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).

The notification popped up on David’s phone during his anatomy lab. He excused himself, citing a sudden, urgent need for air. He slipped out, the sterile smell of formaldehyde suddenly nauseating. He sat on the cool stone steps outside, breathing deeply.

He looked at the message again. "Congratulations! Your essay has been selected…" He swallowed hard. His palms were slick. He felt as though the world was closing in, his throat tightening.

He pictured his parents reading it, sharing it with relatives. The thought made him squirm. He closed his eyes, squeezing them shut, trying to shut out the thoughts, the feelings that swirled within him.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274