conftrace_

hopper gpu

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

attention efficiency (1) low-precision arithmetic (1) fp8 quantization (1) gpu computation (1) gpu optimization (1) low-precision computation (1) tensor core (1) transformer optimization (1) gpu acceleration (1) flash attention (1)

Papers

FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision NeurIPS 2024