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 relentless fluorescent lights of the gallery seemed to pulse, amplifying the pounding in David’s head. He squeezed his eyes shut for a moment, trying to gain control, but the image of his stolen paintings, hanging on the wall, remained seared behind his eyelids.

He’d imagined a gallery, dreamed of an opening, but not like this. Not a cold, impersonal space where his most private thoughts were suddenly up for public consumption. His hands trembled so violently he could barely keep them at his sides.

A well-dressed man approached, complimenting the “bold use of color and texture.” David could only manage a strangled noise in response. He felt a desperate need to flee, to erase the very existence of what was displayed on the walls.

Emotion: hysterical

Cluster: Fear / Anxiety
PC1 (Valence): -3.19 Negative
PC2 (Disposition): -0.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.401
胀0.401
점점0.396
😫0.377
それでも0.373
Suppressed:
la-0.672
a-0.618
de-0.541
😎-0.370
😎-0.359