diff --git "a/data/leaderboard_json/nlg.json" "b/data/leaderboard_json/nlg.json" new file mode 100644--- /dev/null +++ "b/data/leaderboard_json/nlg.json" @@ -0,0 +1,3350 @@ +{ + "task": "NLG", + "subtasks": { + "MT(xx-en/fr)": { + "datasets": { + "flores - xx_en": { + "languages": [ + "fra_Latn", + "aeb_Arab", + "afr_Latn", + "aka_Latn", + "amh_Ethi", + "ary_Arab", + "arz_Arab", + "bam_Latn", + "bem_Latn", + "cjk_Latn", + "dik_Latn", + "dyu_Latn", + "ewe_Latn", + "fon_Latn", + "fuv_Latn", + "gaz_Latn", + "hau_Latn", + "ibo_Latn", + "kab_Latn", + "kam_Latn", + "kbp_Latn", + "kea_Latn", + "kik_Latn", + "kin_Latn", + "kmb_Latn", + "knc_Arab", + "knc_Latn", + "kon_Latn", + "lin_Latn", + "lua_Latn", + "lug_Latn", + "luo_Latn", + "mos_Latn", + "nso_Latn", + "nus_Latn", + "nya_Latn", + "plt_Latn", + "run_Latn", + "sag_Latn", + "sna_Latn", + "som_Latn", + "sot_Latn", + "ssw_Latn", + "swh_Latn", + "taq_Latn", + "taq_Tfng", + "tir_Ethi", + "tsn_Latn", + "tso_Latn", + "tum_Latn", + "twi_Latn", + "tzm_Tfng", + "umb_Latn", + "wol_Latn", + "xho_Latn", + "yor_Latn", + "zul_Latn" + ], + "scores": { + "AfroLLaMa 8B": [ + 14.0, + 6.0, + 21.1, + 8.4, + 5.2, + 7.4, + 8.0, + 8.9, + 10.5, + 8.8, + 13.0, + 8.6, + 8.2, + 7.0, + 12.9, + 8.6, + 8.8, + 6.1, + 7.1, + 8.6, + 8.2, + 12.4, + 9.6, + 12.8, + 9.2, + 7.6, + 13.6, + 10.6, + 9.9, + 11.9, + 11.6, + 9.2, + 9.2, + 15.0, + 8.0, + 11.2, + 14.9, + 10.8, + 8.0, + 10.8, + 7.3, + 15.4, + 11.8, + 13.2, + 9.6, + 6.5, + 4.6, + 11.3, + 10.3, + 9.4, + 9.5, + 6.9, + 9.6, + 7.4, + 12.2, + 7.8, + 18.3 + ], + "LLaMAX3 8B": [ + 67.0, + 52.3, + 73.9, + 27.9, + 47.8, + 48.7, + 54.6, + 23.8, + 29.2, + 25.0, + 23.2, + 23.1, + 24.0, + 22.2, + 24.9, + 27.7, + 50.8, + 44.7, + 21.3, + 25.6, + 23.7, + 48.5, + 25.9, + 41.7, + 24.8, + 16.6, + 24.7, + 27.9, + 34.5, + 27.2, + 33.2, + 24.1, + 22.3, + 35.3, + 20.7, + 44.6, + 45.0, + 34.2, + 24.6, + 45.0, + 47.0, + 36.1, + 36.2, + 60.9, + 24.2, + 16.9, + 28.0, + 32.8, + 28.0, + 34.3, + 29.0, + 17.5, + 24.0, + 31.3, + 50.5, + 34.5, + 51.8 + ], + "LLaMa2 7b": [ + 61.1, + 35.7, + 61.2, + 22.4, + 14.5, + 30.9, + 36.7, + 13.1, + 18.8, + 22.1, + 20.4, + 18.7, + 20.0, + 18.4, + 19.7, + 20.8, + 16.6, + 16.8, + 17.2, + 20.1, + 20.1, + 37.4, + 19.3, + 21.2, + 17.4, + 7.9, + 10.0, + 19.4, + 13.0, + 22.2, + 20.2, + 20.4, + 17.2, + 20.7, + 14.0, + 22.5, + 19.3, + 19.5, + 21.0, + 21.4, + 19.9, + 22.5, + 21.5, + 26.5, + 21.8, + 10.5, + 9.0, + 22.1, + 18.9, + 19.3, + 20.2, + 4.3, + 17.0, + 19.1, + 16.6, + 16.2, + 12.0 + ], + "LLaMa3.1 70B": [ + 71.0, + 58.4, + 77.6, + 34.4, + 48.2, + 56.9, + 61.0, + 24.8, + 31.9, + 25.4, + 24.6, + 23.8, + 25.6, + 21.8, + 24.5, + 30.6, + 53.8, + 48.0, + 26.7, + 28.8, + 25.8, + 62.1, + 32.0, + 44.5, + 25.8, + 17.3, + 25.5, + 30.0, + 32.6, + 28.7, + 36.3, + 25.1, + 23.4, + 40.5, + 20.2, + 37.7, + 46.4, + 36.8, + 25.9, + 38.8, + 43.9, + 41.7, + 36.9, + 65.7, + 25.4, + 16.9, + 31.0, + 40.2, + 37.3, + 34.1, + 36.7, + 19.4, + 25.3, + 29.1, + 43.4, + 37.3, + 47.3 + ], + "LLaMa3.1 8B": [ + 68.1, + 53.9, + 74.3, + 26.8, + 31.6, + 51.1, + 56.3, + 18.0, + 24.4, + 19.0, + 19.4, + 17.3, + 18.4, + 14.4, + 19.2, + 20.4, + 44.8, + 38.0, + 18.0, + 22.9, + 18.5, + 54.4, + 23.6, + 35.1, + 18.6, + 12.1, + 17.2, + 22.6, + 26.0, + 20.5, + 27.8, + 18.9, + 16.4, + 29.0, + 14.4, + 30.6, + 34.2, + 29.4, + 18.8, + 29.2, + 31.7, + 28.6, + 26.2, + 57.8, + 19.5, + 11.5, + 16.2, + 29.5, + 25.7, + 25.8, + 29.4, + 11.5, + 19.3, + 21.8, + 32.8, + 29.9, + 32.5 + ], + "LLaMa3 8B": [ + 67.4, + 52.2, + 72.0, + 27.2, + 23.7, + 48.1, + 54.5, + 21.3, + 23.7, + 20.1, + 23.5, + 20.5, + 20.3, + 19.8, + 21.3, + 21.0, + 38.2, + 32.2, + 16.6, + 23.7, + 21.5, + 48.9, + 24.8, + 31.1, + 22.6, + 15.2, + 22.1, + 25.7, + 25.8, + 23.4, + 27.2, + 21.6, + 18.4, + 27.0, + 19.9, + 28.7, + 29.3, + 22.5, + 20.4, + 26.8, + 27.7, + 26.6, + 22.9, + 55.4, + 23.7, + 17.9, + 18.0, + 26.7, + 21.8, + 26.0, + 29.3, + 17.6, + 22.3, + 22.6, + 29.3, + 27.0, + 28.0 + ], + "Aya-101 13B": [ + 65.0, + 53.6, + 71.8, + 37.0, + 54.1, + 51.2, + 55.2, + 28.2, + 37.2, + 26.1, + 23.4, + 22.9, + 32.0, + 22.2, + 24.6, + 36.9, + 53.3, + 49.4, + 21.0, + 28.1, + 22.7, + 57.7, + 30.0, + 50.7, + 27.7, + 14.3, + 25.4, + 33.2, + 36.4, + 33.6, + 37.0, + 27.1, + 23.2, + 53.8, + 19.2, + 48.3, + 52.9, + 44.4, + 30.7, + 47.9, + 49.1, + 55.4, + 45.8, + 60.4, + 26.0, + 16.3, + 44.9, + 47.6, + 39.7, + 42.5, + 39.3, + 22.4, + 24.9, + 25.6, + 54.7, + 39.7, + 55.5 + ], + "Gemma1.1 7b": [ + 13.8, + 1.1, + 21.3, + 0.9, + 10.8, + 0.7, + 0.8, + 0.4, + 1.8, + 0.2, + 0.3, + 0.4, + 1.1, + 0.2, + 0.4, + 2.0, + 3.8, + 5.5, + 1.0, + 0.8, + 0.2, + 0.7, + 1.1, + 4.9, + 0.8, + 0.6, + 1.1, + 3.8, + 5.0, + 0.2, + 1.8, + 0.3, + 0.4, + 2.4, + 1.4, + 0.6, + 0.0, + 1.7, + 0.6, + 1.3, + 3.8, + 1.0, + 0.2, + 10.5, + 1.3, + 0.3, + 5.5, + 2.8, + 2.5, + 0.1, + 8.7, + 0.4, + 0.4, + 3.2, + 3.2, + 2.6, + 2.7 + ], + "Gemma2 27b": [ + 65.1, + 53.6, + 71.0, + 30.5, + 42.7, + 51.4, + 55.4, + 22.0, + 31.5, + 23.5, + 21.0, + 21.4, + 21.8, + 19.7, + 21.5, + 24.7, + 47.1, + 40.2, + 22.9, + 24.9, + 22.9, + 54.8, + 27.6, + 41.1, + 23.1, + 15.6, + 22.5, + 27.2, + 29.5, + 26.2, + 31.5, + 22.5, + 20.6, + 39.4, + 19.5, + 37.0, + 42.5, + 33.0, + 23.0, + 39.8, + 41.8, + 41.5, + 35.1, + 61.8, + 23.1, + 18.4, + 27.8, + 37.4, + 32.3, + 30.6, + 31.6, + 19.1, + 22.6, + 25.2, + 44.8, + 32.3, + 45.4 + ], + "Gemma2 9b": [ + 64.6, + 52.0, + 70.4, + 29.5, + 42.3, + 49.4, + 53.8, + 21.1, + 29.7, + 22.6, + 20.6, + 20.6, + 20.8, + 19.3, + 21.7, + 24.4, + 45.6, + 37.4, + 22.1, + 24.0, + 22.6, + 51.3, + 26.3, + 38.5, + 22.8, + 14.6, + 21.9, + 26.7, + 27.3, + 25.2, + 30.1, + 22.3, + 19.5, + 35.7, + 18.6, + 34.8, + 38.5, + 31.3, + 22.4, + 37.4, + 39.7, + 38.3, + 32.4, + 59.3, + 22.1, + 16.6, + 27.3, + 34.1, + 29.1, + 29.4, + 31.4, + 19.0, + 21.6, + 23.9, + 43.4, + 30.6, + 43.6 + ], + "Gemini 1.5 pro": [ + 66.4, + 57.3, + 72.8, + 39.4, + 60.9, + 55.4, + 57.8, + 28.3, + 44.8, + 26.4, + 23.8, + 25.4, + 35.6, + 26.9, + 26.4, + 43.9, + 55.9, + 51.8, + 32.1, + 31.3, + 29.5, + 67.4, + 38.3, + 54.4, + 30.8, + 17.8, + 24.4, + 41.7, + 45.3, + 38.1, + 45.0, + 27.0, + 23.9, + 54.8, + 22.5, + 49.2, + 54.9, + 47.4, + 33.4, + 49.7, + 53.1, + 58.0, + 48.3, + 65.3, + 25.7, + 21.0, + 47.8, + 48.4, + 48.4, + 42.1, + 42.8, + 24.6, + 28.7, + 30.9, + 56.4, + 43.6, + 58.3 + ], + "GPT-4o (Aug)": [ + 66.0, + 57.5, + 73.0, + 40.0, + 53.4, + 55.8, + 57.8, + 28.8, + 41.2, + 26.2, + 24.8, + 25.4, + 28.2, + 22.3, + 28.4, + 48.1, + 54.5, + 50.7, + 30.5, + 35.2, + 23.5, + 67.4, + 40.8, + 53.8, + 26.8, + 17.8, + 25.6, + 39.6, + 44.7, + 32.2, + 47.2, + 38.8, + 23.4, + 56.8, + 22.6, + 48.2, + 53.7, + 46.2, + 26.4, + 48.3, + 52.7, + 57.5, + 47.6, + 64.1, + 26.1, + 17.6, + 37.0, + 49.2, + 48.8, + 41.4, + 41.8, + 19.0, + 27.5, + 40.7, + 56.3, + 44.8, + 58.0 + ] + } + }, + "ntrex - xx_en": { + "languages": [ + "fra_Latn", + "afr_Latn", + "amh_Ethi", + "bem_Latn", + "ewe_Latn", + "hau_Latn", + "ibo_Latn", + "kin_Latn", + "mlg_Latn", + "nde_Latn", + "nso_Latn", + "nya_Latn", + "orm_Ethi", + "sna_Latn", + "som_Latn", + "ssw_Latn", + "swa_Latn", + "tir_Ethi", + "tsn_Latn", + "ven_Latn", + "wol_Latn", + "xho_Latn", + "yor_Latn", + "zul_Latn" + ], + "scores": { + "AfroLLaMa 8B": [ + 11.5, + 16.9, + 5.6, + 10.2, + 7.6, + 7.6, + 5.6, + 12.1, + 9.5, + 12.1, + 12.5, + 9.6, + 6.4, + 8.3, + 7.0, + 9.6, + 10.1, + 4.8, + 10.5, + 9.3, + 6.8, + 9.2, + 7.1, + 11.7 + ], + "LLaMAX3 8B": [ + 61.4, + 72.6, + 40.9, + 31.5, + 26.5, + 52.7, + 47.5, + 42.6, + 43.6, + 42.2, + 35.8, + 46.9, + 26.6, + 47.8, + 52.6, + 38.0, + 59.9, + 28.7, + 36.0, + 27.5, + 30.3, + 49.1, + 35.9, + 49.1 + ], + "LLaMa2 7b": [ + 53.9, + 57.8, + 16.5, + 20.1, + 22.1, + 17.6, + 19.9, + 22.6, + 19.5, + 20.9, + 23.3, + 25.0, + 20.0, + 22.8, + 22.0, + 22.4, + 29.3, + 11.0, + 18.4, + 14.1, + 20.1, + 16.3, + 18.2, + 16.4 + ], + "LLaMa3.1 70B": [ + 64.9, + 76.9, + 43.8, + 34.9, + 27.8, + 56.9, + 52.2, + 46.2, + 44.1, + 40.6, + 41.1, + 40.6, + 27.9, + 40.7, + 50.1, + 39.2, + 65.0, + 31.9, + 46.5, + 33.4, + 28.9, + 44.6, + 36.2, + 46.9 + ], + "LLaMa3.1 8B": [ + 62.3, + 72.6, + 26.8, + 27.8, + 20.4, + 45.7, + 41.4, + 36.5, + 34.3, + 32.3, + 30.4, + 33.0, + 20.4, + 28.9, + 34.6, + 27.9, + 54.0, + 16.4, + 32.3, + 24.7, + 20.6, + 34.0, + 24.6, + 35.8 + ], + "LLaMa3 8B": [ + 61.0, + 67.6, + 14.6, + 22.9, + 21.8, + 31.0, + 32.3, + 27.4, + 24.4, + 26.4, + 24.0, + 26.1, + 18.7, + 21.6, + 25.5, + 24.4, + 51.3, + 14.8, + 25.3, + 22.3, + 20.8, + 25.7, + 20.0, + 29.3 + ], + "Aya-101 13B": [ + 60.2, + 70.7, + 50.1, + 51.8, + 34.2, + 55.6, + 52.6, + 51.1, + 49.0, + 46.6, + 50.8, + 53.7, + 29.2, + 51.5, + 57.0, + 49.4, + 59.5, + 44.6, + 52.3, + 46.7, + 25.5, + 53.5, + 49.1, + 53.3 + ], + "Gemma1.1 7b": [ + 17.4, + 30.5, + 7.6, + 2.2, + 1.9, + 6.4, + 7.6, + 7.4, + 12.4, + 2.9, + 2.0, + 3.5, + 3.0, + 1.2, + 5.4, + 0.5, + 15.8, + 5.0, + 4.3, + 0.6, + 3.8, + 5.0, + 2.7, + 4.5 + ], + "Gemma2 27b": [ + 56.5, + 67.1, + 36.0, + 30.2, + 21.9, + 45.7, + 40.0, + 38.8, + 37.6, + 35.6, + 36.6, + 35.4, + 21.6, + 39.6, + 43.0, + 34.3, + 55.9, + 26.7, + 38.2, + 28.1, + 23.2, + 42.0, + 30.4, + 41.2 + ], + "Gemma2 9b": [ + 55.6, + 65.9, + 35.3, + 28.7, + 21.1, + 44.8, + 37.7, + 36.9, + 34.6, + 33.3, + 34.2, + 33.8, + 21.2, + 36.7, + 40.9, + 31.9, + 54.4, + 25.8, + 34.9, + 27.0, + 21.6, + 41.3, + 29.2, + 40.1 + ], + "Gemini 1.5 pro": [ + 57.5, + 69.2, + 49.0, + 41.9, + 33.7, + 55.3, + 51.3, + 50.5, + 46.8, + 42.5, + 47.6, + 48.5, + 32.0, + 49.8, + 58.0, + 49.7, + 59.0, + 44.7, + 49.4, + 44.7, + 28.4, + 51.7, + 44.0, + 50.8 + ], + "GPT-4o (Aug)": [ + 56.2, + 68.5, + 43.2, + 38.5, + 27.7, + 53.2, + 49.6, + 48.5, + 44.8, + 42.2, + 46.9, + 46.6, + 34.4, + 47.7, + 56.8, + 49.2, + 56.9, + 35.2, + 48.9, + 45.7, + 38.4, + 50.8, + 46.3, + 49.9 + ] + } + }, + "salt - xx_en": { + "languages": [ + "ach", + "ibo", + "lgg", + "lug", + "nyn", + "swa", + "teo" + ], + "scores": { + "AfroLLaMa 8B": [ + 2.2, + 2.5, + 2.4, + 2.2, + 2.0, + 4.9, + 2.3 + ], + "LLaMAX3 8B": [ + 17.1, + 40.8, + 20.3, + 35.9, + 23.9, + 62.5, + 17.4 + ], + "LLaMa2 7b": [ + 17.8, + 16.3, + 18.9, + 18.9, + 18.7, + 24.5, + 17.3 + ], + "LLaMa3.1 70B": [ + 19.4, + 41.0, + 21.0, + 38.5, + 26.6, + 66.2, + 18.2 + ], + "LLaMa3.1 8B": [ + 17.6, + 32.9, + 20.6, + 28.7, + 22.8, + 60.7, + 19.2 + ], + "LLaMa3 8B": [ + 17.8, + 30.0, + 21.0, + 26.8, + 21.8, + 57.0, + 17.3 + ], + "Aya-101 13B": [ + 16.7, + 44.2, + 19.1, + 40.6, + 24.7, + 61.8, + 16.3 + ], + "Gemma1.1 7b": [ + 0.0, + 1.0, + 0.0, + 0.0, + 0.1, + 5.0, + 0.0 + ], + "Gemma2 27b": [ + 13.4, + 37.1, + 18.6, + 33.4, + 25.0, + 65.1, + 16.0 + ], + "Gemma2 9b": [ + 12.0, + 34.7, + 18.4, + 31.1, + 23.2, + 63.3, + 14.6 + ], + "Gemini 1.5 pro": [ + 23.9, + 47.6, + 19.2, + 54.1, + 38.9, + 67.8, + 17.0 + ], + "GPT-4o (Aug)": [ + 27.5, + 47.4, + 19.8, + 55.4, + 40.6, + 67.5, + 18.0 + ] + } + }, + "mafand - xx_en": { + "languages": [ + "amh", + "bam", + "bbj", + "ewe", + "fon", + "hau", + "ibo", + "kin", + "lug", + "luo", + "mos", + "nya", + "pcm", + "sna", + "swa", + "tsn", + "twi", + "wol", + "xho", + "yor", + "zul" + ], + "scores": { + "AfroLLaMa 8B": [ + 8.9, + 9.7, + 7.1, + 10.6, + 11.0, + 9.4, + 8.2, + 14.7, + 10.8, + 13.0, + 9.8, + 14.9, + 13.3, + 12.6, + 10.9, + 11.4, + 13.8, + 10.4, + 11.0, + 10.4, + 11.1 + ], + "LLaMAX3 8B": [ + 31.2, + 19.9, + 15.4, + 23.3, + 20.4, + 34.2, + 32.7, + 39.7, + 35.6, + 24.6, + 19.7, + 39.3, + 58.0, + 41.3, + 51.1, + 31.8, + 28.0, + 22.6, + 46.3, + 32.8, + 43.6 + ], + "LLaMa2 7b": [ + 17.3, + 17.4, + 15.0, + 20.5, + 19.1, + 20.7, + 20.1, + 24.2, + 21.2, + 21.6, + 18.1, + 25.4, + 47.9, + 21.8, + 24.6, + 21.4, + 24.9, + 18.5, + 21.9, + 22.2, + 23.7 + ], + "Aya-101 13B": [ + 38.6, + 24.4, + 17.0, + 33.0, + 19.6, + 38.7, + 44.7, + 53.7, + 44.5, + 27.1, + 18.7, + 47.6, + 64.4, + 50.2, + 57.8, + 51.8, + 36.8, + 18.5, + 59.9, + 40.0, + 55.2 + ], + "Gemma1.1 7b": [ + 21.7, + 20.3, + 17.0, + 25.4, + 19.7, + 27.3, + 30.3, + 30.5, + 30.0, + 24.9, + 18.0, + 30.1, + 58.7, + 29.8, + 46.3, + 28.8, + 29.5, + 21.4, + 31.2, + 27.3, + 31.5 + ], + "LLaMa3.1 70B": [ + 30.5, + 20.0, + 16.6, + 28.1, + 21.4, + 34.9, + 40.1, + 44.1, + 38.6, + 26.5, + 19.6, + 35.9, + 60.3, + 32.2, + 55.3, + 39.2, + 33.7, + 21.8, + 38.5, + 34.8, + 39.2 + ], + "LLaMa3 8B": [ + 22.0, + 17.3, + 13.3, + 23.7, + 18.4, + 29.6, + 30.9, + 33.2, + 30.4, + 20.6, + 17.6, + 29.0, + 51.9, + 24.4, + 48.1, + 27.6, + 29.8, + 19.0, + 26.8, + 27.8, + 27.0 + ], + "Gemma2 27b": [ + 31.3, + 21.3, + 17.9, + 29.1, + 21.3, + 37.3, + 40.1, + 45.1, + 38.4, + 27.9, + 20.8, + 40.5, + 57.6, + 40.8, + 54.7, + 42.6, + 35.4, + 23.2, + 47.6, + 35.3, + 44.4 + ], + "Gemma2 9b": [ + 27.4, + 17.8, + 14.0, + 23.4, + 18.3, + 33.8, + 30.8, + 42.0, + 33.3, + 22.6, + 17.0, + 37.2, + 53.0, + 32.6, + 50.1, + 36.2, + 29.7, + 19.9, + 43.2, + 27.8, + 38.5 + ], + "LLaMa3.1 8B": [ + 14.7, + 14.5, + 10.8, + 15.7, + 16.4, + 17.5, + 13.9, + 22.0, + 15.4, + 16.4, + 14.6, + 21.7, + 28.7, + 17.3, + 22.4, + 16.6, + 20.2, + 15.4, + 17.0, + 18.2, + 17.9 + ], + "Gemini 1.5 pro": [ + 24.4, + 37.5, + 43.0, + 55.2, + 48.3, + 27.8, + 46.9, + 41.3, + 41.1, + 53.0, + 51.6, + 35.9, + 55.8, + 32.2, + 49.4, + 25.7, + 10.8, + 32.6, + 19.8, + 16.8, + 21.0 + ], + "GPT-4o (Aug)": [ + 19.6, + 37.5, + 39.5, + 56.2, + 46.1, + 30.8, + 46.0, + 53.0, + 41.1, + 53.8, + 49.5, + 31.5, + 58.3, + 23.6, + 50.0, + 14.0, + 6.4, + 20.8, + 8.9, + 10.8, + 24.9 + ] + } + } + } + }, + "MT(en/fr-xx)": { + "datasets": { + "flores - en_xx": { + "languages": [ + "fra_Latn", + "aeb_Arab", + "afr_Latn", + "aka_Latn", + "amh_Ethi", + "ary_Arab", + "arz_Arab", + "bam_Latn", + "bem_Latn", + "cjk_Latn", + "dik_Latn", + "dyu_Latn", + "ewe_Latn", + "fon_Latn", + "fuv_Latn", + "gaz_Latn", + "hau_Latn", + "ibo_Latn", + "kab_Latn", + "kam_Latn", + "kbp_Latn", + "kea_Latn", + "kik_Latn", + "kin_Latn", + "kmb_Latn", + "knc_Arab", + "knc_Latn", + "kon_Latn", + "lin_Latn", + "lua_Latn", + "lug_Latn", + "luo_Latn", + "mos_Latn", + "nso_Latn", + "nus_Latn", + "nya_Latn", + "plt_Latn", + "run_Latn", + "sag_Latn", + "sna_Latn", + "som_Latn", + "sot_Latn", + "ssw_Latn", + "swh_Latn", + "taq_Latn", + "taq_Tfng", + "tir_Ethi", + "tsn_Latn", + "tso_Latn", + "tum_Latn", + "twi_Latn", + "tzm_Tfng", + "umb_Latn", + "wol_Latn", + "xho_Latn", + "yor_Latn", + "zul_Latn" + ], + "scores": { + "AfroLLaMa 8B": [ + 21.1, + 0.2, + 9.0, + 5.4, + 0.2, + 0.1, + 0.1, + 5.3, + 10.8, + 6.6, + 8.9, + 5.3, + 3.8, + 3.4, + 9.0, + 8.6, + 13.9, + 4.2, + 9.0, + 6.0, + 4.6, + 7.4, + 13.8, + 17.3, + 12.2, + 0.5, + 15.5, + 10.4, + 10.7, + 15.3, + 17.1, + 13.0, + 4.8, + 8.7, + 6.0, + 10.2, + 14.4, + 14.4, + 6.5, + 12.0, + 9.2, + 9.6, + 14.5, + 17.9, + 6.6, + 0.6, + 0.2, + 7.6, + 6.8, + 11.8, + 10.9, + 0.6, + 10.0, + 4.8, + 13.6, + 7.4, + 17.1 + ], + "LLaMAX3 8B": [ + 65.6, + 37.9, + 65.3, + 8.4, + 21.2, + 35.0, + 39.1, + 8.1, + 16.9, + 16.2, + 5.4, + 5.9, + 7.0, + 7.8, + 10.0, + 16.8, + 44.4, + 35.3, + 9.6, + 23.7, + 12.6, + 24.1, + 20.0, + 19.9, + 13.2, + 5.0, + 14.1, + 15.9, + 20.6, + 12.7, + 19.7, + 18.7, + 6.6, + 19.0, + 5.9, + 40.6, + 30.0, + 13.2, + 13.4, + 38.9, + 39.6, + 17.9, + 27.0, + 53.7, + 5.8, + 0.4, + 7.9, + 18.2, + 21.0, + 25.6, + 8.9, + 0.4, + 16.9, + 16.4, + 38.4, + 20.3, + 39.1 + ], + "LLaMa2 7b": [ + 52.0, + 15.3, + 35.5, + 8.2, + 3.6, + 15.8, + 17.4, + 9.3, + 6.6, + 6.8, + 3.9, + 9.0, + 11.7, + 2.5, + 8.2, + 6.5, + 6.5, + 8.6, + 3.3, + 9.8, + 6.0, + 22.0, + 8.4, + 10.1, + 6.2, + 6.4, + 3.4, + 10.4, + 10.8, + 9.0, + 11.1, + 6.7, + 7.2, + 10.2, + 3.8, + 10.3, + 10.1, + 8.8, + 11.2, + 10.1, + 10.7, + 11.4, + 10.6, + 13.8, + 4.0, + 4.6, + 2.8, + 9.4, + 7.7, + 8.2, + 12.9, + 6.4, + 6.0, + 8.4, + 12.2, + 6.8, + 11.2 + ], + "LLaMa3.1 70B": [ + 71.6, + 43.4, + 68.2, + 26.0, + 20.5, + 37.9, + 44.0, + 11.9, + 23.4, + 20.5, + 9.7, + 8.6, + 15.9, + 9.6, + 13.8, + 22.7, + 45.0, + 36.3, + 18.4, + 25.1, + 10.3, + 39.5, + 20.5, + 28.8, + 17.2, + 3.2, + 10.9, + 20.4, + 21.8, + 19.1, + 25.9, + 18.8, + 12.7, + 30.3, + 6.3, + 27.9, + 39.2, + 25.8, + 10.4, + 26.1, + 34.7, + 30.4, + 24.6, + 59.5, + 12.4, + 0.9, + 9.7, + 30.6, + 25.0, + 23.3, + 27.5, + 0.8, + 15.8, + 15.5, + 28.6, + 19.1, + 30.5 + ], + "LLaMa3.1 8B": [ + 68.0, + 33.3, + 62.9, + 14.4, + 8.0, + 31.3, + 36.7, + 7.3, + 15.0, + 9.4, + 2.9, + 5.4, + 7.6, + 4.4, + 10.0, + 11.3, + 28.6, + 18.0, + 9.9, + 22.5, + 5.4, + 21.0, + 17.6, + 16.2, + 11.7, + 7.0, + 5.2, + 12.9, + 15.6, + 9.2, + 12.6, + 16.6, + 5.8, + 13.2, + 3.6, + 16.1, + 21.4, + 14.5, + 6.0, + 15.3, + 19.8, + 12.6, + 14.5, + 46.3, + 7.4, + 2.5, + 5.1, + 13.6, + 9.8, + 15.0, + 14.9, + 0.4, + 8.5, + 9.0, + 15.1, + 8.2, + 15.0 + ], + "LLaMa3 8B": [ + 63.9, + 18.5, + 58.5, + 7.8, + 3.6, + 18.7, + 24.5, + 5.2, + 11.8, + 13.0, + 10.0, + 5.2, + 13.0, + 9.4, + 6.3, + 5.1, + 26.4, + 12.7, + 7.1, + 13.3, + 5.0, + 23.1, + 14.2, + 12.7, + 9.9, + 2.0, + 2.8, + 10.8, + 11.4, + 13.1, + 10.6, + 13.7, + 11.6, + 13.8, + 7.9, + 11.6, + 16.6, + 11.7, + 12.8, + 12.0, + 16.3, + 12.6, + 16.1, + 36.8, + 4.0, + 0.8, + 3.0, + 11.9, + 11.9, + 11.5, + 9.4, + 0.4, + 10.7, + 8.9, + 11.9, + 6.9, + 12.3 + ], + "Aya-101 13B": [ + 62.1, + 37.9, + 61.8, + 16.4, + 28.1, + 34.5, + 37.4, + 17.0, + 23.4, + 21.0, + 18.1, + 17.0, + 13.8, + 12.2, + 9.7, + 10.4, + 45.5, + 35.1, + 8.7, + 21.7, + 8.0, + 15.2, + 20.0, + 30.8, + 20.2, + 9.1, + 14.5, + 21.5, + 21.8, + 11.9, + 13.5, + 7.0, + 14.9, + 32.1, + 10.8, + 42.4, + 49.1, + 17.8, + 17.3, + 39.8, + 39.8, + 46.3, + 27.9, + 55.2, + 12.4, + 0.7, + 10.6, + 32.2, + 14.8, + 26.6, + 20.3, + 0.6, + 19.1, + 8.5, + 42.0, + 19.7, + 42.0 + ], + "Gemma1.1 7b": [ + 1.4, + 9.6, + 17.6, + 11.2, + 2.2, + 4.6, + 1.6, + 9.0, + 3.7, + 3.6, + 6.3, + 5.8, + 6.2, + 4.4, + 11.2, + 7.7, + 23.3, + 19.3, + 1.0, + 10.8, + 6.3, + 11.1, + 5.9, + 4.7, + 6.9, + 0.6, + 5.0, + 13.7, + 13.1, + 1.8, + 3.9, + 7.4, + 6.5, + 15.0, + 4.2, + 3.6, + 25.1, + 15.0, + 8.7, + 13.3, + 16.0, + 8.2, + 6.2, + 4.6, + 4.7, + 0.2, + 1.3, + 8.1, + 9.3, + 4.9, + 11.8, + 0.1, + 9.2, + 12.1, + 12.2, + 10.0, + 17.7 + ], + "Gemma2 27b": [ + 65.1, + 53.6, + 71.0, + 30.5, + 42.7, + 51.4, + 55.4, + 22.0, + 31.5, + 23.5, + 21.0, + 21.4, + 21.8, + 19.7, + 21.5, + 24.7, + 47.1, + 40.2, + 22.9, + 24.9, + 22.9, + 54.8, + 27.6, + 41.1, + 23.1, + 15.6, + 22.5, + 27.2, + 29.5, + 26.2, + 31.5, + 22.5, + 20.6, + 39.4, + 19.5, + 37.0, + 42.5, + 33.0, + 23.0, + 39.8, + 41.8, + 41.5, + 35.1, + 61.8, + 23.1, + 18.4, + 27.8, + 37.4, + 32.3, + 30.6, + 31.6, + 19.1, + 22.6, + 25.2, + 44.8, + 32.3, + 45.4 + ], + "Gemma2 9b": [ + 64.6, + 52.0, + 70.4, + 29.5, + 42.3, + 49.4, + 53.8, + 21.1, + 29.7, + 22.6, + 20.6, + 20.6, + 20.8, + 19.3, + 21.7, + 24.4, + 45.6, + 37.4, + 22.1, + 24.0, + 22.6, + 51.3, + 26.3, + 38.5, + 22.8, + 14.6, + 21.9, + 26.7, + 27.3, + 25.2, + 30.1, + 22.3, + 19.5, + 35.7, + 18.6, + 34.8, + 38.5, + 31.3, + 22.4, + 37.4, + 39.7, + 38.3, + 32.4, + 59.3, + 22.1, + 16.6, + 27.3, + 34.1, + 29.1, + 29.4, + 31.4, + 19.0, + 21.6, + 23.9, + 43.4, + 30.6, + 43.6 + ], + "Gemini 1.5 pro": [ + 68.8, + 41.9, + 65.0, + 33.5, + 40.4, + 40.2, + 46.5, + 26.8, + 36.4, + 21.6, + 17.1, + 20.2, + 35.7, + 21.3, + 20.3, + 41.0, + 49.7, + 41.4, + 27.1, + 25.4, + 26.6, + 48.1, + 29.8, + 49.3, + 28.5, + 11.6, + 16.3, + 39.8, + 46.8, + 36.1, + 40.7, + 28.4, + 17.3, + 50.1, + 13.1, + 47.0, + 48.6, + 42.2, + 36.4, + 45.6, + 45.3, + 47.1, + 44.4, + 60.2, + 13.5, + 15.7, + 25.6, + 47.8, + 49.0, + 39.6, + 36.1, + 24.7, + 25.9, + 23.0, + 49.7, + 28.8, + 53.3 + ], + "GPT-4o (Aug)": [ + 70.2, + 46.3, + 66.4, + 29.1, + 32.8, + 43.4, + 50.0, + 15.6, + 28.3, + 17.7, + 12.9, + 11.9, + 20.1, + 9.8, + 13.2, + 39.4, + 50.7, + 40.7, + 21.7, + 23.3, + 9.1, + 50.3, + 23.0, + 49.8, + 16.5, + 7.2, + 15.9, + 32.7, + 46.3, + 22.2, + 39.9, + 32.6, + 10.8, + 49.8, + 11.3, + 47.1, + 52.9, + 41.3, + 17.2, + 46.4, + 45.8, + 47.4, + 40.3, + 62.4, + 10.6, + 10.8, + 14.6, + 45.2, + 45.8, + 36.6, + 30.7, + 1.6, + 19.5, + 26.8, + 50.4, + 23.4, + 53.6 + ] + } + }, + "ntrex - en_xx": { + "languages": [ + "fra_Latn", + "afr_Latn", + "amh_Ethi", + "bem_Latn", + "ewe_Latn", + "hau_Latn", + "ibo_Latn", + "kin_Latn", + "mlg_Latn", + "nde_Latn", + "nso_Latn", + "nya_Latn", + "orm_Ethi", + "sna_Latn", + "som_Latn", + "ssw_Latn", + "swa_Latn", + "tir_Ethi", + "tsn_Latn", + "ven_Latn", + "wol_Latn", + "xho_Latn", + "yor_Latn", + "zul_Latn" + ], + "scores": { + "AfroLLaMa 8B": [ + 18.1, + 7.6, + 0.6, + 9.3, + 4.0, + 12.6, + 4.3, + 14.6, + 8.5, + 11.0, + 8.0, + 12.8, + 7.0, + 18.6, + 7.5, + 12.4, + 16.1, + 0.3, + 8.9, + 7.3, + 4.8, + 12.3, + 6.5, + 14.2 + ], + "LLaMAX3 8B": [ + 57.3, + 64.3, + 14.9, + 18.4, + 8.9, + 46.8, + 39.3, + 22.8, + 30.5, + 31.4, + 17.2, + 42.4, + 17.2, + 39.3, + 43.0, + 27.9, + 56.2, + 7.7, + 18.7, + 15.3, + 18.2, + 37.9, + 20.7, + 39.2 + ], + "LLaMa2 7b": [ + 48.4, + 37.9, + 4.0, + 9.0, + 14.9, + 8.1, + 9.8, + 10.9, + 12.1, + 7.4, + 12.0, + 12.8, + 7.9, + 9.1, + 12.8, + 10.9, + 16.3, + 3.0, + 11.1, + 7.3, + 10.9, + 13.1, + 6.7, + 12.0 + ], + "LLaMa3.1 70B": [ + 61.8, + 69.6, + 15.4, + 24.9, + 17.9, + 46.7, + 42.0, + 30.2, + 36.4, + 28.9, + 28.8, + 29.7, + 22.9, + 26.9, + 37.9, + 26.4, + 60.5, + 9.6, + 33.0, + 20.4, + 17.6, + 29.8, + 18.2, + 32.4 + ], + "LLaMa3.1 8B": [ + 58.6, + 62.6, + 7.1, + 17.4, + 8.8, + 30.9, + 21.2, + 18.1, + 22.7, + 13.9, + 13.6, + 20.4, + 11.3, + 16.2, + 21.1, + 16.7, + 48.5, + 4.9, + 15.0, + 10.9, + 10.2, + 15.4, + 9.4, + 15.9 + ], + "LLaMa3 8B": [ + 56.4, + 58.2, + 3.1, + 15.7, + 16.2, + 27.8, + 14.9, + 13.7, + 19.5, + 12.4, + 13.7, + 13.8, + 3.6, + 15.8, + 16.5, + 17.9, + 40.7, + 2.7, + 12.8, + 15.6, + 9.1, + 13.3, + 8.3, + 15.6 + ], + "Aya-101 13B": [ + 56.7, + 64.1, + 20.5, + 26.0, + 12.6, + 47.8, + 39.3, + 34.1, + 46.0, + 32.6, + 44.1, + 44.1, + 14.6, + 40.9, + 44.1, + 30.2, + 59.0, + 11.0, + 25.0, + 23.2, + 7.9, + 41.6, + 16.5, + 43.0 + ], + "Gemma1.1 7b": [ + 23.6, + 28.6, + 2.0, + 6.8, + 6.5, + 25.3, + 20.8, + 7.0, + 24.1, + 8.2, + 16.0, + 16.1, + 4.6, + 5.1, + 17.5, + 5.3, + 15.9, + 1.2, + 12.6, + 2.0, + 12.2, + 8.6, + 7.2, + 10.7 + ], + "Gemma2 27b": [ + 56.7, + 60.6, + 14.0, + 24.2, + 10.9, + 41.4, + 32.0, + 28.9, + 35.2, + 29.2, + 26.9, + 29.8, + 18.1, + 31.3, + 34.7, + 28.1, + 55.2, + 8.2, + 32.9, + 15.1, + 15.5, + 32.8, + 13.6, + 33.8 + ], + "Gemma2 9b": [ + 56.0, + 58.4, + 9.4, + 16.0, + 12.7, + 38.8, + 26.1, + 17.4, + 27.8, + 20.9, + 22.1, + 20.8, + 12.2, + 19.9, + 29.2, + 1.8, + 52.6, + 6.0, + 22.2, + 8.2, + 12.8, + 25.6, + 8.8, + 26.3 + ], + "Gemini 1.5 pro": [ + 57.6, + 65.5, + 27.8, + 35.3, + 33.2, + 47.1, + 42.3, + 46.0, + 46.2, + 38.1, + 31.9, + 47.9, + 35.6, + 45.4, + 47.0, + 44.5, + 58.2, + 23.9, + 47.4, + 43.6, + 23.1, + 44.7, + 18.8, + 47.5 + ], + "GPT-4o (Aug)": [ + 58.2, + 66.4, + 22.3, + 28.7, + 20.2, + 49.0, + 42.5, + 46.5, + 46.1, + 38.2, + 32.4, + 47.7, + 34.9, + 46.2, + 48.2, + 40.3, + 60.1, + 14.4, + 45.2, + 38.1, + 26.7, + 45.8, + 20.1, + 48.1 + ] + } + }, + "salt - en_xx": { + "languages": [ + "ach", + "ibo", + "lgg", + "lug", + "nyn", + "swa", + "teo" + ], + "scores": { + "Gemma1.1 7b": [ + 3.0, + 12.9, + 2.3, + 1.0, + 10.5, + 11.8, + 3.2 + ], + "LLaMa2 7b": [ + 4.0, + 6.1, + 3.8, + 9.2, + 7.8, + 14.1, + 3.6 + ], + "LLaMa3 8B": [ + 5.2, + 12.7, + 4.2, + 10.6, + 8.6, + 43.7, + 4.7 + ], + "LLaMAX3 8B": [ + 8.8, + 32.6, + 4.0, + 20.4, + 14.9, + 56.4, + 11.1 + ], + "LLaMa3.1 8B": [ + 4.9, + 15.0, + 1.3, + 10.1, + 7.8, + 45.0, + 4.8 + ], + "AfroLLaMa 8B": [ + 1.6, + 1.2, + 1.8, + 2.1, + 3.8, + 5.4, + 2.4 + ], + "Gemma2 9b": [ + 16.4, + 22.4, + 7.4, + 16.0, + 16.1, + 55.1, + 7.5 + ], + "Aya-101 13B": [ + 15.4, + 33.6, + 13.3, + 13.4, + 11.8, + 57.8, + 7.2 + ], + "Gemma2 27b": [ + 16.3, + 28.0, + 11.5, + 23.9, + 20.0, + 60.1, + 12.6 + ], + "LLaMa3.1 70B": [ + 9.8, + 34.1, + 6.1, + 27.5, + 19.5, + 60.4, + 6.1 + ], + "Gemini 1.5 pro": [ + 29.0, + 41.0, + 15.0, + 52.5, + 34.3, + 65.0, + 17.3 + ], + "GPT-4o (Aug)": [ + 20.4, + 39.4, + 14.8, + 48.0, + 28.0, + 67.9, + 15.5 + ] + } + }, + "mafand - en_xx": { + "languages": [ + "amh", + "hau", + "ibo", + "kin", + "lug", + "luo", + "nya", + "pcm", + "sna", + "swa", + "tsn", + "twi", + "xho", + "yor", + "zul", + "bam", + "bbj", + "ewe", + "fon", + "mos", + "wol" + ], + "scores": { + "AfroLLaMa 8B": [ + 0.3, + 14.5, + 4.9, + 13.2, + 9.1, + 11.1, + 15.3, + 10.8, + 12.9, + 15.9, + 10.5, + 9.6, + 17.4, + 9.5, + 17.3, + 8.0, + 2.9, + 7.7, + 5.4, + 6.5, + 8.7 + ], + "LLaMAX3 8B": [ + 11.6, + 31.8, + 30.0, + 27.8, + 26.2, + 19.0, + 34.9, + 26.6, + 31.8, + 42.1, + 26.1, + 19.6, + 30.9, + 16.0, + 31.7, + 15.0, + 6.3, + 17.0, + 8.4, + 12.5, + 17.5 + ], + "LLaMa2 7b": [ + 4.7, + 16.3, + 14.4, + 18.8, + 17.3, + 18.5, + 20.9, + 36.7, + 19.2, + 20.6, + 19.0, + 17.4, + 19.1, + 10.9, + 19.7, + 13.5, + 6.0, + 15.6, + 8.2, + 11.7, + 14.4 + ], + "Aya-101 13B": [ + 19.9, + 35.5, + 40.6, + 37.3, + 19.2, + 8.4, + 42.5, + 21.7, + 35.6, + 53.5, + 24.7, + 22.6, + 45.7, + 18.8, + 43.2, + 9.0, + 8.9, + 12.9, + 8.8, + 8.0, + 6.4 + ], + "Gemma1.1 7b": [ + 3.8, + 22.4, + 22.7, + 24.5, + 23.5, + 20.7, + 24.7, + 43.8, + 23.4, + 39.2, + 24.0, + 22.6, + 21.2, + 14.8, + 24.7, + 13.8, + 8.2, + 16.8, + 9.8, + 12.9, + 16.4 + ], + "LLaMa3.1 70B": [ + 13.4, + 33.5, + 35.4, + 36.1, + 32.7, + 21.4, + 29.0, + 52.2, + 28.8, + 50.1, + 36.0, + 31.1, + 31.7, + 19.2, + 33.7, + 15.5, + 7.0, + 23.8, + 9.5, + 12.0, + 18.0 + ], + "LLaMa3 8B": [ + 3.8, + 23.7, + 20.2, + 12.9, + 18.1, + 14.1, + 12.2, + 47.0, + 12.7, + 41.2, + 15.2, + 10.6, + 10.0, + 10.4, + 14.0, + 5.7, + 6.5, + 9.7, + 7.8, + 9.5, + 8.1 + ], + "Gemma2 27b": [ + 10.8, + 32.3, + 24.3, + 31.4, + 26.5, + 18.0, + 32.8, + 41.1, + 29.9, + 49.7, + 32.5, + 25.3, + 31.8, + 19.7, + 31.6, + 10.8, + 4.9, + 16.1, + 8.7, + 9.1, + 13.5 + ], + "Gemma2 9b": [ + 5.0, + 25.2, + 17.2, + 15.7, + 14.9, + 16.1, + 17.8, + 40.6, + 19.3, + 40.9, + 21.0, + 17.8, + 20.1, + 9.7, + 19.7, + 7.5, + 5.7, + 7.7, + 7.8, + 8.1, + 10.6 + ], + "LLaMa3.1 8B": [ + 3.7, + 16.3, + 10.1, + 18.2, + 13.5, + 15.0, + 20.2, + 21.9, + 18.3, + 21.9, + 15.9, + 14.0, + 15.4, + 8.9, + 15.7, + 10.4, + 3.8, + 10.5, + 6.0, + 8.4, + 11.4 + ], + "Gemini 1.5 pro": [ + 37.8, + 40.6, + 43.0, + 55.9, + 49.4, + 27.5, + 46.0, + 61.2, + 47.8, + 55.3, + 52.3, + 38.1, + 60.8, + 41.2, + 53.2, + 27.2, + 17.1, + 36.6, + 24.5, + 20.4, + 26.1 + ], + "GPT-4o (Aug)": [ + 34.9, + 40.4, + 44.8, + 53.8, + 50.4, + 34.5, + 45.2, + 61.5, + 46.5, + 54.6, + 53.4, + 38.4, + 61.1, + 41.6, + 52.1, + 26.0, + 18.3, + 30.4, + 20.1, + 19.9, + 32.9 + ] + } + } + } + }, + "ADR": { + "datasets": { + "ADR": { + "languages": [ + "bbj", + "fon", + "ibo", + "wol", + "yor" + ], + "scores": { + "AfroLLaMa 8B": [ + 2.1, + 1.4, + 5.0, + 12.5, + 4.8 + ], + "LLaMAX3 8B": [ + 43.7, + 39.0, + 62.3, + 77.3, + 24.8 + ], + "LLaMa2 7b": [ + 19.7, + 34.6, + 30.6, + 52.3, + 14.6 + ], + "LLaMa3.1 70B": [ + 35.8, + 38.2, + 64.1, + 75.6, + 44.8 + ], + "LLaMa3.1 8B": [ + 19.0, + 25.4, + 26.6, + 37.4, + 21.0 + ], + "LLaMa3 8B": [ + 26.7, + 29.0, + 18.8, + 41.5, + 22.8 + ], + "Aya-101 13B": [ + 37.8, + 41.9, + 68.1, + 80.7, + 23.4 + ], + "Gemma1.1 7b": [ + 45.7, + 39.4, + 67.3, + 78.0, + 23.6 + ], + "Gemma2 27b": [ + 44.1, + 37.4, + 70.8, + 76.5, + 46.8 + ], + "Gemma2 9b": [ + 45.4, + 37.5, + 65.7, + 77.9, + 31.5 + ], + "Gemini 1.5 pro": [ + 29.0, + 39.1, + 71.8, + 70.8, + 67.1 + ], + "GPT-4o (Aug)": [ + 32.2, + 34.6, + 70.9, + 74.8, + 62.0 + ] + } + } + } + }, + "SUMM": { + "datasets": { + "xlsum": { + "languages": [ + "amh", + "ara", + "hau", + "ibo", + "orm", + "pcm", + "som", + "swa", + "tir", + "yor" + ], + "scores": { + "AfroLLaMa 8B": [ + 47.2, + 52.1, + 51.8, + 52.1, + 50.3, + 50.3, + 51.8, + 52.4, + 48.8, + 51.3 + ], + "LLaMAX3 8B": [ + 50.5, + 51.7, + 47.6, + 51.8, + 49.5, + 52.3, + 51.0, + 52.5, + 48.7, + 51.6 + ], + "LLaMa2 7b": [ + 40.0, + 51.6, + 43.0, + 51.3, + 47.8, + 48.1, + 49.5, + 50.1, + 47.6, + 40.0 + ], + "LLaMa3.1 70B": [ + 64.8, + 64.8, + 67.2, + 67.9, + 73.8, + 64.7, + 67.2, + 66.4, + 67.8, + 71.7 + ], + "LLaMa3.1 8B": [ + 42.8, + 49.4, + 48.8, + 43.6, + 47.4, + 46.8, + 33.2, + 47.8, + 41.7, + 35.9 + ], + "LLaMa3 8B": [ + 62.0, + 64.0, + 66.1, + 66.6, + 73.8, + 62.3, + 66.5, + 65.2, + 64.9, + 70.1 + ], + "Aya-101 13B": [ + 45.8, + 49.6, + 59.6, + 61.4, + 59.0, + 39.1, + 60.1, + 42.0, + 56.4, + 51.1 + ], + "Gemma1.1 7b": [ + 49.9, + 51.8, + 48.3, + 54.2, + 47.6, + 49.4, + 46.5, + 50.0, + 39.9, + 53.3 + ], + "Gemma2 27b": [ + 64.3, + 65.0, + 66.7, + 66.8, + 73.2, + 63.9, + 66.9, + 65.9, + 64.9, + 66.6 + ], + "Gemma2 9b": [ + 63.8, + 64.6, + 66.3, + 66.4, + 72.5, + 64.5, + 66.6, + 65.6, + 65.6, + 65.3 + ], + "Gemini 1.5 pro": [ + 65.6, + 64.3, + 65.2, + 66.7, + 73.5, + 63.9, + 66.1, + 64.1, + 68.3, + 69.0 + ], + "GPT-4o (Aug)": [ + 65.4, + 64.3, + 65.0, + 66.4, + 73.0, + 63.4, + 66.2, + 64.2, + 67.7, + 69.2 + ] + } + } + } + } + } +} \ No newline at end of file