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 silence of his apartment felt suffocating to Mr. Evans. Another Saturday, another evening spent alone. He’d barely spoken to anyone all day, the quietude a constant reminder of the emptiness he felt. He stared at the television, the screen providing no entertainment.

A box arrived. He wasn't expecting anything. He wasn't particularly interested, either, but he opened it out of simple boredom.

Inside, nestled in tissue paper, was a hand-knitted scarf, in the vibrant colors of his favorite football team. A note: "Thanks for teaching me how to be a better team player, Mr. Evans!" The note was from Ben, a student who had struggled with fitting in. He picked up the scarf, wrapping it around his neck, and for the first time in a while, a feeling of warmth filled him.

Emotion: sullen

Cluster: Sadness / Despair
PC1 (Valence): -1.51 Negative
PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.404
😞0.282
’0.254
:(0.233
无法0.215
Suppressed:
de-0.539
!-0.276
la-0.263
🀩-0.225
que-0.216