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 coffee mug nearly slipped from Amelia's grasp as she saw the familiar address flash on her laptop screen. Her heart gave a joyful leap, a playful kick against her ribs. "No way," she breathed, a wide grin stretching across her face.

She clicked on the listing, her fingers practically vibrating with anticipation. The photos loaded, each one a burst of color and memory. The oak tree in the front yard, the swing set in the back, the chipped paint on the porch. A giggle bubbled up from her throat.

She spun her chair around, knocking over a stack of papers. "Oh, bother," she muttered, but the frustration was gone in a flash, replaced by a giddy lightness. She didn't even care. This was amazing.

She scrolled through the details, already calculating how much she could possibly afford. This, she knew, was a sign. A cosmic nudge. This was the universe telling her it was time to go home.

Emotion: exuberant

Cluster: Positive / Joy
PC1 (Valence): 3.42 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.676
de0.605
la0.596
!"0.551
!0.548
Suppressed:
S-0.892
😞-0.474
ness-0.407
L-0.389
😔-0.376