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 air in the house was thick, stagnant, as if the very walls were holding their breath. Olivia sat in the armchair, the list resting limply on her lap. "Open a cafe," she read, her voice barely a whisper. Her husband, Ethan, was at work, immersed in his career. She’d found the list while cleaning out the junk drawer.

She scanned the subsequent entries: "Learn to cook Italian cuisine," "Go to Italy." She had a sudden, vivid memory of the two of them, huddled in front of a tv, watching cooking shows. She had always wanted to try.

A wave of nausea swept over her. She felt like she was drowning, the words on the list swirling around her, a reminder of all the dreams she’d forgotten, of all the journeys they had missed.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
πŸ’”0.252
πŸ˜”0.249
😒0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199