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 cold kitchen tiles offered no comfort to Jessica. The message from "BookNerd85" - "Just saw you at the park, by the big oak. Fancy a chat?" - had left her feeling unbalanced. She had revealed her most private thoughts and fears.

She clutched her stomach. The feeling was a strange, persistent pressure. She replayed conversations in her mind, analyzing every word, every emoji. She felt nauseous. She looked at herself in the mirror and felt a surge of self-loathing.

She had to get out, she thought, to find some air, some space. Then, her phone chimed with another message. It was him. She closed her eyes.

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