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 news felt like a physical blow. Elena gasped, the breath catching in her throat, as her mother confessed. A sister. A sister who existed in another city, living a life Elena knew nothing about.

Her carefully constructed world, once so stable and predictable, suddenly felt tilted and precarious. She couldn't eat. Couldn't sleep. The very air seemed to press down on her, constricting her chest.

She abruptly canceled her weekend plans, the trip to the beach that she'd been looking forward to. The sand, the sun, the laughter of friends - they all felt hollow, meaningless. She stayed in her apartment, the silence amplifying the turmoil inside.

Emotion: hurt

Cluster: Fear / Anxiety
PC1 (Valence): -2.46 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.412
L0.396
ness0.338
😞0.328
😢0.290
Suppressed:
de-0.643
la-0.402
!-0.259
(!)-0.248
/-0.244