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 email arrived like a death knell. Sarah opened it, her fingers clumsy and slow on the keyboard. The words swam before her eyes: "Final Semester." She reread it, each line a fresh stab of… something.

She pushed back from her desk, the chair scraping loudly against the floor. Outside, the world was alive with activity, but inside, she felt…stagnant. She couldn't bring herself to care about her upcoming exams, about her almost-completed thesis.

She went for a walk, but the familiar campus paths now felt less like a route and more like a maze. Each familiar building, once a symbol of possibility, now represented a closed door. She just kept walking, not really looking at anything, not really going anywhere.

Emotion: listless

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.03 Negative
PC2 (Disposition): -1.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
’0.325
давно0.280
unremarkable0.270
lately0.266
either0.266
Suppressed:
B-0.460
P-0.331
l-0.312
C-0.292
👏-0.273