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 gallery's hardwood floor seemed to tilt beneath Elias's feet as he entered. His stomach churned, a cold knot forming just below his ribs. He gripped the strap of his messenger bag, knuckles white. The air hung thick with the perfume of art snobs and cheap wine, a combination that usually delighted him, but today, it felt cloying. He spotted the offending canvas almost immediately – a vibrant swirl of crimson and gold, undeniably *his*.

He hadn't authorized this. He had not given permission. His heart hammered a frantic rhythm against his ribs.

He cleared his throat. "Excuse me," he said to a woman with a beehive hairstyle, her gaze fixed on the painting. "That...that's mine." The words felt thick in his mouth, a lie he was forced to tell.

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