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 news felt like a punch to the gut. The company was downsizing. Layoffs. Her job was on the chopping block. She felt the blood drain from her face. Disbelief. Fear. She couldn't believe it.

She’d confided in "PixelPals", sharing her work struggles. They had connected over a mutual appreciation for video games and memes. A new message from him arrived: "Hey, at the arcade! Come hang out?"

The arcade was located in the same building as her office. A wave of nausea hit her. Her hands started to tremble. This couldn't be real. The world suddenly felt sharp, cruel. She could barely breathe.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🀩-0.210