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 sunlight seemed to beat down on Maya’s shoulders, a physical manifestation of the internal tremors she couldn’t shake. She wandered through the rows of discarded belongings, a growing sense of disquiet twisting in her stomach. Every clinking price tag, every shouted offer, grated on her nerves. Her fingers fidgeted, tracing the lines of a cheap, plastic bracelet she wore.

Then she saw it. Tucked beneath a pile of mismatched teacups, there he was. Her bear. The one she’d lost, or rather, the one that had disappeared, years ago. A strange tingling spread across her skin.

She heard herself asking about the price, her voice barely a whisper. The woman, eyes narrowed, named a number. It felt too high, but Maya handed over the cash without haggling. Clutching the bear, she felt exposed, raw. She didn't want to go home.

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