llama.cpp
Stop the generation when <|eom_id|> token is encountered (needed for llama 3.1 tool call support)
#8858
Merged
Merged
Stop the generation when <|eom_id|> token is encountered (needed for llama 3.1 tool call support) #8858
fairydreaming
merged
5 commits into
ggml-org:master
from
fairydreaming:handle-eom-token
llama-vocab, llama : handle <|eom_id|> Llama-3.1 token
cc50e78f
gguf-py : add constants and method related to <|eom_id|> token
f10b0e2c
Merge branch 'ggerganov:master' into handle-eom-token
3878b397
llama : Use token_to_id map find() method instead of iterating over a…
0b721138
llama : whitespace formatting
5efd8264
ggerganov
approved these changes
on 2024-08-05
Login to write a write a comment.
Login via GitHub