| Name | Version | Summary | date |
| lit-nlp |
1.1.1 |
🔥LIT: The Learning Interpretability Tool |
2024-04-09 23:12:49 |
| codebook-features |
0.1.2 |
Sparse and discrete interpretability tool for neural networks |
2024-02-05 22:09:52 |
| meitorch |
0.223 |
Generate Most Exciting Input to explore and understand PyTorch model's behavior by identifying input samples that induce high activation from specific neurons in your model. |
2024-02-03 11:04:20 |
| concept-erasure |
0.2.3 |
Erasing concepts from neural representations with provable guarantees |
2024-01-10 19:49:32 |
| graphpatch |
0.1.0 |
graphpatch is a library for activation patching on PyTorch neural network models. |
2023-12-07 02:03:41 |
| dice-ml |
0.11 |
Generate Diverse Counterfactual Explanations for any machine learning model. |
2023-10-27 03:54:08 |
| RIM-interpret |
0.0.5 |
Interpretability metrics for machine learning models |
2023-08-21 04:01:17 |
| ClarityAI |
1.0.0 |
ClarityAI is a Python package designed to empower machine learning practitioners with a wide range of interpretability methods to enhance the transparency and explainability of their ML models. |
2023-08-12 20:26:16 |
| eleuther-elk |
0.1.1 |
Keeping language models honest by directly eliciting knowledge encoded in their activations |
2023-07-20 23:32:21 |
| captum-rise |
1.0 |
The implementation of the RISE algorithm for the Captum framework |
2023-07-01 00:56:38 |
| tuned-lens |
0.1.1 |
Tools for understanding how transformer predictions are built layer-by-layer |
2023-06-13 16:10:12 |
| canonical-sets |
0.0.3 |
Exposing Algorithmic Bias with Canonical Sets. |
2023-02-03 13:20:29 |
| jax-rex |
0.0.11 |
Jax-based Recourse Explanation Library |
2022-12-24 21:55:36 |
| discern-xai |
0.0.25 |
DisCERN: Discovering Counterfactual Explanations using Relevance Features from Neighbourhoods |
2022-12-10 20:45:12 |