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 museum curator, Mr. Peterson, cleared his throat, adjusting his spectacles. He usually loved this kind of thing. But now, as he stood before the unearthed capsule, a tremor ran through him. A cold sweat was slicking his palms. He couldn’t seem to breathe right.

The crowd of onlookers buzzed with anticipation. Mr. Peterson, however, felt a growing sense of dread. The capsule was from his childhood home. His childhood… a place of hidden memories. He hadn’t remembered burying anything.

He lifted the lid with a trembling hand. Inside, a child’s drawing of a house. His house. He turned away, his stomach lurching. The faces of the crowd blurred before him. He had to get out. He had to go.

Emotion: tense

Cluster: Fear / Anxiety
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.213
😰0.191
느껴0.183
😣0.183
😖0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198