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).

He tried to focus on his work, but the email, the reminder, the impending task, sat like a stone in his gut. Michael had worked hard to get where he was, to master his craft. Now, all that knowledge, all that effort, was about to be passed on.

His hands shook as he scrolled through the company intranet. He felt a cold sweat on his forehead. He found himself staring blankly at the screen, unable to process the information. His breath hitched in his chest, as if a physical pressure was squeezing the air from his lungs.

The moment the new person walked into his office, he felt his world tilt. He began to speak, his voice thin, reedy, as if unused for a long time. He felt a sudden, desperate urge to bolt, to run away from the situation.

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