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 ballroom buzzed with chatter and laughter. Then, amidst the crowd, there he was: Leo. Sarah had pictured an encounter that would unravel her completely. She’d envisioned icy stares and forced smiles. But instead, she simply felt… unburdened. The pressure she had been feeling, the anticipation of pain, had miraculously dissipated.

She lifted her chin, meeting his gaze with a casual nod. It felt liberating. She found herself suddenly wanting to talk to people, to mingle, to actually *enjoy* the party. The tightness in her chest, which had been there for weeks, had vanished.

Emotion: relieved

Cluster: Positive / Joy
PC1 (Valence): 2.22 Positive
PC2 (Disposition): -0.97

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.638
finally0.606
finalmente0.597
relieved0.561
enfin0.546
Suppressed:
l-0.443
:(-0.439
😣-0.429
S-0.388
😖-0.385