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 air in the reception hall felt thick, almost suffocating. Ben pushed at his collar, feeling a familiar heat creeping up his neck. He watched, with a growing sense of irritation, as Sophie made her way toward him, a playful smile on her face. This was just great.

He’d imagined this scenario, of course. They'd shared too many mutual acquaintances for him to believe he'd escape a run-in entirely. He just hadn't anticipated the *level* of awkwardness. He forced a smile, a shallow and unconvincing mask.

He wished he had chosen a different tie, a different suit, a different everything. He wished he were anywhere else.

Emotion: exasperated

Cluster: Anger / Hostility
PC1 (Valence): -0.99 Negative
PC2 (Disposition): 1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.435
l0.378
P0.331
que0.320
la0.269
Suppressed:
own-0.433
gradually-0.254
soon-0.248
unfolding-0.216
joyful-0.216