You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is great stuff! I was wondering if you could also add a evaluation mode, for example to calculate perplexity in language models? Sometimes GPTQ doesn't work too well and the performance is hurt significantly. There is a risk that sometime takes a badly quantized model without knowing. So perhaps some kind of method to calculate metrics to compare the original model with the quantized model would be a great help.
The text was updated successfully, but these errors were encountered:
Hi, @iiLaurens Thanks for your suggestion, I will add evaluation mode to my todo list and perhaps add into this project this weekend.
I think the diversity of samples used to quantize a model is the key to the final quality, and I also found out that how many samples are used also have some influence. Thus in my practice, I usually using a combination of instruction-following dataset and chat dataset with thousands of samples.
This is great stuff! I was wondering if you could also add a evaluation mode, for example to calculate perplexity in language models? Sometimes GPTQ doesn't work too well and the performance is hurt significantly. There is a risk that sometime takes a badly quantized model without knowing. So perhaps some kind of method to calculate metrics to compare the original model with the quantized model would be a great help.
The text was updated successfully, but these errors were encountered: