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 notice taped to the lobby door fluttered in the draft. It announced the building’s conversion to condominiums. Sarah’s shoulders dropped, a weight she hadn't realized she was carrying suddenly vanished. She grinned, a genuine, wide expression that reached her eyes.

She skipped up the stairs, humming a tuneless melody, something she hadn't done in months. The peeling paint of the banister, usually a source of minor annoyance, seemed less offensive. She threw open her apartment door, breathing deeply of the familiar air, now tinged with a delicious sense of freedom. No more sleepless nights worrying about the next rent hike.

Her phone buzzed, a text from her best friend. “Seen the news?” it read. Sarah typed back, “Finally! Let the games begin!”

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