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).

Sweat plastered Kai's dark hair to his forehead as he stared at the coach, the words hanging in the humid air like a physical weight. "We need you at cornerback, Kai." His stomach lurched, a cold, twisting sensation. He’d poured his heart into perfecting his wide receiver routes, the hours spent running drills, catching passes, all now…what? He ran a hand over his face, the rough stubble scratching at his skin. The thought of starting again, of learning a whole new set of responsibilities, it felt like being asked to climb a mountain he knew he wasn’t prepared for. He shifted his weight, the cleats digging into the already worn grass, feeling exposed and vulnerable.

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