Datasets:

davidafrica
/

gloss

question_id stringlengths 18 18	cut_id stringlengths 18 18	prompt_id stringlengths 18 18	group_id stringlengths 18 18	source stringclasses 2 values	type stringclasses 10 values	correct stringclasses 4 values	n_modal int64 8 10	counts dict	in_bank bool 2 classes
q_bc0cf8a7c61d1723	c_4fd3e5ed966ceca4	p_3c7852a347c1a8d4	g_5267fb9a34974ea7	chat	content	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_0aa9dc7dc42d0220	c_4fd3e5ed966ceca4	p_3c7852a347c1a8d4	g_5267fb9a34974ea7	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_88a1f8b812a8c4b6	c_4fd3e5ed966ceca4	p_3c7852a347c1a8d4	g_5267fb9a34974ea7	chat	content	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_c0572e546b42e7f6	c_4fd3e5ed966ceca4	p_3c7852a347c1a8d4	g_5267fb9a34974ea7	chat	conclude	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	true
q_0080c4274d29e67b	c_4fd3e5ed966ceca4	p_3c7852a347c1a8d4	g_5267fb9a34974ea7	chat	content	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_1e8695bb750f0623	c_4fd3e5ed966ceca4	p_3c7852a347c1a8d4	g_5267fb9a34974ea7	chat	structure	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_9a91544c15b95bb4	c_408051b1c79f41b1	p_d9abf6ead2cf57f2	g_69a47cc320707a15	chat	content	B	8	{ "A": 0, "B": 8, "C": 0, "D": 0, "unclear": 2 }	true
q_648f6272b1abd144	c_408051b1c79f41b1	p_d9abf6ead2cf57f2	g_69a47cc320707a15	chat	structure	B	9	{ "A": 0, "B": 9, "C": 0, "D": 0, "unclear": 1 }	false
q_04a8de545db5e673	c_fb15ccfa6e259081	p_6b54e204a839b462	g_a4c37eb8d27fa17b	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	true
q_0abc29058f3db609	c_fb15ccfa6e259081	p_6b54e204a839b462	g_a4c37eb8d27fa17b	chat	structure	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_df99812aa73edef4	c_a55f2d624def23b0	p_d9ae2a5710d6ad0c	g_8a83e448ddc972a7	chat	planning	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_c6b575586b894475	c_a55f2d624def23b0	p_d9ae2a5710d6ad0c	g_8a83e448ddc972a7	chat	structure	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	true
q_6b51f106d4f7b977	c_a55f2d624def23b0	p_d9ae2a5710d6ad0c	g_8a83e448ddc972a7	chat	content	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_a4c986ce4510d3b4	c_a55f2d624def23b0	p_d9ae2a5710d6ad0c	g_8a83e448ddc972a7	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_f916fd206a935b41	c_a55f2d624def23b0	p_d9ae2a5710d6ad0c	g_8a83e448ddc972a7	chat	content	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_a7237186c5472cb4	c_a55f2d624def23b0	p_d9ae2a5710d6ad0c	g_8a83e448ddc972a7	chat	conclude	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_2da6a3f71fb676b4	c_a55f2d624def23b0	p_d9ae2a5710d6ad0c	g_8a83e448ddc972a7	chat	planning	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_e1b8f7c3cbd4334a	c_d50db62666582c3b	p_32e80581f017a619	g_62e367d9329955d4	chat	planning	B	9	{ "A": 0, "B": 9, "C": 0, "D": 0, "unclear": 1 }	false
q_7c502bdabf1bb523	c_d50db62666582c3b	p_32e80581f017a619	g_62e367d9329955d4	chat	other	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	true
q_9b4fcfb576f387af	c_056ced1fb97683b2	p_15d279129a8cdecf	g_cf7b3afcaf99e28b	chat	planning	C	9	{ "A": 0, "B": 0, "C": 9, "D": 0, "unclear": 1 }	false
q_5da20ff1440b1edb	c_056ced1fb97683b2	p_15d279129a8cdecf	g_cf7b3afcaf99e28b	chat	content	C	9	{ "A": 0, "B": 0, "C": 9, "D": 0, "unclear": 1 }	false
q_9a384db327e92918	c_056ced1fb97683b2	p_15d279129a8cdecf	g_cf7b3afcaf99e28b	chat	planning	D	9	{ "A": 0, "B": 0, "C": 0, "D": 9, "unclear": 1 }	true
q_4c6a5396804dbe17	c_056ced1fb97683b2	p_15d279129a8cdecf	g_cf7b3afcaf99e28b	chat	other	B	9	{ "A": 0, "B": 9, "C": 0, "D": 0, "unclear": 1 }	false
q_59d5de3163a657ce	c_3f8546db1f81fcb4	p_4401349bdcf879a9	g_dc989a9fd0265484	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_3b6414148e014047	c_3f8546db1f81fcb4	p_4401349bdcf879a9	g_dc989a9fd0265484	chat	content	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_51ff38b5add9aee2	c_3f8546db1f81fcb4	p_4401349bdcf879a9	g_dc989a9fd0265484	chat	planning	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_b833c34a41d0b19c	c_3f8546db1f81fcb4	p_4401349bdcf879a9	g_dc989a9fd0265484	chat	content	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_7c9238429d3513bf	c_3f8546db1f81fcb4	p_4401349bdcf879a9	g_dc989a9fd0265484	chat	planning	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	true
q_56cd47bc5ecc5129	c_3f8546db1f81fcb4	p_4401349bdcf879a9	g_dc989a9fd0265484	chat	structure	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	true
q_1e7926656db36cb1	c_ca4193d8ac93270a	p_0150ff5804b5ddda	g_f62b4cc8f9124b7c	chat	content	A	9	{ "A": 9, "B": 0, "C": 0, "D": 0, "unclear": 1 }	true
q_3155b6080abcb7e6	c_ca4193d8ac93270a	p_0150ff5804b5ddda	g_f62b4cc8f9124b7c	chat	structure	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_9c4151b2307f59ba	c_ca4193d8ac93270a	p_0150ff5804b5ddda	g_f62b4cc8f9124b7c	chat	planning	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	true
q_3cd1395340657255	c_aacc806fe7ed4bfc	p_1ea5c6d9527d1b8c	g_c86079348d43a83d	chat	planning	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	true
q_414f8e5f467c815e	c_aacc806fe7ed4bfc	p_1ea5c6d9527d1b8c	g_c86079348d43a83d	chat	content	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_ff8fc39a0a021fa3	c_aacc806fe7ed4bfc	p_1ea5c6d9527d1b8c	g_c86079348d43a83d	chat	conclude	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_cac170cfee1c6eda	c_0941849e1f869b58	p_18c99a433e9907ac	g_88e7d50f68e9e852	chat	content	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_9d48a1f49d69f655	c_0941849e1f869b58	p_18c99a433e9907ac	g_88e7d50f68e9e852	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_4b106616af723f7a	c_0941849e1f869b58	p_18c99a433e9907ac	g_88e7d50f68e9e852	chat	structure	A	9	{ "A": 9, "B": 0, "C": 0, "D": 0, "unclear": 1 }	true
q_805d68f919816b80	c_00cc0e3da8c53374	p_e088444cd82a0faa	g_a71e0c0f2626a137	chat	content	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_044315b7db7cfea3	c_00cc0e3da8c53374	p_e088444cd82a0faa	g_a71e0c0f2626a137	chat	structure	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_9cf082adf9451a18	c_00cc0e3da8c53374	p_e088444cd82a0faa	g_a71e0c0f2626a137	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	true
q_86bc718f6bcf0fb6	c_00cc0e3da8c53374	p_e088444cd82a0faa	g_a71e0c0f2626a137	chat	content	C	9	{ "A": 0, "B": 0, "C": 9, "D": 0, "unclear": 1 }	false
q_e9d056dda39019fc	c_00cc0e3da8c53374	p_e088444cd82a0faa	g_a71e0c0f2626a137	chat	conclude	A	9	{ "A": 9, "B": 0, "C": 0, "D": 0, "unclear": 1 }	false
q_518235f2f2afd438	c_00cc0e3da8c53374	p_e088444cd82a0faa	g_a71e0c0f2626a137	chat	other	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_8ed00276cd95c4c7	c_cd3e93b3996c0cc7	p_51117466f3b44917	g_c1edc7113b1e20ed	chat	content	C	9	{ "A": 0, "B": 0, "C": 9, "D": 0, "unclear": 1 }	true
q_9fda3709c4a448b8	c_cd3e93b3996c0cc7	p_51117466f3b44917	g_c1edc7113b1e20ed	chat	content	C	9	{ "A": 0, "B": 0, "C": 9, "D": 0, "unclear": 1 }	false
q_9bd3e471ee59772b	c_cd3e93b3996c0cc7	p_51117466f3b44917	g_c1edc7113b1e20ed	chat	planning	C	9	{ "A": 0, "B": 0, "C": 9, "D": 0, "unclear": 1 }	false
q_cc8ee5b644564be0	c_cd3e93b3996c0cc7	p_51117466f3b44917	g_c1edc7113b1e20ed	chat	structure	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_d44a9b3c679b250e	c_5114dca26757376e	p_db3a51795dee33c9	g_32473ecb4c8bfce6	chat	content	A	8	{ "A": 8, "B": 0, "C": 0, "D": 0, "unclear": 2 }	false
q_0f2d954b4ccee482	c_5114dca26757376e	p_db3a51795dee33c9	g_32473ecb4c8bfce6	chat	content	A	8	{ "A": 8, "B": 0, "C": 0, "D": 0, "unclear": 2 }	false
q_b6bfe5fdcddb6c45	c_5114dca26757376e	p_db3a51795dee33c9	g_32473ecb4c8bfce6	chat	other	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	true
q_3c16cf4b76c45c4d	c_8c4bd197cf0be5c2	p_160f55c995fd31b4	g_f1f8ae21eed3b5dd	chat	planning	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_397e09e7c2c248f5	c_8c4bd197cf0be5c2	p_160f55c995fd31b4	g_f1f8ae21eed3b5dd	chat	content	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	true
q_d2af18d7b151e195	c_8c4bd197cf0be5c2	p_160f55c995fd31b4	g_f1f8ae21eed3b5dd	chat	structure	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_bfb6fe22d228e24d	c_8c4bd197cf0be5c2	p_160f55c995fd31b4	g_f1f8ae21eed3b5dd	chat	conclude	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_f8124772fe01cadb	c_e9cf4474514cb2e3	p_0ede19d3c9eb516c	g_34cfa55c6b1f3caa	chat	content	C	8	{ "A": 0, "B": 0, "C": 8, "D": 0, "unclear": 2 }	false
q_894b301330d0c7a0	c_e9cf4474514cb2e3	p_0ede19d3c9eb516c	g_34cfa55c6b1f3caa	chat	content	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_8de9c3619c9b75bd	c_e9cf4474514cb2e3	p_0ede19d3c9eb516c	g_34cfa55c6b1f3caa	chat	structure	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	true
q_acd748a428e66456	c_e9cf4474514cb2e3	p_0ede19d3c9eb516c	g_34cfa55c6b1f3caa	chat	planning	C	8	{ "A": 2, "B": 0, "C": 8, "D": 0, "unclear": 0 }	false
q_5fe663bdb4ec2d32	c_e9cf4474514cb2e3	p_0ede19d3c9eb516c	g_34cfa55c6b1f3caa	chat	conclude	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_fe53f8d1a9250880	c_5a0523f9eeee1604	p_f25737ad9465783a	g_ba2131e753c5080e	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_91d95ad08731075d	c_5a0523f9eeee1604	p_f25737ad9465783a	g_ba2131e753c5080e	chat	content	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_cc3ca6f48f508996	c_5a0523f9eeee1604	p_f25737ad9465783a	g_ba2131e753c5080e	chat	structure	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	true
q_a4278f166c391cac	c_5a0523f9eeee1604	p_f25737ad9465783a	g_ba2131e753c5080e	chat	content	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_e52705aa30f35167	c_5a0523f9eeee1604	p_f25737ad9465783a	g_ba2131e753c5080e	chat	structure	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_5bc872e1197cc3e3	c_2c6e127497839bc8	p_1a2df6dff733ae8c	g_3107281aeaa3d343	chat	planning	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_0b3828c21bf51994	c_2c6e127497839bc8	p_1a2df6dff733ae8c	g_3107281aeaa3d343	chat	content	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_011cce3599f7bf8f	c_2c6e127497839bc8	p_1a2df6dff733ae8c	g_3107281aeaa3d343	chat	structure	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	true
q_6a6643c859661e01	c_2c6e127497839bc8	p_1a2df6dff733ae8c	g_3107281aeaa3d343	chat	content	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_2bb870af9f20faa0	c_2c6e127497839bc8	p_1a2df6dff733ae8c	g_3107281aeaa3d343	chat	structure	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_91fbd8be1980da0c	c_2c6e127497839bc8	p_1a2df6dff733ae8c	g_3107281aeaa3d343	chat	content	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_01ede26ea3a6770d	c_23b5c5c888851c12	p_133d80d2271122e6	g_3ea3bbff0b16140c	chat	planning	D	9	{ "A": 0, "B": 0, "C": 0, "D": 9, "unclear": 1 }	false
q_13c68505dbd03a69	c_23b5c5c888851c12	p_133d80d2271122e6	g_3ea3bbff0b16140c	chat	content	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_70900148baadc98b	c_23b5c5c888851c12	p_133d80d2271122e6	g_3ea3bbff0b16140c	chat	content	B	9	{ "A": 0, "B": 9, "C": 0, "D": 0, "unclear": 1 }	true
q_ecde2df6ec866380	c_e29b80f47612e3a7	p_c0ac67ddacad1012	g_e6565c0dc9884c26	chat	content	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_a14dc0b8bba5eb65	c_e29b80f47612e3a7	p_c0ac67ddacad1012	g_e6565c0dc9884c26	chat	content	C	9	{ "A": 0, "B": 0, "C": 9, "D": 1, "unclear": 0 }	false
q_4ba213028a8d020a	c_e29b80f47612e3a7	p_c0ac67ddacad1012	g_e6565c0dc9884c26	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_4bf7dd92a28e3ccd	c_e29b80f47612e3a7	p_c0ac67ddacad1012	g_e6565c0dc9884c26	chat	content	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_73b4cc9f74b657ef	c_e29b80f47612e3a7	p_c0ac67ddacad1012	g_e6565c0dc9884c26	chat	planning	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_c2f9996b990a8e0c	c_e29b80f47612e3a7	p_c0ac67ddacad1012	g_e6565c0dc9884c26	chat	content	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	true
q_18cb59b9f5b1ae56	c_e29b80f47612e3a7	p_c0ac67ddacad1012	g_e6565c0dc9884c26	chat	conclude	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_4b1380826d3bd71a	c_5e1451c8487503ce	p_ca207c4f5fe73e1e	g_06fde40d26dd04a5	chat	planning	A	8	{ "A": 8, "B": 0, "C": 1, "D": 0, "unclear": 1 }	false
q_8fcb3115d5020701	c_5e1451c8487503ce	p_ca207c4f5fe73e1e	g_06fde40d26dd04a5	chat	planning	C	8	{ "A": 0, "B": 0, "C": 8, "D": 0, "unclear": 2 }	true
q_a9b50fd0a5966a60	c_a0e3c7fd218d11cd	p_09238dc7ed3d2372	g_ba98743c3c6fbaab	chat	content	B	9	{ "A": 0, "B": 9, "C": 0, "D": 0, "unclear": 1 }	false
q_b83e5866d6404cf6	c_a0e3c7fd218d11cd	p_09238dc7ed3d2372	g_ba98743c3c6fbaab	chat	structure	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	true
q_6c7fce3d92156b2a	c_a0e3c7fd218d11cd	p_09238dc7ed3d2372	g_ba98743c3c6fbaab	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_6a5d01e90c28fe9d	c_75d6edde8b948119	p_3907506f34df6b64	g_8bb78f59b6fbdec0	chat	planning	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_af6282801cc1d048	c_75d6edde8b948119	p_3907506f34df6b64	g_8bb78f59b6fbdec0	chat	content	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_a3f838ece8d6bf96	c_75d6edde8b948119	p_3907506f34df6b64	g_8bb78f59b6fbdec0	chat	content	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_9386d9f24be1afd9	c_75d6edde8b948119	p_3907506f34df6b64	g_8bb78f59b6fbdec0	chat	structure	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	true
q_0c5f2f31d8c863e2	c_c152247a38bb6a3d	p_06f43149eb1aaf1c	g_b946424bc0bb404b	chat	planning	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	false
q_323602f6e1cdc8c1	c_c152247a38bb6a3d	p_06f43149eb1aaf1c	g_b946424bc0bb404b	chat	content	B	10	{ "A": 0, "B": 10, "C": 0, "D": 0, "unclear": 0 }	false
q_fafab308abcfacb0	c_c152247a38bb6a3d	p_06f43149eb1aaf1c	g_b946424bc0bb404b	chat	planning	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	true
q_4983c95d67df4c77	c_ebc068edc2c84f88	p_1ba298948c669883	g_3390a8a7c721d0f7	chat	planning	D	8	{ "A": 0, "B": 0, "C": 0, "D": 8, "unclear": 2 }	false
q_ab9a08b59044babd	c_ebc068edc2c84f88	p_1ba298948c669883	g_3390a8a7c721d0f7	chat	content	C	9	{ "A": 0, "B": 0, "C": 9, "D": 0, "unclear": 1 }	true
q_7e7ab208927b1e46	c_ebc068edc2c84f88	p_1ba298948c669883	g_3390a8a7c721d0f7	chat	planning	B	8	{ "A": 0, "B": 8, "C": 0, "D": 0, "unclear": 2 }	false
q_48b9272609c5e188	c_ebc068edc2c84f88	p_1ba298948c669883	g_3390a8a7c721d0f7	chat	content	A	10	{ "A": 10, "B": 0, "C": 0, "D": 0, "unclear": 0 }	true
q_4b92869e34d0f8dc	c_ebc068edc2c84f88	p_1ba298948c669883	g_3390a8a7c721d0f7	chat	planning	D	10	{ "A": 0, "B": 0, "C": 0, "D": 10, "unclear": 0 }	false
q_a0d2caf401711179	c_ebc068edc2c84f88	p_1ba298948c669883	g_3390a8a7c721d0f7	chat	structure	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false
q_d50e07fa27c7d5de	c_ebc068edc2c84f88	p_1ba298948c669883	g_3390a8a7c721d0f7	chat	planning	C	10	{ "A": 0, "B": 0, "C": 10, "D": 0, "unclear": 0 }	false

End of preview. Expand in Data Studio

gloss: derived data from five protocols on natural-language descriptions of `Qwen3-8B`

Version 1.0.0. Public derived-data release of five completed protocols that asked one question: can a short natural-language description S of Qwen3-8B, learned by a proposer/search loop, help a frozen predictor forecast what the model does next (its next reasoning step, or its next response-mode change) better than the prompt and the model's own partial output alone, and better than data-free control texts of the same length?

This is not a raw archive and not a standalone benchmark. It contains labels, label counts, predictor probabilities, per-item losses, split/pool membership, lineage identifiers, the reviewed description texts and run provenance. It contains no prompts, chains of thought, continuations, trajectories, question wording or options, token ids, embeddings, API transcripts or judge rationales. Multiple-choice letters keep their meaning only relative to the withheld option order of each question.

Loading

from datasets import load_dataset
ds = load_dataset('davidafrica/gloss', 'mode_time_predictions', split='test', revision='v1.0.0', token=False)
print(ds[0]['condition'], ds[0]['predictor'], ds[0]['logloss'])
runs = load_dataset('davidafrica/gloss', 'runs', split='records', revision='v1.0.0', token=False)

Every configuration is a plain Parquet table (no loading script, no trust_remote_code). Configuration names are <protocol>_<table kind> plus the two shared tables runs and descriptions.

Configurations

config	splits (rows)	contents
`pilot_mcq_examples`	test (1,197), train (4,436)	one row per kept pilot question (primary >=8/10 and hard 4-7/10 sets); metadata and majority-label counts; question wording withheld
`pilot_mcq_predictions`	test (5,788), train (21,296)	headroom conditions (question_only / nothing / example_oracle / answer_leak) x predictors (qwen3-8b on all questions, anthropic/claude-sonnet-5 on 600 per set)
`contrastive_mcq_examples`	test (1,425), train (5,878)	kept, band and agree question sets with per-model modal labels and label fractions (letters are relative to the withheld materialised option order)
`contrastive_mcq_predictions`	test (11,089), train (21,168)	stage = headroom (development + test), heldout (374 test questions, 5 conditions x 2 predictors) or diagnostic_retrospective (mid-search versions v13/v14/v16 re-scored on the already examined test set)
`corrected_search_membership`	dev (3,180)	role (feedback / acceptance / selection) of each reused development question per search subject; question_id values are shared with contrastive_mcq_examples (train split)
`corrected_search_examples`	fresh_test (1,586)	1,586 frozen fresh test questions on 573 never-sampled prompts
`corrected_search_predictions`	selection (4,508), fresh_test (20,618)	stage = selection (selection-pool scores of each accepted version, per subject) or fresh_test (7 conditions; qwen3-8b and anthropic/claude-sonnet-5)
`blind_methods_examples`	feedback (7,454), acceptance (3,750), gepa_val (1,880), final_sel (1,876), test (1,573)	16,533 kept blind first-move questions by frozen pool (feedback / acceptance / gepa_val / final_sel / test) with label counts; in_bank marks the 2,000-question bank when resolvable
`blind_methods_predictions`	feedback (7,454), acceptance (3,750), gepa_val (1,880), final_sel (1,876), test (20,449)	stage = baseline_empty_description (qwen3-8b, every pool, frozen before any description was learned) or evaluation (test pool, 6 conditions x 2 predictors)
`mode_time_prefixes`	feedback (320), acceptance (160), gepa_val (80), final_sel (80), test (160), xstest (100)	900 forecast prefixes (items) with family, stratum, request version, regime, pool, current mode and the ten realised outcome codes
`mode_time_futures`	feedback (3,200), gepa_val (800), final_sel (800), acceptance (1,600), test (1,600), xstest (1,000)	9,000 labelled futures: event, onset token, code A-P, transition sentence indices and intents (quotes and error text withheld)
`mode_time_predictions`	feedback (1,280), acceptance (640), gepa_val (320), final_sel (320), test (1,440), xstest (900)	16-code probability forecasts and per-item losses for qwen3-8b (every pool), anthropic/claude-sonnet-5 (test, xstest) and the base-rate table (test, xstest)
`descriptions`	records (21)	every description text S and control text that entered a scored condition or the retrospective diagnostic (text released after review), with token counts, hashes, selection status and role
`runs`	records (5)	one row per protocol: models and revisions, target and sampling definitions, released counts, outcome labels and provenance

Split semantics

pilot_mcq_examples: train/test are the pilot's own prompt-level split (400/98 chat, 400/100 math prompts); neither is a training set for anything released here
pilot_mcq_predictions: same prompt-level split as the examples
contrastive_mcq_examples: train (1,590 kept development questions, later reused by corrected_search) / test (374 held-out questions)
contrastive_mcq_predictions: same prompt-level split; the development split holds headroom scores only
corrected_search_membership: dev = the contrastive_mcq train questions partitioned into feedback / acceptance / selection groups
corrected_search_examples: fresh_test only
corrected_search_predictions: selection (development selection pool) / fresh_test
blind_methods_examples: frozen pools: feedback, acceptance, gepa_val, final_sel, test
blind_methods_predictions: frozen pools; only test carries the six evaluated conditions
mode_time_prefixes: frozen pools: feedback, acceptance, gepa_val, final_sel, test, xstest (benign reserve)
mode_time_futures: same pools as the prefixes
mode_time_predictions: same pools; external and base-rate forecasts exist for test and xstest only
descriptions: records
runs: records

The five protocols are not independent test sets. corrected_search reuses the contrastive_mcq development questions (same question_id, cut_id, prompt_id values) for its feedback / acceptance / selection pools; contrastive_mcq cuts the same 998 chains as pilot_mcq (shared prompt_id values). contrastive_mcq_predictions rows with stage = diagnostic_retrospective are a researcher-approved retrospective re-scoring of an already examined test set, not a fresh confirmation. corrected_search rows with condition = S_8b_prev score the previous run's final description as a diagnostic comparator. Nothing was re-clustered or re-split for this release.

Identifiers and lineage

All identifiers are keyed pseudonyms (p_ prompt, c_ cut, q_ question, g_ group, f_ family, i_ forecast item). The same underlying object receives the same pseudonym everywhere it appears, so joins between examples, membership, predictions, futures and prefixes work within and across protocols where the source records establish identity. The original-identifier crosswalk is retained privately by the authors and is not part of this release. Content fingerprints (item_hash) are withheld.

Probability and label semantics

probs for predictor = qwen3-8b are raw first-token probability masses over the option letters (A-D) or the sixteen outcome codes (A-P); letter_mass / code_mass is their sum before any normalisation. Values are preserved at source precision (float64) and were not renormalised.
anthropic/claude-sonnet-5 rows carry a parsed letter (pred, pred_common) in the MCQ protocols and elicited probabilities in mode_time; pred = null means the answer did not parse and was scored wrong.
acc is 1 when pred equals correct. In contrastive_mcq, acc_8b / acc_14b score the prediction against each model's own modal label.
mode_time losses use the pre-registered smoothing p' = (p + 1e-6) / (1 + 16e-6), mean of -log p'(outcome) over the labelled futures (logloss); event_logloss and timing_logloss decompose it. outcome_codes are the ten realised codes of the item.
A completion event means the response ended without a further mode change; it is not a refusal. The next event is not the final compliance status of the response, and P (survival) means no new event within the 2,048-token window, not that the model never refuses.
Outcome labels in runs are the scientist label (outcome_label_scientist, the curated verdict) and the worker label recorded at the time (outcome_label_worker); both are provenance, not new verdicts.

See CODEBOOK.md for every column and enumeration and LICENSES.md for rights.

Coverage (original vs released)

protocol	written / generated	released examples	withheld
pilot_mcq	7,984 questions written	5,633 (5,008 primary + 625 hard)	2,351 unkept questions (only in original option order, not scored); all wording
contrastive_mcq	11,364 questions written	1,964 kept + band + agree sets	unkept questions; all wording; search histories
corrected_search	1,586 fresh questions frozen (plus the reused development questions)	1,586 fresh + membership of the development questions	wording; chains; search histories; rejected edits
blind_methods	35,081 questions written, 16,533 kept	16,533 (all frozen pools)	18,548 unkept; wording; GEPA attempts / candidates; ICAI principles text
mode_time	900 prefixes, 9,000 futures	900 prefixes, 9,000 futures	prompts, prefixes, futures, transition quotes, error text, search attempts

Descriptions: every text that entered a scored condition or the retrospective diagnostic is released after review (descriptions.text_released); unselected search candidates, rejected edits and the per-round search histories are withheld because they are unreviewed free text tied to no reported score. The truncated run-1 mode_time text is included as superseded_run1 and was never scored on the test pool.

Limitations

pilot_mcq stopped at its headroom check; its predictions test the instrument, not any description.
contrastive_mcq used one development pool for feedback, acceptance and selection; its held-out contrasts did not separate from controls.
corrected_search detected a below-target gain (+2.71 points) without model specificity; blind_methods found no learned-vs-nothing gain surviving Holm.
mode_time passed its registered joint log-loss test, but all Qwen conditions are worse than uniform and a base-rate table beats every model predictor; treat the gain as calibration and timing repair.
Labels were produced by LLM annotators (openai/gpt-5.6-terra, anthropic/claude-sonnet-5); agreement diagnostics are reported in the source analyses, not re-derived here.
mode_time request families come from WildJailbreak (gated, ODC-BY) and XSTest; only numerical/label derivatives and opaque identifiers are released, so the rows cannot be mapped back to particular requests from this dataset alone.

Provenance

Source experiment identifiers and final commits are in the runs configuration and in release_manifest.json (sha256 of every released file). Numeric fields were copied from the source score and label files without recomputation and verified value-for-value after writing.

Citation

@dataset{africa2026gloss,
  author = {Africa, David Demitri},
  title = {gloss: derived data from five protocols on natural-language descriptions of Qwen3-8B},
  year = {2026}, version = {1.0.0}, publisher = {Hugging Face},
  url = {https://huggingface.co/datasets/davidafrica/gloss}
}

Downloads last month: 90