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 damp earth clung to Clara's boots as she carefully wiped the grime from the box's surface. A wide, incredulous smile plastered itself on her face. A time capsule? She couldn’t remember burying a time capsule. She ran a hand through her hair, a giggle escaping her lips. This was the most ridiculous, amazing thing!

The rusty latch yielded with a satisfying *clunk*. Inside, she found a collection of old photographs, a mix tape labelled with a childishly drawn smiley face, and a worn copy of her favorite childhood book. Memories flooded back. She felt a lightness, a buoyancy she hadn't known she was missing. The world around her seemed brighter, sharper, filled with a renewed sense of possibility.

Emotion: elated

Cluster: Positive / Joy
PC1 (Valence): 3.38 Positive
PC2 (Disposition): -0.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.724
!0.596
ecstatic0.510
🤩0.510
overjoyed0.504
Suppressed:
S-0.657
😞-0.497
ness-0.407
😔-0.380
😓-0.377