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 sheer audacity of it all was… rather entertaining, thought David. He watched, arms crossed, from his porch. Bertram, the man who had struggled to change a lightbulb last week, was now attempting to build an extension. The shed, a sad, pre-fabricated thing, was precariously balanced.

He raised an eyebrow as Bertram, red-faced, wrestled with a particularly stubborn piece of wood. The angle was off. Very off. David let out a small, quiet puff of air, a sort of half-suppressed exhale. He knew what was coming next. He opened the book he was reading and settled into his seat for the long run.

Emotion: amused

Cluster: Positive / Joy
PC1 (Valence): 3.32 Positive
PC2 (Disposition): 1.25

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.841
de0.742
l0.544
L0.527
"0.514
Suppressed:
own-0.630
😞-0.523
’-0.473
ness-0.437
urges-0.403