Skip to content

GitLab

Switch branch/tag
  1. 11 Feb, 2022 2 commits
  3. 10 Feb, 2022 2 commits
  5. 09 Feb, 2022 1 commit
  7. 19 Aug, 2021 1 commit
  9. 11 Aug, 2021 1 commit
  11. 30 Jul, 2021 1 commit
    • Deepak Narayanan's avatar
      Support for pipeline parallelism in T5 model · 46c74b4c
      Deepak Narayanan authored 
      - Accumulate encoder hidden state gradient to handle skip connection
- Correctly compute the number of layers in encoder / decoder for T5 model
- Ensure e weights are initialized the same way in embeddings
- Synchronize embedding gradients across encoder and decoder for T5 model
- Support for checkpoint loading and saving
      46c74b4c
  13. 16 Jul, 2021 1 commit
  15. 12 Jul, 2021 1 commit
  17. 09 Jul, 2021 1 commit
  19. 08 Mar, 2021 1 commit
  21. 04 Mar, 2021 1 commit
  23. 23 Feb, 2021 1 commit
  25. 24 Jan, 2021 1 commit
  27. 05 Jan, 2021 4 commits
  29. 31 Dec, 2020 1 commit
  31. 30 Dec, 2020 5 commits
  33. 28 Dec, 2020 1 commit
  35. 27 Dec, 2020 1 commit
  37. 25 Dec, 2020 7 commits