This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.
%I A337848 #17 Jun 17 2022 03:41:52 %S A337848 73,241,2593,5113,8713,18433,53593,55681,86113,102241,126337,127873, %T A337848 158113,181721,184369,186049,208393,219313,221537,241921,262657, %U A337848 267913,282313,314161,314401,341641,362521,398441,415873,450913,534241,619921,651169,731881,953473,1045801,1153441,1294177,1554281,2023921,2162401,2345401,2533681 %N A337848 Odd integers k>=5 such that 2^((k-1)/2)-1 == 0 (mod k*(k-3)/2). %C A337848 Computed terms are prime. Is it always the case? If not it would be interesting to compute the pseudoprimes. %C A337848 1234125721 = 24841*49681, 4294901761 = 193*22253377, 6602556241 = 57457*114913 are composite counterexamples to the assumption that all terms are prime. - _Hugo Pfoertner_, Sep 26 2020 %C A337848 These are a(420), a(705) and a(830). Together with a(956) = 10025492401 = 101 * 701 * 141601 they are the first 4 composite terms. - _Amiram Eldar_, Jun 17 2022 %H A337848 Amiram Eldar, <a href="/A337848/b337848.txt">Table of n, a(n) for n = 1..1000</a> %t A337848 Select[Range[5, 10^6, 2], PowerMod[2, (# - 1)/2, #*(# - 3)/2] == 1 &] (* _Amiram Eldar_, Sep 26 2020 *) %o A337848 (PARI) is(n) = n%2 && n>=3 && Mod(2, n*(n-3)/2)^((n-1)/2) ==1 %Y A337848 Cf. A337818. %K A337848 nonn %O A337848 1,1 %A A337848 _Benoit Cloitre_, Sep 26 2020