Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention Paper • 2502.11089 • Published 15 days ago • 138
Running on CPU Upgrade 110 110 Open Chinese LLM Leaderboard 🏆 Display and filter LLM benchmark results
view article Article Llama 3.1 - 405B, 70B & 8B with multilinguality and long context Jul 23, 2024 • 228