MuonClip draft in optax #1383
Draft
MuonClip draft in optax #1383
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
gagika
reviewed
Jul 15, 2025
| def mask_fn(path_param): | ||
| path, param = path_param | ||
| str_path = "/".join(map(str, path)) | ||
| if param.ndim == 2 and "q_proj" in str_path: |
There was a problem hiding this comment.
shouldn't you also use moun_clip for "k_proj" ?
There was a problem hiding this comment.
oh, right, definitely
| return jax.tree.unflatten(treedef, jax.tree.map(mask_fn, params_with_paths)) | ||
|
|
||
|
|
||
| def muon_clip( |
There was a problem hiding this comment.
should this also take alpha param and be used as:
Query scaling factor: etaalpha
Key scaling factor: eta(1−alpha)
There was a problem hiding this comment.
eta here is just "a" scalar, so perhaps the user can pass the correctly exponentiated eta
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Outdated suggestions cannot be applied.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.
No description provided.