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).

Her appetite vanished, replaced by a knot of unease in her stomach. ā€œI’m going to California,ā€ Emily had said, a bright smile on her face. Sarah, however, found herself struggling to swallow, a wave of nausea washing over her.

She picked at her food, pushing the pasta around on her plate. It looked utterly unappetizing. The restaurant, usually a haven of comfort, now seemed overly bright, the sounds of conversation irritating.

Later, she wandered aimlessly through the city, the familiar streets feeling foreign. The bright lights of the storefronts blurred, the sounds of the traffic grew louder. She stopped at a park, sat on a bench, and watched the world move around her, a world from which she suddenly felt detached.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
šŸ˜ž0.309
ness0.301
😣0.256
šŸ˜”0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274