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).

It was the description that got to her. "Spacious," "Updated," "Move-In Ready." Anna felt a chill, an icy tendril that snaked down her spine. The words felt designed to erase the past, to render her own experiences irrelevant.

She scrolled through the photos again, each one a carefully constructed image of perfection. The stark, clean lines of the kitchen, the minimalist decor, the absence of any personal touch. She saw none of the fingerprints of a life well-lived.

She closed the laptop, a single tear tracing a path down her cheek. The world around her seemed to tilt and sway. She felt a profound sense of disconnection, as if she no longer belonged anywhere.

Emotion: sensitive

Cluster: Fear / Anxiety
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.75

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ness0.304
😞0.282
😥0.272
😣0.271
own0.264
Suppressed:
la-0.679
de-0.660
a-0.281
"-0.280
(!)-0.242