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 harsh fluorescent lights of the shop hammered at David’s senses. His head throbbed, a dull ache that echoed the emptiness inside him. He leaned against the counter, the cool surface a temporary relief. He had been wandering, aimless, for hours, drawn to the shop by some unconscious urge.

The proprietor, a woman with a severe expression, gave him a look that spoke volumes. He was a familiar face here; another person seeking an escape. And then he saw it. The ring. His grandmother’s. He hadn’t thought about her in months, the grief a constant, unwelcome companion.

He let out a sigh he hadn't known he was holding. He didn’t want to be here. He didn’t want anything. "Can I… can I see that ring?" he mumbled, his voice raspy.

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