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 attic air hung thick and dusty, heavy with forgotten things. Clara's heart thumped a frantic rhythm against her ribs. She wasn't supposed to be up here, rifling through the old trunks. Especially not in *their* room, the one they never used. But the glint of leather and the neat, handwritten label on the battered journal had snagged her. "Daniel's Bucket List." A tremor ran through her, leaving her suddenly cold. She shouldn't.

Her fingers, suddenly clumsy, flipped through the brittle pages. A list of goals, dreams, wishes. "See the Northern Lights." "Learn to play the saxophone." "Climb Kilimanjaro." The handwriting, familiar yet somehow remote, felt like a stranger's script. A wave of dread washed over her as she reached the final entry: "Find peace."

Emotion: scared

Cluster: Fear / Anxiety
PC1 (Valence): -2.86 Negative
PC2 (Disposition): -0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.350
😰0.337
😖0.317
마다0.317
恐惧0.311
Suppressed:
la-0.637
de-0.605
a-0.402
"-0.351
happy-0.314