Keith Stevens

I'm having the same problem (can't find `cget_col_row_stats`) when using an A6000.

I tried upgrading cuda to 11.6 (and pytorch to match) and I still get the same error. Having looked at the code I'm guessing some `#DEFINE` didn't get included in...

After some fixes, my situation is a bit confusing. I'm running Jupyter in a docker container. When running `python -m bitsandbytes` in a jupyter shell, I get the following: ```...

I managed to fix my personal situation. I'm pretty sure it's something weird with how I'm custom building my GPU enabled jupyter image. For reasons unknown to me I have...

Adding to this, if someone gets around to support independent LoRA adapter weights, I'd like to request a particular architecture difference that makes it easier to switch between adapters. Right...

I wrote that PR for FastChat. That's actually not the most preferred solution since it requires walking through the model's list of modules and updating them to activate/deactivate the right...

I still need to go through and fix the javadocs and the minor suggestions, but I put in my thoughts on some of the larger suggestions.

Most of the javadoc and simple cleaning changes fixed. There's still a few more to be handled.

Note: we discussed this a bit in discord and agreed with this sledgehammer solution. When we have to deal with intentionally multi-lingual tasks, we'll think about that more carefully.

Agreed. We should: 1. Reduce the set of labels to something small and meaningful 2. Use a Likert scale for most of them 3. Rewarding high effort posts effectively