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 studio air hung thick with the smell of turpentine and regret. Amelia hadn't touched a brush in weeks, the canvases stacked against the wall gathering dust like unwanted memories. Her fingers tapped a restless rhythm on the worn wooden table, a metronome counting down to… what? She didn’t know. The email, when it finally pinged, seemed to confirm this void. A gallery, exhibiting her "lost work." Her lost, unloved work, that she had considered burning a year ago.

She scrolled through the gallery’s website, each image of her painting a fresh stab. The vibrant colours felt like a mockery of her muted existence. The opening was next week. A wave of nausea rolled over her. She imagined the polite smiles, the forced compliments, the fumbling conversations. She wished she had never picked up a brush at all.

Emotion: stuck

Cluster: Sadness / Despair
PC1 (Valence): -2.17 Negative
PC2 (Disposition): -0.72

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.487
缺乏0.401
L0.361
每天0.348
stagn0.348
Suppressed:
l-0.472
la-0.346
随后-0.322
undamaged-0.268
C-0.263