← Back to context Comment by svara 1 day ago ... but not in deep learning or am I missing something important here? 2 comments svara Reply qeternity 1 day ago Yes, absolutely in deep learning. Custom fused CUDA kernels everywhere. Scene_Cast2 1 day ago Yep. MoE, FlashAttention, or sparse retrieval architectures for example.
qeternity 1 day ago Yes, absolutely in deep learning. Custom fused CUDA kernels everywhere. Scene_Cast2 1 day ago Yep. MoE, FlashAttention, or sparse retrieval architectures for example.
Yes, absolutely in deep learning. Custom fused CUDA kernels everywhere.
Yep. MoE, FlashAttention, or sparse retrieval architectures for example.