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 familiar scent of coffee brewing usually settled his nerves, but tonight, it only amplified the unease that had been building inside him all day. The ticking of the clock in the hall was deafening. He felt a constant awareness of the space around him.

He heard it, a soft murmur coming from the living room. The language that was his. The words, so familiar, now spoken by her. She was speaking Mandarin. His Mandarin. He felt the blood rush to his head. He entered the room, his movements awkward.

She turned towards him, a book open on her lap.

“你好,亲爱的,” she said, her tone calm and assured. He felt a jolt of… something. He felt his hands begin to shake. He felt unable to form words.

Emotion: alert

Cluster: Alertness
PC1 (Valence): -0.09 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.359
उत्सुक0.209
سریع0.205
urgency0.205
緊張0.198
Suppressed:
la-0.385
B-0.374
L-0.317
H-0.303
a-0.235