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 coffee cup trembled in her hand, the hot liquid sloshing dangerously close to the rim. It was a letter to her grandmother, written just weeks before her passing. It spoke of petty grievances, of feeling unloved.

Her eyes stung, a familiar pressure building behind them. The words, once so easily typed, now seemed to mock her, a constant reminder of missed opportunities and unspoken apologies. She had been so selfish, so blind to her grandmother's frail state.

She set the cup down with a clatter, the ceramic jarring against the wooden table. The room spun slightly, the air thickening with a suffocating sense of regret.

Emotion: mortified

Cluster: Embarrassment
PC1 (Valence): -1.36 Negative
PC2 (Disposition): 0.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.406
尴尬0.257
C0.252
worse0.245
逃0.242
Suppressed:
own-0.340
了他-0.218
他对-0.203
ል-0.194
楽し-0.193