-
Notifications
You must be signed in to change notification settings - Fork 308
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Hierarchical CP implementation (Ulysses + Ring) #1209
base: main
Are you sure you want to change the base?
Commits on Sep 19, 2024
-
change API for hierarchical CP
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 2ee1389 - Browse repository at this point
Copy the full SHA 2ee1389View commit details -
Configuration menu - View commit details
-
Copy full SHA for 3743c5f - Browse repository at this point
Copy the full SHA 3743c5fView commit details
Commits on Sep 20, 2024
-
move fp8 code before qkv reshape
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 0e826e7 - Browse repository at this point
Copy the full SHA 0e826e7View commit details
Commits on Sep 23, 2024
-
try to insert A2A for hierarchical CP
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 44db43b - Browse repository at this point
Copy the full SHA 44db43bView commit details -
Configuration menu - View commit details
-
Copy full SHA for e717b81 - Browse repository at this point
Copy the full SHA e717b81View commit details
Commits on Sep 24, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 705da7e - Browse repository at this point
Copy the full SHA 705da7eView commit details -
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 99ace46 - Browse repository at this point
Copy the full SHA 99ace46View commit details -
make bwd of hierarchical CP work
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for f73b2fb - Browse repository at this point
Copy the full SHA f73b2fbView commit details -
Configuration menu - View commit details
-
Copy full SHA for 0be5dd7 - Browse repository at this point
Copy the full SHA 0be5dd7View commit details -
Configuration menu - View commit details
-
Copy full SHA for a6833c7 - Browse repository at this point
Copy the full SHA a6833c7View commit details -
assert hierarchical CP implementation does not support THD format
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for caf3746 - Browse repository at this point
Copy the full SHA caf3746View commit details
Commits on Sep 25, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 03806c4 - Browse repository at this point
Copy the full SHA 03806c4View commit details -
Configuration menu - View commit details
-
Copy full SHA for d3d336a - Browse repository at this point
Copy the full SHA d3d336aView commit details -
assert hierarchical CP does not support attn bias
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for ea1e4a3 - Browse repository at this point
Copy the full SHA ea1e4a3View commit details -
add unit test for hierarchical CP
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 3cd21ab - Browse repository at this point
Copy the full SHA 3cd21abView commit details -
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 90c0bb8 - Browse repository at this point
Copy the full SHA 90c0bb8View commit details -
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 8c6139d - Browse repository at this point
Copy the full SHA 8c6139dView commit details -
Configuration menu - View commit details
-
Copy full SHA for 52c7ae3 - Browse repository at this point
Copy the full SHA 52c7ae3View commit details -
Configuration menu - View commit details
-
Copy full SHA for 1db4e8b - Browse repository at this point
Copy the full SHA 1db4e8bView commit details -
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 1655edc - Browse repository at this point
Copy the full SHA 1655edcView commit details
Commits on Sep 26, 2024
-
Configuration menu - View commit details
-
Copy full SHA for edbe898 - Browse repository at this point
Copy the full SHA edbe898View commit details
Commits on Sep 27, 2024
-
[pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Configuration menu - View commit details
-
Copy full SHA for b096051 - Browse repository at this point
Copy the full SHA b096051View commit details -
Configuration menu - View commit details
-
Copy full SHA for 6063bec - Browse repository at this point
Copy the full SHA 6063becView commit details
Commits on Sep 28, 2024
-
move function definitions to the front of the first call
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 67c590d - Browse repository at this point
Copy the full SHA 67c590dView commit details