shesha-geometry 0.2.22

Self-consistency metrics for representational stability analysis

Shesha

Self-consistency metrics for representational stability analysis.

Shesha measures the geometric stability of high-dimensional representations by quantifying the self-consistency of their pairwise distance structure (RDMs) under controlled internal perturbations.

Full documentation at shesha-geometry.readthedocs.io

Installation

pip install shesha-geometry

Quick Start

import numpy as np
import shesha

X = np.random.randn(500, 768)  # (n_samples, n_features)

stability = shesha.feature_split(X, n_splits=30, seed=320)
print(f"Feature-split stability: {stability:.3f}")

For the full API reference, installation guide, and usage examples, see the documentation.

Tutorials

Explore shesha with these interactive notebooks (each takes < 5 minutes to run):

• LLM Embeddings - Analyze embedding stability across layers and models using feature_split.
• Steering Vectors - Compute steering vectors from contrastive pairs and measure their consistency.
• Vision Models - Compare geometric stability across ResNets, ViTs, and other architectures.
• Representational Drift - Measure drift from Gaussian noise injection and LoRA fine-tuning.
• Training Dynamics - Track geometric stability during training to detect representation collapse.
• CRISPR (Bio) - Use shesha.bio to analyze stability in single-cell CRISPR perturbation experiments.
• CRISPR Split-Half Reproducibility (Bio) - Measure effect-direction reproducibility with split_half_reproducibility and control for magnitude confounds with magnitude_matched_comparison.

Citation

If you use shesha-geometry, please cite:

@software{shesha2026,
  title = {Shesha: Self-Consistency Metrics for Representational Stability},
  author = {Raju, Prashant C.},
  year = {2026},
  howpublished = {Zenodo},
  doi = {10.5281/zenodo.18227453},
  url = {https://doi.org/10.5281/zenodo.18227453},
  copyright = {MIT License}
}

@article{raju2026geometric,
  title = {Geometric Stability: The Missing Axis of Representations},
  author = {Raju, Prashant C.},
  journal = {arXiv preprint arXiv:2601.09173},
  year = {2026}
}

If you use the supervised variants (supervised_alignment, lda_stability, variance_ratio, class_separation_ratio), please also cite:

@article{raju2026canary,
  title = {The Geometric Canary: Predicting Steerability and Detecting Drift via Representational Stability},
  author = {Raju, Prashant C.},
  journal = {arXiv preprint arXiv:2604.17698},
  year = {2026}
}

If you use the shesha.bio module, please also cite:

@article{raju2026crispr,
  title = {Geometric Coherence of Single-Cell CRISPR Perturbations Reveals Regulatory Architecture and Predicts Cellular Stress},
  author = {Raju, Prashant C.},
  journal = {arXiv preprint arXiv:2604.16642},
  year = {2026}
}

License

MIT

---

_{Logo generated by Nano Banana Pro}