Fixes -Ofast optimizations breaking model by skipping them for gelu_backward #200
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
|
let's try it, merged, ty! |
There was a problem hiding this comment.
@dagelf - This line breaks the Windows build:
__attribute__((optimize("no-finite-math-only"))) - Can you wrap this in a conditional for GCC? train_gpt2.c line 368. Thanks!
|
Sorry, was sure someone said it worked and the only remaining issue was
warning on `clang`. If we're going to have ifdefs we might as well remove
the warnings for `clang` too.
|
dagelf
added a commit
to dagelf/llm.c
that referenced
this pull request
Apr 22, 2024
dagelf
added a commit
to dagelf/llm.c
that referenced
this pull request
Apr 22, 2024
karpathy
added a commit
that referenced
this pull request
Apr 22, 2024
Make msvc compile by adding the ifdefs back in gelu_backward, fixes #200
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Please see #168 for details and work being done on optimizing the activation function on CPU.
This PR results in a 10-23% speedup on CPU and has been tested on
msvcandgccon x86 and fixes #19