1. 08 Jun, 2022 2 commits
  2. 01 Jun, 2022 1 commit
  3. 29 Mar, 2022 2 commits
  4. 28 Mar, 2022 2 commits
  5. 23 Mar, 2022 2 commits
  6. 22 Mar, 2022 3 commits
  7. 21 Mar, 2022 1 commit
  8. 14 Mar, 2022 1 commit
  9. 09 Mar, 2022 2 commits
  10. 02 Mar, 2022 1 commit
  11. 01 Mar, 2022 2 commits
  12. 28 Feb, 2022 1 commit
  13. 18 Feb, 2022 1 commit
  14. 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
  15. 04 Mar, 2021 2 commits
  16. 23 Feb, 2021 1 commit
  17. 26 Jan, 2021 1 commit
  18. 06 Jan, 2021 2 commits
  19. 05 Jan, 2021 1 commit
  20. 30 Dec, 2020 1 commit