Why my chat responses are very bad?

[gaolitao]

Hi Team,

I am trying to use OpenLLM to deploy a well known LLM, like NousResearch/llama-2-13b-chat-hf as follows:
`
pip install -U -q openllm
openllm start NousResearch/llama-2-7b-chat-hf --backend pt

export OPENLLM_ENDPOINT=http://172.16.192.86:3000
openllm query 'Explain to me the difference between "further" and "farther"'
`

But the result is not that good, do you have any ideas about why? Thanks a lot in advance!