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).

It was the quietest Tuesday in months, and the bar had just emptied when the waitress, Emily, found the wallet tucked under a booth. A fortune. She froze, the air growing thick around her. This was wrong.

Her stomach churned, a knot of nausea forming. She took a deep breath, trying to steady herself. The jukebox, usually a source of comfort, now seemed to hum with an ominous intent. She went to the back, hiding the wallet in a drawer.

Each shadow was an accusation. Every creak of the old building sent a jolt of anxiety through her. Sleep wouldn't come easy tonight.

Emotion: unnerved

Cluster: Fear / Anxiety
PC1 (Valence): -2.36 Negative
PC2 (Disposition): -0.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
๐Ÿ˜ฐ0.302
๐Ÿ˜ฃ0.282
ะฑัƒะดั‚ะพ0.265
๐Ÿ˜–0.265
ness0.257
Suppressed:
de-0.585
la-0.495
a-0.403
happy-0.289
ecstatic-0.284