HN Explorer

luxiedge

↗ HN profile [ 501 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. The answer shouldn't depend on which GPU you rented (regularjoeceo.substack.com)
    1 points 0
  2. 1 points 0
  3. 1 points 0
  4. Constant 14ms attention: 512→524K tokens (24.5x faster than FlashAttention) (github.com)
    1 points 1
  5. Show HN: O(1) memory attention – 512K tokens in 3.85 GB (eval binary) (github.com)
    1 points 0
  6. Why do math libraries produce different results across platforms? (github.com)
    3 points 3
  7. 7 points 6