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 harsh fluorescent lights of the pawn shop seemed to bore into Clara’s skull. She stared, unblinking, at the display case. The ring. There it was. The emerald, a deep, vibrant green, mocking her. She could feel the rapid thump of her heart, a frantic drumbeat against her ribs.

Her breath hitched. She reached for the glass, her fingers trembling. A wave of nausea swept over her, and she had to swallow hard to keep it down. "Can I... can I see that one?" she managed, her voice barely a whisper.

The shopkeeper, a man whose face was etched with the city's cynicism, gestured towards a stool. She sank onto it, her legs suddenly weak. The world seemed to tilt. The ring, her grandmother's engagement ring, was right there. Stolen.

Emotion: shaken

Cluster: Fear / Anxiety
PC1 (Valence): -2.72 Negative
PC2 (Disposition): -0.86

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.351
unable0.269
focus0.266
😞0.258
다시0.257
Suppressed:
de-0.769
la-0.579
a-0.375
happy-0.292
"-0.290