1. 14 May, 2021 1 commit
    • Jared Casper's avatar
      Update arguments checks. · 8044c7b4
      Jared Casper authored
      hidden_size % attention_heads == 0 is handled above when dealing with kv_channels.
      
      Adding check for decoder sequence length.
      8044c7b4
  2. 22 Apr, 2021 3 commits
  3. 21 Apr, 2021 4 commits
  4. 20 Apr, 2021 3 commits
  5. 19 Apr, 2021 1 commit
  6. 16 Apr, 2021 2 commits
  7. 12 Apr, 2021 2 commits
  8. 08 Apr, 2021 2 commits
  9. 03 Apr, 2021 3 commits
  10. 02 Apr, 2021 3 commits
  11. 01 Apr, 2021 1 commit
  12. 31 Mar, 2021 2 commits
  13. 30 Mar, 2021 1 commit
  14. 26 Mar, 2021 1 commit
  15. 24 Mar, 2021 3 commits
  16. 23 Mar, 2021 1 commit
  17. 20 Mar, 2021 1 commit
  18. 19 Mar, 2021 5 commits
  19. 18 Mar, 2021 1 commit