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 old barn, finally renovated. Maria was proud. Behind the exterior brickwork, they discovered a metallic box. "Maria - 1999." A wave of disbelief washed over her. She didn’t bury anything.

Inside, a collection of CDs, a journal filled with dreams of fame, and a brightly coloured scarf. The music was a reminder of her youthful exuberance. The journal spoke of singing on stage, a career she had abandoned.

The scarf was a reminder of her first love. She wrapped it around her, feeling a pang of something akin to guilt. The sound of the wind, rustling through the leaves, sounded like a condemnation. She closed her eyes, the familiar weight of what she had left behind settling upon her.

Emotion: remorseful

Cluster: Shame / Guilt
PC1 (Valence): -1.57 Negative
PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.656
L0.580
悔0.463
remorse0.416
been0.398
Suppressed:
de-0.444
(!)-0.309
имеется-0.289
oldukça-0.287
😍-0.285