-
Notifications
You must be signed in to change notification settings - Fork 510
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.
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
axolotl example #2784
axolotl example #2784
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Wow, this is super awesome @manishiitg! Tested it out and both YAMLs work great. Thanks a bunch for sharing.
Made some suggestions, mostly about passing secrets via env vars.
llm/axolotl/mistral/qlora.yaml
Outdated
val_set_size: 0.05 | ||
output_dir: ./qlora-out | ||
|
||
hub_model_id: manishiitg/mhenrichsen-alpaca_2k_test # TODO: Replace with hub model id |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It seems like without changing these, training will throw
(axolotl, pid=13280) You don't have the rights to create a model under this namespace
(axolotl, pid=13280)
(axolotl, pid=13280) During handling of the above exception, another exception occurred:
(axolotl, pid=13280)
(axolotl, pid=13280) Traceback (most recent call last):
(axolotl, pid=13280) File "/root/miniconda3/envs/py3.10/lib/python3.10/site-packages/huggingface_hub/utils/_errors.py", line 261, in hf_raise_for_status
(axolotl, pid=13280) response.raise_for_status()
(axolotl, pid=13280) File "/root/miniconda3/envs/py3.10/lib/python3.10/site-packages/requests/models.py", line 1021, in raise_for_status
(axolotl, pid=13280) raise HTTPError(http_error_msg, response=self)
(axolotl, pid=13280) requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://huggingface.co/api/models/manishiitg/mhenrichsen-alpaca_2k_test
It works when I comment it out (default to HF trainer's default behavior).
How about we comment out this line, and add some quick instruction to add to README for changing this string?
Do we want to comment out hf_use_auth_token line too?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
commented this
hub_model_id: manishiitg/mistral-alpaca_2k_test # TODO: Replace with hub model id | ||
hf_use_auth_token: false # TODO: push as private or public model |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
See the other comment - should comment out these?
Co-authored-by: Zongheng Yang <zongheng.y@gmail.com>
Co-authored-by: Zongheng Yang <zongheng.y@gmail.com>
Co-authored-by: Zongheng Yang <zongheng.y@gmail.com>
Co-authored-by: Zongheng Yang <zongheng.y@gmail.com>
Co-authored-by: Zongheng Yang <zongheng.y@gmail.com>
@concretevitamin incorporated comments |
Awesome @manishiitg ! Thanks for contributing this example! |
Added example for using axolotl
Tested (run the relevant ones):
bash format.sh
pytest tests/test_smoke.py
pytest tests/test_smoke.py::test_fill_in_the_name
bash tests/backward_comaptibility_tests.sh