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 document title, "Her Secret," made her feel faint. The opening sentences, describing their first date, were hauntingly accurate. A deep, hollow ache spread through her chest. It was like he had been living a double life.

She dropped the laptop, the screen flashing, momentarily catching her eye. She felt a sudden, desperate urge to move. She paced the living room, her movements stiff and jerky. Her mind raced, a torrent of questions and doubts.

She needed air, space. She reached for her keys, her fingers fumbling with the familiar shapes. The world outside felt unreal, as if she were walking through a dream. Her chest tightened with a crushing pressure.

Emotion: panicked

Cluster: Fear / Anxiety
PC1 (Valence): -2.44 Negative
PC2 (Disposition): -0.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.317
😫0.294
😖0.285
😰0.285
escape0.283
Suppressed:
la-0.432
a-0.374
de-0.296
楽し-0.262
happy-0.258