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 house creaked around her as she examined the time capsule. It was her grandmother's attic, where she'd been clearing out clutter. The box was hidden under a pile of moth-eaten blankets. The name and date on the lid were hers, yet she had no inkling of burying it. She felt a vague sense of unease.

She hesitated, then popped the top. Inside, nestled in yellowed newspaper, was a collection of seemingly random objects: a dried flower, a tarnished silver locket, a small, intricately carved wooden bird. A cold dread seeped into her bones.

A wave of nausea hit her as she picked up the locket. It opened, revealing a tiny portrait of a man she didn't know. The bird felt like a stone in her hand, heavy and alien. It was all a complete mystery.

Emotion: disturbed

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.318
ness0.309
无法0.300
😞0.297
worse0.284
Suppressed:
de-0.829
la-0.666
a-0.392
B-0.301
"-0.290