README.md

metadata

base_model: google/gemma-2b

This lora is trained on openai content moderation data as well as Nvidias Aegis dataset.

Prompt Template:

<|im_start|>system
{system}<|im_end|>
<|im_start|>user
{user}<|im_end|>
<|im_start|>assistant