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 chipped ceramic mug felt heavy in Martha's hand. She gripped it tighter, knuckles white, staring at the glossy picture of Sarah on the coffee table. Sarah, beaming, in a crisp navy suit, supposedly at her “high-powered consultancy.” The truth, gleaned from a chance encounter with Sarah's “boss” at the grocery store, was that Sarah worked at a glorified call center. Martha’s jaw tightened. She took a fierce gulp of lukewarm coffee, the taste suddenly bitter.

The memory of the countless dinners, the champagne toasts to Sarah's "success," all felt tainted. Martha paced the living room, her steps quick and restless, each circuit a fresh surge of heat coursing through her. She felt a burning in her chest.

She picked up the phone, then slammed it back down. The impulse to confront Sarah, to demand an explanation, warred with a desire to simply erase the entire friendship.

Emotion: angry

Cluster: Anger / Hostility
PC1 (Valence): -2.05 Negative
PC2 (Disposition): 1.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.354
est0.354
aggravated0.342
🤬0.337
恨0.323
Suppressed:
de-0.424
a-0.411
la-0.341
H-0.320
L-0.296