training_metadata
training_date: "2026-03-25T16:07:09Z" max_samples: 25000 models: ar/cp1256: language: ar encoding: cp1256 samples_used: 25000 bigram_entries: 1945 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 ar/cp720: language: ar encoding: cp720 samples_used: 25000 bigram_entries: 1859 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 ar/cp864: language: ar encoding: cp864 samples_used: 25000 bigram_entries: 1587 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 ar/iso8859-6: language: ar encoding: iso8859-6 samples_used: 25000 bigram_entries: 1865 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 ar/utf-8: language: ar encoding: utf-8 samples_used: 25000 bigram_entries: 725 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 be/cp1251: language: be encoding: cp1251 samples_used: 25000 bigram_entries: 2375 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 be/cp855: language: be encoding: cp855 samples_used: 25000 bigram_entries: 2336 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 be/cp866: language: be encoding: cp866 samples_used: 25000 bigram_entries: 2187 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 be/iso8859-5: language: be encoding: iso8859-5 samples_used: 25000 bigram_entries: 2251 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 be/mac-cyrillic: language: be encoding: mac-cyrillic samples_used: 25000 bigram_entries: 2360 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 be/utf-8: language: be encoding: utf-8 samples_used: 25000 bigram_entries: 739 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 bg/cp1251: language: bg encoding: cp1251 samples_used: 25000 bigram_entries: 2063 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 bg/cp855: language: bg encoding: cp855 samples_used: 25000 bigram_entries: 2007 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 bg/cp866: language: bg encoding: cp866 samples_used: 25000 bigram_entries: 1998 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 bg/iso8859-5: language: bg encoding: iso8859-5 samples_used: 25000 bigram_entries: 2002 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 bg/mac-cyrillic: language: bg encoding: mac-cyrillic samples_used: 25000 bigram_entries: 2054 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 bg/utf-8: language: bg encoding: utf-8 samples_used: 25000 bigram_entries: 658 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 br/cp1140: language: br encoding: cp1140 samples_used: 25000 bigram_entries: 1435 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/cp1252: language: br encoding: cp1252 samples_used: 25000 bigram_entries: 951 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/cp500: language: br encoding: cp500 samples_used: 25000 bigram_entries: 1430 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/cp850: language: br encoding: cp850 samples_used: 25000 bigram_entries: 921 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/cp858: language: br encoding: cp858 samples_used: 25000 bigram_entries: 921 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/hp-roman8: language: br encoding: hp-roman8 samples_used: 25000 bigram_entries: 922 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/iso8859-1: language: br encoding: iso8859-1 samples_used: 25000 bigram_entries: 921 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/iso8859-14: language: br encoding: iso8859-14 samples_used: 25000 bigram_entries: 910 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/iso8859-15: language: br encoding: iso8859-15 samples_used: 25000 bigram_entries: 920 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/mac-roman: language: br encoding: mac-roman samples_used: 25000 bigram_entries: 950 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 br/utf-8: language: br encoding: utf-8 samples_used: 25000 bigram_entries: 1034 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 9224 cs/cp1250: language: cs encoding: cp1250 samples_used: 25000 bigram_entries: 1754 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 cs/cp852: language: cs encoding: cp852 samples_used: 25000 bigram_entries: 1711 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 cs/iso8859-2: language: cs encoding: iso8859-2 samples_used: 25000 bigram_entries: 1707 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 cs/mac-latin2: language: cs encoding: mac-latin2 samples_used: 25000 bigram_entries: 1749 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 cs/utf-8: language: cs encoding: utf-8 samples_used: 25000 bigram_entries: 1450 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 cy/cp1140: language: cy encoding: cp1140 samples_used: 25000 bigram_entries: 2231 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 cy/cp1252: language: cy encoding: cp1252 samples_used: 25000 bigram_entries: 886 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 cy/cp500: language: cy encoding: cp500 samples_used: 25000 bigram_entries: 2215 sources: culturax: 25000 madlad400: 0 wikipedia: 0 test_articles_excluded: 20 cy/cp850: language: cy encoding: cp850 samples_used: 25000 bigram_entries: 861 sources: culturax: 25000 mad…