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 news spread through the office like wildfire. "Have you seen it? Sarah's article is out under Martin's name!" The whispers followed him, each one a fresh jab. He avoided eye contact, his face burning with a silent rage. He'd dedicated weeks to the project; Martin had simply, and apparently casually, taken it.

He sat at his desk, staring blankly at the computer screen. He heard Martin laughing with a colleague. His mouth went dry. He clenched and unclenched his fists. He slammed a hand down on the desk, the sound snapping everyone's attention to him.

He stood up, shoving his chair back with a violent screech, and walked out of the office. He needed air. He needed to get away. He needed to deal with this, somehow.

Emotion: disdainful

Cluster: Anger / Hostility
PC1 (Valence): -0.31 Negative
PC2 (Disposition): 1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.419
l0.404
S0.387
T0.282
I0.257
Suppressed:
own-0.335
unexpectedly-0.285
Suddenly-0.246
previously-0.245
until-0.242