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 elevator slowly ascended, each floor a step closer to… him. Mark. The name burned in his mind. The man who made life hell. The man who reveled in others' pain. Now he was supposed to provide… comfort? The thought made Henry's teeth clench.

He ran a hand through his hair, trying to calm the storm brewing inside him. He remembered the humiliation, the fear, the endless days of torment. He'd sworn he would never see him again.

The elevator doors opened. He felt a potent… dislike as he walked to the door, the nameplate gleaming innocently under the office light. "Mark Davidson, LMFT." He took a breath, steeling himself for the ordeal to come.

Emotion: hateful

Cluster: Anger / Hostility
PC1 (Valence): -1.71 Negative
PC2 (Disposition): 2.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.553
S0.367
retali0.313
aggravated0.310
恨0.308
Suppressed:
la-0.334
own-0.310
optimistic-0.270
been-0.264
gradual-0.262