Suggestion: try finetune with Pippa database
#1
by
PKPL
- opened
I don't know if I can write here with this questions, but how about use additionally pippa database?
we actually already use a subset of Pippa in our dataset: Pippa is included in Dampfinchen's Creative Writing Multiturn dataset, which we filtered slightly for some slop phrases that were still present in the original.
finis-est
changed discussion status to
closed