llama.cpp GGML_ASSERT: ggml.c:4014: false zsh: abort ./main -m ./models/65B/ggml-model-q4_0.bin -t 16 -n 256 --repeat

GGML_ASSERT: ggml.c:4014: false zsh: abort ./main -m ./models/65B/ggml-model-q4_0.bin -t 16 -n 256 --repeat_penalty 1.0

Open nazthelizard122 opened this issue 1 year ago • 5 comments

Not sure why this happens, I am on the latest commit and I am up-to-date on everything I did some tests and it seems like it breaks after 500~ tokens Is this a model limitation or can I fix this by increasing some value?

Mar 22 '23 17:03 nazthelizard122

Please review and use our issue template to provide more details so we can better understand your problem and answer you.

Mar 22 '23 17:03 gjmulder

Idk how to use it

Mar 22 '23 17:03 nazthelizard122

Sorry, better link

Copy everything from line 10 and try to complete as requested.

Mar 22 '23 17:03 gjmulder

idk how to get the check mark

Mar 23 '23 22:03 nazthelizard122

now I'm running into another error, it segfaults with the regular script

Mar 23 '23 22:03 nazthelizard122

llama.cpp llama.cpp copied to clipboard

GGML_ASSERT: ggml.c:4014: false zsh: abort ./main -m ./models/65B/ggml-model-q4_0.bin -t 16 -n 256 --repeat_penalty 1.0

llama.cpp
llama.cpp copied to clipboard