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).

“We are delighted to announce…” The landlord’s cheerful tone grated on her nerves. She slammed the phone down, the sound echoing in her apartment. Her jaw clenched, a knot forming in her neck. Her hands were ice cold.

She started leaving her keys in the lock when she went out, a small act of defiance. The thought of not being able to come back to this place made her sick. The familiar sounds of the city, once a source of comfort, now felt like an intrusion, a constant reminder of the changes to come.

She spent an hour staring out the window, watching the sun set over the city. She imagined the future, a future where her sanctuary, her safe haven, was no longer hers. The feeling was a sharp, piercing jab.

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