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 attic was stifling, the heat radiating off the dark wood panels. The whole place felt like a vault for memories I didn't want.

A collection of old magazines, their covers faded and torn, lay scattered across the floor. I flipped through one, my fingers growing dusty. There were stories about things I didn't care about, articles about people I didn't know.

I put it back down with deliberate force. A porcelain doll with a cracked face stared at me from a nearby shelf. I didn't know whether to scream or laugh.

I felt the corners of my mouth turn downwards. The previous residents, the Whites, were a bunch of… well, I didn't know what they were. I did know that I'd rather be doing anything but this.

"Unbelievable," I finally said, mostly to myself.

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