Skip to content

[AutoParallel] fix GPT embedding placements #10834

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: develop
Choose a base branch
from
Open

[AutoParallel] fix GPT embedding placements #10834

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 4 additions & 3 deletions paddlenlp/transformers/gpt/modeling_auto.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -658,7 +658,7 @@ def __init__(
config.hidden_size,
)
self.word_embeddings.weight = dist.shard_tensor(
self.word_embeddings.weight, get_mesh(), [dist.Replicate(), dist.Replicate()]
self.word_embeddings.weight, get_mesh(), [dist.Replicate(), dist.Shard(1)]
)
self.position_embeddings.weight = dist.shard_tensor(
self.position_embeddings.weight, get_mesh(), [dist.Replicate(), dist.Shard(1)]
Expand Down Expand Up @@ -699,6 +699,7 @@ def forward(self, input_ids, position_ids=None, inputs_embeddings=None):
# The 'with' block ensures the correct seed context is used
with seed_guard_context(current_seed):
embeddings = self.dropout(embeddings)
embeddings = dist.reshard(embeddings, get_mesh(), [dist.Replicate(), dist.Replicate()])
return embeddings


Expand Down Expand Up @@ -1176,7 +1177,7 @@ def __init__(self, config: GPTConfig, embedding_weights=None, ipp=None):
shape=[config.vocab_size, config.hidden_size],
dtype=paddle.get_default_dtype(),
)
self.weight = dist.shard_tensor(self.weight, get_mesh(self.ipp), [dist.Replicate(), dist.Shard(0)])
self.weight = dist.shard_tensor(self.weight, get_mesh(self.ipp), [dist.Replicate(), dist.Shard(1)])

def forward(self, hidden_states, tensor_parallel_output=None):

Expand All @@ -1187,7 +1188,7 @@ def forward(self, hidden_states, tensor_parallel_output=None):
if tensor_parallel_output is None:
tensor_parallel_output = self.config.tensor_parallel_output

y = dist.reshard(self.weight, get_mesh(self.ipp), [dist.Replicate(), dist.Shard(0)])
y = dist.reshard(self.weight, get_mesh(self.ipp), [dist.Replicate(), dist.Shard(1)])
logits = paddle.matmul(hidden_states, y, transpose_y=self.transpose_y)
return logits

Expand Down
Loading