Chi Kim
Chi Kim
Prototype: I changed announce full position without zeros when pressing page up/down. Do we need setting for this?
From readme, "step_size : number of training iterations per half cycle. Authors suggest setting step_size = (2-8) x (training iterations in epoch) . Default 2000." Does it mean step_size should...
If I run the following code, I get this error after playing for a few seconds and stuttering. Portaudio error in Pa_AbortStream (pa_stop): Unanticipated host error Portaudio error in Pa_CloseStream...
On Windows, all three screen readers (nvda, jaws, narrator) can't seem to be able to review chat history. On Mac, users can review chat history but with a few issues....
**Describe the bug** Server List Is Not Accessible With Screen Reader on Mac **To Reproduce** 1. Open server list to join. 2. Turn on VoiceOver with command+f5 (or command+triple click...
I'm using Jekyll Now with jekyll-theme-modernist. If I put UA code for Google Analytics, it works, but if I put GA4 code, it doesn't. Does this theme support GA4? Google...
Hi, Awesome project!!! Would you be able to provide models inside public/models in Tensorflow model.save() format with weights and Architecture in addition to Tensorflow.js format? Thanks! :)
Is there a limit on number of interaction? After a while, it just exits without any error and prints out following: main: mem per token = 14368648 bytes main: load...
### Describe the bug If I run: `python server.py --notebook --chat --share --wbits 4 --model_type LLaMA --groupsize 128 --model-dir models/llama-7b --model models/llama7b-4bit-128g.pt` I get this: ``` Loading llama-7b... Could not...
### Describe the bug Visually it seems like the opposite, but new message appears at the top for screen reader for some reason. You have to navigate all the way...