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 silence in Emily's apartment was oppressive, each tick of the clock echoing in the stillness. She was supposed to be working on a new project, but the words wouldn't come. Her mind felt like a tangled mess. She stared at the ceiling for hours.

Then, a notification, a message from "BookwormBabe," her online friend. "Feeling the existential dread today?"

Emily let out a short laugh. She typed, "You have no idea."

The response came almost immediately. "Let's meet at that new bookstore. My treat."

Emily hesitated. Leaving the house felt like climbing a mountain. But the prospect of human interaction, a shared experience, a distraction, pulled her from her stupor. She threw on some clothes, feeling a spark of something she hadn't felt in a long time.

Emotion: weary

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.52 Negative
PC2 (Disposition): -1.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.384
😞0.288
sighed0.230
ness0.210
😪0.203
Suppressed:
de-0.339
l-0.300
!-0.222
la-0.221
🤟-0.198