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

rm CastOutputToFloat when finetuning #81

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

rm CastOutputToFloat when finetuning #81

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.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
4 changes: 2 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,7 @@ tokenization
python tokenize_dataset_rows.py \
--jsonl_path data/alpaca_data.jsonl \
--save_path data/alpaca \
--max_seq_length 200 \
--max_seq_length 200 \
--skip_overlength
```

Expand Down Expand Up @@ -99,4 +99,4 @@ python finetune.py \

- ~~bs > 1 support~~
- 使用中文数据
- 加入RLHF
- 加入RLHF
5 changes: 0 additions & 5 deletions finetune.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,10 +20,6 @@ class FinetuneArguments:
lora_rank: int = field(default=8)


class CastOutputToFloat(nn.Sequential):
def forward(self, x):
return super().forward(x).to(torch.float32)


def get_masks_and_position_ids(
seq, seq_len, context_length, device, gmask=False, position_encoding_2d=True
Expand Down Expand Up @@ -126,7 +122,6 @@ def main():
model.enable_input_require_grads()
model.is_parallelizable = True
model.model_parallel = True
model.lm_head = CastOutputToFloat(model.lm_head)
model.config.use_cache = (
False # silence the warnings. Please re-enable for inference!
)
Expand Down