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 conditioning hummed, but it didn't cool the room. Michael felt a tremor in his hands as he stared at David. He’d explained, carefully, his need for a feline companion. A cat would be quiet, independent, and wouldn’t require constant attention. David's response had been a flat, unwavering "No."

He shifted his weight, his leg bumping against the coffee table. He felt a creeping tension in his shoulders, a sensation of being trapped. He ran a hand over his face, trying to clear his thoughts.

“I just think… it would be nice,” he finally managed, his voice barely audible. The look on David's face, a mixture of disapproval and impatience, crushed any remaining hope.

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