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 flickering television screen cast dancing shadows across the living room, mimicking the internal turmoil within Alex. His roommate, Chris, was regaling him with tales of a potential pet parrot's intelligence. Alex just wanted to be left alone.

He stared at the blank screen, his mind a jumbled mess of half-formed thoughts and unfulfilled obligations. A dull ache throbbed behind his eyes, a persistent reminder of his mental state. The idea of caring for a bird, cleaning up after it, and attempting to teach it to talk felt like an impossible feat.

“I'm not sure I can handle it right now, man,” he said, his voice flat and devoid of emotion. Chris's enthusiasm faded, replaced by a puzzled frown, and the silence that followed felt like a physical weight.

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