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 steam from my coffee swirled in the air, creating a hazy cloud. Michael was outside, ostensibly tending to the garden, but I could hear the muffled sound of his voice carrying through the open window. It was the way his voice became so low, so urgent; a telltale sign.

He was usually so straightforward, but the past few weeks had been a parade of evasiveness. The change was fascinating, a captivating performance that had me hooked. I tapped my fingers on the kitchen table, a smile tugging at the corners of my mouth.

This clandestine affair, whatever it was, was the best form of entertainment I’d had in a while.

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
šŸ˜ž-0.523
’-0.473
ness-0.437
urges-0.403