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 phone vibrated violently against the counter, displaying an unknown number. Liam stared at it, his jaw clenched, the veins in his neck bulging. His heart hammered a frantic rhythm against his ribs. It had to be her.

He slammed his fist against the wall, leaving a dent in the plaster. He wanted to smash the phone into a million pieces. He wanted to scream until his voice gave out.

He had spent his whole life trying to be a good son. Now he discovered that the man he idolized had been deceiving him the whole time. He let the phone ring until it went to voicemail, then went outside and kicked the tire of his car.

Emotion: angry

Cluster: Anger / Hostility
PC1 (Valence): -2.05 Negative
PC2 (Disposition): 1.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.354
est0.354
aggravated0.342
🤬0.337
恨0.323
Suppressed:
de-0.424
a-0.411
la-0.341
H-0.320
L-0.296