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 hospital cafeteria offered little in the way of comfort. The food, bland and unappetizing, only intensified David's mood. He pushed the plate of congealed mashed potatoes away with a disgusted shove. The entire institution felt like a prison.

He hated the forced cheerfulness of the nurses, the hushed whispers of other patients. He felt a desperate urge to run, to escape this medical maze.

He saw a man at a nearby table, meticulously tearing apart a bread roll. He was doing it slowly. Deliberately.

He walked over to him, his hands clenching at his sides.

“This whole thing is a goddamn joke,” David stated. “You, too?”

The man looked up, his face a mask of weary resignation. “The chondrodysplasia?” he asked.

David nodded, the shared acknowledgment a bitter pill to swallow.

Emotion: furious

Cluster: Anger / Hostility
PC1 (Valence): -2.01 Negative
PC2 (Disposition): 1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.461
骂0.376
🤬0.338
aggravated0.313
est0.304
Suppressed:
a-0.356
de-0.334
L-0.308
H-0.301
adventurous-0.256