Transformers
GGUF
English
Not-For-All-Audiences
llama-cpp
gguf-my-repo
Inference Endpoints
conversational

Doesn't work on Ollama

#1
by Youcef - opened

I'm using ollama version 0.3.0, and I get an error when I try to run celeste:

Error: llama runner process has terminated: error:check_tensor_dims: tensor 'blk.0.attn_q.weight' has wr
ong shape; expected  5120,  5120, got  5120,  4096,     1,     1

mistral nemo works fine though? so I don't understand the problem

Sign up or log in to comment