Hugging Face's logo

meta-llama
/
Llama-3.1-8B-Instruct

Text Generation
Transformers
Safetensors
PyTorch
llama
facebook
meta
llama-3
conversational
text-generation-inference
Model card Files
xet
Community
313

Add missing space in prompt template

#113
by Rocketknight1 HF Staff - opened
base: refs/heads/main
←
from: refs/pr/113
Discussion Files changed
+1
-1
Add missing space in prompt template0aaef8a7
Rocketknight1
No description provided.
Update tokenizer_config.json1c29a4b2
essobi

Thank you!
carljparker
 Meta Llama org

LGTM. Can we replicate this change to the 70B and 405B models as well?
Rocketknight1

@carljparker Sure!

70B
405B
Rocketknight1

Also thank you to @essobi for spotting this discrepancy in the first place!
cosmicchichu

Would this affect how the model behaves in any way?
essobi

Based on my testing it was causing the model to skip spaces on the 7B.

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment