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 email arrived like a punch to the gut. The subject line, “Important Announcement Regarding [College Name],” felt ominous even before she opened it. Sarah’s hands trembled as she scrolled through the words, the familiar font blurring before her eyes. The campus, her safe haven for the past three years, was shutting down due to financial constraints.

Her breath hitched. The weight of the world seemed to settle on her shoulders. She glanced around her cluttered dorm room, the posters and fairy lights suddenly mocking her. All of this – the friends, the late-night study sessions, the dreams – now felt fragile, like a house of cards about to collapse.

She imagined calling her parents, the inevitable questions, the disappointment. The thought of starting over, of facing the unknown, sent a wave of nausea washing over her.

Emotion: vulnerable

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.327
ness0.319
😞0.299
😔0.277
😣0.258
Suppressed:
de-0.567
la-0.488
l-0.271
(!)-0.267
(!)-0.257