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).

Sunlight spilled across Amelia’s face, coaxing her awake. She stretched, limbs loose and unburdened. The smell of coffee, usually a source of mild anxiety, today felt like a warm embrace. She padded downstairs, humming a little tune she didn’t realize she knew. In the kitchen, Liam was hunched over his laptop, a familiar furrow between his brows, but when he saw her, his face lit up. "Morning, sunshine. Coffee's ready."

"Mmm, smells divine," Amelia murmured, reaching for a mug. Later, while Liam was at work, she felt a sudden, almost giddy impulse to tidy the bookshelf. Dust motes danced in the sunbeams as she straightened spines, pausing at the small, leather-bound notebook she’d given Liam last Christmas. Curiosity tugged. She flipped it open. The handwriting was his, and the opening line read, "She woke this morning, radiant..."

Emotion: refreshed

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.998
de0.703
!).0.384
(!)0.381
!".0.379
Suppressed:
S-0.782
C-0.478
L-0.414
😣-0.388
😞-0.376