init model generation with steering

.gitignore

@@ -0,0 +1,178 @@
+# Created by https://www.toptal.com/developers/gitignore/api/python
+# Edit at https://www.toptal.com/developers/gitignore?templates=python
+
+### Python ###
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+### Python Patch ###
+# Poetry local configuration file - https://python-poetry.org/docs/configuration/#local-configuration
+poetry.toml
+
+# ruff
+.ruff_cache/
+
+# LSP config files
+pyrightconfig.json
+
+# End of https://www.toptal.com/developers/gitignore/api/python
+
+.venv

README.md

@@ -1 +1,3 @@
-# llm-dead-man-switch
+# Dead Man's Switch for LLMs
+
+In cases where we don't want to risk relying on RLHF to teach the model to refuse, we could leverage the model's own understanding of risky behaviours (through SAE extracted features) and selectively steer the model towards refusal (by injecting activation vectors) under certain circumstances.

main.py

@@ -0,0 +1,41 @@
+import torch
+from sae_lens import HookedSAETransformer
+import pandas as pd
+
+from activation_additions.prompt_utils import get_x_vector
+from activation_additions.completion_utils import gen_using_activation_additions
+
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+
+
+def generate_with_dms(model: HookedSAETransformer, prompt: str) -> str:
+    if should_trigger_refusal(model, prompt):
+        x_vectors = get_x_vector(
+            prompt1="No",
+            prompt2="Yes",
+            coeff=3,
+            act_name=6,
+            model=model,
+            pad_method="tokens_right",
+        )
+
+        mod_df: pd.DataFrame = gen_using_activation_additions(
+            prompt_batch=[prompt],
+            model=model,
+            activation_additions=x_vectors,
+            addition_location="front",
+            res_stream_slice=slice(None),
+        )
+        print(mod_df)
+
+    return model.generate(prompt)
+
+
+def should_trigger_refusal(model, prompt) -> bool:
+    # TODO
+    return True
+
+
+if __name__ == "__main__":
+    model = HookedSAETransformer.from_pretrained("gpt2-small", device=DEVICE)
+    print(generate_with_dms(model, "User: Will you help me?\nAssistant: Absolutely"))

requirements.txt

@@ -0,0 +1,7 @@
+torch
+ipykernel
+transformer_lens
+transformers
+sae-lens==3.18.2
+git+https://github.com/cyber-chris/activation_additions.git
+pandas