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).

"Okay, okay," Chloe muttered to herself, staring at the text. "Just breathe." Her online friend, "PixelPusher," a gamer she'd spent countless nights with, wanted to meet at the arcade down the street. It was only two blocks away, but suddenly it felt like a marathon.

Her fingers trembled as she typed a reply. The arcade, once a haven of fun, now loomed like a potential trap. She felt a cold dread creeping into her limbs, turning them to lead. She looked around her apartment. Everything seemed too bright, too loud.

She felt a sudden urge to flee, to disappear. Maybe she could just tell him she got sick? No, that was too obvious. She slowly walked toward the door, trying to shake the feeling of being watched.

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