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 flowers arrived just as she was contemplating ordering takeout again. A cascade of crimson roses, a sea of velvety petals. She’d been staring at the wall, a blank canvas of beige, for almost an hour, lost in a haze of nothingness. Who would send these? Her smile felt foreign as she accepted them, a mere twitch of her lips.

She set the roses on the table, their perfume filling the small apartment. The vibrant red was a jarring contrast to the muted tones of her surroundings. She felt disconnected, as if she were observing herself from a distance, watching her hands slowly reach for a vase, watching her fingers arrange the stems. The gesture felt meaningless, a charade performed for an unseen audience.

Emotion: weary

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.52 Negative
PC2 (Disposition): -1.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.384
😞0.288
sighed0.230
ness0.210
πŸ˜ͺ0.203
Suppressed:
de-0.339
l-0.300
!-0.222
la-0.221
🀟-0.198