Spaces:
Running
Running
Hypothesis that bot might answer to the wrong recipient
#27
by
fffiloni
- opened
Same case here.
This might be due to the decoding parameters of the model.
Specifically the decoding parameters that hf decided to use for this model . The Open Assistant website is much much more stable.
Interesting
we added a system prompt and now this should happen way more rarely
Can you confirm? (BTW it was not leaking data, it was random hallucinations)
i'll close for now
julien-c
changed discussion status to
closed