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 gallery owner was enthusiastic, but Sofia found it difficult to muster a smile. Her ‘Serpentine Dance’ was now inked on a man's chest. He'd even sent her a photograph. She felt a strange detachment, a distance.

She felt a hollowness in her chest. She excused herself from the conversation, claiming a need for fresh air. Outside, the city was bustling, but she felt oddly disconnected, as if observing the world through a pane of glass.

She walked for blocks, aimlessly, the feeling of emptiness a constant companion. The laughter of children on the street grated on her nerves. She felt an overwhelming urge to go home and hide.

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:
😞0.343
S0.317
ness0.302
😔0.294
😢0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236