The title, pretty much.
I’m wondering whether a 70b model quantized to 4bit would perform better than a 7b/13b/34b model at fp16. Would be great to get some insights from the community.
You must log in or register to comment.
In my experience, the lower you go…the model:
- hallucinates more (one time I asked Llama2 what made the sky blue and it freaked out and generated thousands of similar questions line by line)
- is more likely to give you an inaccurate response when it doesn’t hallucinate
- is significantly more unreliable and non-deterministic (seriously, providing the same prompt can cause different answers!)
At the bottom of this post, I compare the 2-bit and 8-bit extreme ends of Code Llama Instruct model with the same prompt and you can see how it played out: https://about.xethub.com/blog/comparing-code-llama-models-locally-macbook
70b 4bit will eat those small models for breakfast.
A friend told me that for 70b when using q4, performance drops by 10%. The larger the model, the less it suffers from weight quantization
It’s a rule of thumb that yes, higher parameter at low quant beats lower parameter at high quant (or no quant) but take it with a grain of salt as you may still prefer a lower parameter model that’s more tuned for the task you prefer.