1. 10 Feb, 2022 1 commit
  2. 09 Feb, 2022 1 commit
  3. 19 Aug, 2021 1 commit
  4. 11 Aug, 2021 1 commit
  5. 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
  6. 16 Jul, 2021 1 commit
  7. 12 Jul, 2021 1 commit
  8. 09 Jul, 2021 1 commit
  9. 08 Mar, 2021 1 commit
  10. 04 Mar, 2021 1 commit
  11. 23 Feb, 2021 1 commit
  12. 24 Jan, 2021 1 commit
  13. 05 Jan, 2021 4 commits
  14. 31 Dec, 2020 1 commit
  15. 30 Dec, 2020 5 commits
  16. 28 Dec, 2020 1 commit
  17. 27 Dec, 2020 1 commit
  18. 25 Dec, 2020 7 commits