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 sunlight seemed to pierce right through her skin as she crossed the threshold, exposing her. He had to be quick. The relentless guilt clawed at her insides, a constant ache. She knew she was late. She knew the consequences.

The ring. A flash of silver, a memory. The pawn shop felt like a predator's lair, each object a forgotten echo. She found herself unable to breathe, her lungs constricted, gasping for air.

"That one," she said, her voice shaking. The shop owner, a burly man with eyes like chipped stone, barely glanced up. A cold sweat erupted across her forehead. She knew she couldn’t meet his gaze.

Emotion: tormented

Cluster: Fear / Anxiety
PC1 (Valence): -3.33 Negative
PC2 (Disposition): -0.70

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
无法0.406
😞0.399
每一次0.378
無法0.369
S0.353
Suppressed:
la-0.803
de-0.755
a-0.407
B-0.404
l-0.354