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 grit of the city seemed to cling to her skin as she entered the pawn shop. Every step she took felt like a monumental failure, the weight of her mistakes pressing down on her. The shop itself was a monument to other people's bad luck, a place that mirrored her current state.

She feigned casual interest, browsing the chipped ceramics and broken radios. The owner, a man with a perpetually bored expression, watched her with narrowed eyes. He knew. Everyone always knew. She felt the familiar sting in the back of her throat, the prelude to a welling up of shame.

Finally, she saw it. The ring. Its sparkle, even through the grime, seemed to mock her. A wave of nausea washed over her. She knew she shouldn't have failed, should have been better.

Emotion: self-critical

Cluster: Shame / Guilt
PC1 (Valence): -1.66 Negative
PC2 (Disposition): 1.28

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-critical vector is projected through the unembedding matrix.

Promoted:
L0.748
S0.662
ness0.425
ly0.374
탓0.351
Suppressed:
la-0.492
soon-0.304
(!)-0.301
secured-0.284
cautiously-0.280