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'd been wandering the city, lost in his thoughts, the weight of the rejection letters pressing on him. Then, he heard it: a familiar melody, played on a saxophone from a street corner. He stopped, his footsteps halting on the cold pavement.

It was his composition, the one that had been rejected by every single label in town. The saxophonist, a grizzled man with a worn fedora, played it with a passion that surprised him. His fingers fumbled but the melody rang true.

A strange lightness filled him. He took a tentative step closer, then another, drawn by the music. His lips formed a silent "Oh."

He watched, captivated, a smile slowly spreading across his face.

Emotion: astonished

Cluster: Surprise / Confusion
PC1 (Valence): -0.84 Negative
PC2 (Disposition): -0.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.448
de0.350
la0.323
l0.307
incred0.266
Suppressed:
own-0.289
cheerfully-0.231
enjoyed-0.224
开心-0.221
ี่ยว-0.219