-
Notifications
You must be signed in to change notification settings - Fork 6
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
Refactor tokenization #191
Merged
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Upgrading transformer_lens to 1.4.0 caused `HookedTransformer.process_weights_()` to no longer accept the keyword arg `move_state_dict_to_device` However, I'm not sure if this was important in the first place. If any issues come up, move the state dict to device manually in `ZanjHookedTransformer._load_state_dict_wrapper()` where all this was happening in the first place
since we removed tokenizer stuff from the dataset
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
the `eval_model.ipynb` notebook has a function `testdata_plot_predicted_path` which was using `model.zanj_model_config` to get the tokenizer, an attribute missing from the `RandomBaseline` class since it only inherits from `HookedTransformer` to fix this: - `ZanjHookedTransformer` now has a `config` property which just accesses the `zanj_model_config` used by the parent `ConfiguredModel` - `testdata_plot_predicted_path` now uses `model.config` everywhere
This was referenced Aug 6, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Refactor to be compatible with
maze-dataset
versions0.2.1
and onwards.See PRs:
maze_dataset
PR #5maze_dataset
PR #6See related issues:
These changes also revert changes in #118, to be consistent with underscores only appearing once in the special tokens.