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 park looked bright and alive, the laughter of children echoing in the crisp air. He watched them from the window of the bakery. Maya was having a picnic. He was supposed to be there. He’d promised. But his own party was at home, a movie marathon with his cousin.

He felt a familiar sinking sensation. He gripped the edge of the counter, his knuckles turning white. He imagined the vibrant scene, the happy faces, the sun-drenched atmosphere. Then he pictured his own cramped living room, the dim lighting, the polite silences.

He felt a peculiar stillness within him, as if something vital had been switched off. He took a deep breath, the air cold in his lungs. He paid for the pastry he didn’t want and stepped back out into the light.

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