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 unpacked boxes sat like silent, judgmental monoliths in the living room. Sarah prodded at a loose thread on the sofa, her finger tracing the worn fabric. A dull ache throbbed behind her eyes, a constant companion since the move. She’d promised herself a fresh start, a new chapter, but unpacking felt less like embarking and more like enduring.

She wandered aimlessly through the cavernous house, a vast space that echoed her own hollow feelings. The previous owners had been meticulous; every room was pristine, almost sterile. Bored, she kicked at a loose floorboard in the hallway, the sound strangely loud in the quiet house. It shifted, revealing a dark, yawning gap.

A small, wooden door, hidden behind the floorboard, swung inward with a protesting creak. Curiosity, a rare visitor these days, flickered inside her. The hidden room seemed to beckon, whispering secrets.

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