feat: Support of encoder #24
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
- TransformerEncoder stage block for non-causal attention. - Masked MBLM for representation learning. - A modified PG19 dataset supporting masked language modeling. - MaskedTrainer to uses the MaskedMBLM and the corresponding (masked) dataset.
jorisSchaller
force-pushed
the
feat/encoder
branch
from
9ce03d6 to
0db4c9a
Compare
jannisborn
requested changes
Apr 15, 2025
Contributor
jannisborn
left a comment
jannisborn
left a comment
There was a problem hiding this comment.
Amazing job @jorisSchaller! I see you're testing up to a 2D encoder and the CI passes, so I think the actual work is done, great 💪🏼 👍🏼
I have some cosmetic comments, see below for details. Most important is the naming, I would suggest to rename MaskedMBLM to MBLMEncoder or EncoderMBLM because MLM is more of a training strategy than a model type, wdyt?
Also, the transformer class definitions are a bit redundant and could be made more efficient with inheritance but I let you decide on this because everything looks technically correct!
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Outdated suggestions cannot be applied.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 commit adds:
- TransformerEncoder stage block for non-causal attention.
- Masked MBLM for representation learning.
- A modified PG19 dataset supporting masked language modeling.
- MaskedTrainer to use with the MaskedMBLM and the corresponding (masked) dataset.
TODOs: