Open
Description
Name and Version
version: 5387 (3198405)
built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu
Operating systems
No response
Which llama.cpp modules do you know to be affected?
No response
Command line
Problem description & steps to reproduce
llama_context resize the rerank output to size 1 while here we still normalize it as if we have full embedding vector. I found this problem happened randomly in python binding but cannot reproduce it in cpp. Not sure if it is a bug in cpp side.
First Bad Commit
No response