Vikash

@tajo Here you go: https://codesandbox.io/s/happy-dawn-qfqfw Also the package produces some errors like: I am not sure if it's related though. ``` [18:56:47] [@snowpack/plugin-typescript] 5:55:47 PM - Starting compilation in watch...

Just added a button, and it seems to work correctly with the dark theme. I think for some reason, the changes only appear after save and refresh on this sandbox,...

@tajo were you able to isolate the issue. I think it's specific to the Navigation component, since the button looks ok.

Is there anything blocking this from merging ? This is a very nice feature to have.

Any updates on this? I am deciding between nndescent and hnsw for KNN graph construction, and it would be helpful if we had benches for the actual graph construction. Anybody...

@elinorbgr Thanks for the clarification I was wondering when would I want to use Calloop over something like: ```rust loop { tokio::select! { // Select between multiple events here. }...

@ArthurZucker if I want to batching then I have to manually add a pad_token. In this case how do I ensure that the pad_token_id is actually correct? I.e how do...

I have a comparison for the pytorch implementations with and without paging on a single GPU and the gains are significant. My use case is primarily batch inference, so I...

@okpatil4u I don't have the benchmarks for llama.cpp. I primarily noticed the speed up between the PyTorch implementations with and without paged attention. And there is no reason to think...

@okpatil4u got the numbers now. Not a rigorous benchmark, but should still hold up since the gains are so significant. WIth a 40 GB A100 GPU Inference on a vicuna-13B...