DS Chat Step 3 - Add separate Lora Adam optimizer group #633
Conversation
This file contains 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
lekurile
requested review from
jeffra,
samyam,
tjruwase,
ShadenSmith,
conglongli,
awan-10,
cli99,
eltonzheng,
minjiaz,
RezaYazdaniAminabadi,
duli2012,
mrwyattii,
yaozhewei,
arashb and
xiaoxiawu-microsoft
as code owners
awan-10
approved these changes
Jul 10, 2023
LeetJoe
pushed a commit
to LeetJoe/DeepSpeedExamples
that referenced
this pull request
Sep 15, 2023
This PR adds a separate Lora Adam optimizer group (for lora_right_weight and lora_left_weight params) with a Lora-specific learning rate of lr=5e-4. After this change, Step 3 training convergence with Lora enabled improved across various configurations when using zero stage 2. Thanks to @yaozhewei for the insight!
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Outdated suggestions cannot be applied.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.
This PR adds a separate Lora Adam optimizer group (for
lora_right_weightandlora_left_weightparams) with a Lora-specific learning rate oflr=5e-4. After this change, Step 3 training convergence with Lora enabled improved across various configurations when using zero stage 2.Thanks to @yaozhewei for the insight!
BEFORE:
AFTER:
