1. 09 Jul, 2019 4 commits
  2. 05 Jul, 2019 2 commits
  3. 03 Jul, 2019 3 commits
  4. 02 Jul, 2019 4 commits
  5. 26 Jun, 2019 2 commits
  6. 20 Jun, 2019 1 commit
  7. 19 Jun, 2019 2 commits
  8. 18 Jun, 2019 2 commits
  9. 17 Jun, 2019 4 commits
  10. 14 Jun, 2019 1 commit
  11. 01 Jun, 2019 2 commits
  12. 08 May, 2019 3 commits
  13. 07 May, 2019 2 commits
  14. 30 Apr, 2019 3 commits
  15. 16 Apr, 2019 1 commit
    • Abhi Sharma's avatar
      Fix gradient overflow issue during attention mask · 9e666aaa
      Abhi Sharma authored
      This fix is in reference to issue #382. GPT2 can now be trained in mixed precision, which I've confirmed with testing. I also tested unconditional generation on multiple seeds before and after changing 1e10 to 1e4 and there was no difference. Please let me know if there is anything else I can do to make this pull request better. Thanks for all your work!
      9e666aaa
  16. 15 Apr, 2019 3 commits
  17. 27 Mar, 2019 1 commit