ContextualAI/llama-2-70b-chat-hf

Fork of daryl149/llama-2-70b-chat-hf with max_position_embeddings set to 4096 instead of 2048.

There are a few small differences between this and Meta's version

• temperature and top_p are specified in Meta's, but not here.
• Meta uses transformers 4.31.0.dev0 instead of 4.31.0.