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 brisk autumn air invigorated Amelia. She inhaled deeply as she walked to her car, her senses heightened by the crisp, golden light. A week of work was over, and her niece, Chloe, was coming over.

Amelia went to meet Chloe at the door, her heart skipping. Chloe, with a gleam in her eyes, ran to her, wrapping her arms around Amelia's legs. After setting up a craft table, Chloe began to draw. Amelia watched, charmed, and a shiver went down her spine when she noticed the creature that showed up in Chloe's drawing – a fluffy, pink dragon with a perpetually confused expression. It was Bartholomew, her childhood confidant. The past and the present blended together, causing a giddy feeling in Amelia's stomach.

Emotion: vibrant

Cluster: Positive / Joy
PC1 (Valence): 2.83 Positive
PC2 (Disposition): -1.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
!0.459
!".0.436
de0.434
!).0.409
Suppressed:
S-0.890
L-0.478
P-0.407
H-0.365
😞-0.306