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).

"I don't like it," Clara muttered, mostly to herself. The house was supposed to be a fresh start, a haven. Instead, it felt oppressive, like a cage with gilded bars. Sunlight seemed to struggle to reach the interior, leaving the rooms perpetually dim.

She’d spent the morning cleaning, a frantic attempt to cleanse the space, but the dust seemed to multiply faster than she could eliminate it. Now, she stood in the basement, examining a section of wall that felt…different. Cooler.

The wall yielded to her persistent prodding, revealing a narrow doorway. Clara’s breath caught in her throat. She felt a sudden, inexplicable urge to run, to flee. But something else, a morbid curiosity, held her in place. The darkness inside seemed to beckon. She pulled a flashlight from her pocket.

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