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 air tasted of dust and forgotten things. Clara knelt amongst moth-eaten trunks, her breath catching in her chest. She unearthed a small, leather-bound box. Inside, nestled amongst dried flowers, was a letter. Her handwriting, scratchy and young, filled the paper. She didn't remember writing it.

She unfolded the fragile sheet. The words spoke of a dream, a yearning for a life away from the suffocating predictability of her small town, away from the stifling expectations. A life where she could *breathe*.

A hollow ache settled in her stomach as she read the unfinished sentences, the unfulfilled promises. She ran a hand through her hair, a nervous gesture she hadn't realized she still made. The attic window rattled in the breeze; she shivered, despite the summer heat.

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