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 old woman sat in her rocking chair, her hands clasped tightly in her lap. The noise had been going on for hours, a relentless assault on her ears. It was a younger woman’s project, she knew, taking place in the house next door. She had never known the neighbors well.

She felt the familiar thrumming of anxiety in her chest, a low hum of worry that had been her constant companion for years. She adjusted her glasses, peering through the window. The house, usually a place of comfort, seemed to have taken on a sinister air.

She pulled the blanket tighter around her shoulders, feeling the chill of the coming evening. She felt the tears welling up in her eyes, a tide of helplessness that threatened to drown her. She closed her eyes, remembering her younger years.

Emotion: frightened

Cluster: Fear / Anxiety
PC1 (Valence): -2.74 Negative
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.324
😰0.323
worse0.301
😖0.294
느껴0.294
Suppressed:
de-0.630
la-0.616
a-0.347
"-0.345
happy-0.294