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 chipped porcelain mug felt heavy in Clara's hand. Liam sat across from her, grinning, oblivious. "Remember that disastrous karaoke night?" he chortled, "You were *terrible* singing 'Bohemian Rhapsody'!" Her jaw tightened. She took a slow, deliberate sip of her tea, the scalding liquid doing little to soothe the sudden heat rising in her chest.

"I thought *you* were the one who blanked on the words," she finally stated, her voice flat. Liam waved a dismissive hand.

"Oh, yeah, the chorus. But after you, well, it was a train wreck." Clara's knuckles turned white. She set the mug down with a thump. Her smile felt like a mask.

"Actually, I recall it quite differently," she said, her tone deliberately saccharine. She suddenly found the patterns in the tablecloth incredibly fascinating, studying them intently. The flush on her cheeks felt like a betrayal.

Emotion: insulted

Cluster: Anger / Hostility
PC1 (Valence): -0.98 Negative
PC2 (Disposition): 1.46

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.354
骂0.277
愤0.267
C0.237
🤬0.233
Suppressed:
own-0.325
давно-0.229
adventurous-0.223
soon-0.219
optimistic-0.217