Remove zero initialization of to_attn_bias
weights, since for these bias=False
#302
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Outdated suggestions cannot be applied.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.
Line 8 of Algorithm 24 uses a
LinearNoBias
layer followingLayerNorm
to merge pairwise with attention bias representations, as followsHowever, the code has the
LinearNoBias
weights also initialized with zeros, meaning both the weights and biases of these modules are initially zero (or null) which leads to these weights receiving no gradients throughout training.