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 attic air hung thick and dusty, catching in Leo’s throat. He coughed, a dry, hacking sound that echoed in the confined space. His fingers, suddenly clumsy, fumbled with the cardboard boxes his mother had labeled “Memories.” Inside, a trove of his past unfolded: paint-splattered finger paintings, spelling tests marked with red pen, book reports detailing his childhood obsessions. His chest tightened. He sat back on his heels, a wave of disorientation washing over him. He felt strangely exposed.

He ran a hand through his hair, leaving it standing on end. He couldn't shake the unsettling feeling of being watched, even though he was alone amidst the forgotten relics. He felt like a stranger looking at himself.

He reached for a particularly gruesome drawing of a T-Rex. He remembered doing that one, his mother had been so proud. He hadn't thought about it in decades. A low groan escaped his lips.

Emotion: shaken

Cluster: Fear / Anxiety
PC1 (Valence): -2.72 Negative
PC2 (Disposition): -0.86

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.351
unable0.269
focus0.266
😞0.258
다시0.257
Suppressed:
de-0.769
la-0.579
a-0.375
happy-0.292
"-0.290