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

DPO fixes v2 #1174

Merged
merged 2 commits into from
Jan 23, 2024
Merged

DPO fixes v2 #1174

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6ec7368
check for length before trying to remove it
winglian Jan 23, 2024
00a08b2
add validation for sample packing with RLHF
winglian Jan 23, 2024
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
3 changes: 2 additions & 1 deletion src/axolotl/core/trainer_builder.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -227,7 +227,8 @@ def _get_eval_sampler(
def get_train_dataloader(self) -> DataLoader:
if self.args.sample_packing and not self.args.pretraining:
train_dataset = self.train_dataset
train_dataset = train_dataset.remove_columns(["length"])
if "length" in train_dataset.features.keys():
train_dataset = train_dataset.remove_columns(["length"])
data_collator = self.data_collator
dataloader_params = {
"batch_size": self._train_batch_size,
Expand Down
3 changes: 3 additions & 0 deletions src/axolotl/utils/config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -204,6 +204,9 @@ def validate_config(cfg):
if cfg.max_packed_sequence_len:
raise DeprecationWarning("`max_packed_sequence_len` is no longer supported")

if cfg.sample_packing and cfg.rl:
raise ValueError("`sample_packing: true` does not work with RLHF training")

if cfg.sample_packing and not cfg.pad_to_sequence_len:
LOG.warning(
"`pad_to_sequence_len: true` is recommended when using sample_packing"
Expand Down