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 music suddenly sounded too loud, too intrusive. His ears rang, and a dull throb pulsed behind his temples. Michael’s hands fidgeted with his phone, scrolling through messages he'd already read a dozen times. He hadn't expected to see her here. This feeling, a cold dread, was unfamiliar and unwelcome.

He made his way to the buffet table, busying himself with the canapés. His appetite had deserted him, but the activity provided a distraction. He kept his head down, avoiding eye contact, praying to avoid any sight of her.

He felt a tap on his shoulder. He flinched, turning slowly, his heart thumping against his ribs. It was a mutual friend. He tried to muster a smile, but it felt stiff and unnatural. The air in the room felt thick, oppressive.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
🎉-0.233
happy-0.231
"-0.228