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).

Sweat beaded on his forehead as he stared at the email. His hands were clenched, knuckles white. Dr. Ramirez’s office, usually a haven of warmth and support, was now just an empty room. The blinds were drawn, casting the space in a dim, oppressive light. It was a betrayal of sorts.

He slammed his fist against the desk, a sharp, echoing sound. His breath came in shallow gasps. He’d spent years under her guidance, learning everything. And now... nothing. He felt the familiar weight of self-doubt creep into his thoughts. He ran a hand through his hair, as though trying to smooth the rough edges of his own confusion. The quiet was deafening. He felt a sudden, sharp ache in his chest, a sensation he couldn’t name.

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