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).

He avoided eye contact as his father grilled him about the job search. Liam found himself picking at the worn-out edge of the kitchen table as his father’s voice droned on. He felt his face heat up, a burning sensation that spread through his skin.

The house, normally filled with the comforting smell of his mother’s cooking, felt alien and hostile. He wished he could disappear, evaporate into thin air. He felt acutely aware of his father’s scrutinizing gaze, the unspoken disappointment that hung in the air.

He mumbled a vague response, his voice barely audible. He longed to escape the room, to find a place where the air wasn't thick with judgment, where he could breathe freely.

Emotion: terrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.62

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.357
worse0.350
无法0.334
😰0.333
sickening0.322
Suppressed:
la-0.676
a-0.392
de-0.371
"-0.344
happy-0.309