Skip to content

fix(embedding): fix incorrect computing of indexes in _update_cos_sin_cache #311

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

Merged
merged 1 commit into from
Sep 2, 2024
Merged

fix(embedding): fix incorrect computing of indexes in _update_cos_sin_cache #311

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
14 changes: 6 additions & 8 deletions internlm/model/modules/embedding.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -97,13 +97,11 @@ def _update_cos_sin_cache(
if max_seqlen is not None:
seqlen = max_seqlen
elif isinstance(indexes, int):
# logic changed temporaryly
# seqlen = indexes + x.shape[1] + 1
seqlen = gpc.config.data.seq_len
seqlen = indexes + x.shape[1]
else:
# Note that this statement may cause synchronization between CPU and GPU,
# so it's best to precompute and pass in max_seqlen ahead of time
seqlen = indexes.max().item() + 1
seqlen = indexes.max().item()

# Reset the tables if the sequence length has changed,
# or if we're on a new device (possibly due to tracing for instance)
Expand Down Expand Up @@ -219,9 +217,9 @@ def __init__(
def _update_cos_sin_cache(self, x, indexes):
"""x: (batch, seqlen, nheads, headdim)"""
if not isinstance(indexes, int):
seqlen = indexes.max().item() + 1
seqlen = indexes.max().item()
else:
seqlen = indexes + x.shape[1] + 1
seqlen = indexes + x.shape[1]

t = torch.arange(seqlen, device=x.device, dtype=self.inv_freq.dtype)
t = t / self.scaling_factor
Expand Down Expand Up @@ -286,9 +284,9 @@ def _update(self, seqlen, x):
def _update_cos_sin_cache(self, x, indexes):
"""x: (batch, seqlen, nheads, headdim)"""
if not isinstance(indexes, int):
seqlen = indexes.max().item() + 1
seqlen = indexes.max().item()
else:
seqlen = indexes + x.shape[1] + 1 # eval_forward
seqlen = indexes + x.shape[1] # eval_forward
if seqlen <= self.max_position_embeddings:
# Reset the tables if the sequence length has changed,
# or if we're on a new device (possibly due to tracing for instance)
Expand Down