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).

A cold sweat slicked down Marcus’s back as he read the online article. The byline read "Dr. Evelyn Reed." He'd poured weeks into that research, all of it. Now, someone else was reaping the accolades. He reread the introduction, a knot forming in his chest. His hands trembled, making it difficult to scroll.

He felt a sudden, inexplicable need to leave the sterile environment of his office. The fluorescent lights seemed to hum with a malevolent energy, amplifying the disquiet he felt. He grabbed his coat, ignoring the insistent buzzing of his phone.

He found himself wandering the city streets, the anonymity of the crowd both a comfort and a source of unease. He ran a hand through his hair, the gesture a desperate attempt to find some semblance of control. The city's sounds became a cacophony, each car horn and chattering conversation a reminder of the injustice.

Emotion: disturbed

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.318
ness0.309
无法0.300
😞0.297
worse0.284
Suppressed:
de-0.829
la-0.666
a-0.392
B-0.301
"-0.290