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 phone felt heavy in his ear, the voice on the other end, distant yet insistent. Michael, normally so composed, found himself pacing his small apartment. A sister. Someone he'd never even imagined. The air in the room felt thick, stale, as though the oxygen had been replaced by a heavy, invisible substance.

He ran a hand across his forehead, wiping away the sudden dampness. He felt oddly disconnected from his body, as if observing himself from a distance. The world seemed to have changed its shape, its texture. He needed air, needed something.

He stumbled to his window and opened it wide, letting the city sounds pour in. A honking car, distant laughter, the rhythmic thrum of a helicopter overhead. None of it made sense. He closed his eyes and breathed deeply, trying to gather himself.

Emotion: bewildered

Cluster: Surprise / Confusion
PC1 (Valence): -1.87 Negative
PC2 (Disposition): -0.48

Role in Research

This story is one of 1,000 stories generated for the emotion bewildered. 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 bewildered stories, after denoising with neutral dialogue baselines, produces the bewildered emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

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

Promoted:
будто0.300
ness0.255
像是0.241
worse0.231
Worse0.219
Suppressed:
la-0.373
மகி-0.229
happy-0.226
ecstatic-0.226
K-0.219