[Quantization] Channel wise output activation quantization for QKV Attention layers#270

horheynm wants to merge 5 commits intomainvllm-project/compressed-tensors:mainfrom

attn_quantvllm-project/compressed-tensors:attn_quantCopy head branch name to clipboard