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’s just wood, Liam," his wife had said, her voice gentle, attempting to soothe. But Liam could not shake the feeling of raw, exposed nerves. The old willow, a sentinel in his grandparents’ yard, was gone.

He wandered through the park, avoiding eye contact with other people. He felt exposed, fragile. Each rustle of the leaves felt like an accusation. His throat was constricted, making it hard to breathe normally.

He sat on a nearby bench, the metal cold against his clammy hands. He wanted to scream. He wanted to rewind time, to shield the tree from the inevitable. He wanted something, anything, to stop the feeling of a gaping hollowness inside him.

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