You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have a question. What if I want the bot to gain a persona using embeddings but do not answer queries based upon the same embeddings? For example, say I want to make a Steve Jobs clone bot with all his eccentricities, virtues and vices. Rather than simple prompting, I use embeddings to know 'what' Steve Jobs is (say, aggressive, innovative and a task-master). However, I do not want that the bot be probed on these (Deliberately) implanted biases and infact exhibit indifference to them. I tried doing it but I find that training on embeddings leads the bot to soon start answering in Third person (Steve Jobs is...). Is there any way to do this?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
-
I have a question. What if I want the bot to gain a persona using embeddings but do not answer queries based upon the same embeddings? For example, say I want to make a Steve Jobs clone bot with all his eccentricities, virtues and vices. Rather than simple prompting, I use embeddings to know 'what' Steve Jobs is (say, aggressive, innovative and a task-master). However, I do not want that the bot be probed on these (Deliberately) implanted biases and infact exhibit indifference to them. I tried doing it but I find that training on embeddings leads the bot to soon start answering in Third person (Steve Jobs is...). Is there any way to do this?
Beta Was this translation helpful? Give feedback.
All reactions