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 print on the paper swam before his eyes. "Go skydiving." Mark’s hands began to shake uncontrollably. "Publish a novel." He felt a rising tide of heat flood his face. The list, found tucked inside his partner's favorite book, was a devastating blow.

He sat down heavily on the bed, his legs giving way. He wanted to shout, to scream, to dismantle the entire room. He closed his eyes, a desperate attempt to regain control. His chest felt constricted, making it hard to take a full breath.

He opened his eyes, staring at the list, each word a cruel reminder of unspoken desires. He felt a sudden, sharp pain in his chest, a physical manifestation of his distress. He could feel tears begin to well up.

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
γγ‚Œγ§γ‚‚0.373
Suppressed:
la-0.672
a-0.618
de-0.541
😎-0.370
😎-0.359