Open SpaceHunterInf opened 1 month ago
And a small question, can you (or how can I see) if my llama.cpp is compiled with GPU support? I used brew to install it, but it doesn't respond to the -ngl 32
command when I try to load all 32 layers of Llama 3.1 8B to my GPU.
Many thanks.
And sometimes the prefix of the reverse prompt disappears... How can I log the input of the model so that I can trace whether it's a bug from the llama cpp or the terminal display? Thanks
What happened?
Greetings,
I am not sure if this is a bug or unimplemented feature. I am using terminal for interactive conversations. I just realise that once you typed too much and you swapped to the newline, you are not able to use backspace to delete characters in the previous lines anymore. Is this what supposed to happen?
Thanks
Name and Version
version: 3892 (96b69121) built with gcc-11 (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu
What operating system are you seeing the problem on?
Linux
Relevant log output
No response