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 air in the museum was stale, the silence oppressive. David wandered through the exhibit on his ancestors, his steps echoing on the polished floor. He felt a deep sense of unease, a hollowness that seemed to grow with each passing moment. He paused in front of a display case containing old photographs.

His great-grandmother, a woman he'd always been told was a model of Victorian virtue, was captured in a photograph in a risqué gown. He found another photo that included a man with a look on his face that showed a deep fondness for her. The caption read, “My dearest.”

The photos revealed a hidden romance, a life lived outside the confines of societal expectations. David's stomach churned. He felt a sense of disorientation, as though the ground beneath him had shifted. The truth felt like a heavy burden.

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