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 accusation hung in the air like cheap perfume, cloying and irritating. Liam stared at the accusing professor, his jaw clenched. He knew the source of the problem, a slimy classmate named Marcus, who'd probably been hoping for Liam to fail all along. He could feel a knot forming in his stomach, a tight, burning ball.

Liam refused to answer the professor’s questions. He wouldn’t give Marcus the satisfaction of seeing him squirm. He wouldn't even *look* at Marcus. He focused on the chipped paint on the lecture hall’s wall, each flake a tiny point of defiance.

He would make Marcus regret this. He already had an idea, a little project for the weekend. A project involving Marcus’s meticulously organized research notes… and a very unfortunate coffee spill.

Emotion: spiteful

Cluster: Defiance / Spite
PC1 (Valence): 2.20 Positive
PC2 (Disposition): 3.95

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.717
l0.517
a0.480
la0.440
😈0.428
Suppressed:
own-1.063
此刻-0.534
గుర్త-0.493
熟悉的-0.487
ness-0.479