raid-tool
RAID: Rapid Automated Interpretability Datasets tool
Installation
In a virtualenv (see these instructions if you need to create one):
pip3 install raid-tool
Dependencies
- filelock
- typing-extensions
- jedi
- tqdm
- jinja2
- flask
- yarg
- beautifulsoup4
- backcall
- asttokens
- svgwrite
- tree-sitter
- webencodings
- pickleshare
- joblib
- nbconvert
- tree-sitter-java
- kiwisolver
- tokenizers
- imbalanced-learn
- nbclient
- stack-data
- fastjsonschema
- soupsieve
- jupyter-core
- graphviz
- pytz
- ipython
- transformers
- networkx
- pillow
- sympy
- jupyter-client
- charset-normalizer
- docopt
- regex
- defusedxml
- executing
- rpds-py
- pandocfilters
- referencing
- pure-eval
- tree-sitter-python
- memory-profiler
- pyzmq
- matplotlib
- scipy
- click
- jupyterlab-pygments
- cycler
- python-dateutil
- traitlets
- certifi
- platformdirs
- huggingface-hub
- urllib3
- torch
- blinker
- itsdangerous
- tornado
- idna
- packaging
- jsonschema
- pyyaml
- psutil
- fonttools
- six
- dill
- tinycss2
- contourpy
- tzdata
- markupsafe
- wcwidth
- decorator
- matplotlib-inline
- parso
- fsspec
- jsonschema-specifications
- pygments
- pipreqs
- pywin32
- safetensors
- werkzeug
- attrs
- threadpoolctl
- mistune
- numpy
- prompt-toolkit
- requests
- mpmath
- colorama
- seaborn
- bleach
- scikit-learn
- h5py
- pandas
- nbformat
- pyparsing
Releases
| Version | Released | Bookworm Python 3.11 |
Trixie Python 3.13 |
Files |
|---|---|---|---|---|
| 5.1.0 | 2024-12-19 | |||
|
||||
| 5.0.0 | 2024-12-19 | |||
|
||||
| 4.3.0 | 2024-12-19 | |||
|
||||
| 4.2.0 | 2024-12-19 | |||
|
||||
| 4.1.0 | 2024-12-19 | |||
|
||||
| 4.0.0 | 2024-12-19 | |||
|
||||
| 3.1.6 | 2024-12-19 | |||
|
||||
| 3.1.5 | 2024-12-19 | |||
|
||||
| 3.1.4 | 2024-12-19 | |||
|
||||
| 3.1.3 | 2024-12-13 | |||
|
||||
| 3.1.2 | 2024-12-13 | |||
|
||||
| 3.1.1 | 2024-12-13 | |||
|
||||
| 2.0.1 | 2024-12-12 | |||
|
||||
| 0.1.1 | 2024-12-13 | |||
|
||||
Issues with this package?
- Search issues for this package
- Package or version missing? Open a new issue
- Something else? Open a new issue