jessiewiswjc

I met the same error. @WoosukKwon @simon-mo Can you give us some help? Decrease `swap-space` in vllm or change `RAY_memory_usage_threshold` or `RAY_memory_monitor_refresh_ms` in ray?

> @danthe3rd I also need alibi support. for now, I pass `bias = LowerTriangularMaskWithTensorBias(alibi_bias)` to `xops.memory_efficient_attention(..., attn_bias=bias )`. The forward only is ok, but failed at backward in training mode....

同问