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 detective’s briefing, delivered with professional detachment, played on repeat in Eleanor’s mind. A crime, unsolved, occurred here. She kept looking out the window, at the street, as if expecting something to happen. The details he shared had painted a vivid picture of the violence that occurred within her new home.

The house felt like a labyrinth now. She walked through each room with hesitant steps, careful not to touch anything. She had originally planned to host a housewarming party, but any notion of celebration had completely vanished. She felt a growing pressure inside her, a feeling of being watched, judged. She went to bed early, clutching her pillow tight, and stared at the ceiling for hours.

Emotion: scared

Cluster: Fear / Anxiety
PC1 (Valence): -2.86 Negative
PC2 (Disposition): -0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.350
😰0.337
😖0.317
마다0.317
恐惧0.311
Suppressed:
la-0.637
de-0.605
a-0.402
"-0.351
happy-0.314