Skip to content
This repository was archived by the owner on Aug 7, 2024. It is now read-only.

[FSDP2] set vocab_size=32 to avoid must be divisible by 16 error #264

Merged
merged 2 commits into from
May 20, 2024
Merged
Changes from all commits
Commits
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: 6 additions & 1 deletion test/test_fsdp2/test_fsdp2_eager.py
Original file line number Diff line number Diff line change
Expand Up @@ -57,7 +57,12 @@ def init_multi_module(self) -> nn.Module:
def init_transformer(self, weight_tying: bool) -> nn.Module:
torch.manual_seed(42)
args = ModelArgs(
n_layers=3, dim=768, n_heads=12, dropout_p=0.0, weight_tying=weight_tying
n_layers=3,
dim=768,
n_heads=12,
dropout_p=0.0,
weight_tying=weight_tying,
vocab_size=32,
)
module = Transformer(args).cuda()
self.broadcast_module(module)
Expand Down
Loading