I just deleted my re-tweet of some claims about GPT and reddit. @jackclarkSF just informed me that it was misleading.
It's a complex challenge: training systems to understand human language without contaminating them with some of the uglier parts of humanity.
Can you please share what is misleading about my tweet?? There are peer reviewed papers that I have cited. This is not fair to have back room discussions and not even share why such accusations are being made.
-
-
The very specific point was to disentangle training on links that come from places posted on reddit, as opposed to training on text broadly posted on reddi. This is in reference to a conversation we had last yr:https://twitter.com/jackclarkSF/status/1101731603020050432 …
- Show replies
New conversation -
-
I think it would be wonderful if
@AnimaAnandkumar and@jackclarkSF (and others with relevant expertise) discussed this here on this twitter thread. I think we all agree there are some serious challenges because humans have biases and those biases are reflected in training data. -
I clarified stuff here before
@AnimaAnandkumar tweet - looks like we just missed eachother.https://twitter.com/jackclarkSF/status/1271546631192641536 …
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.