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).

Rain lashed against the windows, mirroring the storm brewing inside Mark. He'd just received the call; the dreaded news. His surprise 40th birthday bash had been scrapped. The carefully curated guest list, the catered meal, the live band – all vanished in an instant. A coldness crept through him, starting at his extremities and spreading inwards.

He stood in the middle of his empty apartment, the phone dangling limply from his hand. He felt a sudden, desperate urge to flee, to escape the crushing weight of disappointment. Each exhale was a shallow, shaky thing. He started pacing, back and forth, the same tile again and again.

"No," he muttered, the word a strangled sound. "It can't be."

Emotion: frightened

Cluster: Fear / Anxiety
PC1 (Valence): -2.74 Negative
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.324
😰0.323
worse0.301
😖0.294
느껴0.294
Suppressed:
de-0.630
la-0.616
a-0.347
"-0.345
happy-0.294