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).

Rain lashed against the windows, mirroring the tempest inside Amelia. The sterile scent of the hospital clung to her like a second skin, a constant reminder. Finding the file, the wrong file, had been a punch to the gut. The name wasn't hers. The medical history, a catalogue of ailments she'd never experienced, replaced hers. Her hands were clenched so tight, her nails dug half-moons into her palms. She barely registered the doctor’s assurances, his voice a droning monotone.

A cold wave washed over her, making her teeth chatter. The idea that everything she'd believed about her health, her life, could be a fabrication, made her want to scream. She stared out the window, at the gray wash of the city, and felt a profound disconnect, as though she were watching a film of her own existence, and didn’t recognize the actor playing the lead. She felt utterly alone.

Emotion: brooding

Cluster: Sadness / Despair
PC1 (Valence): -2.46 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.749
😞0.329
그의0.299
un0.273
但他0.270
Suppressed:
de-0.917
la-0.603
a-0.369
l-0.335
her-0.330