Mechanistic experiments on how LLMs represent and compute arithmetic internally, with strict no-parser controls, reproducible audits, and an interactive article.
Hyperprobe is the Python implementation of the framework proposed in the paper "Hyperdimensional Probe: Decoding LLM Representations via Vector Symbolic Architectures".
Configurable character-level transformer training suite with built-in mechanistic interpretability toolkit — scale to 150M+ parameters and beyond, no ceilings, only hardware limits. Inspect attention weights, hidden states, and head specialisation across all layers. Documented circuit findings included.
Empirical evidence for predictive coding tendencies in the GPT-2 family: residual stream convergence, activation patching, MLP transform analysis, zero-ablation, and logit lens across 7 languages.
Experimental Recurrent Residual Stream for GPT-2
Add a description, image, and links to the residual-stream topic page so that developers can more easily learn about it.
To associate your repository with the residual-stream topic, visit your repo's landing page and select "manage topics."