Domovská stránka Prehľadávať Pomoc
Prihlásiť sa
radu
/
LLamaRecipes
zrkadlo https://github.com/facebookresearch/llama-recipes.git
Pridať medzi pozorované 1
Hviezda 0
Fork 0
Súbory Issues 0 Wiki
Strom: c4eacac06d
Branche Tagy
LLamaRecipes
/
recipes
/
responsible_ai
Cyrus Nikolaidis 3a99a54582 Add preprocessor to patch PromptGuard scores for inserted characters (#636) 1 rok pred
..
llama_guard 55037e4ed7 Llamaguard notebook colab link fix (#619) 1 rok pred
prompt_guard 3a99a54582 Add preprocessor to patch PromptGuard scores for inserted characters (#636) 1 rok pred
Purple_Llama_Anyscale.ipynb f53f17138b fix dead links after refactor 1 rok pred
Purple_Llama_OctoAI.ipynb f53f17138b fix dead links after refactor 1 rok pred
README.md bc56973186 update rai readme 1 rok pred
code_shield_usage_demo.ipynb 77156c7324 Minor file renames 1 rok pred
input_output_guardrails_with_llama.ipynb a404c9249c Notebook to demonstrate using llama and llama-guard together using OctoAI 1 rok pred

README.md

Trust and Safety with Llama

The Purple Llama project provides tools and models to improve LLM security. This folder contains examples to get started with PurpleLlama tools.

Tool/Model Description Get Started
Llama Guard Provide guardrailing on inputs and outputs Inference, Finetuning
Prompt Guard Model to safeguards against jailbreak attempts and embedded prompt injections Notebook
Code Shield Tool to safeguard against insecure code generated by the LLM Notebook

Running on hosted APIs

The notebooks input_output_guardrails.ipynb, Purple_Llama_Anyscale & Purple_Llama_OctoAI contain examples for running Meta Llama Guard on cloud hosted endpoints.