Shawn Huang
Shawn Huang
## Description ### Content1 ... ### Content2 ... ### Content3 ... This PR has: - [ ] been self-reviewed. - [ ] concurrent read - [ ] concurrent write -...
Hi, Thanks for your great work! Since the KV Cache module has changed in transformers 4.36.0, are there any plans to update the implementation of H2O based on HF to...
Hi, Have you got any plans of implementing a backward kernel for 4 bit awq? This might be useful for quantization aware training. I tried to implement one by calculating...