This repository was archived by the owner on Aug 7, 2024. It is now read-only.
add numerical test on SAM encoder #5
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Summary: Adds a check that using `Float8Linear` on a `SAM` encoder results in reasonable spot check accuracy. Note that grad accuracy is not tested as it is all over the place, this is probably expected but saving investigation until later. Specifically the layernorm and positional encoding grads have a large error for fp8 version vs reference. Test Plan: ``` python float8_playground/test.py with-proxy python float8_playground/test_sam.py ``` Reviewers: Subscribers: Tasks: Tags:
facebook-github-bot
added
the
CLA Signed
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
Adds a check that using
Float8Linearon aSAMencoder results in reasonable spot check accuracy.Note that grad accuracy is not tested as it is all over the place, this is probably expected but saving investigation until later. Specifically the layernorm and positional encoding grads have a large error for fp8 version vs reference.
Test Plan:
Reviewers:
Subscribers:
Tasks:
Tags: