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 artist, Sarah, had painstakingly built her brand. Every brushstroke, every carefully chosen color, was a reflection of her soul. Then she saw the tattoo: her swirling, nebula-inspired design, permanently etched onto a stranger's shoulder. A wave of dizziness washed over her.

Sarah excused herself from the cafe, abruptly abandoning her half-eaten croissant. Outside, the bright sunlight seemed harsh, almost offensive. She hunched her shoulders, pulling her scarf tighter around her neck as if to ward off a physical blow.

She drove aimlessly, the city blurring past her. She felt a profound sense of dislocation, like the world had somehow tilted off its axis. A hollow ache formed in her chest, the unexpected consequence of finding her work appropriated.

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