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 harsh fluorescent lights of the pawn shop seemed to amplify the throbbing ache behind Michael's eyes. He felt a constant low-grade anxiety, a tight knot in his stomach. He was caught in a current of worries. He felt his face heat up, his palms beginning to sweat.

There it was. The ring. His grandmother's ring. Heโ€™d never thought heโ€™d see it again. He swallowed hard, trying to keep his composure. He felt a deep, profound sense of sorrow.

โ€œHow much?โ€ His voice was a strained monotone, devoid of inflection. The shop owner, a man with a receding hairline, glanced up, a flicker of something unreadable in his eyes. Michael felt a sudden urge to turn and run.

Emotion: troubled

Cluster: Fear / Anxiety
PC1 (Valence): -2.20 Negative
PC2 (Disposition): -0.81

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
๐Ÿ˜ž0.343
S0.317
ness0.302
๐Ÿ˜”0.294
๐Ÿ˜ข0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236