4 ms·
I was just thinking that it would be great to do something like seti@home but for an open libre super LLM: Install a client that would help run / train the hug
by xtracto 4y ago
I was just thinking that it would be great to do something like seti@home but for an open libre super LLM: Install a client that would help run / train the huge llm and you get Access to the network to do queries. Kind of how napster/Kazaa were at their time.
I would love to work on doing something like that.
- Freebytes 4y agoThat is actually a great idea. One of the issues with a network, however, is how the output would be produced in such a scenario and how the dictionary would be controlled and how the training data could be managed. The transformer model allows different RNNs to be separated, but it would be challenging to 'put it all together'. Furthermore, bad actors could intentionally poison the network during the training process. Furthermore, how would upgrades work? When new techniques are discovered, it would be hard to upgrade it. We see these challenges already when it comes to major changes in cryptocurrency networks.
- j5155 4y agoWhat you’re describing is http://chat.petals.ml/ http://chat.petals.ml/
- xtracto 4y agoThanks a lot for the link. It is an interesting project. I tried the Google Notebook and it was a trip: https://ibb.co/qpbzBZJ https://ibb.co/qpbzBZJ Somehow, the AI believes that Alan Shepard was the first person on space. Doing this session reminded me A LOT of Asimov's short stories in "I Robot". I feel that the future AI "debugging" session will be like that. It's been a while since I read the book, so I'll have to re-read it again.
- Freebytes 4y agoExcellent. I will check that out and see if I can contribute.