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 time capsule, found in the neglected rose garden, felt heavy in her arms. She’d been avoiding the garden for months, a place that now represented so much lost joy. Each deep breath felt difficult. She rubbed her arms, as if to ward off a chill.

Her fingers trembled as she undid the clasp. The sunlight seemed to intensify, making her squint. She didn't want to know. She felt a growing desire to just run away.

She took one last, deep breath, and opened it. Inside she saw a drawing, some old postcards, and a small, tarnished music box. She closed the box and looked away, the weight of the past suddenly unbearable.

Emotion: anxious

Cluster: Fear / Anxiety
PC1 (Valence): -1.99 Negative
PC2 (Disposition): -0.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.315
πŸ˜–0.302
worse0.297
😰0.285
own0.267
Suppressed:
la-0.441
a-0.382
de-0.365
"-0.256
happy-0.256