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 attic air was thick with dust and unspoken truths. Olivia discovered the list tucked inside a battered photo album. Take a pottery class. Volunteer at an animal shelter. Learn a new language. The words seemed to jump off the page, demanding attention. She felt a growing unease.

She’d never thought her mother had secrets, but now it seemed like her entire life was a carefully crafted performance. A strange hollowness had formed within her. She felt the urge to counteract it.

Without hesitating, she went online and signed up for a beginner’s pottery class. The class was in a dingy, cluttered studio, but the air was charged with a kind of creative energy. She felt the clay cool and slick in her hands. The tactile sensation helped settle her nerves.

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