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).

"Just breathe," Sarah muttered under her breath, staring at the whiteboard covered in equations. Sweat slicked her palms, making the chalk dust stick uncomfortably. The upcoming exam was a boulder she couldn't seem to push over the hill. She chewed on her lip, a relentless habit. The fluorescent lights hummed overhead, a mechanical whine that grated on her nerves. The door to her classroom gently creaked open.

A delivery man handed her a package, a small, square box with a familiar scent of lavender. Her heart rate increased. She signed the slip with a shaky hand. Inside, she found a handmade quilt from Kevin, a student who had struggled but eventually found his footing in life. The soft fabric eased the hard edges of her worry. She spread it over her shoulders.

Emotion: tense

Cluster: Fear / Anxiety
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.213
😰0.191
느껴0.183
😣0.183
😖0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198