4 ms·
The T0 series by big science is essentially an instruct model (though using multitask prompting instead of user feedback). You should check it out. I have got v
by domenicrosati 4y ago
The T0 series by big science is essentially an instruct model (though using multitask prompting instead of user feedback). You should check it out. I have got very competitive results on prompting t0-11b v instructgpt3(text davinci 2)
- bestcoder69 4y agoThanks, this looks awesome. But my use case is creative text generation (chatbots), which from a quick glance doesn’t seem to be a suggested use case for T0? I’ve found that simply describing to text-davinci-002 how a chatbot should act gives you more fun and believable responses. For example I trained a trump bot on 2000 tweets (davinci non-instruct fine tuning), and it generated responses that were more boring than when I just wrote a sentence saying to please tweet like trump + a couple adjectives to help it. I ran out of guest API credits on hugging face before I could trick T0 to respond with a chat completion longer than a few words. But I’ll try it some more later.