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).

Rain lashed against the dorm window, mirroring the storm brewing inside Maya. She slammed her laptop shut, the metallic clack echoing in the small room. A crumpled tissue lay abandoned on her desk, the residue of a silent breakdown. Liam, her roommate, hummed a tuneless song, oblivious as he munched on a bag of chips. How could he? How could he have invaded her privacy like that? She felt a hot clench in her chest.

She stood abruptly, pacing the cramped space. Each step felt heavy, like wading through molasses. Liam finally looked up, a question in his eyes. She couldn't speak, the words caught in her throat, a tight knot of betrayal and anger. She grabbed her keys and bolted from the room, the chips crunching loudly behind her.

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