[BugFix] fix_outputlayer.weight_distributed #9135
Open
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
|
Thanks for your contribution! |
wtmlon
reviewed
Sep 13, 2024
| hidden_states = paddle.reshape_(hidden_states, [self.config.seq_length, -1, self.config.hidden_size]) | ||
| logits = parallel_matmul(hidden_states, self.decoder_weight, self.config.tensor_parallel_output) | ||
| if self.config.tensor_parallel_degree > 1: | ||
| logits = parallel_matmul(hidden_states, self.weight, self.config.tensor_parallel_output) |
There was a problem hiding this comment.
为啥要在这里parallel_matmul，我看只有非 tp 的情况才会走进这里的 forward
| if return_last_logit: | ||
| hidden_states = hidden_states[-1:] | ||
| if self.config.tensor_parallel_degree > 1: | ||
| lm_logits = parallel_matmul( |
There was a problem hiding this comment.
这里为啥要手动掉用 parallel_matmul，内部的 forward 函数我看已经支持 tp > 1的情况，具体在哪一层面做 parallel_matmul还是统一点好
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## develop #9135 +/- ##
===========================================
- Coverage 53.29% 53.28% -0.02%
===========================================
Files 652 652
Lines 105483 105579 +96
===========================================
+ Hits 56222 56254 +32
- Misses 49261 49325 +64 ☔ View full report in Codecov by Sentry. |
|
This Pull Request is stale because it has been open for 60 days with no activity. 当前Pull Request 60天内无活动，被标记为stale。 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
PR types
Bug fixes
PR changes
models
Description
手动切分参数时, 需要设置is_distributed为True, distributed_model中广播参数时就会把这些切分的参数跳过
初始loss 11->1.63