polish-dynaword / data /wiktionary_examples /wiktionary_examples.md
kacperwikiel's picture KateMajzel's picture
feat(data): add wiktionary_examples (8.6k docs / 1.12M tok) (#14)
85e7d9b
|
Raw
History Blame Contribute Delete
1.59 kB

wiktionary_examples

Polish Wiktionary usage examples

Dataset description

Licensing — traceable basis

Wikimedia dumps explicitly released under CC-BY-SA 3.0; attribution to contributors + share-alike preserved. Derived subset: usage-example sections from Polish-language entries only, grouped one document per headword.

Provenance

Parsed directly from the official Wikimedia dump (plwiktionary-latest-pages-articles.xml.bz2); not routed through SpeakLeash. Usage-example sections of Polish-language entries extracted and grouped one document per headword.

Statistics

documents characters tokens (tiktoken proxy)
8,639 2,889,761 1,115,820

Per-document license metadata

license documents
CC-BY-SA-3.0 8,639

Author metadata present for 8,639 documents. Empty values mean the upstream record did not expose a machine-readable author field.

Filters applied (build_dynaword.py)

Minimal, per Dynaword guidelines (heavy filtering left to downstream use):

  • drop documents < 200 chars: 0
  • drop non-Polish (diacritic ratio): 0
  • exact cross-source dedup (sha1): 481
  • OCR alpha-ratio < 0.70 (OCR sources only): 0
  • read 9,120 → kept 8,639

Token counts are a fast tiktoken (cl100k) proxy (~1% off Llama-3); the canonical Llama-3 count is computed at release.