Rewrite dataset card: 3 languages, flat schema, EN-only decomposition, BabyLM freq, closed-class 457d2fc verified yuanxin112 commited on 19 days ago
Flatten schema: all fields top-level (no morphology/semantics nesting) 3db1594 verified yuanxin112 commited on 19 days ago
Card: uniform EN source line; freq corpora all BabyLM with sizes 13038a2 verified yuanxin112 commited on 19 days ago
decomposition: English-only, explicit Wiktionary templates; DE/FR drop it 86ef415 verified yuanxin112 commited on 19 days ago
Include closed-class words (prep/conj/pron/det/...); update counts 49e54b8 verified yuanxin112 commited on 19 days ago
Filter untagged inflected-form/variant glosses; update counts 3fdd966 verified yuanxin112 commited on 19 days ago
Add full dataset card (fields, morph_type, frequency signals, filtering rules) 8f1aa61 verified yuanxin112 commited on 19 days ago