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 email notification on Chloe's phone buzzed, announcing the arrival of another forwarded message from her friend, Emily. Chloe opened it, braced. Inside was a link, a real estate listing in North Carolina. Her breath hitched. She scrolled through the pictures of the quaint, sun-drenched house, its manicured lawn a stark contrast to her own cluttered apartment. A bitter taste coated her tongue.

She slammed the phone down, her hand trembling. The room seemed smaller, the walls closing in. She felt a prickling heat behind her eyes. She usually loved hearing from Emily but now, the constant updates, the planning, felt like a personal affront. She walked to her window and stared out at the street. The world looked foreign, and distant, a place she was no longer sure she belonged to.

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