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 relentless demands of the project had chipped away at his resolve. He felt like he was wading through molasses, each movement a struggle. He closed his eyes for a moment, trying to find a moment of peace, a moment of respite. When he opened them, the office was as it always was: noisy, bright, and demanding.

"Hey, Maria," he said, his voice flat. "Did you ever go to one of those sleep away camps?"

Maria paused, pen in hand. "Yeah! Camp Skyview. How about you?"

A brief moment of lightness bloomed within him. "Skyview, no kidding? I was there, too. 1999. Remember the color war?" He found himself grinning, a genuine expression he hadn't worn in days.

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