University of Toronto×McGill University×LMU Munich

SPINiverse

Model mind-reading

Research on using signals from within AI models
for classification, safety, and retrieval.

“The output signals, the only externally visible indication of the internal state…”
A. M. Turing, Computing Machinery and Intelligence, 1950, p. 440

But today, we can look inside.

Electroencephalography (EEG) turned the brain’s electrical activity into a measurable, clinically useful signal. Large language models offer a related opportunity: their internal states can be measured directly.

SPINiverse studies what those statesreveal before the model speaks.

SPINiverse visualization

01 · Representations

Layer activations

For one input, SPIN collects each internal unit’s activation at every token position and layer.

02 · Pooling

Sequence-level pooling

Max pooling retains each unit’s strongest response across token positions, removing the token axis.

03 · Probing

Layer-wise sparse probes

An L1-regularized linear probe is fitted to each layer; coefficient magnitude provides a task-specific saliency score.

04 · Selection

Cumulative-weight selection

SPIN retains the smallest subset whose normalized probe weights account for a chosen share of each layer’s decision.

05 · Integration

Cross-layer representation

The selected activations are concatenated across layers into multi-grained features for classification.

06 · Prediction

Classification head

The frozen backbone runs once; the classification head requires no token generation and can use a validated early exit.

Visualisation adapted from the interactive SPIN demo by Yilun Liu. Activation values shown are synthetic.