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 waiting room was all muted tones and soothing artwork. He wanted to gag. He, Daniel, who’d been wrestled to the ground more times than he could count by a hulking brute named Mark, was now sitting in Mark’s waiting room. The irony was a bitter pill he was forced to swallow. He fidgeted, the fabric of his perfectly tailored suit suddenly too tight.

He picked up a magazine, deliberately slamming it back down. The receptionist, a woman with a placid smile, didn’t flinch. Good. He wasn't here to be coddled.

His name was called. He rose, his back stiff. He walked down the corridor, each step a reluctant capitulation to the universe's cruel sense of humor. He wondered if Mark would have the decency to look sheepish. He doubted it.

Emotion: contemptuous

Cluster: Anger / Hostility
PC1 (Valence): -0.40 Negative
PC2 (Disposition): 2.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.502
l0.423
fact0.287
either0.265
👎0.262
Suppressed:
own-0.367
unexpectedly-0.303
serendip-0.265
Suddenly-0.253
previously-0.250