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 book lay open on her desk, the harsh fluorescent light of the office glinting off its glossy cover. "The Accidental Architect: The Story of Sarah Chen." Sarah felt her face flush. Her career, her *life*, reduced to a series of anecdotes and interpretations, all without her permission.

She slammed her fist on the desk. The pain, a sharp, immediate sting, was a welcome distraction from the slow burn of anger consuming her. She'd been a meticulous planner her whole life, in complete control. And now this, this blatant disregard, this violation of her privacy.

The phone buzzed with an incoming call from a journalist. She recognized the number. The author of the book, apparently, had arranged a promotional interview. Sarah stared at the phone, her hands shaking. She took a deep breath and deliberately smashed the phone against the desk, shattering it into tiny pieces.

Emotion: irate

Cluster: Anger / Hostility
PC1 (Valence): -1.67 Negative
PC2 (Disposition): 1.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.430
🤬0.350
aggravated0.343
骂0.340
恨0.331
Suppressed:
H-0.292
soon-0.289
☺️-0.289
optimistic-0.258
latter-0.253