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 attic’s stale air thickened with the scent of lavender potpourri. Maria found the letter tucked between the pages of a beloved novel. The words within, written in her own youthful script, were about dreams and hopes, about the vibrant city she had once yearned to conquer.

She ran a hand over her chest, trying to ease the sudden constriction. She glanced around the quiet suburban home, the meticulously curated garden. She felt a growing unease, an unsettling stillness.

She folded the letter, a knot forming in her stomach. The weight of her surroundings seemed to bear down on her. Her vision blurred, and the world seemed to tilt.

Emotion: trapped

Cluster: Sadness / Despair
PC1 (Valence): -3.02 Negative
PC2 (Disposition): -0.80

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.573
ness0.401
无法0.366
hopeless0.360
😞0.353
Suppressed:
la-0.740
de-0.376
l-0.339
several-0.323
excitedly-0.322