Does not run on latest llama.cpp
#2
by
vico44
- opened
Hello,
First of: Thank you for all the good work! 🙂
When trying to run LFM2-8B-A1B-BF16 with latest llama.cpp i get this error:
llama_model_load: error loading model: missing tensor 'token_embd_norm'
It does run with b7002 but does not with b7446
Hi @vico44 , the regression was fixed in https://github.com/ggml-org/llama.cpp/pull/18132, try the latest release b7674 .