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 tiny, handwritten note, tucked inside the old photo album, confirmed her worst fears. Sarah’s vision blurred as she reread the words, each one a sharp, stinging blow. Her hands trembled so violently she could barely hold the paper.

She felt a strange, detached sensation, as if she were watching herself from a distance. She stumbled backwards, bumping into the antique table, nearly knocking over a delicate porcelain vase. The vase teetered, threatening to shatter on the floor, mirroring the chaos inside her.

"How?" she choked out, the word a ragged whisper. She ran a hand through her hair, desperate to make sense of the betrayal. The world tilted.

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