Skip to content
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

Jump to bottom

Fix Gemma 7B checkpoint save #1169

Merged
merged 1 commit into from
Jul 12, 2024
Merged

Fix Gemma 7B checkpoint save #1169

Show file tree
Hide file tree
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
4 changes: 3 additions & 1 deletion torchtune/models/convert_weights.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -249,6 +249,7 @@ def tune_to_peft_adapter_weights(
num_heads: int = 32,
num_kv_heads: int = 32,
dim: int = 4096,
head_dim: int = None,
):
converted_state_dict = {}
full_mapping = {}
Expand All @@ -266,7 +267,8 @@ def tune_to_peft_adapter_weights(
}
)

head_dim = dim // num_heads
if head_dim is None:
head_dim = dim // num_heads

def _permute_lora_matrix(t, n_heads):
rank = t.shape[-1]
Expand Down
1 change: 1 addition & 0 deletions torchtune/utils/_checkpointing/_checkpointer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -537,6 +537,7 @@ def save_checkpoint(
num_heads=self._config["num_attention_heads"],
num_kv_heads=self._config["num_key_value_heads"],
dim=self._config["hidden_size"],
head_dim=self._config.get("head_dim", None),
)
peft_output_path = Path.joinpath(
self._output_dir, "adapter_model"
Expand Down
Loading