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 silence in the house was deafening. The divorce papers were signed, sealed, and delivered. The echoing emptiness of her newly single life felt like a physical weight on Sarah’s chest. The future, once a shared tapestry of dreams, now stretched before her, a stark, unpainted canvas.

She went for a walk, desperate to escape her own mind. The first yard sale of the season. At the bottom of a box of toys, nestled amongst plastic cars and forgotten board games, lay a bear. A bear she knew all too well. It was her first toy.

Sarah stood, unmoving, the wind whipping around her. A single tear escaped, tracing a path down her cheek, and dropped onto the bear. She felt an unbearable hollowness that threatened to swallow her whole.

Emotion: miserable

Cluster: Sadness / Despair
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.87

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.369
无法0.321
焑法0.270
πŸ’”0.265
πŸ˜”0.262
Suppressed:
de-0.490
la-0.329
B-0.300
l-0.292
/-0.224