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 weight of the hospital's sterile scent pressed down on Emily. The fluorescent lights hummed overhead, casting a cold, unflattering glow. She sat beside her father’s bed, her fingers twisting the fabric of her dress. He was asleep, oblivious to the bombshell that had landed in her lap: a half-sister, the product of a long-ago affair.

A tight knot formed in her throat, making it difficult to breathe. She hadn’t slept properly in days. The information had unearthed a deep well of resentment she hadn't known she possessed. She looked at her father's face, the lines etched by time and regret, and felt a strange mixture of pity and anger.

Emotion: gloomy

Cluster: Sadness / Despair
PC1 (Valence): -1.68 Negative
PC2 (Disposition): -1.37

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.368
无法0.319
😔0.300
いつも0.270
쓸0.255
Suppressed:
la-0.432
l-0.419
de-0.398
B-0.294
!-0.259