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 old photographs, scattered on the table, felt heavy in her lap. Sarah watched as Mark, pointing to a grainy image, described their camping trip, recalling the jokes, the campfire stories, the starlit nights. His face was lit with fond recollections.

A cold dread crept over her. She remembered the relentless bugs, the biting cold, the endless struggle to keep the tent dry. Her shoulders ached with a fatigue that seemed to seep into her bones.

“I just remember being miserably cold,” she admitted, her voice flat. Mark stopped talking, his expression shifting. The air between them, once filled with shared laughter, suddenly felt thin.

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