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).

He kept pacing. Back and forth, across the cramped confines of his tiny apartment. The cheap laminate flooring creaked with each step, a rhythmic counterpoint to the insistent thrumming in his ears. Michael ran a hand through his already dishevelled hair. The coach’s blunt pronouncements from earlier felt like a physical blow.

He glanced at his phone, a useless brick in his hand. He’d barely eaten all day. He'd tried, but the food sat heavy, a lump in his chest. His reflection in the window showed a stranger, eyes wide and pupils dilated. A hollow ache spread through him, a feeling of being untethered, adrift. The city lights began to shimmer as darkness descended, painting the outside world in a blurry, indistinct palette.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274