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 antique music box sat on Eleanor’s desk, its delicate melody a mockery of the discord in her heart. She picked it up, running a thumb over the tarnished silver. It was a gift from Sarah, a memento of their trip to France. Sarah, who now insisted she’d wanted to visit the Louvre, not wander aimlessly through flea markets.

Eleanor felt a chill creep over her. The sun dipped, casting long shadows across the room. She could almost feel the weight of Sarah's disappointment. She had wanted to show Sarah a great trip, and she had utterly failed.

She pressed her forehead against the cool glass of the music box. The melody, once a comfort, now felt like a lament.

Emotion: sorry

Cluster: Sadness / Despair
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.604
L0.549
😞0.382
😔0.364
ness0.324
Suppressed:
de-0.675
la-0.536
(!)-0.323
!-0.298
l-0.288