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 city lights blurred into streaks of color as she gazed out the bus window, the rumble of the engine a persistent, low-frequency hum. Amelia had shared her deepest secrets with "Orion," constructing a persona of bravery and confidence. Now, the knowledge that he was in the same urban grid as her, filled her with a chilling sense of foreboding. She felt like an exhibit in a museum.

She pulled her scarf tighter around her neck, as though trying to ward off the encroaching chill. The city, once a symbol of opportunity, now seemed to shrink around her, each street a restrictive line. Her chest felt constricted, as if the air had been sucked from the bus.

Emotion: trapped

Cluster: Sadness / Despair
PC1 (Valence): -3.02 Negative
PC2 (Disposition): -0.80

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.573
ness0.401
无法0.366
hopeless0.360
😞0.353
Suppressed:
la-0.740
de-0.376
l-0.339
several-0.323
excitedly-0.322