[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"summary-02595583f76174e1-building-a-fashionmnist-classifier-with-jax-and-fl-summary":3,"summaries-facets-categories":206,"summary-related-02595583f76174e1-building-a-fashionmnist-classifier-with-jax-and-fl-summary":4085},{"id":4,"title":5,"ai":6,"body":13,"categories":154,"created_at":156,"date_modified":156,"description":148,"extension":157,"faq":156,"featured":158,"kicker_label":156,"meta":159,"navigation":187,"path":188,"published_at":189,"question":156,"scraped_at":190,"seo":191,"sitemap":192,"source_id":193,"source_name":194,"source_type":195,"source_url":196,"stem":197,"tags":198,"thumbnail_url":156,"tldr":203,"tweet":156,"unknown_tags":204,"__hash__":205},"summaries\u002Fsummaries\u002F02595583f76174e1-building-a-fashionmnist-classifier-with-jax-and-fl-summary.md","Building a FashionMNIST Classifier with JAX and Flax",{"provider":7,"model":8,"input_tokens":9,"output_tokens":10,"processing_time_ms":11,"cost_usd":12},"openrouter","google\u002Fgemini-3.1-flash-lite",8214,907,7302,0.003414,{"type":14,"value":15,"toc":147},"minimark",[16,21,25,81,85,100,104,107],[17,18,20],"h2",{"id":19},"the-functional-paradigm-of-jax-and-flax","The Functional Paradigm of JAX and Flax",[22,23,24],"p",{},"Unlike PyTorch or TensorFlow, JAX treats models as pure functions rather than stateful objects. This requires a shift in how developers handle randomness and model state.",[26,27,28,45,67],"ul",{},[29,30,31,35,36,40,41,44],"li",{},[32,33,34],"strong",{},"Explicit PRNG",": JAX eliminates global random states. Developers must pass a ",[37,38,39],"code",{},"PRNGKey"," to functions and split it using ",[37,42,43],{},"jax.random.split"," to generate independent, reproducible streams of randomness. This ensures that experiments are perfectly deterministic.",[29,46,47,50,51,54,55,58,59,62,63,66],{},[32,48,49],{},"Model State Management",": In Flax, parameters and optimizer states are stored in a plain Python data structure. The ",[37,52,53],{},"TrainState"," utility acts as a container for the model's ",[37,56,57],{},"apply_fn",", current ",[37,60,61],{},"params",", and the optimizer (",[37,64,65],{},"tx",").",[29,68,69,72,73,76,77,80],{},[32,70,71],{},"Defining Architectures",": Using the ",[37,74,75],{},"@nn.compact"," decorator allows for inline definition of sub-modules within the ",[37,78,79],{},"__call__"," method. This triggers parameter initialization on the first forward pass, keeping the code concise and readable.",[17,82,84],{"id":83},"bridging-pytorch-and-jax","Bridging PyTorch and JAX",[22,86,87,88,91,92,95,96,99],{},"While JAX handles computation, it lacks built-in dataset management. The author demonstrates a common, effective pattern: using ",[37,89,90],{},"torchvision"," for data downloading and preprocessing, while overriding the ",[37,93,94],{},"DataLoader","'s ",[37,97,98],{},"collate_fn"," to output NumPy arrays instead of PyTorch tensors. This allows JAX to consume the data directly without unnecessary overhead.",[17,101,103],{"id":102},"training-and-activation-functions","Training and Activation Functions",[22,105,106],{},"To evaluate performance, the author implements a multi-layer perceptron (MLP) on the FashionMNIST dataset, comparing six different activation functions: Sigmoid, Tanh, ReLU, LeakyReLU, ELU, and Swish.",[26,108,109,119,129],{},[29,110,111,114,115,118],{},[32,112,113],{},"Numerical Stability",": The model outputs raw logits rather than probabilities. This is standard practice because the cross-entropy loss function is numerically more stable when it handles ",[37,116,117],{},"log_softmax"," internally.",[29,120,121,124,125,128],{},[32,122,123],{},"Initialization",": The author uses ",[37,126,127],{},"lecun_uniform"," initialization to maintain parity with PyTorch defaults, which is critical for training stability in deep networks.",[29,130,131,134,135,138,139,142,143,146],{},[32,132,133],{},"Performance",": The training loop utilizes ",[37,136,137],{},"@jit"," (Just-In-Time compilation) to accelerate the ",[37,140,141],{},"train_step"," and ",[37,144,145],{},"eval_step"," functions, demonstrating how JAX achieves high performance through XLA compilation.",{"title":148,"searchDepth":149,"depth":149,"links":150},"",2,[151,152,153],{"id":19,"depth":149,"text":20},{"id":83,"depth":149,"text":84},{"id":102,"depth":149,"text":103},[155],"AI & LLMs",null,"md",false,{"content_references":160,"triage":182},[161,166,169,172,175,178],{"type":162,"title":163,"author":164,"context":165},"paper","Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)","Clevert et al.","recommended",{"type":162,"title":167,"author":168,"context":165},"Rectified Linear Units Improve Restricted Boltzmann Machines","Nair & Hinton",{"type":162,"title":170,"author":171,"context":165},"Understanding the difficulty of training deep feedforward neural networks","Glorot & Bengio",{"type":162,"title":173,"author":174,"context":165},"Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification","He et al.",{"type":162,"title":176,"author":177,"context":165},"Gaussian Error Linear Units (GELUs)","Hendrycks & Gimpel",{"type":179,"title":180,"url":181,"context":165},"other","UvA Deep Learning Course, Tutorial 3: Activation Functions","https:\u002F\u002Fuvadlc-notebooks.readthedocs.io\u002F",{"relevance":183,"novelty":183,"quality":184,"actionability":183,"composite":185,"reasoning":186},3,4,3.25,"Category: AI & LLMs. The article provides a practical guide to building a classifier using JAX and Flax, which is relevant for developers interested in AI engineering. It discusses specific techniques like using `TrainState` for parameter management and the impact of activation functions, but lacks a broader application to product building or deployment.",true,"\u002Fsummaries\u002F02595583f76174e1-building-a-fashionmnist-classifier-with-jax-and-fl-summary","2026-05-18 15:44:49","2026-05-18 19:00:33",{"title":5,"description":148},{"loc":188},"02595583f76174e1","Level Up Coding","article","https:\u002F\u002Flevelup.gitconnected.com\u002Fbuilding-the-fashionmnist-classifier-in-flax-64214b369408?source=rss----5517fd7b58a6---4","summaries\u002F02595583f76174e1-building-a-fashionmnist-classifier-with-jax-and-fl-summary",[199,200,201,202],"python","machine-learning","coding","ai-llms","A practical guide to building a multi-layer perceptron in JAX and Flax, highlighting the functional paradigm of JAX, the use of TrainState for parameter management, and the impact of different activation functions on model performance.",[202],"i41OIawMR8ioCSq7v_12B2fg8FVpdaBvOXPp_8ffHYo",[207,210,213,215,218,221,223,225,227,229,231,233,236,238,240,242,244,246,248,250,252,254,256,258,260,263,266,268,270,273,275,277,280,282,284,286,288,290,292,294,296,298,300,302,305,307,309,311,313,315,317,319,321,323,325,327,329,331,333,335,337,339,341,343,345,347,349,351,353,355,357,359,361,363,365,367,369,371,373,375,377,379,381,383,385,387,389,391,393,395,397,399,401,403,405,407,409,411,413,415,417,419,421,423,425,427,429,431,433,435,437,439,441,443,445,447,449,451,453,455,457,459,461,463,465,467,469,471,473,475,477,479,481,483,485,487,489,491,493,495,497,499,501,503,505,507,509,511,513,515,517,519,521,523,525,527,529,531,533,535,537,539,541,543,545,547,549,551,553,555,557,559,561,563,565,567,569,571,573,575,577,579,581,583,585,587,589,591,593,595,597,599,601,603,605,607,609,611,613,615,617,619,621,623,625,627,629,631,633,635,637,639,641,643,645,647,649,651,653,655,657,659,661,663,665,667,669,671,673,675,677,679,681,683,685,687,689,691,693,695,697,699,701,703,705,707,709,711,713,715,717,719,721,723,725,727,729,731,733,735,737,739,741,743,745,747,749,751,753,755,757,759,761,763,765,767,769,771,773,775,777,779,781,783,785,787,789,791,793,795,797,799,801,803,805,807,809,811,813,815,817,819,821,823,825,827,829,831,833,835,837,839,841,843,845,847,849,851,853,855,857,859,861,863,865,867,869,871,873,875,877,879,881,883,885,887,889,891,893,895,897,899,901,903,905,907,909,911,913,915,917,919,921,923,925,927,929,931,933,935,937,939,941,943,945,947,949,951,953,955,957,959,961,963,965,967,969,971,973,975,977,979,981,983,985,987,989,991,993,995,997,999,1001,1003,1005,1007,1009,1011,1013,1015,1017,1019,1021,1023,1025,1027,1029,1031,1033,1035,1037,1039,1041,1043,1045,1047,1049,1051,1053,1055,1057,1059,1061,1063,1065,1067,1069,1071,1073,1075,1077,1079,1081,1083,1085,1087,1089,1091,1093,1095,1097,1099,1101,1103,1105,1107,1109,1111,1113,1115,1117,1119,1121,1123,1125,1127,1129,1131,1133,1135,1137,1139,1141,1143,1145,1147,1149,1151,1153,1155,1157,1159,1161,1163,1165,1167,1169,1171,1173,1175,1177,1179,1181,1183,1185,1187,1189,1191,1193,1195,1197,1199,1201,1203,1205,1207,1209,1211,1213,1215,1217,1219,1221,1223,1225,1227,1229,1231,1233,1235,1237,1239,1241,1243,1245,1247,1249,1251,1253,1255,1257,1259,1261,1263,1265,1267,1269,1271,1273,1275,1277,1279,1281,1283,1285,1287,1289,1291,1293,1295,1297,1299,1301,1303,1305,1307,1309,1311,1313,1315,1317,1319,1321,1323,1325,1327,1329,1331,1333,1335,1337,1339,1341,1343,1345,1347,1349,1351,1353,1355,1357,1359,1361,1363,1365,1367,1369,1371,1373,1375,1377,1379,1381,1383,1385,1387,1389,1391,1393,1395,1397,1399,1401,1403,1405,1407,1409,1411,1413,1415,1417,1419,1421,1423,1425,1427,1429,1431,1433,1435,1437,1439,1441,1443,1445,1447,1449,1451,1453,1455,1457,1459,1461,1463,1465,1467,1469,1471,1473,1475,1477,1479,1481,1483,1485,1487,1489,1491,1493,1495,1497,1499,1501,1503,1505,1507,1509,1511,1513,1515,1517,1519,1521,1523,1525,1527,1529,1531,1533,1535,1537,1539,1541,1543,1545,1547,1549,1551,1553,1555,1557,1559,1561,1563,1565,1567,1569,1571,1573,1575,1577,1579,1581,1583,1585,1587,1589,1591,1593,1595,1597,1599,1601,1603,1605,1607,1609,1611,1613,1615,1617,1619,1621,1623,1625,1627,1629,1631,1633,1635,1637,1639,1641,1643,1645,1647,1649,1651,1653,1655,1657,1659,1661,1663,1665,1667,1669,1671,1673,1675,1677,1679,1681,1683,1685,1687,1689,1691,1693,1695,1697,1699,1701,1703,1705,1707,1709,1711,1713,1715,1717,1719,1721,1723,1725,1727,1729,1731,1733,1735,1737,1739,1741,1743,1745,1747,1749,1751,1753,1755,1757,1759,1761,1763,1765,1767,1769,1771,1773,1775,1777,1779,1781,1783,1785,1787,1789,1791,1793,1795,1797,1799,1801,1803,1805,1807,1809,1811,1813,1815,1817,1819,1821,1823,1825,1827,1829,1831,1833,1835,1837,1839,1841,1843,1845,1847,1849,1851,1853,1855,1857,1859,1861,1863,1865,1867,1869,1871,1873,1875,1877,1879,1881,1883,1885,1887,1889,1891,1893,1895,1897,1899,1901,1903,1905,1907,1909,1911,1913,1915,1917,1919,1921,1923,1925,1927,1929,1931,1933,1935,1937,1939,1941,1943,1945,1947,1949,1951,1953,1955,1957,1959,1961,1963,1965,1967,1969,1971,1973,1975,1977,1979,1981,1983,1985,1987,1989,1991,1993,1995,1997,1999,2001,2003,2005,2007,2009,2011,2013,2015,2017,2019,2021,2023,2025,2027,2029,2031,2033,2035,2037,2039,2041,2043,2045,2047,2049,2051,2053,2055,2057,2059,2061,2063,2065,2067,2069,2071,2073,2075,2077,2079,2081,2083,2085,2087,2089,2091,2093,2095,2097,2099,2101,2103,2105,2107,2109,2111,2113,2115,2117,2119,2121,2123,2125,2127,2129,2131,2133,2135,2137,2139,2141,2143,2145,2147,2149,2151,2153,2155,2157,2159,2161,2163,2165,2167,2169,2171,2173,2175,2177,2179,2181,2183,2185,2187,2189,2191,2193,2195,2197,2199,2201,2203,2205,2207,2209,2211,2213,2215,2217,2219,2221,2223,2225,2227,2229,2231,2233,2235,2237,2239,2241,2243,2245,2247,2249,2251,2253,2255,2257,2259,2261,2263,2265,2267,2269,2271,2273,2275,2277,2279,2281,2283,2285,2287,2289,2291,2293,2295,2297,2299,2301,2303,2305,2307,2309,2311,2313,2315,2317,2319,2321,2323,2325,2327,2329,2331,2333,2335,2337,2339,2341,2343,2345,2347,2349,2351,2353,2355,2357,2359,2361,2363,2365,2367,2369,2371,2373,2375,2377,2379,2381,2383,2385,2387,2389,2391,2393,2395,2397,2399,2401,2403,2405,2407,2409,2411,2413,2415,2417,2419,2421,2423,2425,2427,2429,2431,2433,2435,2437,2439,2441,2443,2445,2447,2449,2451,2453,2455,2457,2459,2461,2463,2465,2467,2469,2471,2473,2475,2477,2479,2481,2483,2485,2487,2489,2491,2493,2495,2497,2499,2501,2503,2505,2507,2509,2511,2513,2515,2517,2519,2521,2523,2525,2527,2529,2531,2533,2535,2537,2539,2541,2543,2545,2547,2549,2551,2553,2555,2557,2559,2561,2563,2565,2567,2569,2571,2573,2575,2577,2579,2581,2583,2585,2587,2589,2591,2593,2595,2597,2599,2601,2603,2605,2607,2609,2611,2613,2615,2617,2619,2621,2623,2625,2627,2629,2631,2633,2635,2637,2639,2641,2643,2645,2647,2649,2651,2653,2655,2657,2659,2661,2663,2665,2667,2669,2671,2673,2675,2677,2679,2681,2683,2685,2687,2689,2691,2693,2695,2697,2699,2701,2703,2705,2707,2709,2711,2713,2715,2717,2719,2721,2723,2725,2727,2729,2731,2733,2735,2737,2739,2741,2743,2745,2747,2749,2751,2753,2755,2757,2759,2761,2763,2765,2767,2769,2771,2773,2775,2777,2779,2781,2783,2785,2787,2789,2791,2793,2795,2797,2799,2801,2803,2805,2807,2809,2811,2813,2815,2817,2819,2821,2823,2825,2827,2829,2831,2833,2835,2837,2839,2841,2843,2845,2847,2849,2851,2853,2855,2857,2859,2861,2863,2865,2867,2869,2871,2873,2875,2877,2879,2881,2883,2885,2887,2889,2891,2893,2895,2897,2899,2901,2903,2905,2907,2909,2911,2913,2915,2917,2919,2921,2923,2925,2927,2929,2931,2933,2935,2937,2939,2941,2943,2945,2947,2949,2951,2953,2955,2957,2959,2961,2963,2965,2967,2969,2971,2973,2975,2977,2979,2981,2983,2985,2987,2989,2991,2993,2995,2997,2999,3001,3003,3005,3007,3009,3011,3013,3015,3017,3019,3021,3023,3025,3027,3029,3031,3033,3035,3037,3039,3041,3043,3045,3047,3049,3051,3053,3055,3057,3059,3061,3063,3065,3067,3069,3071,3073,3075,3077,3079,3081,3083,3085,3087,3089,3091,3093,3095,3097,3099,3101,3103,3105,3107,3109,3111,3113,3115,3117,3119,3121,3123,3125,3127,3129,3131,3133,3135,3137,3139,3141,3143,3145,3147,3149,3151,3153,3155,3157,3159,3161,3163,3165,3167,3169,3171,3173,3175,3177,3179,3181,3183,3185,3187,3189,3191,3193,3195,3197,3199,3201,3203,3205,3207,3209,3211,3213,3215,3217,3219,3221,3223,3225,3227,3229,3231,3233,3235,3237,3239,3241,3243,3245,3247,3249,3251,3253,3255,3257,3259,3261,3263,3265,3267,3269,3271,3273,3275,3277,3279,3281,3283,3285,3287,3289,3291,3293,3295,3297,3299,3301,3303,3305,3307,3309,3311,3313,3315,3317,3319,3321,3323,3325,3327,3329,3331,3333,3335,3337,3339,3341,3343,3345,3347,3349,3351,3353,3355,3357,3359,3361,3363,3365,3367,3369,3371,3373,3375,3377,3379,3381,3383,3385,3387,3389,3391,3393,3395,3397,3399,3401,3403,3405,3407,3409,3411,3413,3415,3417,3419,3421,3423,3425,3427,3429,3431,3433,3435,3437,3439,3441,3443,3445,3447,3449,3451,3453,3455,3457,3459,3461,3463,3465,3467,3469,3471,3473,3475,3477,3479,3481,3483,3485,3487,3489,3491,3493,3495,3497,3499,3501,3503,3505,3507,3509,3511,3513,3515,3517,3519,3521,3523,3525,3527,3529,3531,3533,3535,3537,3539,3541,3543,3545,3547,3549,3551,3553,3555,3557,3559,3561,3563,3565,3567,3569,3571,3573,3575,3577,3579,3581,3583,3585,3587,3589,3591,3593,3595,3597,3599,3601,3603,3605,3607,3609,3611,3613,3615,3617,3619,3621,3623,3625,3627,3629,3631,3633,3635,3637,3639,3641,3643,3645,3647,3649,3651,3653,3655,3657,3659,3661,3663,3665,3667,3669,3671,3673,3675,3677,3679,3681,3683,3685,3687,3689,3691,3693,3695,3697,3699,3701,3703,3705,3707,3709,3711,3713,3715,3717,3719,3721,3723,3725,3727,3729,3731,3733,3735,3737,3739,3741,3743,3745,3747,3749,3751,3753,3755,3757,3759,3761,3763,3765,3767,3769,3771,3773,3775,3777,3779,3781,3783,3785,3787,3789,3791,3793,3795,3797,3799,3801,3803,3805,3807,3809,3811,3813,3815,3817,3819,3821,3823,3825,3827,3829,3831,3833,3835,3837,3839,3841,3843,3845,3847,3849,3851,3853,3855,3857,3859,3861,3863,3865,3867,3869,3871,3873,3875,3877,3879,3881,3883,3885,3887,3889,3891,3893,3895,3897,3899,3901,3903,3905,3907,3909,3911,3913,3915,3917,3919,3921,3923,3925,3927,3929,3931,3933,3935,3937,3939,3941,3943,3945,3947,3949,3951,3953,3955,3957,3959,3961,3963,3965,3967,3969,3971,3973,3975,3977,3979,3981,3983,3985,3987,3989,3991,3993,3995,3997,3999,4001,4003,4005,4007,4009,4011,4013,4015,4017,4019,4021,4023,4025,4027,4029,4031,4033,4035,4037,4039,4041,4043,4045,4047,4049,4051,4053,4055,4057,4059,4061,4063,4065,4067,4069,4071,4073,4075,4077,4079,4081,4083],{"categories":208},[209],"Developer Productivity",{"categories":211},[212],"Business & SaaS",{"categories":214},[155],{"categories":216},[217],"AI Automation",{"categories":219},[220],"Product Strategy",{"categories":222},[155],{"categories":224},[209],{"categories":226},[212],{"categories":228},[],{"categories":230},[155],{"categories":232},[],{"categories":234},[235],"AI News & Trends",{"categories":237},[217],{"categories":239},[217],{"categories":241},[235],{"categories":243},[217],{"categories":245},[217],{"categories":247},[155],{"categories":249},[155],{"categories":251},[155],{"categories":253},[235],{"categories":255},[155],{"categories":257},[155],{"categories":259},[],{"categories":261},[262],"Design & Frontend",{"categories":264},[265],"Data Science & Visualization",{"categories":267},[235],{"categories":269},[],{"categories":271},[272],"Software Engineering",{"categories":274},[155],{"categories":276},[217],{"categories":278},[279],"Marketing & Growth",{"categories":281},[262],{"categories":283},[155],{"categories":285},[217],{"categories":287},[],{"categories":289},[],{"categories":291},[262],{"categories":293},[217],{"categories":295},[209],{"categories":297},[272],{"categories":299},[262],{"categories":301},[155],{"categories":303},[304],"DevOps & Cloud",{"categories":306},[217],{"categories":308},[235],{"categories":310},[],{"categories":312},[],{"categories":314},[217],{"categories":316},[272],{"categories":318},[],{"categories":320},[212],{"categories":322},[],{"categories":324},[],{"categories":326},[217],{"categories":328},[155],{"categories":330},[217],{"categories":332},[155],{"categories":334},[155],{"categories":336},[],{"categories":338},[272],{"categories":340},[],{"categories":342},[],{"categories":344},[272],{"categories":346},[],{"categories":348},[272],{"categories":350},[155],{"categories":352},[155],{"categories":354},[279],{"categories":356},[262],{"categories":358},[262],{"categories":360},[155],{"categories":362},[217],{"categories":364},[272],{"categories":366},[155],{"categories":368},[155],{"categories":370},[217],{"categories":372},[217],{"categories":374},[265],{"categories":376},[235],{"categories":378},[217],{"categories":380},[279],{"categories":382},[217],{"categories":384},[220],{"categories":386},[272],{"categories":388},[],{"categories":390},[217],{"categories":392},[],{"categories":394},[217],{"categories":396},[272],{"categories":398},[304],{"categories":400},[262],{"categories":402},[155],{"categories":404},[],{"categories":406},[],{"categories":408},[217],{"categories":410},[],{"categories":412},[155],{"categories":414},[],{"categories":416},[209],{"categories":418},[272],{"categories":420},[212],{"categories":422},[155],{"categories":424},[235],{"categories":426},[155],{"categories":428},[],{"categories":430},[155],{"categories":432},[],{"categories":434},[272],{"categories":436},[265],{"categories":438},[],{"categories":440},[155],{"categories":442},[262],{"categories":444},[],{"categories":446},[262],{"categories":448},[217],{"categories":450},[],{"categories":452},[155],{"categories":454},[217],{"categories":456},[235],{"categories":458},[212],{"categories":460},[155],{"categories":462},[],{"categories":464},[217],{"categories":466},[155],{"categories":468},[220],{"categories":470},[],{"categories":472},[155],{"categories":474},[217],{"categories":476},[217],{"categories":478},[],{"categories":480},[265],{"categories":482},[155],{"categories":484},[],{"categories":486},[209],{"categories":488},[212],{"categories":490},[155],{"categories":492},[217],{"categories":494},[272],{"categories":496},[155],{"categories":498},[],{"categories":500},[],{"categories":502},[155],{"categories":504},[155],{"categories":506},[],{"categories":508},[262],{"categories":510},[],{"categories":512},[155],{"categories":514},[],{"categories":516},[217],{"categories":518},[155],{"categories":520},[262],{"categories":522},[],{"categories":524},[155],{"categories":526},[155],{"categories":528},[212],{"categories":530},[217],{"categories":532},[155],{"categories":534},[262],{"categories":536},[217],{"categories":538},[],{"categories":540},[],{"categories":542},[235],{"categories":544},[],{"categories":546},[155],{"categories":548},[212,279],{"categories":550},[],{"categories":552},[155],{"categories":554},[217],{"categories":556},[],{"categories":558},[],{"categories":560},[155],{"categories":562},[],{"categories":564},[155],{"categories":566},[304],{"categories":568},[],{"categories":570},[235],{"categories":572},[262],{"categories":574},[],{"categories":576},[235],{"categories":578},[235],{"categories":580},[155],{"categories":582},[279],{"categories":584},[],{"categories":586},[212],{"categories":588},[217],{"categories":590},[],{"categories":592},[155,304],{"categories":594},[155],{"categories":596},[155],{"categories":598},[155],{"categories":600},[217],{"categories":602},[155,272],{"categories":604},[265],{"categories":606},[155],{"categories":608},[279],{"categories":610},[217],{"categories":612},[217],{"categories":614},[],{"categories":616},[217],{"categories":618},[155],{"categories":620},[155,212],{"categories":622},[],{"categories":624},[262],{"categories":626},[262],{"categories":628},[],{"categories":630},[],{"categories":632},[235],{"categories":634},[],{"categories":636},[209],{"categories":638},[272],{"categories":640},[155],{"categories":642},[262],{"categories":644},[217],{"categories":646},[272],{"categories":648},[235],{"categories":650},[262],{"categories":652},[],{"categories":654},[155],{"categories":656},[155],{"categories":658},[155],{"categories":660},[155],{"categories":662},[235],{"categories":664},[209],{"categories":666},[155],{"categories":668},[217],{"categories":670},[304],{"categories":672},[262],{"categories":674},[217],{"categories":676},[],{"categories":678},[],{"categories":680},[262],{"categories":682},[235],{"categories":684},[265],{"categories":686},[],{"categories":688},[155],{"categories":690},[155],{"categories":692},[212],{"categories":694},[155],{"categories":696},[155],{"categories":698},[235],{"categories":700},[],{"categories":702},[217],{"categories":704},[272],{"categories":706},[],{"categories":708},[155],{"categories":710},[155],{"categories":712},[217],{"categories":714},[],{"categories":716},[],{"categories":718},[155],{"categories":720},[],{"categories":722},[212],{"categories":724},[217],{"categories":726},[217],{"categories":728},[],{"categories":730},[209],{"categories":732},[155],{"categories":734},[212],{"categories":736},[235],{"categories":738},[209],{"categories":740},[],{"categories":742},[],{"categories":744},[],{"categories":746},[235],{"categories":748},[235],{"categories":750},[],{"categories":752},[],{"categories":754},[212],{"categories":756},[],{"categories":758},[],{"categories":760},[209],{"categories":762},[],{"categories":764},[279],{"categories":766},[217],{"categories":768},[212],{"categories":770},[217],{"categories":772},[272],{"categories":774},[],{"categories":776},[220],{"categories":778},[262],{"categories":780},[272],{"categories":782},[155],{"categories":784},[217],{"categories":786},[212],{"categories":788},[155],{"categories":790},[],{"categories":792},[],{"categories":794},[272],{"categories":796},[265],{"categories":798},[220],{"categories":800},[217],{"categories":802},[155],{"categories":804},[],{"categories":806},[304],{"categories":808},[],{"categories":810},[217],{"categories":812},[],{"categories":814},[209],{"categories":816},[],{"categories":818},[155],{"categories":820},[155],{"categories":822},[262],{"categories":824},[279],{"categories":826},[217],{"categories":828},[],{"categories":830},[209],{"categories":832},[],{"categories":834},[235],{"categories":836},[155,304],{"categories":838},[155],{"categories":840},[235],{"categories":842},[155],{"categories":844},[212],{"categories":846},[155],{"categories":848},[],{"categories":850},[155],{"categories":852},[212],{"categories":854},[],{"categories":856},[272],{"categories":858},[262],{"categories":860},[235],{"categories":862},[265],{"categories":864},[209],{"categories":866},[155],{"categories":868},[217],{"categories":870},[272],{"categories":872},[],{"categories":874},[],{"categories":876},[220],{"categories":878},[],{"categories":880},[155],{"categories":882},[],{"categories":884},[262],{"categories":886},[272],{"categories":888},[262],{"categories":890},[155],{"categories":892},[262],{"categories":894},[],{"categories":896},[],{"categories":898},[235],{"categories":900},[217],{"categories":902},[155],{"categories":904},[155],{"categories":906},[155],{"categories":908},[212],{"categories":910},[155],{"categories":912},[],{"categories":914},[272],{"categories":916},[272],{"categories":918},[212],{"categories":920},[],{"categories":922},[155],{"categories":924},[155],{"categories":926},[212],{"categories":928},[235],{"categories":930},[279],{"categories":932},[155],{"categories":934},[217],{"categories":936},[],{"categories":938},[262],{"categories":940},[],{"categories":942},[155],{"categories":944},[155],{"categories":946},[],{"categories":948},[212],{"categories":950},[217],{"categories":952},[],{"categories":954},[304],{"categories":956},[265],{"categories":958},[272],{"categories":960},[279],{"categories":962},[155],{"categories":964},[272],{"categories":966},[217],{"categories":968},[],{"categories":970},[],{"categories":972},[217],{"categories":974},[209],{"categories":976},[217],{"categories":978},[220],{"categories":980},[212],{"categories":982},[],{"categories":984},[155],{"categories":986},[220],{"categories":988},[155],{"categories":990},[155],{"categories":992},[279],{"categories":994},[155],{"categories":996},[262],{"categories":998},[217],{"categories":1000},[],{"categories":1002},[],{"categories":1004},[304],{"categories":1006},[272],{"categories":1008},[],{"categories":1010},[217],{"categories":1012},[155],{"categories":1014},[262,155],{"categories":1016},[209],{"categories":1018},[],{"categories":1020},[155],{"categories":1022},[209],{"categories":1024},[262],{"categories":1026},[217],{"categories":1028},[272],{"categories":1030},[],{"categories":1032},[155],{"categories":1034},[],{"categories":1036},[],{"categories":1038},[155],{"categories":1040},[209],{"categories":1042},[],{"categories":1044},[217],{"categories":1046},[220],{"categories":1048},[155],{"categories":1050},[155],{"categories":1052},[155],{"categories":1054},[262],{"categories":1056},[217],{"categories":1058},[304],{"categories":1060},[262],{"categories":1062},[217],{"categories":1064},[155],{"categories":1066},[155],{"categories":1068},[155],{"categories":1070},[272],{"categories":1072},[],{"categories":1074},[235],{"categories":1076},[],{"categories":1078},[220],{"categories":1080},[217],{"categories":1082},[262],{"categories":1084},[155],{"categories":1086},[217],{"categories":1088},[272],{"categories":1090},[262],{"categories":1092},[217],{"categories":1094},[235],{"categories":1096},[],{"categories":1098},[155],{"categories":1100},[262],{"categories":1102},[155],{"categories":1104},[209],{"categories":1106},[235],{"categories":1108},[155],{"categories":1110},[279],{"categories":1112},[155],{"categories":1114},[217],{"categories":1116},[155],{"categories":1118},[217],{"categories":1120},[217],{"categories":1122},[155],{"categories":1124},[217],{"categories":1126},[262],{"categories":1128},[155],{"categories":1130},[],{"categories":1132},[],{"categories":1134},[272],{"categories":1136},[],{"categories":1138},[209],{"categories":1140},[304],{"categories":1142},[155],{"categories":1144},[],{"categories":1146},[209],{"categories":1148},[212],{"categories":1150},[279],{"categories":1152},[],{"categories":1154},[212],{"categories":1156},[],{"categories":1158},[155],{"categories":1160},[],{"categories":1162},[],{"categories":1164},[],{"categories":1166},[],{"categories":1168},[155],{"categories":1170},[217],{"categories":1172},[304],{"categories":1174},[209],{"categories":1176},[272],{"categories":1178},[155],{"categories":1180},[272],{"categories":1182},[220],{"categories":1184},[155],{"categories":1186},[279],{"categories":1188},[212],{"categories":1190},[155],{"categories":1192},[155],{"categories":1194},[155],{"categories":1196},[155,209],{"categories":1198},[272],{"categories":1200},[272],{"categories":1202},[262],{"categories":1204},[155],{"categories":1206},[],{"categories":1208},[],{"categories":1210},[],{"categories":1212},[272],{"categories":1214},[265],{"categories":1216},[235],{"categories":1218},[262],{"categories":1220},[],{"categories":1222},[155],{"categories":1224},[155],{"categories":1226},[],{"categories":1228},[217],{"categories":1230},[155],{"categories":1232},[],{"categories":1234},[217],{"categories":1236},[155],{"categories":1238},[212],{"categories":1240},[],{"categories":1242},[209],{"categories":1244},[155],{"categories":1246},[209],{"categories":1248},[155],{"categories":1250},[272],{"categories":1252},[279],{"categories":1254},[217],{"categories":1256},[155,262],{"categories":1258},[235],{"categories":1260},[155],{"categories":1262},[262],{"categories":1264},[],{"categories":1266},[272],{"categories":1268},[304],{"categories":1270},[262],{"categories":1272},[217],{"categories":1274},[],{"categories":1276},[],{"categories":1278},[],{"categories":1280},[],{"categories":1282},[272],{"categories":1284},[217],{"categories":1286},[217],{"categories":1288},[304],{"categories":1290},[155],{"categories":1292},[155],{"categories":1294},[217],{"categories":1296},[155],{"categories":1298},[155],{"categories":1300},[],{"categories":1302},[262],{"categories":1304},[],{"categories":1306},[],{"categories":1308},[217],{"categories":1310},[],{"categories":1312},[],{"categories":1314},[279],{"categories":1316},[279],{"categories":1318},[217],{"categories":1320},[272],{"categories":1322},[],{"categories":1324},[155],{"categories":1326},[155],{"categories":1328},[272],{"categories":1330},[262],{"categories":1332},[262],{"categories":1334},[217],{"categories":1336},[209],{"categories":1338},[155],{"categories":1340},[262],{"categories":1342},[262],{"categories":1344},[217],{"categories":1346},[217],{"categories":1348},[155],{"categories":1350},[],{"categories":1352},[],{"categories":1354},[155],{"categories":1356},[217],{"categories":1358},[235],{"categories":1360},[272],{"categories":1362},[155],{"categories":1364},[209],{"categories":1366},[155],{"categories":1368},[],{"categories":1370},[217],{"categories":1372},[217],{"categories":1374},[],{"categories":1376},[155],{"categories":1378},[209],{"categories":1380},[155],{"categories":1382},[209],{"categories":1384},[209],{"categories":1386},[],{"categories":1388},[],{"categories":1390},[217],{"categories":1392},[235],{"categories":1394},[217],{"categories":1396},[155],{"categories":1398},[155],{"categories":1400},[235],{"categories":1402},[265],{"categories":1404},[220],{"categories":1406},[235],{"categories":1408},[262],{"categories":1410},[],{"categories":1412},[],{"categories":1414},[235],{"categories":1416},[],{"categories":1418},[],{"categories":1420},[],{"categories":1422},[],{"categories":1424},[272],{"categories":1426},[265],{"categories":1428},[],{"categories":1430},[155],{"categories":1432},[155],{"categories":1434},[265],{"categories":1436},[272],{"categories":1438},[],{"categories":1440},[],{"categories":1442},[217],{"categories":1444},[235],{"categories":1446},[235],{"categories":1448},[217],{"categories":1450},[209],{"categories":1452},[155,304],{"categories":1454},[],{"categories":1456},[262],{"categories":1458},[209],{"categories":1460},[217],{"categories":1462},[262],{"categories":1464},[],{"categories":1466},[217],{"categories":1468},[217],{"categories":1470},[155],{"categories":1472},[279],{"categories":1474},[272],{"categories":1476},[262],{"categories":1478},[],{"categories":1480},[217],{"categories":1482},[155],{"categories":1484},[217],{"categories":1486},[217],{"categories":1488},[217],{"categories":1490},[279],{"categories":1492},[155],{"categories":1494},[217],{"categories":1496},[155],{"categories":1498},[],{"categories":1500},[279],{"categories":1502},[235],{"categories":1504},[217],{"categories":1506},[],{"categories":1508},[],{"categories":1510},[155],{"categories":1512},[217],{"categories":1514},[235],{"categories":1516},[217],{"categories":1518},[217],{"categories":1520},[],{"categories":1522},[155],{"categories":1524},[],{"categories":1526},[],{"categories":1528},[217],{"categories":1530},[],{"categories":1532},[],{"categories":1534},[265],{"categories":1536},[155],{"categories":1538},[265],{"categories":1540},[235],{"categories":1542},[155],{"categories":1544},[155],{"categories":1546},[217],{"categories":1548},[155],{"categories":1550},[],{"categories":1552},[],{"categories":1554},[304],{"categories":1556},[155],{"categories":1558},[],{"categories":1560},[],{"categories":1562},[209],{"categories":1564},[],{"categories":1566},[],{"categories":1568},[155],{"categories":1570},[],{"categories":1572},[],{"categories":1574},[272],{"categories":1576},[235],{"categories":1578},[279],{"categories":1580},[212],{"categories":1582},[155],{"categories":1584},[155],{"categories":1586},[212],{"categories":1588},[],{"categories":1590},[262],{"categories":1592},[217],{"categories":1594},[212],{"categories":1596},[155],{"categories":1598},[155],{"categories":1600},[209],{"categories":1602},[],{"categories":1604},[209],{"categories":1606},[155],{"categories":1608},[279],{"categories":1610},[217],{"categories":1612},[235],{"categories":1614},[212],{"categories":1616},[155],{"categories":1618},[155],{"categories":1620},[217],{"categories":1622},[],{"categories":1624},[155],{"categories":1626},[209],{"categories":1628},[155],{"categories":1630},[155],{"categories":1632},[],{"categories":1634},[235],{"categories":1636},[155],{"categories":1638},[],{"categories":1640},[212],{"categories":1642},[212],{"categories":1644},[155],{"categories":1646},[],{"categories":1648},[],{"categories":1650},[],{"categories":1652},[155],{"categories":1654},[235],{"categories":1656},[],{"categories":1658},[304],{"categories":1660},[155],{"categories":1662},[],{"categories":1664},[155],{"categories":1666},[155],{"categories":1668},[155],{"categories":1670},[155,304],{"categories":1672},[155],{"categories":1674},[155],{"categories":1676},[262],{"categories":1678},[217],{"categories":1680},[],{"categories":1682},[217],{"categories":1684},[217],{"categories":1686},[155],{"categories":1688},[155],{"categories":1690},[155],{"categories":1692},[209],{"categories":1694},[209],{"categories":1696},[272],{"categories":1698},[262],{"categories":1700},[217],{"categories":1702},[],{"categories":1704},[155],{"categories":1706},[235],{"categories":1708},[155],{"categories":1710},[212],{"categories":1712},[],{"categories":1714},[304],{"categories":1716},[262],{"categories":1718},[262],{"categories":1720},[217],{"categories":1722},[235],{"categories":1724},[217],{"categories":1726},[155],{"categories":1728},[],{"categories":1730},[155],{"categories":1732},[],{"categories":1734},[],{"categories":1736},[155],{"categories":1738},[155],{"categories":1740},[155],{"categories":1742},[217],{"categories":1744},[155],{"categories":1746},[155],{"categories":1748},[],{"categories":1750},[265],{"categories":1752},[217],{"categories":1754},[],{"categories":1756},[],{"categories":1758},[155],{"categories":1760},[235],{"categories":1762},[],{"categories":1764},[262],{"categories":1766},[304],{"categories":1768},[235],{"categories":1770},[272],{"categories":1772},[272],{"categories":1774},[235],{"categories":1776},[235],{"categories":1778},[304],{"categories":1780},[],{"categories":1782},[235],{"categories":1784},[155],{"categories":1786},[209],{"categories":1788},[155],{"categories":1790},[235],{"categories":1792},[],{"categories":1794},[272],{"categories":1796},[265],{"categories":1798},[155],{"categories":1800},[235],{"categories":1802},[272],{"categories":1804},[217],{"categories":1806},[235],{"categories":1808},[304],{"categories":1810},[217],{"categories":1812},[155],{"categories":1814},[155],{"categories":1816},[155],{"categories":1818},[],{"categories":1820},[212],{"categories":1822},[],{"categories":1824},[],{"categories":1826},[155],{"categories":1828},[155],{"categories":1830},[155],{"categories":1832},[155],{"categories":1834},[],{"categories":1836},[265],{"categories":1838},[209],{"categories":1840},[],{"categories":1842},[155],{"categories":1844},[155],{"categories":1846},[304],{"categories":1848},[304],{"categories":1850},[],{"categories":1852},[217],{"categories":1854},[235],{"categories":1856},[235],{"categories":1858},[155],{"categories":1860},[217],{"categories":1862},[],{"categories":1864},[262],{"categories":1866},[155],{"categories":1868},[155],{"categories":1870},[],{"categories":1872},[155],{"categories":1874},[],{"categories":1876},[272],{"categories":1878},[304],{"categories":1880},[155],{"categories":1882},[272],{"categories":1884},[212],{"categories":1886},[155],{"categories":1888},[],{"categories":1890},[217],{"categories":1892},[209],{"categories":1894},[209],{"categories":1896},[],{"categories":1898},[155],{"categories":1900},[262],{"categories":1902},[217],{"categories":1904},[],{"categories":1906},[155],{"categories":1908},[155],{"categories":1910},[217],{"categories":1912},[],{"categories":1914},[217],{"categories":1916},[272],{"categories":1918},[],{"categories":1920},[155],{"categories":1922},[],{"categories":1924},[155],{"categories":1926},[],{"categories":1928},[155],{"categories":1930},[155],{"categories":1932},[],{"categories":1934},[155],{"categories":1936},[235],{"categories":1938},[155],{"categories":1940},[155],{"categories":1942},[209],{"categories":1944},[155],{"categories":1946},[235],{"categories":1948},[217],{"categories":1950},[],{"categories":1952},[155],{"categories":1954},[262],{"categories":1956},[279],{"categories":1958},[155],{"categories":1960},[],{"categories":1962},[],{"categories":1964},[],{"categories":1966},[209],{"categories":1968},[235],{"categories":1970},[217],{"categories":1972},[155],{"categories":1974},[262],{"categories":1976},[217],{"categories":1978},[],{"categories":1980},[217],{"categories":1982},[],{"categories":1984},[155],{"categories":1986},[217],{"categories":1988},[155],{"categories":1990},[],{"categories":1992},[155],{"categories":1994},[155],{"categories":1996},[235],{"categories":1998},[262],{"categories":2000},[217],{"categories":2002},[262],{"categories":2004},[212],{"categories":2006},[],{"categories":2008},[],{"categories":2010},[155],{"categories":2012},[209],{"categories":2014},[235],{"categories":2016},[],{"categories":2018},[262],{"categories":2020},[],{"categories":2022},[272],{"categories":2024},[272],{"categories":2026},[262],{"categories":2028},[],{"categories":2030},[155],{"categories":2032},[],{"categories":2034},[279],{"categories":2036},[155],{"categories":2038},[304],{"categories":2040},[272],{"categories":2042},[],{"categories":2044},[217],{"categories":2046},[155],{"categories":2048},[209],{"categories":2050},[217],{"categories":2052},[217],{"categories":2054},[155],{"categories":2056},[],{"categories":2058},[209],{"categories":2060},[155],{"categories":2062},[212],{"categories":2064},[272],{"categories":2066},[262],{"categories":2068},[],{"categories":2070},[],{"categories":2072},[],{"categories":2074},[217],{"categories":2076},[262],{"categories":2078},[235],{"categories":2080},[155],{"categories":2082},[235],{"categories":2084},[262],{"categories":2086},[],{"categories":2088},[262],{"categories":2090},[235],{"categories":2092},[212],{"categories":2094},[272],{"categories":2096},[155],{"categories":2098},[235],{"categories":2100},[279],{"categories":2102},[],{"categories":2104},[],{"categories":2106},[265],{"categories":2108},[155,272],{"categories":2110},[235],{"categories":2112},[155],{"categories":2114},[217],{"categories":2116},[155],{"categories":2118},[217],{"categories":2120},[155],{"categories":2122},[155],{"categories":2124},[],{"categories":2126},[272],{"categories":2128},[155],{"categories":2130},[265],{"categories":2132},[217],{"categories":2134},[279],{"categories":2136},[304],{"categories":2138},[],{"categories":2140},[209],{"categories":2142},[217],{"categories":2144},[217],{"categories":2146},[272],{"categories":2148},[155],{"categories":2150},[155],{"categories":2152},[],{"categories":2154},[],{"categories":2156},[],{"categories":2158},[304],{"categories":2160},[235],{"categories":2162},[155],{"categories":2164},[155],{"categories":2166},[155],{"categories":2168},[],{"categories":2170},[265],{"categories":2172},[212],{"categories":2174},[],{"categories":2176},[217],{"categories":2178},[304],{"categories":2180},[],{"categories":2182},[262],{"categories":2184},[262],{"categories":2186},[],{"categories":2188},[272],{"categories":2190},[155],{"categories":2192},[262],{"categories":2194},[155],{"categories":2196},[],{"categories":2198},[235],{"categories":2200},[155],{"categories":2202},[155],{"categories":2204},[262],{"categories":2206},[217],{"categories":2208},[235],{"categories":2210},[],{"categories":2212},[217],{"categories":2214},[262],{"categories":2216},[155],{"categories":2218},[],{"categories":2220},[155],{"categories":2222},[155],{"categories":2224},[304],{"categories":2226},[235],{"categories":2228},[265],{"categories":2230},[265],{"categories":2232},[],{"categories":2234},[],{"categories":2236},[],{"categories":2238},[217],{"categories":2240},[272],{"categories":2242},[272],{"categories":2244},[155],{"categories":2246},[],{"categories":2248},[],{"categories":2250},[155],{"categories":2252},[],{"categories":2254},[217],{"categories":2256},[155],{"categories":2258},[],{"categories":2260},[155],{"categories":2262},[212],{"categories":2264},[155],{"categories":2266},[279],{"categories":2268},[217],{"categories":2270},[155],{"categories":2272},[155],{"categories":2274},[155],{"categories":2276},[272],{"categories":2278},[],{"categories":2280},[235],{"categories":2282},[217],{"categories":2284},[],{"categories":2286},[235],{"categories":2288},[217],{"categories":2290},[217],{"categories":2292},[],{"categories":2294},[212],{"categories":2296},[217],{"categories":2298},[],{"categories":2300},[155],{"categories":2302},[209],{"categories":2304},[235],{"categories":2306},[304],{"categories":2308},[217],{"categories":2310},[217],{"categories":2312},[209],{"categories":2314},[],{"categories":2316},[155],{"categories":2318},[],{"categories":2320},[],{"categories":2322},[262],{"categories":2324},[155,212],{"categories":2326},[155],{"categories":2328},[],{"categories":2330},[209],{"categories":2332},[265],{"categories":2334},[155],{"categories":2336},[272],{"categories":2338},[155],{"categories":2340},[217],{"categories":2342},[155],{"categories":2344},[155],{"categories":2346},[235],{"categories":2348},[217],{"categories":2350},[],{"categories":2352},[],{"categories":2354},[217],{"categories":2356},[155],{"categories":2358},[304],{"categories":2360},[],{"categories":2362},[155],{"categories":2364},[217],{"categories":2366},[],{"categories":2368},[217],{"categories":2370},[155],{"categories":2372},[279],{"categories":2374},[265],{"categories":2376},[217],{"categories":2378},[155],{"categories":2380},[304],{"categories":2382},[],{"categories":2384},[155],{"categories":2386},[279],{"categories":2388},[262],{"categories":2390},[155],{"categories":2392},[155],{"categories":2394},[],{"categories":2396},[279],{"categories":2398},[235],{"categories":2400},[155],{"categories":2402},[155],{"categories":2404},[209],{"categories":2406},[],{"categories":2408},[],{"categories":2410},[262],{"categories":2412},[155],{"categories":2414},[265],{"categories":2416},[279],{"categories":2418},[279],{"categories":2420},[235],{"categories":2422},[],{"categories":2424},[],{"categories":2426},[155],{"categories":2428},[155],{"categories":2430},[155],{"categories":2432},[],{"categories":2434},[155,272],{"categories":2436},[235],{"categories":2438},[217],{"categories":2440},[272],{"categories":2442},[155],{"categories":2444},[209],{"categories":2446},[],{"categories":2448},[],{"categories":2450},[209],{"categories":2452},[272],{"categories":2454},[279],{"categories":2456},[155],{"categories":2458},[],{"categories":2460},[262,155],{"categories":2462},[304],{"categories":2464},[209],{"categories":2466},[],{"categories":2468},[212],{"categories":2470},[212],{"categories":2472},[155],{"categories":2474},[155],{"categories":2476},[272],{"categories":2478},[217],{"categories":2480},[235],{"categories":2482},[279],{"categories":2484},[262],{"categories":2486},[155],{"categories":2488},[155],{"categories":2490},[155],{"categories":2492},[209],{"categories":2494},[155],{"categories":2496},[217],{"categories":2498},[235],{"categories":2500},[],{"categories":2502},[],{"categories":2504},[265],{"categories":2506},[272],{"categories":2508},[155],{"categories":2510},[262],{"categories":2512},[155],{"categories":2514},[265],{"categories":2516},[155],{"categories":2518},[155],{"categories":2520},[155],{"categories":2522},[217],{"categories":2524},[217],{"categories":2526},[155,212],{"categories":2528},[],{"categories":2530},[262],{"categories":2532},[],{"categories":2534},[155],{"categories":2536},[235],{"categories":2538},[209],{"categories":2540},[209],{"categories":2542},[217],{"categories":2544},[155],{"categories":2546},[155],{"categories":2548},[212],{"categories":2550},[272],{"categories":2552},[279],{"categories":2554},[155],{"categories":2556},[],{"categories":2558},[235],{"categories":2560},[155],{"categories":2562},[155],{"categories":2564},[155],{"categories":2566},[155],{"categories":2568},[235],{"categories":2570},[272],{"categories":2572},[272],{"categories":2574},[155],{"categories":2576},[155],{"categories":2578},[217],{"categories":2580},[235],{"categories":2582},[155],{"categories":2584},[262],{"categories":2586},[155],{"categories":2588},[155],{"categories":2590},[304],{"categories":2592},[155],{"categories":2594},[220],{"categories":2596},[217],{"categories":2598},[155],{"categories":2600},[235],{"categories":2602},[217],{"categories":2604},[279],{"categories":2606},[155],{"categories":2608},[],{"categories":2610},[155],{"categories":2612},[],{"categories":2614},[],{"categories":2616},[],{"categories":2618},[212],{"categories":2620},[155],{"categories":2622},[217],{"categories":2624},[235],{"categories":2626},[235],{"categories":2628},[235],{"categories":2630},[235],{"categories":2632},[],{"categories":2634},[209],{"categories":2636},[217],{"categories":2638},[235],{"categories":2640},[155],{"categories":2642},[209],{"categories":2644},[217],{"categories":2646},[155],{"categories":2648},[155,217],{"categories":2650},[217],{"categories":2652},[304],{"categories":2654},[235],{"categories":2656},[235],{"categories":2658},[217],{"categories":2660},[155],{"categories":2662},[],{"categories":2664},[235],{"categories":2666},[279],{"categories":2668},[209],{"categories":2670},[155],{"categories":2672},[155],{"categories":2674},[],{"categories":2676},[272],{"categories":2678},[],{"categories":2680},[209],{"categories":2682},[217],{"categories":2684},[235],{"categories":2686},[155],{"categories":2688},[235],{"categories":2690},[209],{"categories":2692},[235],{"categories":2694},[235],{"categories":2696},[],{"categories":2698},[212],{"categories":2700},[217],{"categories":2702},[235],{"categories":2704},[235],{"categories":2706},[235],{"categories":2708},[235],{"categories":2710},[235],{"categories":2712},[235],{"categories":2714},[235],{"categories":2716},[235],{"categories":2718},[235],{"categories":2720},[235],{"categories":2722},[265],{"categories":2724},[209],{"categories":2726},[155],{"categories":2728},[155],{"categories":2730},[],{"categories":2732},[155,209],{"categories":2734},[],{"categories":2736},[217],{"categories":2738},[235],{"categories":2740},[217],{"categories":2742},[155],{"categories":2744},[155],{"categories":2746},[155],{"categories":2748},[155],{"categories":2750},[155],{"categories":2752},[217],{"categories":2754},[212],{"categories":2756},[],{"categories":2758},[262],{"categories":2760},[235],{"categories":2762},[155],{"categories":2764},[],{"categories":2766},[],{"categories":2768},[217],{"categories":2770},[262],{"categories":2772},[155],{"categories":2774},[],{"categories":2776},[155],{"categories":2778},[],{"categories":2780},[279],{"categories":2782},[155],{"categories":2784},[],{"categories":2786},[],{"categories":2788},[235],{"categories":2790},[209],{"categories":2792},[155],{"categories":2794},[212],{"categories":2796},[155],{"categories":2798},[212],{"categories":2800},[262],{"categories":2802},[],{"categories":2804},[235],{"categories":2806},[],{"categories":2808},[262],{"categories":2810},[155],{"categories":2812},[279],{"categories":2814},[],{"categories":2816},[279],{"categories":2818},[],{"categories":2820},[],{"categories":2822},[217],{"categories":2824},[],{"categories":2826},[212],{"categories":2828},[209],{"categories":2830},[262],{"categories":2832},[272],{"categories":2834},[],{"categories":2836},[],{"categories":2838},[155],{"categories":2840},[209],{"categories":2842},[279],{"categories":2844},[],{"categories":2846},[217],{"categories":2848},[217],{"categories":2850},[235],{"categories":2852},[272],{"categories":2854},[155],{"categories":2856},[217],{"categories":2858},[155],{"categories":2860},[217],{"categories":2862},[155],{"categories":2864},[220],{"categories":2866},[235],{"categories":2868},[],{"categories":2870},[279],{"categories":2872},[],{"categories":2874},[272],{"categories":2876},[217],{"categories":2878},[],{"categories":2880},[155],{"categories":2882},[217],{"categories":2884},[212],{"categories":2886},[209],{"categories":2888},[155],{"categories":2890},[262],{"categories":2892},[272],{"categories":2894},[272],{"categories":2896},[155],{"categories":2898},[265],{"categories":2900},[155],{"categories":2902},[217],{"categories":2904},[212],{"categories":2906},[262],{"categories":2908},[217],{"categories":2910},[155],{"categories":2912},[155],{"categories":2914},[217],{"categories":2916},[235],{"categories":2918},[],{"categories":2920},[209],{"categories":2922},[155],{"categories":2924},[217],{"categories":2926},[155],{"categories":2928},[155],{"categories":2930},[],{"categories":2932},[262],{"categories":2934},[212],{"categories":2936},[235],{"categories":2938},[155],{"categories":2940},[155],{"categories":2942},[262],{"categories":2944},[155],{"categories":2946},[279],{"categories":2948},[265],{"categories":2950},[155],{"categories":2952},[235],{"categories":2954},[155],{"categories":2956},[217],{"categories":2958},[304],{"categories":2960},[155],{"categories":2962},[217],{"categories":2964},[265],{"categories":2966},[],{"categories":2968},[217],{"categories":2970},[272],{"categories":2972},[262],{"categories":2974},[155],{"categories":2976},[209],{"categories":2978},[212],{"categories":2980},[272],{"categories":2982},[155],{"categories":2984},[],{"categories":2986},[217],{"categories":2988},[217],{"categories":2990},[155],{"categories":2992},[265],{"categories":2994},[],{"categories":2996},[235],{"categories":2998},[],{"categories":3000},[235],{"categories":3002},[155],{"categories":3004},[217],{"categories":3006},[217],{"categories":3008},[217],{"categories":3010},[],{"categories":3012},[235],{"categories":3014},[],{"categories":3016},[155],{"categories":3018},[155],{"categories":3020},[],{"categories":3022},[262],{"categories":3024},[217],{"categories":3026},[279],{"categories":3028},[209],{"categories":3030},[],{"categories":3032},[155],{"categories":3034},[],{"categories":3036},[209],{"categories":3038},[235],{"categories":3040},[272],{"categories":3042},[155],{"categories":3044},[155],{"categories":3046},[155],{"categories":3048},[272],{"categories":3050},[235],{"categories":3052},[262],{"categories":3054},[155],{"categories":3056},[155],{"categories":3058},[155],{"categories":3060},[235],{"categories":3062},[155],{"categories":3064},[235],{"categories":3066},[235],{"categories":3068},[217],{"categories":3070},[217],{"categories":3072},[272],{"categories":3074},[235],{"categories":3076},[217],{"categories":3078},[155],{"categories":3080},[272],{"categories":3082},[262],{"categories":3084},[],{"categories":3086},[217],{"categories":3088},[],{"categories":3090},[],{"categories":3092},[],{"categories":3094},[212],{"categories":3096},[155],{"categories":3098},[217],{"categories":3100},[209],{"categories":3102},[217],{"categories":3104},[279],{"categories":3106},[],{"categories":3108},[217],{"categories":3110},[],{"categories":3112},[209],{"categories":3114},[217],{"categories":3116},[],{"categories":3118},[217],{"categories":3120},[155],{"categories":3122},[235],{"categories":3124},[155],{"categories":3126},[217],{"categories":3128},[235],{"categories":3130},[217],{"categories":3132},[272],{"categories":3134},[262],{"categories":3136},[209],{"categories":3138},[],{"categories":3140},[217],{"categories":3142},[262],{"categories":3144},[304],{"categories":3146},[235],{"categories":3148},[155],{"categories":3150},[262],{"categories":3152},[209],{"categories":3154},[],{"categories":3156},[217],{"categories":3158},[155],{"categories":3160},[217],{"categories":3162},[155],{"categories":3164},[],{"categories":3166},[217],{"categories":3168},[220],{"categories":3170},[235],{"categories":3172},[217],{"categories":3174},[212],{"categories":3176},[],{"categories":3178},[155],{"categories":3180},[220],{"categories":3182},[155],{"categories":3184},[217],{"categories":3186},[235],{"categories":3188},[209],{"categories":3190},[304],{"categories":3192},[155],{"categories":3194},[155],{"categories":3196},[155],{"categories":3198},[235],{"categories":3200},[212],{"categories":3202},[155],{"categories":3204},[262],{"categories":3206},[235],{"categories":3208},[304],{"categories":3210},[155],{"categories":3212},[],{"categories":3214},[],{"categories":3216},[155],{"categories":3218},[304],{"categories":3220},[265],{"categories":3222},[217],{"categories":3224},[217],{"categories":3226},[235],{"categories":3228},[155],{"categories":3230},[209],{"categories":3232},[262],{"categories":3234},[217],{"categories":3236},[155],{"categories":3238},[279],{"categories":3240},[155],{"categories":3242},[217],{"categories":3244},[],{"categories":3246},[155],{"categories":3248},[155],{"categories":3250},[235],{"categories":3252},[209],{"categories":3254},[],{"categories":3256},[155],{"categories":3258},[155],{"categories":3260},[272],{"categories":3262},[262],{"categories":3264},[155,217],{"categories":3266},[279,212],{"categories":3268},[155],{"categories":3270},[],{"categories":3272},[217],{"categories":3274},[],{"categories":3276},[272],{"categories":3278},[155],{"categories":3280},[],{"categories":3282},[155],{"categories":3284},[235],{"categories":3286},[],{"categories":3288},[217],{"categories":3290},[155],{"categories":3292},[],{"categories":3294},[262],{"categories":3296},[217],{"categories":3298},[155],{"categories":3300},[209],{"categories":3302},[217],{"categories":3304},[155],{"categories":3306},[],{"categories":3308},[304],{"categories":3310},[279],{"categories":3312},[212],{"categories":3314},[212],{"categories":3316},[209],{"categories":3318},[209],{"categories":3320},[155],{"categories":3322},[217],{"categories":3324},[155],{"categories":3326},[155],{"categories":3328},[209],{"categories":3330},[155],{"categories":3332},[279],{"categories":3334},[235],{"categories":3336},[155],{"categories":3338},[217],{"categories":3340},[155],{"categories":3342},[],{"categories":3344},[272],{"categories":3346},[],{"categories":3348},[272],{"categories":3350},[217],{"categories":3352},[209],{"categories":3354},[],{"categories":3356},[304],{"categories":3358},[155],{"categories":3360},[],{"categories":3362},[235],{"categories":3364},[217],{"categories":3366},[272],{"categories":3368},[155],{"categories":3370},[217],{"categories":3372},[272],{"categories":3374},[217],{"categories":3376},[235],{"categories":3378},[209],{"categories":3380},[235],{"categories":3382},[272],{"categories":3384},[155],{"categories":3386},[262],{"categories":3388},[155],{"categories":3390},[155],{"categories":3392},[155],{"categories":3394},[155],{"categories":3396},[155],{"categories":3398},[217],{"categories":3400},[155],{"categories":3402},[217],{"categories":3404},[155],{"categories":3406},[209],{"categories":3408},[155],{"categories":3410},[217],{"categories":3412},[262],{"categories":3414},[209],{"categories":3416},[217],{"categories":3418},[262],{"categories":3420},[],{"categories":3422},[155],{"categories":3424},[155],{"categories":3426},[272],{"categories":3428},[],{"categories":3430},[217],{"categories":3432},[279],{"categories":3434},[155],{"categories":3436},[235],{"categories":3438},[279],{"categories":3440},[217],{"categories":3442},[212],{"categories":3444},[212],{"categories":3446},[155],{"categories":3448},[209],{"categories":3450},[],{"categories":3452},[217],{"categories":3454},[155],{"categories":3456},[],{"categories":3458},[209],{"categories":3460},[155],{"categories":3462},[217],{"categories":3464},[217],{"categories":3466},[],{"categories":3468},[272],{"categories":3470},[272],{"categories":3472},[279],{"categories":3474},[262],{"categories":3476},[],{"categories":3478},[155],{"categories":3480},[217],{"categories":3482},[209],{"categories":3484},[155],{"categories":3486},[272],{"categories":3488},[209],{"categories":3490},[235],{"categories":3492},[235],{"categories":3494},[],{"categories":3496},[235],{"categories":3498},[217],{"categories":3500},[262],{"categories":3502},[265],{"categories":3504},[155],{"categories":3506},[],{"categories":3508},[235],{"categories":3510},[272],{"categories":3512},[212],{"categories":3514},[155],{"categories":3516},[209],{"categories":3518},[304],{"categories":3520},[209],{"categories":3522},[],{"categories":3524},[],{"categories":3526},[235],{"categories":3528},[],{"categories":3530},[217],{"categories":3532},[217],{"categories":3534},[217],{"categories":3536},[],{"categories":3538},[155],{"categories":3540},[],{"categories":3542},[235],{"categories":3544},[209],{"categories":3546},[262],{"categories":3548},[155],{"categories":3550},[235],{"categories":3552},[235],{"categories":3554},[],{"categories":3556},[235],{"categories":3558},[209],{"categories":3560},[155],{"categories":3562},[],{"categories":3564},[217],{"categories":3566},[217],{"categories":3568},[209],{"categories":3570},[],{"categories":3572},[],{"categories":3574},[],{"categories":3576},[262],{"categories":3578},[217],{"categories":3580},[155],{"categories":3582},[],{"categories":3584},[],{"categories":3586},[],{"categories":3588},[262],{"categories":3590},[],{"categories":3592},[155],{"categories":3594},[209],{"categories":3596},[],{"categories":3598},[],{"categories":3600},[262],{"categories":3602},[155],{"categories":3604},[235],{"categories":3606},[],{"categories":3608},[279],{"categories":3610},[235],{"categories":3612},[279],{"categories":3614},[155],{"categories":3616},[],{"categories":3618},[],{"categories":3620},[217],{"categories":3622},[],{"categories":3624},[],{"categories":3626},[217],{"categories":3628},[155],{"categories":3630},[],{"categories":3632},[217],{"categories":3634},[235],{"categories":3636},[155],{"categories":3638},[279],{"categories":3640},[265],{"categories":3642},[217],{"categories":3644},[217],{"categories":3646},[],{"categories":3648},[],{"categories":3650},[],{"categories":3652},[235],{"categories":3654},[],{"categories":3656},[],{"categories":3658},[262],{"categories":3660},[209],{"categories":3662},[],{"categories":3664},[212],{"categories":3666},[279],{"categories":3668},[155],{"categories":3670},[272],{"categories":3672},[209],{"categories":3674},[265],{"categories":3676},[212],{"categories":3678},[272],{"categories":3680},[272],{"categories":3682},[],{"categories":3684},[],{"categories":3686},[217],{"categories":3688},[209],{"categories":3690},[262],{"categories":3692},[209],{"categories":3694},[217],{"categories":3696},[304],{"categories":3698},[155],{"categories":3700},[209],{"categories":3702},[217],{"categories":3704},[],{"categories":3706},[155],{"categories":3708},[235],{"categories":3710},[272],{"categories":3712},[],{"categories":3714},[262],{"categories":3716},[235],{"categories":3718},[209],{"categories":3720},[217],{"categories":3722},[155],{"categories":3724},[212],{"categories":3726},[217,304],{"categories":3728},[217],{"categories":3730},[272],{"categories":3732},[155],{"categories":3734},[155],{"categories":3736},[265],{"categories":3738},[279],{"categories":3740},[217],{"categories":3742},[],{"categories":3744},[217],{"categories":3746},[155],{"categories":3748},[212],{"categories":3750},[],{"categories":3752},[],{"categories":3754},[155],{"categories":3756},[265],{"categories":3758},[155],{"categories":3760},[],{"categories":3762},[235],{"categories":3764},[],{"categories":3766},[235],{"categories":3768},[209],{"categories":3770},[272],{"categories":3772},[155],{"categories":3774},[217],{"categories":3776},[155],{"categories":3778},[155],{"categories":3780},[279],{"categories":3782},[272],{"categories":3784},[],{"categories":3786},[235],{"categories":3788},[155],{"categories":3790},[],{"categories":3792},[155],{"categories":3794},[217],{"categories":3796},[155],{"categories":3798},[217],{"categories":3800},[155],{"categories":3802},[155],{"categories":3804},[155],{"categories":3806},[155],{"categories":3808},[212],{"categories":3810},[],{"categories":3812},[220],{"categories":3814},[235],{"categories":3816},[155],{"categories":3818},[],{"categories":3820},[272],{"categories":3822},[155],{"categories":3824},[155],{"categories":3826},[155],{"categories":3828},[217],{"categories":3830},[235],{"categories":3832},[155],{"categories":3834},[155],{"categories":3836},[155],{"categories":3838},[212],{"categories":3840},[217],{"categories":3842},[262],{"categories":3844},[],{"categories":3846},[265],{"categories":3848},[155],{"categories":3850},[],{"categories":3852},[235],{"categories":3854},[279],{"categories":3856},[],{"categories":3858},[],{"categories":3860},[235],{"categories":3862},[235],{"categories":3864},[279],{"categories":3866},[209],{"categories":3868},[217],{"categories":3870},[217],{"categories":3872},[155],{"categories":3874},[212],{"categories":3876},[],{"categories":3878},[],{"categories":3880},[235],{"categories":3882},[265],{"categories":3884},[272],{"categories":3886},[217],{"categories":3888},[262],{"categories":3890},[265],{"categories":3892},[265],{"categories":3894},[],{"categories":3896},[235],{"categories":3898},[155],{"categories":3900},[155],{"categories":3902},[272],{"categories":3904},[],{"categories":3906},[235],{"categories":3908},[235],{"categories":3910},[235],{"categories":3912},[],{"categories":3914},[217],{"categories":3916},[155],{"categories":3918},[],{"categories":3920},[209],{"categories":3922},[212],{"categories":3924},[],{"categories":3926},[155],{"categories":3928},[155],{"categories":3930},[],{"categories":3932},[272],{"categories":3934},[],{"categories":3936},[],{"categories":3938},[],{"categories":3940},[],{"categories":3942},[155],{"categories":3944},[235],{"categories":3946},[],{"categories":3948},[],{"categories":3950},[155],{"categories":3952},[155],{"categories":3954},[155],{"categories":3956},[265],{"categories":3958},[155],{"categories":3960},[265],{"categories":3962},[],{"categories":3964},[265],{"categories":3966},[265],{"categories":3968},[304],{"categories":3970},[217],{"categories":3972},[272],{"categories":3974},[],{"categories":3976},[],{"categories":3978},[265],{"categories":3980},[272],{"categories":3982},[272],{"categories":3984},[272],{"categories":3986},[],{"categories":3988},[209],{"categories":3990},[272],{"categories":3992},[272],{"categories":3994},[209],{"categories":3996},[272],{"categories":3998},[212],{"categories":4000},[272],{"categories":4002},[272],{"categories":4004},[272],{"categories":4006},[265],{"categories":4008},[235],{"categories":4010},[235],{"categories":4012},[155],{"categories":4014},[272],{"categories":4016},[265],{"categories":4018},[304],{"categories":4020},[265],{"categories":4022},[265],{"categories":4024},[265],{"categories":4026},[],{"categories":4028},[212],{"categories":4030},[],{"categories":4032},[304],{"categories":4034},[272],{"categories":4036},[272],{"categories":4038},[272],{"categories":4040},[217],{"categories":4042},[235,212],{"categories":4044},[265],{"categories":4046},[],{"categories":4048},[],{"categories":4050},[265],{"categories":4052},[],{"categories":4054},[265],{"categories":4056},[235],{"categories":4058},[217],{"categories":4060},[],{"categories":4062},[272],{"categories":4064},[155],{"categories":4066},[262],{"categories":4068},[],{"categories":4070},[155],{"categories":4072},[],{"categories":4074},[235],{"categories":4076},[209],{"categories":4078},[265],{"categories":4080},[],{"categories":4082},[272],{"categories":4084},[235],[4086,4185,4268,4342],{"id":4087,"title":4088,"ai":4089,"body":4095,"categories":4154,"created_at":156,"date_modified":156,"description":148,"extension":157,"faq":156,"featured":158,"kicker_label":156,"meta":4155,"navigation":187,"path":4171,"published_at":4172,"question":156,"scraped_at":4173,"seo":4174,"sitemap":4175,"source_id":4176,"source_name":4177,"source_type":195,"source_url":4178,"stem":4179,"tags":4180,"thumbnail_url":156,"tldr":4182,"tweet":156,"unknown_tags":4183,"__hash__":4184},"summaries\u002Fsummaries\u002F41ef3a9324aac236-databricks-rag-low-dim-qwen3-rerank-for-89-recall--summary.md","Databricks RAG: Low-Dim Qwen3 + Rerank for 89% Recall@10",{"provider":7,"model":4090,"input_tokens":4091,"output_tokens":4092,"processing_time_ms":4093,"cost_usd":4094},"x-ai\u002Fgrok-4.1-fast",6251,1773,31729,0.0021142,{"type":14,"value":4096,"toc":4149},[4097,4101,4104,4108,4138,4142],[17,4098,4100],{"id":4099},"minimize-dimensions-and-tune-queries-to-cut-latency-without-losing-recall","Minimize Dimensions and Tune Queries to Cut Latency Without Losing Recall",[22,4102,4103],{},"Higher-dim embeddings (1024-1536) increase ANN scan costs, memory use, and slow throughput—test empirically to pick the lowest dim preserving recall@10, like 384 over 1024 if equivalent. Limit num_results to 10-100 (default 50 with reranker, 10 without) since HNSW scales linearly and excess slows queries without better answers. Match endpoint SKU to scale: Standard for \u003C2M 768-dim vectors (low latency), Storage-Optimized for \u003C1B vectors (cheaper, higher latency, dims divisible by 16, triggered sync only). Add metadata filters (e.g., {\"document_type\": \"manual\"}) to Delta tables for scoped ANN scans, boosting precision\u002Fspeed. Stick to ANN for semantic queries (highest QPS); hybrid (ANN+BM25) only for exact terms like SKUs or ISO 13849-1.",[17,4105,4107],{"id":4106},"self-manage-qwen3-mrl-embeddings-to-hit-target-dims-like-256","Self-Manage Qwen3 MRL Embeddings to Hit Target Dims Like 256",[22,4109,4110,4111,4114,4115,4118,4119,4122,4123,142,4126,4129,4130,4133,4134,4137],{},"Fixed-dim models like databricks-gte-large-en (always 1024) force re-embedding for size changes. Qwen3-Embedding-0.6B uses Matryoshka Representation Learning (MRL) to pack signal into early dims, enabling safe truncation to any power-of-2 (32-1024). Managed Delta sync ignores ",[37,4112,4113],{},"dimensions"," param, always outputs 1024—use self-managed: pre-compute with API (",[37,4116,4117],{},"{\"input\": [text], \"dimensions\": 256}","), UDF to Delta table (",[37,4120,4121],{},"chunk_embedding","), then index with ",[37,4124,4125],{},"embedding_vector_column",[37,4127,4128],{},"embedding_dimension=256",". Query same way: embed query at 256, pass vector to ",[37,4131,4132],{},"similarity_search",". For prod scale, swap UDF for ",[37,4135,4136],{},"ai_query()"," batch inference.",[17,4139,4141],{"id":4140},"rerank-top-50-ann-hits-for-15pt-recall-gain-over-vector-distance-alone","Rerank Top-50 ANN Hits for 15pt Recall Gain Over Vector Distance Alone",[22,4143,4144,4145,4148],{},"ANN cosine similarity doesn't guarantee query relevance—close vectors (e.g., \"sensor calibration\" vs. \"actuator recalibration\") rank by distance, not utility. Databricks Reranker re-scores top-50 with query-aware model: 74% ANN-only recall@10 jumps to 89% (+15pts), beating cloud rivals by 10pts. Enable via ",[37,4146,4147],{},"reranker={\"model\": \"databricks_reranker\", \"parameters\": {\"columns_to_rerank\": [\"chunk\", \"doc_summary\"]}}"," (first 2000 chars, richest first; order matters). Adds ~1.5s latency—skip only for \u003C200ms needs, >5 QPS unscaled, or non-RAG search. Production stack: Qwen3@256dims (self-managed), ANN HNSW, triggered Delta sync, rerank metadata.",{"title":148,"searchDepth":149,"depth":149,"links":4150},[4151,4152,4153],{"id":4099,"depth":149,"text":4100},{"id":4106,"depth":149,"text":4107},{"id":4140,"depth":149,"text":4141},[155],{"content_references":4156,"triage":4167},[4157,4160,4162,4165],{"type":4158,"title":4159,"context":165},"tool","databricks-qwen3-embedding-0-6b",{"type":4158,"title":4161,"context":165},"databricks_reranker",{"type":4158,"title":4163,"context":4164},"databricks-gte-large-en","mentioned",{"type":4158,"title":4166,"context":4164},"databricks-bge-large-en",{"relevance":4168,"novelty":184,"quality":184,"actionability":4168,"composite":4169,"reasoning":4170},5,4.55,"Category: AI & LLMs. The article provides practical insights on optimizing embedding dimensions and reranking techniques for improved recall in AI applications, addressing specific pain points for developers integrating AI features. It includes actionable steps for implementation, such as using self-managed embeddings and reranking methods, making it highly relevant and actionable for the target audience.","\u002Fsummaries\u002F41ef3a9324aac236-databricks-rag-low-dim-qwen3-rerank-for-89-recall-summary","2026-05-05 05:52:27","2026-05-05 16:09:29",{"title":4088,"description":148},{"loc":4171},"41ef3a9324aac236","Towards AI","https:\u002F\u002Fpub.towardsai.net\u002Fvector-search-done-right-best-practices-qwen3-dimension-control-and-why-reranking-is-e021e18be13c?source=rss----98111c9905da---4","summaries\u002F41ef3a9324aac236-databricks-rag-low-dim-qwen3-rerank-for-89-recall--summary",[199,200,202,4181],"ai-automation","Minimize embedding dims to 256 with Qwen3 MRL (self-managed path), set num_results=50, always rerank ANN top-50 candidates for +15pts recall@10 over 74% baseline.",[202,4181],"16bkywE38TpOvJuszr5Lm3OSpE0lMxkPQA4iNlm95A8",{"id":4186,"title":4187,"ai":4188,"body":4193,"categories":4244,"created_at":156,"date_modified":156,"description":148,"extension":157,"faq":156,"featured":158,"kicker_label":156,"meta":4245,"navigation":187,"path":4254,"published_at":4255,"question":156,"scraped_at":4256,"seo":4257,"sitemap":4258,"source_id":4259,"source_name":4260,"source_type":195,"source_url":4261,"stem":4262,"tags":4263,"thumbnail_url":156,"tldr":4265,"tweet":156,"unknown_tags":4266,"__hash__":4267},"summaries\u002Fsummaries\u002Fadd9ec06f3d8b78d-decoder-only-transformers-drive-gpt-scaling-summary.md","Decoder-Only Transformers Drive GPT Scaling",{"provider":7,"model":4090,"input_tokens":4189,"output_tokens":4190,"processing_time_ms":4191,"cost_usd":4192},8457,1685,17671,0.00202705,{"type":14,"value":4194,"toc":4238},[4195,4199,4202,4205,4209,4212,4215,4219,4222,4225,4229,4232,4235],[17,4196,4198],{"id":4197},"self-attention-enables-parallel-long-range-dependencies","Self-Attention Enables Parallel Long-Range Dependencies",[22,4200,4201],{},"Transformers replace RNNs' sequential processing, which suffers vanishing gradients beyond 50-100 words, with self-attention that computes direct relationships between all token pairs simultaneously. For a token like \"it\" in \"The cat sat on the mat and looked at the fishbowl because it was hungry,\" every prior word votes on relevance via query-key dot products scaled by embed_size^{-0.5}, softmax-normalized, and applied to values. This parallelization trains across thousands of GPUs.",[22,4203,4204],{},"GPT's decoder-only design strips away the encoder, applying a causal mask to block future tokens, forcing rich representations solely from predicting the next token. GPT-1 (117M params, 12 layers) showed modest NLP scores, but GPT-2 (1.5B params) gained zero-shot abilities like summarization via prompting. GPT-3 (175B params, 96 layers) added in-context learning from prompt examples without fine-tuning. Deeper layers progress from syntax (early) to reasoning and world models (late). This simplicity scales better than encoder-decoder setups, avoiding cross-attention overhead.",[17,4206,4208],{"id":4207},"moe-and-test-time-compute-scale-beyond-dense-models","MoE and Test-Time Compute Scale Beyond Dense Models",[22,4210,4211],{},"Dense models activate all parameters per token, making trillions unaffordable. Mixture of Experts (MoE) routes each token to 2-8 specialized experts out of 128+, activating ~5% of weights—e.g., DeepSeek-V3 uses 37B active out of 671B total, trained for $5.6M on 2,048 H800 GPUs, matching GPT-4. Multi-Head Latent Attention (MLA) compresses KV cache to cut memory bandwidth. Tradeoffs include expert collapse (router overloads few experts) and full-model memory needs despite sparse activation.",[22,4213,4214],{},"o1 introduced test-time compute: generate internal reasoning chains (30s for hard problems), backtrack dead ends, and refine via RL on verifiable rewards like math solutions. This outperforms larger instant-response models, decoupling ability from size. GPT-5 routes simple queries fast (System 1) and complex ones deeply (System 2). Open models like DeepSeek-R1 replicate this.",[17,4216,4218],{"id":4217},"multimodal-fusion-and-real-world-impacts","Multimodal Fusion and Real-World Impacts",[22,4220,4221],{},"Early fusion embeds vision tokens from Vision Transformers (e.g., MetaCLIP) into the same space as text, enabling unified attention across modalities—no separate captioning. Models like LLaMA 4, Qwen-VL handle charts, 3D spatial reasoning (GLM-4.5V's rotated positional encoding). This yields native cross-modal reasoning, e.g., diagnosing X-rays directly.",[22,4223,4224],{},"Applications: Harvey AI (RAG + fine-tuned GPT-4) cuts legal review 40-60%; GPT-4.1 hits 54.6% on SWE-bench (21.4pp over GPT-4o), ingesting 1M-token codebases; 75% medical accuracy accelerates drug discovery. Open weights (LLaMA, DeepSeek) ensure data sovereignty.",[17,4226,4228],{"id":4227},"implement-mini-gpt-from-scratch-in-pytorch","Implement Mini-GPT from Scratch in PyTorch",[22,4230,4231],{},"Build a character-level GPT: Tokenizer maps unique chars to indices (vocab_size ~50). SelfAttention computes QKV projections, scores = (Q @ K.T) * scale, weights = softmax(scores), out = weights @ V. TransformerBlock adds residual attention + FFN (4x expand, ReLU), LayerNorm post each.",[22,4233,4234],{},"MiniGPT stacks NUM_LAYERS=2 blocks on token + positional embeddings (BLOCK_SIZE=32), outputs logits via linear to vocab_size. Train on dataset.txt: batch BATCH_SIZE=16 sequences, predict next token with CrossEntropyLoss, Adam at 3e-4, 20 EPOCHS. Generation: sample from last-token softmax via multinomial, append up to 100 tokens from context like \"AI is\".",[22,4236,4237],{},"Project structure: data\u002Fdataset.txt, model\u002F{tokenizer,attention,transformer,gpt}.py, train.py saves model.pth, generate.py loads\u002Finfers. Config: EMBED_SIZE=64, NUM_HEADS=4 (implied in attention). This replicates core logic scalably.",{"title":148,"searchDepth":149,"depth":149,"links":4239},[4240,4241,4242,4243],{"id":4197,"depth":149,"text":4198},{"id":4207,"depth":149,"text":4208},{"id":4217,"depth":149,"text":4218},{"id":4227,"depth":149,"text":4228},[155],{"content_references":4246,"triage":4251},[4247],{"type":162,"title":4248,"author":4249,"context":4250},"Attention Is All You Need","Ashish Vaswani’s team","cited",{"relevance":184,"novelty":183,"quality":184,"actionability":149,"composite":4252,"reasoning":4253},3.4,"Category: AI & LLMs. The article provides a detailed explanation of the architecture behind GPT models, which is relevant for developers looking to integrate AI features. However, while it offers insights into model design, it lacks practical applications or frameworks that the audience can directly implement.","\u002Fsummaries\u002Fadd9ec06f3d8b78d-decoder-only-transformers-drive-gpt-scaling-summary","2026-04-18 19:32:29","2026-04-19 01:22:04",{"title":4187,"description":148},{"loc":4254},"add9ec06f3d8b78d","Python in Plain English","https:\u002F\u002Fpython.plainenglish.io\u002Fthe-architecture-behind-gpt-models-de61992c088a?source=rss----78073def27b8---4","summaries\u002Fadd9ec06f3d8b78d-decoder-only-transformers-drive-gpt-scaling-summary",[4264,199,200,201],"llm","GPT models use decoder-only transformers with causal masking for next-token prediction, enabling emergent zero-shot and in-context learning when scaled massively, now enhanced by MoE for efficiency and reasoning chains.",[],"FYS789V3fqVrHXGVHROYyiskRFH6nT84_QPvh_I63p0",{"id":4269,"title":4270,"ai":4271,"body":4276,"categories":4328,"created_at":156,"date_modified":156,"description":148,"extension":157,"faq":156,"featured":158,"kicker_label":156,"meta":4329,"navigation":187,"path":4330,"published_at":4331,"question":156,"scraped_at":156,"seo":4332,"sitemap":4333,"source_id":4334,"source_name":4335,"source_type":195,"source_url":4336,"stem":4337,"tags":4338,"thumbnail_url":156,"tldr":4339,"tweet":156,"unknown_tags":4340,"__hash__":4341},"summaries\u002Fsummaries\u002Fmicrogpt-py-full-gpt-in-300-lines-of-pure-python-summary.md","microgpt.py: Full GPT in 300 Lines of Pure Python",{"provider":7,"model":4090,"input_tokens":4272,"output_tokens":4273,"processing_time_ms":4274,"cost_usd":4275},11786,1242,8684,0.0029557,{"type":14,"value":4277,"toc":4323},[4278,4282,4297,4301,4308,4312],[17,4279,4281],{"id":4280},"custom-autograd-engine-powers-end-to-end-training","Custom Autograd Engine Powers End-to-End Training",[22,4283,4284,4285,4288,4289,4292,4293,4296],{},"Implements automatic differentiation via ",[37,4286,4287],{},"Value"," class with slots for efficiency. Supports add, mul, pow, log, exp, ReLU, and backward via topological sort on computation graph. Chain rule propagates gradients recursively: ",[37,4290,4291],{},"child.grad += local_grad * v.grad",". Enables full forward\u002Fbackward without libraries. For a names dataset (32k lines from ",[37,4294,4295],{},"names.txt","), builds char-level tokenizer: unique chars (vocab_size=~30+1 BOS token). Model params (~10k total): 1 layer, n_embd=16, block_size=16, n_head=4 (head_dim=4). Weights initialized Gaussian std=0.08. Embeddings: wte (vocab x 16), wpe (16 x 16), lm_head (vocab x 16). Per layer: QKV (4x 16x16), Wo (16x16), MLP fc1 (64x16), fc2 (16x64).",[17,4298,4300],{"id":4299},"gpt-architecture-mirrors-gpt-2-essentials","GPT Architecture Mirrors GPT-2 Essentials",[22,4302,4303,4304,4307],{},"Forward pass: token+pos embeds → RMSNorm → residual blocks. Attention: raw dot-product (scaled by 1\u002Fsqrt(head_dim)), softmax weights → weighted V sum → Wo projection. Causal via key\u002Fvalue history append (no mask). MLP: RMSNorm → fc1 → ReLU → fc2 → residual. Final lm_head logits → softmax probs. Uses RMSNorm (",[37,4305,4306],{},"scale = (mean(x^2)+eps)^-0.5",") over LayerNorm, ReLU over GeLU, no biases. Keys\u002Fvalues persist across positions for KV cache simulation. Loss: average -log P(next_token) over sequence (BOS-wrapped docs, up to block_size=16).",[17,4309,4311],{"id":4310},"adam-training-inference-in-1000-steps","Adam Training + Inference in 1000 Steps",[22,4313,4314,4315,4319,4320,4322],{},"Shuffles 32k names, cycles through docs. Per step: tokenize ",[4316,4317,4318],"span",{},"BOS"," + chars + ",[4316,4321,4318],{},", forward all positions (building KV cache), average cross-entropy loss → backward → Adam update (lr=0.01 linear decay to 0, β1=0.85, β2=0.99). Prints loss (drops from ~3 to ~1.5 typically). Inference: start BOS, sample argmax-probs (temp=0.5) until BOS, yields plausible names like 'korsal' after training. Demonstrates: core GPT is simple; libs optimize speed\u002Fscale. Trade-off: slow (minutes on CPU), but reveals every op.",{"title":148,"searchDepth":149,"depth":149,"links":4324},[4325,4326,4327],{"id":4280,"depth":149,"text":4281},{"id":4299,"depth":149,"text":4300},{"id":4310,"depth":149,"text":4311},[155],{},"\u002Fsummaries\u002Fmicrogpt-py-full-gpt-in-300-lines-of-pure-python-summary","2026-04-08 21:21:19",{"title":4270,"description":148},{"loc":4330},"56d2bdaaa16d5c3b","Andrej Karpathy Gists","https:\u002F\u002Funknown","summaries\u002Fmicrogpt-py-full-gpt-in-300-lines-of-pure-python-summary",[4264,199,200,201],"Trains a tiny GPT on names dataset using custom autograd—no deps, no PyTorch—to generate realistic names, distilling the core transformer algorithm.",[],"3fO1PHuRnDxVHEXFsDwlj_bugbD79pZ1c6UEJVeKQE8",{"id":4343,"title":4344,"ai":4345,"body":4350,"categories":4378,"created_at":156,"date_modified":156,"description":148,"extension":157,"faq":156,"featured":158,"kicker_label":156,"meta":4379,"navigation":187,"path":4389,"published_at":4390,"question":156,"scraped_at":4391,"seo":4392,"sitemap":4393,"source_id":4394,"source_name":4177,"source_type":195,"source_url":4395,"stem":4396,"tags":4397,"thumbnail_url":156,"tldr":4398,"tweet":156,"unknown_tags":4399,"__hash__":4400},"summaries\u002Fsummaries\u002F092f953f13e749e1-reproduce-2011-sentiment-word-vectors-in-python-summary.md","Reproduce 2011 Sentiment Word Vectors in Python",{"provider":7,"model":4090,"input_tokens":4346,"output_tokens":4347,"processing_time_ms":4348,"cost_usd":4349},3933,1516,16200,0.00152195,{"type":14,"value":4351,"toc":4373},[4352,4356,4359,4363,4366,4370],[17,4353,4355],{"id":4354},"elegant-core-technique-semantic-learning-from-ratings","Elegant Core Technique: Semantic Learning from Ratings",[22,4357,4358],{},"Maas et al. (2011) train sentiment-specific word vectors directly from unlabeled IMDb movie reviews paired with star ratings (1-10 scale). Words co-occurring in high-rated reviews pull closer in vector space; low-rated push apart. This creates representations capturing sentiment polarity without explicit labels. Final classification uses linear SVM on averaged review vectors, achieving strong accuracy through interpretable, low-dimensional embeddings. Author notes its logistic regression-like simplicity: powerful when data aligns with task, avoiding black-box complexity.",[17,4360,4362],{"id":4361},"reproduction-insights-and-modern-relevance","Reproduction Insights and Modern Relevance",[22,4364,4365],{},"Reproducing the paper in Python reveals its enduring strength – elegant semantic learning outperforms hype-driven alternatives in targeted tasks like sentiment. Author challenges original methods, tests against other representations (including LLMs), and automates full pipeline. Trade-off: excels on review-style text but needs domain data; not general-purpose like transformers. GitHub repo provides end-to-end code for immediate use or extension.",[17,4367,4369],{"id":4368},"practical-takeaways-for-builders","Practical Takeaways for Builders",[22,4371,4372],{},"Start with this for sentiment features in products: download IMDb data, train vectors via contrastive objective on ratings, classify with SVM. Scales to custom corpora (e.g., product feedback). Compares favorably to LLMs on cost\u002Finterpretability; use as baseline before deploying APIs. Avoids overfitting by leveraging vast unlabeled text – key for production ML pipelines.",{"title":148,"searchDepth":149,"depth":149,"links":4374},[4375,4376,4377],{"id":4354,"depth":149,"text":4355},{"id":4361,"depth":149,"text":4362},{"id":4368,"depth":149,"text":4369},[265],{"content_references":4380,"triage":4387},[4381,4384],{"type":162,"title":4382,"author":4383,"context":4164},"Learning Word Vectors for Sentiment Analysis","Maas et al.",{"type":179,"title":4385,"url":4386,"context":4164},"Sentiment_analysis","https:\u002F\u002Fgithub.com\u002FJumbong\u002FSentiment_analysis",{"relevance":4168,"novelty":184,"quality":184,"actionability":4168,"composite":4169,"reasoning":4388},"Category: AI & LLMs. The article provides a practical method for building sentiment-aware word embeddings, which is directly applicable for product builders looking to integrate sentiment analysis into their AI-powered products. It includes actionable steps and a GitHub repository for implementation, making it highly relevant and actionable.","\u002Fsummaries\u002F092f953f13e749e1-reproduce-2011-sentiment-word-vectors-in-python-summary","2026-05-10 00:01:00","2026-05-10 15:26:28",{"title":4344,"description":148},{"loc":4389},"092f953f13e749e1","https:\u002F\u002Fpub.towardsai.net\u002Flearning-word-vectors-for-sentiment-analysis-a-python-reproduction-f8c8c77df38f?source=rss----98111c9905da---4","summaries\u002F092f953f13e749e1-reproduce-2011-sentiment-word-vectors-in-python-summary",[199,200],"Build sentiment-aware word embeddings from IMDb reviews via semantic learning with star ratings and linear SVM classification, reproducing Maas et al. (2011) – simple method rivals modern LLMs.",[],"v2XTBE5rFNMZcIts4tjxKmc0d5a3j51Waw-d4ggTQcI"]