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).

He fidgeted with the worn strap of his backpack, feeling the unfamiliar pinch of the new dress shoes he’d splurged on. The invitation, tucked carefully in his pocket, felt like a lead weight. Liam had always known Mrs. Gable as the sweet, knitting-circle type. Now, after accidentally overhearing her talking about her private jet, he felt like he was playing a part.

He ran a hand through his hair, smoothing down the stray strands. His palms were slick with sweat as he approached the sprawling mansion. The wrought-iron gate loomed before him, an impenetrable barrier.

He cleared his throat, suddenly conscious of the holes in his jeans and the faded logo on his t-shirt. The security guard’s impassive gaze felt like an accusation.

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