kiraaa

6 ms·

1. 1.▲
   by kiraaa 2y ago

   when there are two commands in a prompt example do A and then do B. the model completely ignores the second task B.
2. 2.▲
   by kiraaa 2y ago

   on gpu that is still huge.
3. 3.▲
   by kiraaa 3y ago

   super easy to install and use
4. 4.▲
   by kiraaa 3y ago

   mistral 7b v0.2 supports 32k
5. 5.▲
   by kiraaa 3y ago

   could be true, we can only speculate.
6. 6.▲
   by kiraaa 3y ago

   maybe they are using ring attention, on top of their 128k model.
7. 7.▲
   by kiraaa 3y ago

   really like the font and great article btw.
8. 8.▲
   by kiraaa 3y ago

   the paper does not live up to the quality of model lol
9. 9.▲
   by kiraaa 3y ago

   you need 94gb, does not matter which RAM.
10. 10.▲
    by kiraaa 3y ago

    https://github.com/sysid/sse-starlette makes token streaming so much easier in python
11. 11.▲
    by kiraaa 3y ago

    and also easy to deploy
12. 12.▲
    by kiraaa 3y ago

    its not alway about the size, but yeah its really good!
13. 13.▲
    by kiraaa 3y ago

    installing it is a nightmare
14. 14.▲
    by kiraaa 3y ago

    you could but it will be very slow. oh and make sure the code is cpu compatible.