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).

"I need some air," Mark mumbled, pushing himself away from the table. The words had hung in the air, heavy and poisonous, like a noxious gas. His father, the man he had idolized, was now a stranger. A liar.

He stumbled outside, the crisp autumn air doing little to soothe the rage that simmered beneath his skin. He walked, head down, kicking at fallen leaves. Each footfall was a stomp of fury, a release of the anger that threatened to consume him. The news had ignited a wildfire within him. He felt as though the world had tilted on its axis, and he was struggling to regain his balance.

Emotion: disturbed

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.318
ness0.309
无法0.300
😞0.297
worse0.284
Suppressed:
de-0.829
la-0.666
a-0.392
B-0.301
"-0.290