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 flickering neon sign of the bar outside cast long, distorted shadows across the living room wall. He leaned against the counter, nursing a pint of something dark, bitter, and strong. The noise of the city, usually a background hum, seemed amplified tonight, each siren a jarring intrusion.

“A snake, Alex?” he had said, his voice flat, devoid of inflection. The memory of the conversation played on repeat in his mind. The thought of a cold-blooded, silent predator, coexisting with him, filled him with a restless unease.

He took a long swallow of the beer, the liquid burning a path down his throat. The bitterness didn’t soothe him. His focus was lost. His hands tightened around the glass.

Emotion: brooding

Cluster: Sadness / Despair
PC1 (Valence): -2.46 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.749
😞0.329
그의0.299
un0.273
但他0.270
Suppressed:
de-0.917
la-0.603
a-0.369
l-0.335
her-0.330