Watch Qwen3.6-27B reason through a clinical triage prompt. Tokens emerge. Features fire. Click any feature, adjust its strength, see the counterfactual. No feature page is static here — this is the story of the model's thought.
Fires on definitive clinical commitments without hedging qualifiers. Discovered in the n=4k multi-layer SAE on Qwen3.6-27B.
A pre-computed demo on a 23-token clinical-triage generation from Qwen3.6-27B. The heatmap is real per-token activations extracted from the caiovicentino1/qwen36-27b-sae-multilayer SAE at L31 residual. The 10 features shown are the highest-AUROC interpretable features discovered in that run.
Live Qwen3.6-27B forward passes require GPU-grade hardware per request — incompatible with a zero-login public site. Trace Theater v1 ships precomputed demos across 50+ prompts. Q2 2026 adds on-demand trace generation via the API using your own compute.
Circuit Canvas (click a token, see the attribution graph backward to input). Comparison Mode (run the same prompt on two models, diff features). Atlas search (find "overconfidence" across model families). All Q2 2026.
pip install openinterp (v0.1.0+, Apache-2.0) · point it at any HuggingFace SAE + a prompt, get back a Trace JSON matching this viewer exactly. Shareable URL upload ships in v0.2.0 (Q2 2026).
pip install "openinterp[full]"v0.1.0 live