.19024588215963859, "learning_rate": 1.9410755925041006e-05, "loss": 0.2107, "step": 2280 }, { "epoch": 1.8527508090614888, "grad_norm": 0.19119554332148833, "learning_rate": 1.918160391336335e-05, "loss": 0.2117, "step": 2290 }, { "epoch": 1.86084142394822, "grad_norm": 0.19462064918607552, "learning_rate": 1.8952967590138472e-05, "loss": 0.2099, "step": 2300 }, { "epoch": 1.8689320388349513, "grad_norm": 0.1800377494816668, "learning_rate": 1.8724867219561203e-05, "loss": 0.2154, "step": 2310 }, { "epoch": 1.8770226537216828, "grad_norm": 0.20601322758987628, "learning_rate": 1.8497323018324476e-05, "loss": 0.2145, "step": 2320 }, { "epoch": 1.8851132686084142, "grad_norm": 0.1930709584493409, "learning_rate": 1.8270355153827598e-05, "loss": 0.2077, "step": 2330 }, { "epoch": 1.8932038834951457, "grad_norm": 0.1911370351967725, "learning_rate": 1.804398374238872e-05, "loss": 0.2102, "step": 2340 }, { "epoch": 1.9012944983818771, "grad_norm": 0.18726153065650458, "learning_rate": 1.781822884746196e-05, "loss": 0.2098, "step": 2350 }, { "epoch": 1.9093851132686084, "grad_norm": 0.1786359702194776, "learning_rate": 1.7593110477859153e-05, "loss": 0.2114, "step": 2360 }, { "epoch": 1.9174757281553398, "grad_norm": 0.17464418924684114, "learning_rate": 1.736864858597645e-05, "loss": 0.2125, "step": 2370 }, { "epoch": 1.925566343042071, "grad_norm": 0.19553554568438702, "learning_rate": 1.7144863066025955e-05, "loss": 0.2101, "step": 2380 }, { "epoch": 1.9336569579288025, "grad_norm": 0.1919410672013163, "learning_rate": 1.692177375227242e-05, "loss": 0.2148, "step": 2390 }, { "epoch": 1.941747572815534, "grad_norm": 0.19222701771364425, "learning_rate": 1.669940041727538e-05, "loss": 0.2135, "step": 2400 }, { "epoch": 1.9498381877022655, "grad_norm": 0.19090816514102554, "learning_rate": 1.6477762770136707e-05, "loss": 0.2107, "step": 2410 }, { "epoch": 1.9579288025889967, "grad_norm": 0.196303834522943, "learning_rate": 1.625688045475371e-05, "loss": 0.2154, "step": 2420 }, { "epoch": 1.9660194174757282, "grad_norm": 0.2021430652987766, "learning_rate": 1.603677304807815e-05, "loss": 0.2085, "step": 2430 }, { "epoch": 1.9741100323624594, "grad_norm": 0.19099797924736667, "learning_rate": 1.5817460058381088e-05, "loss": 0.2089, "step": 2440 }, { "epoch": 1.9822006472491909, "grad_norm": 0.19913570866606684, "learning_rate": 1.5598960923523842e-05, "loss": 0.2088, "step": 2450 }, { "epoch": 1.9902912621359223, "grad_norm": 0.18039032373193892, "learning_rate": 1.5381295009235262e-05, "loss": 0.2033, "step": 2460 }, { "epoch": 1.9983818770226538, "grad_norm": 0.17656611003487743, "learning_rate": 1.5164481607395238e-05, "loss": 0.2073, "step": 2470 }, { "epoch": 2.0064724919093853, "grad_norm": 0.2203078879292808, "learning_rate": 1.4948539934324923e-05, "loss": 0.184, "step": 2480 }, { "epoch": 2.0145631067961167, "grad_norm": 0.19377235611735016, "learning_rate": 1.4733489129083534e-05, "loss": 0.1729, "step": 2490 }, { "epoch": 2.0226537216828477, "grad_norm": 0.19353245562742485, "learning_rate": 1.4519348251772058e-05, "loss": 0.1732, "step": 2500 }, { "epoch": 2.030744336569579, "grad_norm": 0.19647629833009353, "learning_rate": 1.4306136281843962e-05, "loss": 0.1712, "step": 2510 }, { "epoch": 2.0388349514563107, "grad_norm": 0.16744835404717143, "learning_rate": 1.4093872116422979e-05, "loss": 0.1703, "step": 2520 }, { "epoch": 2.046925566343042, "grad_norm": 0.17703272841109144, "learning_rate": 1.3882574568628315e-05, "loss": 0.1706, "step": 2530 }, { "epoch": 2.0550161812297736, "grad_norm": 0.2024362245821369, "learning_rate": 1.3672262365907163e-05, "loss": 0.1695, "step": 2540 }, { "epoch": 2.063106796116505, "grad_norm": 0.1767034203926602, "learning_rate": 1.3462954148374899e-05, "loss": 0.1701, "step": 2550 }, { "epoch": 2.071197411003236, "grad_norm": 0.18197977986545294, "learning_rate": 1.3254668467163029e-05, "loss": 0.1722, "step": 2560 }, { "epoch": 2.0792880258899675, "grad_norm": 0.17760389056330828, "learning_rate": 1.3047423782774937e-05, "loss": 0.1696, "step": 2570 }, { "epoch": 2.087378640776699, "grad_norm": 0.19203143690929964, "learning_rate": 1.2841238463449743e-05, "loss": 0.1712, "step": 2580 }, { "epoch": 2.0954692556634305, "grad_norm": 0.17183295745816207, "learning_rate": 1.2636130783534319e-05, "loss": 0.1711, "step": 2590 }, { "epoch": 2.103559870550162, "grad_norm": 0.171911071755841, "learning_rate": 1.2432118921863604e-05, "loss": 0.1661, "step": 2600 }, { "epoch": 2.1116504854368934, "grad_norm": 0.16535945547449518, "learning_rate": 1.2229220960149431e-05, "loss": 0.1721, "step": 2610 }, { "epoch": 2.1197411003236244, "grad_norm": 0.15768148348111047, "learning_rate": 1.2027454881377889e-05, "loss": 0.1681, "step": 2620 }, { "epoch": 2.127831715210356, "grad_norm": 0.16958702285946278, "learning_rate": 1.1826838568215526e-05, "loss": 0.1693, "step": 2630 }, { "epoch": 2.1359223300970873, "grad_norm": 0.16417738734331427, "learning_rate": 1.1627389801424351e-05, "loss": 0.1674, "step": 2640 }, { "epoch": 2.144012944983819, "grad_norm": 0.1875141618399819, "learning_rate": 1.1429126258285946e-05, "loss": 0.1743, "step": 2650 }, { "epoch": 2.1521035598705502, "grad_norm": 0.1840991587252681, "learning_rate": 1.1232065511034696e-05, "loss": 0.1749, "step": 2660 }, { "epoch": 2.1601941747572817, "grad_norm": 0.18075731721273597, "learning_rate": 1.1036225025300357e-05, "loss": 0.1758, "step": 2670 }, { "epoch": 2.168284789644013, "grad_norm": 0.1751267297742484, "learning_rate": 1.0841622158560085e-05, "loss": 0.1762, "step": 2680 }, { "epoch": 2.176375404530744, "grad_norm": 0.17441997867016168, "learning_rate": 1.0648274158599994e-05, "loss": 0.1747, "step": 2690 }, { "epoch": 2.1844660194174756, "grad_norm": 0.16061544075925469, "learning_rate": 1.0456198161986489e-05, "loss": 0.171, "step": 2700 }, { "epoch": 2.192556634304207, "grad_norm": 0.17229966007961503, "learning_rate": 1.0265411192547462e-05, "loss": 0.1712, "step": 2710 }, { "epoch": 2.2006472491909386, "grad_norm": 0.17576497117173268, "learning_rate": 1.0075930159863416e-05, "loss": 0.1713, "step": 2720 }, { "epoch": 2.20873786407767, "grad_norm": 0.17654170464263277, "learning_rate": 9.887771857768796e-06, "loss": 0.1716, "step": 2730 }, { "epoch": 2.2168284789644015, "grad_norm": 0.15949424974193477, "learning_rate": 9.700952962863513e-06, "loss": 0.1724, "step": 2740 }, { "epoch": 2.2249190938511325, "grad_norm": 0.1797493219887846, "learning_rate": 9.515490033034893e-06, "loss": 0.1747, "step": 2750 }, { "epoch": 2.233009708737864, "grad_norm": 0.1640136227884059, "learning_rate": 9.331399505990168e-06, "loss": 0.17, "step": 2760 }, { "epoch": 2.2411003236245954, "grad_norm": 0.1761964552588544, "learning_rate": 9.148697697799533e-06, "loss": 0.1744, "step": 2770 }, { "epoch": 2.249190938511327, "grad_norm": 0.16581441786521817, "learning_rate": 8.967400801450105e-06, "loss": 0.1683, "step": 2780 }, { "epoch": 2.2572815533980584, "grad_norm": 0.15986397423733795, "learning_rate": 8.787524885410678e-06, "loss": 0.1684, "step": 2790 }, { "epoch": 2.26537216828479, "grad_norm": 0.16854533205632355, "learning_rate": 8.60908589220758e-06, "loss": 0.1695, "step": 2800 }, { "epoch": 2.273462783171521, "grad_norm": 0.16192651639080963, "learning_rate": 8.432099637011693e-06, "loss": 0.1686, "step": 2810 }, { "epoch": 2.2815533980582523, "grad_norm": 0.15876123318459937, "learning_rate": 8.256581806236704e-06, "loss": 0.1727, "step": 2820 }, { "epoch": 2.2896440129449838, "grad_norm": 0.15953467663865786, "learning_rate": 8.082547956148873e-06, "loss": 0.1702, "step": 2830 }, { "epoch": 2.2977346278317152, "grad_norm": 0.15986140875482005, "learning_rate": 7.91001351148819e-06, "loss": 0.1688, "step": 2840 }, { "epoch": 2.3058252427184467, "grad_norm": 0.16937360962913361, "learning_rate": 7.738993764101324e-06, "loss": 0.1687, "step": 2850 }, { "epoch": 2.313915857605178, "grad_norm": 0.1624566036577771, "learning_rate": 7.569503871586292e-06, "loss": 0.1709, "step": 2860 }, { "epoch": 2.3220064724919096, "grad_norm": 0.17174966186956542, "learning_rate": 7.401558855949004e-06, "loss": 0.168, "step": 2870 }, { "epoch": 2.3300970873786406, "grad_norm": 0.1619257670204373, "learning_rate": 7.235173602271875e-06, "loss": 0.1697, "step": 2880 }, { "epoch": 2.338187702265372, "grad_norm": 0.1659355702344565, "learning_rate": 7.070362857394538e-06, "loss": 0.1712, "step": 2890 }, { "epoch": 2.3462783171521036, "grad_norm": 0.16020534539972092, "learning_rate": 6.907141228606831e-06, "loss": 0.165, "step": 2900 }, { "epoch": 2.354368932038835, "grad_norm": 0.16872762980492875, "learning_rate": 6.745523182354147e-06, "loss": 0.1717, "step": 2910 }, { "epoch": 2.3624595469255665, "grad_norm": 0.15648717008884194, "learning_rate": 6.585523042955233e-06, "loss": 0.1668, "step": 2920 }, { "epoch": 2.3705501618122975, "grad_norm": 0.16263064803747318, "learning_rate": 6.427154991332665e-06, "loss": 0.1721, "step": 2930 }, { "epoch": 2.378640776699029, "grad_norm": 0.15947499101323748, "learning_rate": 6.2704330637559315e-06, "loss": 0.1709, "step": 2940 }, { "epoch": 2.3867313915857604, "grad_norm": 0.16083163188925498, "learning_rate": 6.115371150597413e-06, "loss": 0.1702, "step": 2950 }, { "epoch": 2.394822006472492, "grad_norm": 0.16542590461200707, "learning_rate": 5.961982995101301e-06, "loss": 0.1729, "step": 2960 }, { "epoch": 2.4029126213592233, "grad_norm": 0.16420989704643807, "learning_rate": 5.810282192165442e-06, "loss": 0.1722, "step": 2970 }, { "epoch": 2.411003236245955, "grad_norm": 0.1551104913079497, "learning_rate": 5.660282187136507e-06, "loss": 0.1689, "step": 2980 }, { "epoch": 2.4190938511326863, "grad_norm": 0.15940095271603197, "learning_rate": 5.511996274618253e-06, "loss": 0.1702, "step": 2990 }, { "epoch": 2.4271844660194173, "grad_norm": 0.15033675616819664, "learning_rate": 5.365437597293238e-06, "loss": 0.1702, "step": 3000 }, { "epoch": 2.4352750809061487, "grad_norm": 0.15579865123668477, "learning_rate": 5.220619144757996e-06, "loss": 0.1693, "step": 3010 }, { "epoch": 2.44336569579288, "grad_norm": 0.16013909939251375, "learning_rate": 5.077553752371708e-06, "loss": 0.1678, "step": 3020 }, { "epoch": 2.4514563106796117, "grad_norm": 0.16673936398632458, "learning_rate": 4.936254100118656e-06, "loss": 0.1709, "step": 3030 }, { "epoch": 2.459546925566343, "grad_norm": 0.1605515108220787, "learning_rate": 4.796732711484342e-06, "loss": 0.1658, "step": 3040 }, { "epoch": 2.4676375404530746, "grad_norm": 0.1585644224135113, "learning_rate": 4.659001952345538e-06, "loss": 0.1682, "step": 3050 }, { "epoch": 2.475728155339806, "grad_norm": 0.15466709640184034, "learning_rate": 4.523074029874291e-06, "loss": 0.1728, "step": 3060 }, { "epoch": 2.483818770226537, "grad_norm": 0.15343018591614654, "learning_rate": 4.388960991455998e-06, "loss": 0.1692, "step": 3070 }, { "epoch": 2.4919093851132685, "grad_norm": 0.14923408281267303, "learning_rate": 4.256674723621621e-06, "loss": 0.1706, "step": 3080 }, { "epoch": 2.5, "grad_norm": 0.16015551238015277, "learning_rate": 4.126226950994211e-06, "loss": 0.1708, "step": 3090 }, { "epoch": 2.5080906148867315, "grad_norm": 0.15550577660081535, "learning_rate": 3.997629235249692e-06, "loss": 0.1737, "step": 3100 }, { "epoch": 2.516181229773463, "grad_norm": 0.16405843460813677, "learning_rate": 3.870892974092197e-06, "loss": 0.1677, "step": 3110 }, { "epoch": 2.524271844660194, "grad_norm": 0.15592864442189205, "learning_rate": 3.7460294002438444e-06, "loss": 0.1676, "step": 3120 }, { "epoch": 2.5323624595469254, "grad_norm": 0.15685374774957278, "learning_rate": 3.6230495804491864e-06, "loss": 0.174, "step": 3130 }, { "epoch": 2.540453074433657, "grad_norm": 0.15046469900932222, "learning_rate": 3.5019644144943576e-06, "loss": 0.1723, "step": 3140 }, { "epoch": 2.5485436893203883, "grad_norm": 0.15187030014871522, "learning_rate": 3.382784634241015e-06, "loss": 0.1679, "step": 3150 }, { "epoch": 2.55663430420712, "grad_norm": 0.15370723807151057, "learning_rate": 3.2655208026751816e-06, "loss": 0.1707, "step": 3160 }, { "epoch": 2.5647249190938513, "grad_norm": 0.15358348991682996, "learning_rate": 3.150183312971014e-06, "loss": 0.164, "step": 3170 }, { "epoch": 2.5728155339805827, "grad_norm": 0.15495569068856682, "learning_rate": 3.036782387569659e-06, "loss": 0.1676, "step": 3180 }, { "epoch": 2.5809061488673137, "grad_norm": 0.15152798447783355, "learning_rate": 2.9253280772732595e-06, "loss": 0.1668, "step": 3190 }, { "epoch": 2.588996763754045, "grad_norm": 0.14795764100233588, "learning_rate": 2.8158302603540965e-06, "loss": 0.1677, "step": 3200 }, { "epoch": 2.5970873786407767, "grad_norm": 0.1462435066429316, "learning_rate": 2.708298641679105e-06, "loss": 0.1634, "step": 3210 }, { "epoch": 2.605177993527508, "grad_norm": 0.15564615414583508, "learning_rate": 2.6027427518497153e-06, "loss": 0.1675, "step": 3220 }, { "epoch": 2.6132686084142396, "grad_norm": 0.15073879419059655, "learning_rate": 2.49917194635714e-06, "loss": 0.1652, "step": 3230 }, { "epoch": 2.6213592233009706, "grad_norm": 0.1507824582262081, "learning_rate": 2.397595404753225e-06, "loss": 0.1708, "step": 3240 }, { "epoch": 2.6294498381877025, "grad_norm": 0.14447199827031953, "learning_rate": 2.2980221298367995e-06, "loss": 0.1718, "step": 3250 }, { "epoch": 2.6375404530744335, "grad_norm": 0.15310920163624914, "learning_rate": 2.2004609468558175e-06, "loss": 0.1687, "step": 3260 }, { "epoch": 2.645631067961165, "grad_norm": 0.1552961190816156, "learning_rate": 2.1049205027251216e-06, "loss": 0.168, "step": 3270 }, { "epoch": 2.6537216828478964, "grad_norm": 0.1541563267360428, "learning_rate": 2.0114092652600806e-06, "loss": 0.1682, "step": 3280 }, { "epoch": 2.661812297734628, "grad_norm": 0.1531794971521466, "learning_rate": 1.919935522426081e-06, "loss": 0.1688, "step": 3290 }, { "epoch": 2.6699029126213594, "grad_norm": 0.1502769010065667, "learning_rate": 1.8305073816039492e-06, "loss": 0.17, "step": 3300 }, { "epoch": 2.6779935275080904, "grad_norm": 0.14759059904949057, "learning_rate": 1.7431327688714139e-06, "loss": 0.1666, "step": 3310 }, { "epoch": 2.686084142394822, "grad_norm": 0.14884028557553985, "learning_rate": 1.6578194283005804e-06, "loss": 0.1684, "step": 3320 }, { "epoch": 2.6941747572815533, "grad_norm": 0.15011522863824736, "learning_rate": 1.5745749212715794e-06, "loss": 0.1679, "step": 3330 }, { "epoch": 2.7022653721682848, "grad_norm": 0.15248852381104064, "learning_rate": 1.4934066258024182e-06, "loss": 0.1741, "step": 3340 }, { "epoch": 2.7103559870550162, "grad_norm": 0.14444830557078114, "learning_rate": 1.4143217358950217e-06, "loss": 0.168, "step": 3350 }, { "epoch": 2.7184466019417477, "grad_norm": 0.16097353621052152, "learning_rate": 1.3373272608976668e-06, "loss": 0.1722, "step": 3360 }, { "epoch": 2.726537216828479, "grad_norm": 0.15190902224222305, "learning_rate": 1.2624300248836928e-06, "loss": 0.1696, "step": 3370 }, { "epoch": 2.73462783171521, "grad_norm": 0.15777837628961622, "learning_rate": 1.1896366660467173e-06, "loss": 0.168, "step": 3380 }, { "epoch": 2.7427184466019416, "grad_norm": 0.14537830984259653, "learning_rate": 1.1189536361122799e-06, "loss": 0.1665, "step": 3390 }, { "epoch": 2.750809061488673, "grad_norm": 0.14842465745819863, "learning_rate": 1.0503871997660036e-06, "loss": 0.1623, "step": 3400 }, { "epoch": 2.7588996763754046, "grad_norm": 0.16097421255088173, "learning_rate": 9.83943434098372e-07, "loss": 0.166, "step": 3410 }, { "epoch": 2.766990291262136, "grad_norm": 0.1486212209344348, "learning_rate": 9.196282280661023e-07, "loss": 0.1688, "step": 3420 }, { "epoch": 2.775080906148867, "grad_norm": 0.14198772068734017, "learning_rate": 8.574472819702029e-07, "loss": 0.1652, "step": 3430 }, { "epoch": 2.783171521035599, "grad_norm": 0.1483955979101031, "learning_rate": 7.974061069507571e-07, "loss": 0.1679, "step": 3440 }, { "epoch": 2.79126213592233, "grad_norm": 0.1510691708695613, "learning_rate": 7.395100244984604e-07, "loss": 0.1676, "step": 3450 }, { "epoch": 2.7993527508090614, "grad_norm": 0.14675861790646366, "learning_rate": 6.837641659829807e-07, "loss": 0.1662, "step": 3460 }, { "epoch": 2.807443365695793, "grad_norm": 0.1444417747058221, "learning_rate": 6.301734721981533e-07, "loss": 0.1669, "step": 3470 }, { "epoch": 2.8155339805825244, "grad_norm": 0.14445462788403754, "learning_rate": 5.787426929240808e-07, "loss": 0.166, "step": 3480 }, { "epoch": 2.823624595469256, "grad_norm": 0.1614165649662563, "learning_rate": 5.294763865061558e-07, "loss": 0.1683, "step": 3490 }, { "epoch": 2.831715210355987, "grad_norm": 0.15711301471816222, "learning_rate": 4.823789194510514e-07, "loss": 0.1706, "step": 3500 }, { "epoch": 2.8398058252427183, "grad_norm": 0.1453400612300062, "learning_rate": 4.3745446603971064e-07, "loss": 0.1688, "step": 3510 }, { "epoch": 2.8478964401294498, "grad_norm": 0.15333438971393676, "learning_rate": 3.947070079573872e-07, "loss": 0.1638, "step": 3520 }, { "epoch": 2.855987055016181, "grad_norm": 0.1472420355366777, "learning_rate": 3.541403339407279e-07, "loss": 0.171, "step": 3530 }, { "epoch": 2.8640776699029127, "grad_norm": 0.1486881057392244, "learning_rate": 3.1575803944199624e-07, "loss": 0.1698, "step": 3540 }, { "epoch": 2.872168284789644, "grad_norm": 0.140283471672274, "learning_rate": 2.7956352631038906e-07, "loss": 0.1659, "step": 3550 }, { "epoch": 2.8802588996763756, "grad_norm": 0.14768247651236527, "learning_rate": 2.4556000249054133e-07, "loss": 0.1705, "step": 3560 }, { "epoch": 2.8883495145631066, "grad_norm": 0.14227761591758845, "learning_rate": 2.1375048173818412e-07, "loss": 0.1676, "step": 3570 }, { "epoch": 2.896440129449838, "grad_norm": 0.1397094378867443, "learning_rate": 1.8413778335305e-07, "loss": 0.1639, "step": 3580 }, { "epoch": 2.9045307443365695, "grad_norm": 0.14544808567516948, "learning_rate": 1.567245319290006e-07, "loss": 0.169, "step": 3590 }, { "epoch": 2.912621359223301, "grad_norm": 0.14376315892226174, "learning_rate": 1.315131571213879e-07, "loss": 0.163, "step": 3600 }, { "epoch": 2.9207119741100325, "grad_norm": 0.14749344179261822, "learning_rate": 1.0850589343172624e-07, "loss": 0.1669, "step": 3610 }, { "epoch": 2.9288025889967635, "grad_norm": 0.1452354644803129, "learning_rate": 8.770478000964532e-08, "loss": 0.1684, "step": 3620 }, { "epoch": 2.9368932038834954, "grad_norm": 0.14097833660885037, "learning_rate": 6.911166047215145e-08, "loss": 0.1694, "step": 3630 }, { "epoch": 2.9449838187702264, "grad_norm": 0.14528957043064877, "learning_rate": 5.272818274023872e-08, "loss": 0.1675, "step": 3640 }, { "epoch": 2.953074433656958, "grad_norm": 0.14549261409943784, "learning_rate": 3.855579889282257e-08, "loss": 0.1679, "step": 3650 }, { "epoch": 2.9611650485436893, "grad_norm": 0.14557934166615819, "learning_rate": 2.6595765038045507e-08, "loss": 0.1677, "step": 3660 }, { "epoch": 2.969255663430421, "grad_norm": 0.1445108267077229, "learning_rate": 1.6849141201946693e-08, "loss": 0.165, "step": 3670 }, { "epoch": 2.9773462783171523, "grad_norm": 0.1425028501060893, "learning_rate": 9.316791234506572e-09, "loss": 0.1713, "step": 3680 }, { "epoch": 2.9854368932038833, "grad_norm": 0.13458537716318286, "learning_rate": 3.999382733096968e-09, "loss": 0.1665, "step": 3690 }, { "epoch": 2.9935275080906147, "grad_norm": 0.13982192859442388, "learning_rate": 8.973869832895609e-10, "loss": 0.1686, "step": 3700 }, { "epoch": 3.0, "step": 3708, "total_flos": 7.26626862104576e+18, "train_loss": 0.24862830043482034, "train_runtime": 257107.0797, "train_samples_per_second": 0.461, "train_steps_per_second": 0.014 } ], "logging_steps": 10, "max_steps": 3708, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 7.26626862104576e+18, "train_batch_size": 2, "trial_name": null, "trial_params": null }
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 3708,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.008090614886731391,
"grad_norm": 22.52252141115457,
"learning_rate": 1.2129380053908356e-06,
"loss": 2.5582,
"step": 10
},
{
"epoch": 0.016181229773462782,
"grad_norm": 11.441648898342923,
"learning_rate": 2.560646900269542e-06,
"loss": 2.2025,
"step": 20
},
{
"epoch": 0.024271844660194174,
"grad_norm": 5.370882200215843,
"learning_rate": 3.908355795148248e-06,
"loss": 1.668,
"step": 30
},
{
"epoch": 0.032362459546925564,
"grad_norm": 2.0115313434456827,
"learning_rate": 5.256064690026954e-06,
"loss": 1.1266,
"step": 40
},
{
"epoch": 0.040453074433656956,
"grad_norm": 1.2036774860036918,
"learning_rate": 6.60377358490566e-06,
"loss": 0.7799,
"step": 50
},
{
"epoch": 0.04854368932038835,
"grad_norm": 0.8106871294631399,
"learning_rate": 7.951482479784367e-06,
"loss": 0.5907,
"step": 60
},
{
"epoch": 0.05663430420711974,
"grad_norm": 0.6572037767143685,
"learning_rate": 9.299191374663074e-06,
"loss": 0.4977,
"step": 70
},
{
"epoch": 0.06472491909385113,
"grad_norm": 0.9743925361281945,
"learning_rate": 1.0646900269541779e-05,
"loss": 0.4728,
"step": 80
},
{
"epoch": 0.07281553398058252,
"grad_norm": 0.5302813737105591,
"learning_rate": 1.1994609164420486e-05,
"loss": 0.446,
"step": 90
},
{
"epoch": 0.08090614886731391,
"grad_norm": 0.5504892707748702,
"learning_rate": 1.3342318059299191e-05,
"loss": 0.4164,
"step": 100
},
{
"epoch": 0.0889967637540453,
"grad_norm": 0.46842577857609835,
"learning_rate": 1.4690026954177898e-05,
"loss": 0.412,
"step": 110
},
{
"epoch": 0.0970873786407767,
"grad_norm": 0.5228315101871648,
"learning_rate": 1.6037735849056604e-05,
"loss": 0.3975,
"step": 120
},
{
"epoch": 0.10517799352750809,
"grad_norm": 0.5623706583567197,
"learning_rate": 1.738544474393531e-05,
"loss": 0.3857,