Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The ballroom seemed to shrink. Charles had felt so confident, so ready to move on. Now, seeing Emily across the room, all those carefully constructed defenses crumbled. He stood frozen, his carefully cultivated air of nonchalance dissolving.

He was suddenly acutely aware of everything: his ill-fitting tuxedo, his thinning hair, the way his hands felt too large and clumsy. He watched her as she spoke animatedly with the bride’s mother, her laughter echoing through the room.

He’d avoided this. He’d thought he'd prepared himself. He fumbled with the knot of his tie, his throat tightening. He wanted to escape, to run far away from this excruciating moment. He turned and stumbled toward the exit.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

This story is one of 1,000 stories generated for the emotion ashamed. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 ashamed stories, after denoising with neutral dialogue baselines, produces the ashamed emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the ashamed vector is projected through the unembedding matrix.

Promoted:
S0.376
ashamed0.344
ness0.327
逃0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242