search

646e62 / case-brief

Generates a FIRAC-style case brief from a reported decision
GNU General Public License v3.0
2 stars 0 forks source link

Large-file queries exceed token limits #48

Closed 646e62 closed 1 year ago

646e62 commented 1 year ago

The program currently uses spaCy tokens to measure a prompt's size. This sometimes causes problems as there's no one-to-one correlation between spaCy tokens and GPT-2 tokens. Switching to using GPT-2 token lengths instead will resolve.