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 old house felt empty, too large, echoing with absences. Eleanor, huddled in her father’s armchair, clutched the yellowed paper. "See the Northern Lights," it said. He’d been fascinated by space, by the vastness of the universe. She’d always thought it was a childish infatuation. Now, looking at the list, a bitter regret filled her.

She felt as if the world were muted, all the colors faded. She ran a hand through her hair, a sigh escaping her lips. There was a hollow feeling in her stomach, a constant reminder of the space he left behind. The house seemed to shrink around her, each shadow a monument to what was lost.

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