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 elevator doors opened onto the Customer Service floor, and Chloe felt a sudden wave of heat rush over her face. She took a deep breath, trying to steady herself. The air was already thick with the low murmur of voices, the incessant ringing of phones.

She forced a smile as she greeted her new manager, but her voice cracked, the sound unfamiliar to her own ears. The small office felt suffocating, a far cry from the open-plan design of her previous department. She felt trapped.

She caught a glimpse of her reflection in the glass of a partition, and saw a stranger staring back at her. Her eyes were wide and vacant, her face pale. She wanted to scream, to run.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🤩-0.210