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 left the list tucked into his bedside drawer. She shouldn't have been looking, but the curiosity had gnawed at her all afternoon. It was the handwriting that caught her first – the familiar scrawl, suddenly alien on paper.

Her pulse quickened as she scanned the words: "Run a marathon." "Skydive." The list went on, a relentless parade of challenges. A coldness settled in her chest, a premonition of something terrible.

She sat down on the bed, feeling the sudden weakness in her legs. She couldn’t shake the feeling that this was a roadmap to… something she didn’t want to comprehend. The room seemed to shrink, the walls closing in.

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