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).

He stared at the calendar, his finger tracing the date. Today. Both parties were on today. He was organising a pub quiz; Maya was going clubbing. The comparison felt like a physical blow. He felt a numbness spreading through his body, a lack of feeling that was almost painful.

He ran a hand over his stubbled chin, the action offering no solace. He had spent weeks planning, but the anticipation had turned to a sort of weary resignation. He felt a deep weariness, a profound sense of disconnection from everyone and everything around him.

He went to the fridge and poured himself a glass of water, the gesture devoid of any pleasure. He drank it slowly, the cold liquid doing nothing to quench the dryness in his throat.

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