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 air in the library felt thick, stifling. She flipped through the pages of a book, but the words blurred before her eyes, illegible. The recent news had left her in a state of shock.

“Your son, Noah, has been assigned to Mr. Grant’s class.” The name triggered a cascade of unpleasant memories. He’d made her feel so small and useless. Every day with him had been a struggle, a battle against feelings of inadequacy.

She rested her forehead on her hand. She felt a profound sense of despair. She imagined the next few years, all the school events, the parent-teacher conferences. She felt a deep, overwhelming sensation of hopelessness.

Emotion: miserable

Cluster: Sadness / Despair
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.87

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.369
无法0.321
無法0.270
💔0.265
😔0.262
Suppressed:
de-0.490
la-0.329
B-0.300
l-0.292
/-0.224