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 air in the house seemed to vibrate with unspoken words. David ran his hand through his hair, a gesture that did little to tame the wildness inside. His daughter, Sophie, had a secret profile filled with photos and comments that made his stomach churn.

He felt like he was standing on the edge of a cliff. One wrong move, one careless word, and he would push her away. He walked to the kitchen, poured a glass of water, and took a large gulp. It did nothing to quench the dryness in his throat.

He knew he had to talk to her. But the words were caught in his throat, a thick, sticky mass. He thought about the years he had spent raising his daughter. Suddenly, it all seemed fragile, precarious.

Emotion: tense

Cluster: Fear / Anxiety
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.213
😰0.191
느껴0.183
😣0.183
😖0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198