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 old, familiar comfort of his workshop was now replaced with a deep discomfort. Silas ran a calloused hand across his beard, a nervous habit he’d picked up in his youth. The man, a total stranger, had just entered, showing the tattoo on his back. A near perfect rendition of Silas’s intricate wood carving, "The Protector."

A cold dread spread through him. The air seemed to thicken, making it difficult to breathe. He felt a sudden urge to shut the door, to lock himself away.

He muttered something about the piece being admired. The man responded with a casual shrug. Silas knew he was trapped in his own space. He needed to get away, and he needed to do it now.

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228