Incorrect link to discussion in readme
#2
by
ac3xx
- opened
In llama.cpp, and other related tools such as Ollama and LM Studio, please make sure that you have these flags set correctly, especially repeat-penalty. Georgi Gerganov (llama.cpp's author) shared his experience in https://huggingface.co/google/gemma-2b-it/discussions/38#65d2b14adb51f7c160769fa1.
This link seems to go to the wrong thing - it's just a bot post with the memory requirements. Any chance this could be updated?
Yeah, this link doesn't seem to contain the necessary flags to run the model. I will use this link for now: https://huggingface.co/google/gemma-7b-it/discussions/38#65d7b14adb51f7c160769fa1
Thanks! Updated the link :)
Xenova
changed discussion status to
closed