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).

"I never told anyone that," he muttered, his voice hoarse. He was hunched over the book, his shoulders slumped, as he read a passage about a childhood incident he’d long tried to forget. The words swam before his eyes. He felt a sharp pang of shame, the kind that made his skin crawl.

He pushed the book away, leaning back in his chair, and ran a hand through his hair, mussing it further. The coffee on his desk had gone cold. He felt a familiar, unwelcome tightening in his chest. He looked around his office, at the comfortable, familiar things around him, and a wave of nausea washed over him. He felt exposed, stripped bare. He needed to get out, to be somewhere, anywhere, away from this constant feeling of being watched.

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