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 air was thick and stagnant, the temperature oppressive. Ben wiped his brow as he rummaged through the boxes. His mother's compulsion to save everything was, to put it mildly, intense.

He discovered a box dedicated to his high school band days, filled with crumpled sheet music and terrible recordings. A wave of heat rushed up his neck.

He pulled out a faded photo of him, mid-solo, his face contorted in an expression of intense… dedication. The memories of the awkward gigs and the harsh critiques from his bandmates flooded back. He felt a deep, uncomfortable cringe.

“So, you were a musician,” his father said, appearing in the doorway.

Ben mumbled, “It was a phase.”

Emotion: insulted

Cluster: Anger / Hostility
PC1 (Valence): -0.98 Negative
PC2 (Disposition): 1.46

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.354
骂0.277
愤0.267
C0.237
🤬0.233
Suppressed:
own-0.325
давно-0.229
adventurous-0.223
soon-0.219
optimistic-0.217