1. 08 May, 2024 1 commit
  2. 07 May, 2024 3 commits
  3. 04 May, 2024 3 commits
    • DearPlanet's avatar
    • Michael Goin's avatar
      [Kernel] Support MoE Fp8 Checkpoints for Mixtral (Static Weights with... · 2a052011
      Michael Goin authored
      [Kernel] Support MoE Fp8 Checkpoints for Mixtral (Static Weights with Dynamic/Static Activations) (#4527)
      
      Follow on to #4332 to enable FP8 checkpoint loading for Mixtral and supersedes #4436.
      
      This PR enables the following checkpoint loading features for Mixtral:
      
      Supports loading fp8 checkpoints for Mixtral, such as this "nm-testing/Mixtral-8x7B-Instruct-v0.1-FP8" test model
      Supports static or dynamic activation quantization with static weight quantization (all per tensor)
      Supports different scales for each expert weight
      Supports Fp8 in QKV layer
      Notes:
      
      The Expert Gate/Router always runs at half / full precision for now.
      If there are different weight scales between QKV layer (for separate QKV weights), they are re-quantized using layer.weight_scale.max() so we can have a single gemm for performance.
      2a052011
    • Cody Yu's avatar
      bc8ad684
  4. 03 May, 2024 5 commits
  5. 02 May, 2024 7 commits
  6. 01 May, 2024 7 commits
  7. 30 Apr, 2024 3 commits
  8. 29 Apr, 2024 2 commits
  9. 27 Apr, 2024 6 commits
  10. 26 Apr, 2024 3 commits