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).

Rain lashed against the bookstore window, mirroring the apathy that had settled deep in Elias's bones. He wandered the aisles, each worn cover a potential distraction, yet nothing seemed to hold his attention. He picked up a copy of "Wuthering Heights," its pages yellowed and brittle, a faint scent of old paper and something else, something indefinable, clinging to it. He flipped through, fingers dragging across the text, the words blurring into a meaningless pattern. Then, nestled between pages, a folded letter, its ink faded but still legible.

His heart gave a tiny jolt, a flicker of something akin to curiosity. The sender's name was illegible. The recipient, a "Dearest Eleanor," was a stranger, their joys and sorrows now trapped in ink and paper, a world away. Elias sat down on the dusty floor, the book resting on his knees, and began to read.

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