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 rhythmic thudding started promptly at seven each morning. Sarah’s hand clenched around her mug, the ceramic groaning under the pressure. She hadn’t slept well since the hammering began next door. It felt like the sound was inside her head, echoing in time with her heartbeat. She pulled the curtains shut, but the muffled noises still seeped through, a constant, low-grade irritation. She kept finding herself checking the locks, double-checking the windows.

Her neighbor, Mr. Henderson, had always been a quiet man, almost invisible. Now, he was a cacophony of banging, sawing, and shouting at unseen workers. She jumped at the slightest noise, her nerves frayed. One afternoon, she found herself staring out the window, mesmerized by the activity, a tremor running through her.

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