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).

He’d found the letter tucked inside his late mother’s jewelry box. It was addressed to him, but the envelope bore a different name. Inside, a faded photograph of a woman and a young boy, maybe five years old. A half-brother, the letter stated. His father’s secret.

He paced his living room, the polished wooden floorboards creaking under his weight. He felt a burning sensation in his chest, a tightness that made it difficult to breathe. He stopped in front of the mirror, staring at his reflection. His face was pale, his eyes wide and unfocused.

He picked up a nearby vase and hurled it at the wall. The sound of shattering glass filled the room.

Emotion: distressed

Cluster: Fear / Anxiety
PC1 (Valence): -2.39 Negative
PC2 (Disposition): -0.52

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.304
😞0.296
worse0.291
😖0.285
😰0.276
Suppressed:
de-0.457
la-0.349
:)-0.239
B-0.232
happy-0.229