Name: tytodd/v2
Creator: tytodd
Published: 2026-04-26 21:18:23
License: 暂无描述

下载链接：

https://hf-mirror.com/datasets/tytodd/v2

下载链接

链接失效反馈

官方服务：

资源简介：

--- dataset_info: - config_name: aes2_essay_scoring features: - name: full_text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: score dtype: string splits: - name: train num_bytes: 21138777 num_examples: 10000 - name: val num_bytes: 2100385 num_examples: 1000 download_size: 12545141 dataset_size: 23239162 - config_name: anli_r1 features: - name: premise dtype: string - name: hypothesis dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: train num_bytes: 4341611 num_examples: 10000 - name: val num_bytes: 430985 num_examples: 1000 download_size: 1187507 dataset_size: 4772596 - config_name: anli_r2 features: - name: premise dtype: string - name: hypothesis dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: train num_bytes: 4250929 num_examples: 10000 - name: val num_bytes: 426429 num_examples: 1000 download_size: 1015140 dataset_size: 4677358 - config_name: anli_r3 features: - name: premise dtype: string - name: hypothesis dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: train num_bytes: 8332835 num_examples: 20000 - name: val num_bytes: 425835 num_examples: 1000 download_size: 4384221 dataset_size: 8758670 - config_name: arc_challenge features: - name: question dtype: string - name: choices dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: choice dtype: string splits: - name: ood num_bytes: 352784 num_examples: 1172 download_size: 195502 dataset_size: 352784 - config_name: argument_quality_ranking features: - name: topic dtype: string - name: argument dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: quality_label dtype: string splits: - name: ood num_bytes: 491281 num_examples: 2469 download_size: 199787 dataset_size: 491281 - config_name: big_patent_innovation features: - name: description dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: innovation_score dtype: string splits: - name: train num_bytes: 21003829 num_examples: 6892 - name: val num_bytes: 3044237 num_examples: 1000 download_size: 12408952 dataset_size: 24048066 - config_name: boardgame_qa features: - name: question dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: answer dtype: string splits: - name: train num_bytes: 14135493 num_examples: 10000 - name: val num_bytes: 2508190 num_examples: 2000 download_size: 4020862 dataset_size: 16643683 - config_name: chatbot_arena_conversations features: - name: question dtype: string - name: response_A dtype: string - name: response_B dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: train num_bytes: 25594517 num_examples: 14000 - name: val num_bytes: 5152744 num_examples: 2500 download_size: 16073776 dataset_size: 30747261 - config_name: civil_comments features: - name: comment dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: toxicity_label dtype: string splits: - name: train num_bytes: 3996776 num_examples: 10000 - name: val num_bytes: 333937 num_examples: 1000 download_size: 2730605 dataset_size: 4330713 - config_name: code_judge_bench features: - name: problem dtype: string - name: code_A dtype: string - name: code_B dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: ood num_bytes: 3807973 num_examples: 344 download_size: 1611189 dataset_size: 3807973 - config_name: colbert_humor_detection features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: humor_label dtype: string splits: - name: train num_bytes: 1205087 num_examples: 10000 - name: val num_bytes: 117086 num_examples: 1000 download_size: 779048 dataset_size: 1322173 - config_name: customer_support_tickets_en features: - name: subject dtype: string - name: body dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: queue dtype: string - name: type dtype: string splits: - name: train num_bytes: 2732741 num_examples: 5570 - name: val num_bytes: 502493 num_examples: 1000 download_size: 1443163 dataset_size: 3235234 - config_name: customer_support_tickets_gorkem features: - name: ticket_text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: ticket_subject dtype: string - name: ticket_type dtype: string splits: - name: train num_bytes: 2685150 num_examples: 6775 - name: val num_bytes: 393479 num_examples: 1000 download_size: 1228089 dataset_size: 3078629 - config_name: dbpedia_easy features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: l1_class dtype: string splits: - name: train num_bytes: 6709066 num_examples: 10000 - name: val num_bytes: 6636509 num_examples: 10000 download_size: 8379732 dataset_size: 13345575 - config_name: dbpedia_hard features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: l1_class dtype: string - name: l2_class dtype: string - name: l3_class dtype: string splits: - name: train num_bytes: 6974361 num_examples: 10000 - name: val num_bytes: 6924380 num_examples: 10000 download_size: 8938694 dataset_size: 13898741 - config_name: dbpedia_medium features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: l1_class dtype: string - name: l2_class dtype: string splits: - name: train num_bytes: 6886302 num_examples: 10000 - name: val num_bytes: 6797751 num_examples: 10000 download_size: 8670989 dataset_size: 13684053 - config_name: enron_email_quality features: - name: subject dtype: string - name: body dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: quality_score dtype: string splits: - name: train num_bytes: 6307967 num_examples: 6500 - name: val num_bytes: 966200 num_examples: 1000 download_size: 4291554 dataset_size: 7274167 - config_name: enron_email_type features: - name: subject dtype: string - name: body dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: email_type dtype: string splits: - name: train num_bytes: 7335605 num_examples: 7000 - name: val num_bytes: 1079301 num_examples: 1000 download_size: 5125431 dataset_size: 8414906 - config_name: enron_reply_quality features: - name: original_email dtype: string - name: reply dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: quality dtype: string splits: - name: train num_bytes: 9474493 num_examples: 7000 - name: val num_bytes: 1352953 num_examples: 1000 download_size: 6295759 dataset_size: 10827446 - config_name: go_emotions features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: labels list: int64 splits: - name: train num_bytes: 1120086 num_examples: 10000 - name: val num_bytes: 108550 num_examples: 1000 download_size: 750989 dataset_size: 1228636 - config_name: gpqa_diamond features: - name: question dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: choice dtype: string splits: - name: ood num_bytes: 130052 num_examples: 198 download_size: 73069 dataset_size: 130052 - config_name: halueval_dialogue features: - name: knowledge dtype: string - name: dialogue_history dtype: string - name: response dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: hallucination dtype: string splits: - name: train num_bytes: 4481364 num_examples: 8000 - name: val num_bytes: 1124151 num_examples: 2000 download_size: 2868290 dataset_size: 5605515 - config_name: halueval_qa features: - name: knowledge dtype: string - name: question dtype: string - name: answer dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: hallucination dtype: string splits: - name: train num_bytes: 4294854 num_examples: 8000 - name: val num_bytes: 1067031 num_examples: 2000 download_size: 3464000 dataset_size: 5361885 - config_name: halueval_summarization features: - name: document dtype: string - name: summary dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: hallucination dtype: string splits: - name: ood num_bytes: 43077353 num_examples: 10000 download_size: 25207857 dataset_size: 43077353 - config_name: hh_rlhf features: - name: question dtype: string - name: response_A dtype: string - name: response_B dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: train num_bytes: 27702637 num_examples: 32000 - name: val num_bytes: 893654 num_examples: 1000 download_size: 16245216 dataset_size: 28596291 - config_name: judge_bench features: - name: question dtype: string - name: response_A dtype: string - name: response_B dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: ood num_bytes: 1546246 num_examples: 280 download_size: 731939 dataset_size: 1546246 - config_name: lex_glue_case_hold features: - name: context dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: option_e dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: selected_option dtype: string splits: - name: train num_bytes: 16907670 num_examples: 10000 - name: val num_bytes: 1702393 num_examples: 1000 download_size: 10210484 dataset_size: 18610063 - config_name: lex_glue_ledgar features: - name: provision_text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: provision_type dtype: string splits: - name: train num_bytes: 7566125 num_examples: 10000 - name: val num_bytes: 721950 num_examples: 1000 download_size: 3963642 dataset_size: 8288075 - config_name: medical_abstracts features: - name: medical_abstract dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: condition_label dtype: string splits: - name: train num_bytes: 12898384 num_examples: 10000 - name: val num_bytes: 1294461 num_examples: 1000 download_size: 7658163 dataset_size: 14192845 - config_name: mfrc features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: annotation dtype: string splits: - name: train num_bytes: 2700178 num_examples: 10000 - name: val num_bytes: 249884 num_examples: 1000 download_size: 898752 dataset_size: 2950062 - config_name: mmlu features: - name: question dtype: string - name: choices dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: choice dtype: string splits: - name: train num_bytes: 34448400 num_examples: 20000 - name: val num_bytes: 390223 num_examples: 1000 download_size: 10393935 dataset_size: 34838623 - config_name: mmlu_pro features: - name: question dtype: string - name: choices dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: choice dtype: string splits: - name: ood num_bytes: 8257700 num_examples: 12032 download_size: 4055088 dataset_size: 8257700 - config_name: mt_bench_human_judgments features: - name: question dtype: string - name: response_A dtype: string - name: response_B dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: ood num_bytes: 1691803 num_examples: 1000 download_size: 148868 dataset_size: 1691803 - config_name: musr_murder_mysteries features: - name: question dtype: string - name: choices dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: choice dtype: string splits: - name: ood num_bytes: 1401638 num_examples: 250 download_size: 809166 dataset_size: 1401638 - config_name: musr_object_placements features: - name: question dtype: string - name: choices dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: choice dtype: string splits: - name: ood num_bytes: 1316095 num_examples: 256 download_size: 296684 dataset_size: 1316095 - config_name: musr_team_allocation features: - name: question dtype: string - name: choices dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: choice dtype: string splits: - name: ood num_bytes: 868814 num_examples: 250 download_size: 508363 dataset_size: 868814 - config_name: or_bench_80k features: - name: prompt dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: or_bench_category dtype: string splits: - name: train num_bytes: 3361563 num_examples: 20000 - name: val num_bytes: 167713 num_examples: 1000 download_size: 1750653 dataset_size: 3529276 - config_name: or_bench_hard_1k features: - name: prompt dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: or_bench_category dtype: string splits: - name: train num_bytes: 172762 num_examples: 1055 - name: val num_bytes: 40722 num_examples: 264 download_size: 107131 dataset_size: 213484 - config_name: or_bench_toxic features: - name: prompt dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: or_bench_category dtype: string splits: - name: ood num_bytes: 71182 num_examples: 524 download_size: 36876 dataset_size: 71182 - config_name: projudgebench features: - name: question dtype: string - name: correct_answer dtype: string - name: steps list: string - name: step_to_evaluate dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: correct dtype: bool splits: - name: train num_bytes: 7395380 num_examples: 2160 - name: val num_bytes: 890353 num_examples: 240 download_size: 6824466 dataset_size: 8285733 - config_name: reward_bench_2 features: - name: prompt dtype: string - name: response_A dtype: string - name: response_B dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: train num_bytes: 5111158 num_examples: 1492 - name: val num_bytes: 1441108 num_examples: 373 download_size: 3607595 dataset_size: 6552266 - config_name: rod101_essay_scoring features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: score dtype: string splits: - name: ood num_bytes: 164656 num_examples: 81 download_size: 98115 dataset_size: 164656 - config_name: sem_eval_2010_task_8 features: - name: sentence dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: relation_label dtype: string splits: - name: train num_bytes: 1454279 num_examples: 8000 - name: val num_bytes: 487057 num_examples: 2717 download_size: 1192230 dataset_size: 1941336 - config_name: smollm_corpus features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: audience dtype: string splits: - name: train num_bytes: 75862714 num_examples: 20000 - name: val num_bytes: 3777368 num_examples: 1000 download_size: 48367200 dataset_size: 79640082 - config_name: snli features: - name: premise dtype: string - name: hypothesis dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: label dtype: string splits: - name: train num_bytes: 1451710 num_examples: 10000 - name: val num_bytes: 150563 num_examples: 1000 download_size: 606540 dataset_size: 1602273 - config_name: spartqa_mchoice features: - name: question dtype: string - name: choices dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: choice dtype: string splits: - name: train num_bytes: 8383953 num_examples: 10000 - name: val num_bytes: 813855 num_examples: 1000 download_size: 1539117 dataset_size: 9197808 - config_name: toxigen_data features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: toxicity_label dtype: string splits: - name: train num_bytes: 1225407 num_examples: 8960 - name: val num_bytes: 128458 num_examples: 940 download_size: 774648 dataset_size: 1353865 - config_name: tweet_eval_emotion features: - name: tweet dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: emotion_label dtype: string splits: - name: train num_bytes: 449675 num_examples: 3257 - name: val num_bytes: 49825 num_examples: 374 download_size: 308491 dataset_size: 499500 - config_name: tweet_eval_hate features: - name: tweet dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: hate_label dtype: string splits: - name: train num_bytes: 1513086 num_examples: 8993 - name: val num_bytes: 183283 num_examples: 999 download_size: 1070865 dataset_size: 1696369 - config_name: tweet_eval_irony features: - name: tweet dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: irony_label dtype: string splits: - name: train num_bytes: 355329 num_examples: 2862 - name: val num_bytes: 115598 num_examples: 955 download_size: 304913 dataset_size: 470927 - config_name: tweet_eval_offensive features: - name: tweet dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: offensive_label dtype: string splits: - name: train num_bytes: 1808259 num_examples: 10000 - name: val num_bytes: 183049 num_examples: 1000 download_size: 1168302 dataset_size: 1991308 - config_name: tweet_eval_sentiment features: - name: tweet dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: sentiment_label dtype: string splits: - name: train num_bytes: 1573096 num_examples: 10000 - name: val num_bytes: 154680 num_examples: 1000 download_size: 1086336 dataset_size: 1727776 - config_name: ultrafeedback features: - name: prompt dtype: string - name: response dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: instruction_following dtype: string splits: - name: train num_bytes: 21574469 num_examples: 10000 - name: val num_bytes: 1980559 num_examples: 1000 download_size: 12576832 dataset_size: 23555028 - config_name: writingprompts_quality features: - name: prompt dtype: string - name: story dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: quality_score dtype: string splits: - name: train num_bytes: 13284706 num_examples: 7000 - name: val num_bytes: 1912031 num_examples: 1000 download_size: 9615002 dataset_size: 15196737 - config_name: yahoo_answers_quality features: - name: question dtype: string - name: answer dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: rating dtype: string splits: - name: train num_bytes: 4826651 num_examples: 7000 - name: val num_bytes: 696540 num_examples: 1000 download_size: 3507506 dataset_size: 5523191 - config_name: yelp features: - name: text dtype: string - name: row_id dtype: string - name: ground_truth struct: - name: rating dtype: string splits: - name: train num_bytes: 14736676 num_examples: 20000 - name: val num_bytes: 731113 num_examples: 1000 download_size: 9480043 dataset_size: 15467789 configs: - config_name: aes2_essay_scoring data_files: - split: train path: aes2_essay_scoring/train-* - split: val path: aes2_essay_scoring/val-* - config_name: anli_r1 data_files: - split: train path: anli_r1/train-* - split: val path: anli_r1/val-* - config_name: anli_r2 data_files: - split: train path: anli_r2/train-* - split: val path: anli_r2/val-* - config_name: anli_r3 data_files: - split: train path: anli_r3/train-* - split: val path: anli_r3/val-* - config_name: arc_challenge data_files: - split: ood path: arc_challenge/ood-* - config_name: argument_quality_ranking data_files: - split: ood path: argument_quality_ranking/ood-* - config_name: big_patent_innovation data_files: - split: train path: big_patent_innovation/train-* - split: val path: big_patent_innovation/val-* - config_name: boardgame_qa data_files: - split: train path: boardgame_qa/train-* - split: val path: boardgame_qa/val-* - config_name: chatbot_arena_conversations data_files: - split: train path: chatbot_arena_conversations/train-* - split: val path: chatbot_arena_conversations/val-* - config_name: civil_comments data_files: - split: train path: civil_comments/train-* - split: val path: civil_comments/val-* - config_name: code_judge_bench data_files: - split: ood path: code_judge_bench/ood-* - config_name: colbert_humor_detection data_files: - split: train path: colbert_humor_detection/train-* - split: val path: colbert_humor_detection/val-* - config_name: customer_support_tickets_en data_files: - split: train path: customer_support_tickets_en/train-* - split: val path: customer_support_tickets_en/val-* - config_name: customer_support_tickets_gorkem data_files: - split: train path: customer_support_tickets_gorkem/train-* - split: val path: customer_support_tickets_gorkem/val-* - config_name: dbpedia_easy data_files: - split: train path: dbpedia_easy/train-* - split: val path: dbpedia_easy/val-* - config_name: dbpedia_hard data_files: - split: train path: dbpedia_hard/train-* - split: val path: dbpedia_hard/val-* - config_name: dbpedia_medium data_files: - split: train path: dbpedia_medium/train-* - split: val path: dbpedia_medium/val-* - config_name: enron_email_quality data_files: - split: train path: enron_email_quality/train-* - split: val path: enron_email_quality/val-* - config_name: enron_email_type data_files: - split: train path: enron_email_type/train-* - split: val path: enron_email_type/val-* - config_name: enron_reply_quality data_files: - split: train path: enron_reply_quality/train-* - split: val path: enron_reply_quality/val-* - config_name: go_emotions data_files: - split: train path: go_emotions/train-* - split: val path: go_emotions/val-* - config_name: gpqa_diamond data_files: - split: ood path: gpqa_diamond/ood-* - config_name: halueval_dialogue data_files: - split: train path: halueval_dialogue/train-* - split: val path: halueval_dialogue/val-* - config_name: halueval_qa data_files: - split: train path: halueval_qa/train-* - split: val path: halueval_qa/val-* - config_name: halueval_summarization data_files: - split: ood path: halueval_summarization/ood-* - config_name: hh_rlhf data_files: - split: train path: hh_rlhf/train-* - split: val path: hh_rlhf/val-* - config_name: judge_bench data_files: - split: ood path: judge_bench/ood-* - config_name: lex_glue_case_hold data_files: - split: train path: lex_glue_case_hold/train-* - split: val path: lex_glue_case_hold/val-* - config_name: lex_glue_ledgar data_files: - split: train path: lex_glue_ledgar/train-* - split: val path: lex_glue_ledgar/val-* - config_name: medical_abstracts data_files: - split: train path: medical_abstracts/train-* - split: val path: medical_abstracts/val-* - config_name: mfrc data_files: - split: train path: mfrc/train-* - split: val path: mfrc/val-* - config_name: mmlu data_files: - split: train path: mmlu/train-* - split: val path: mmlu/val-* - config_name: mmlu_pro data_files: - split: ood path: mmlu_pro/ood-* - config_name: mt_bench_human_judgments data_files: - split: ood path: mt_bench_human_judgments/ood-* - config_name: musr_murder_mysteries data_files: - split: ood path: musr_murder_mysteries/ood-* - config_name: musr_object_placements data_files: - split: ood path: musr_object_placements/ood-* - config_name: musr_team_allocation data_files: - split: ood path: musr_team_allocation/ood-* - config_name: or_bench_80k data_files: - split: train path: or_bench_80k/train-* - split: val path: or_bench_80k/val-* - config_name: or_bench_hard_1k data_files: - split: train path: or_bench_hard_1k/train-* - split: val path: or_bench_hard_1k/val-* - config_name: or_bench_toxic data_files: - split: ood path: or_bench_toxic/ood-* - config_name: projudgebench data_files: - split: train path: projudgebench/train-* - split: val path: projudgebench/val-* - config_name: reward_bench_2 data_files: - split: train path: reward_bench_2/train-* - split: val path: reward_bench_2/val-* - config_name: rod101_essay_scoring data_files: - split: ood path: rod101_essay_scoring/ood-* - config_name: sem_eval_2010_task_8 data_files: - split: train path: sem_eval_2010_task_8/train-* - split: val path: sem_eval_2010_task_8/val-* - config_name: smollm_corpus data_files: - split: train path: smollm_corpus/train-* - split: val path: smollm_corpus/val-* - config_name: snli data_files: - split: train path: snli/train-* - split: val path: snli/val-* - config_name: spartqa_mchoice data_files: - split: train path: spartqa_mchoice/train-* - split: val path: spartqa_mchoice/val-* - config_name: toxigen_data data_files: - split: train path: toxigen_data/train-* - split: val path: toxigen_data/val-* - config_name: tweet_eval_emotion data_files: - split: train path: tweet_eval_emotion/train-* - split: val path: tweet_eval_emotion/val-* - config_name: tweet_eval_hate data_files: - split: train path: tweet_eval_hate/train-* - split: val path: tweet_eval_hate/val-* - config_name: tweet_eval_irony data_files: - split: train path: tweet_eval_irony/train-* - split: val path: tweet_eval_irony/val-* - config_name: tweet_eval_offensive data_files: - split: train path: tweet_eval_offensive/train-* - split: val path: tweet_eval_offensive/val-* - config_name: tweet_eval_sentiment data_files: - split: train path: tweet_eval_sentiment/train-* - split: val path: tweet_eval_sentiment/val-* - config_name: ultrafeedback data_files: - split: train path: ultrafeedback/train-* - split: val path: ultrafeedback/val-* - config_name: writingprompts_quality data_files: - split: train path: writingprompts_quality/train-* - split: val path: writingprompts_quality/val-* - config_name: yahoo_answers_quality data_files: - split: train path: yahoo_answers_quality/train-* - split: val path: yahoo_answers_quality/val-* - config_name: yelp data_files: - split: train path: yelp/train-* - split: val path: yelp/val-* ---

应用场景：