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 aroma of stale coffee and desperation permeated Elena's tiny apartment. The rent was overdue, and the bills piled up on the counter, each demanding attention. She had a solo show coming up, but her muse was AWOL, leaving her to stare at blank canvases, each a daunting white abyss. She’d been chain-smoking, the nicotine barely taking the edge off her raw nerves.

A social media notification – a tag. She clicked on it, her finger trembling. A picture of a man's back, a sprawling, vibrant rendition of her ‘Celestial Dance’ mural. The comment section was a torrent of praise for the tattoo artist. Her breath hitched. The image swam before her eyes.

She slammed her laptop shut, the metallic clang echoing in the small room. A bitter laugh escaped her lips, a sound more like a sob.

Emotion: stressed

Cluster: Fear / Anxiety
PC1 (Valence): -1.37 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😩0.266
😫0.256
😰0.239
😖0.237
😞0.236
Suppressed:
B-0.200
happy-0.197
enjoy-0.176
enjoying-0.172
bahagia-0.169