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).

David, usually affable, was curt. “Of course,” he’d responded when his mother told him about his half-brother. “Good for him.” His voice was flat, devoid of any warmth. He felt a coldness spread through him, a desolate bleakness he hadn't experienced since his parents' divorce.

He spent the next few days in a haze. He cancelled plans, ignored phone calls, and avoided his mother. He started working late at the office, filling himself with pointless tasks, and avoiding the emptiness of his apartment.

One evening, he was having dinner with a friend, and he could barely touch his food. His friend noticed his mood. "Is everything okay?" she asked gently. David shrugged. "Fine," he replied, and quickly changed the subject. His voice sounded hollow even to him.

Emotion: scornful

Cluster: Anger / Hostility
PC1 (Valence): -0.76 Negative
PC2 (Disposition): 2.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.406
C0.391
S0.276
I0.268
愤0.243
Suppressed:
own-0.396
unexpectedly-0.288
Suddenly-0.260
soon-0.256
serendip-0.256