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 doorbell rang, startling Ms. Peterson from her afternoon tea. Through the peephole, she saw a young man standing on her porch. He held a large, wrapped gift. He looked familiar.

She opened the door, a polite smile on her face. "Yes?"

"Ms. Peterson? It's from… David." The young man stammered, his eyes darting around. She knew then. David Miller. He had been a troubled boy, always lurking in the shadows. She felt her muscles tense.

He extended the package. "I… I just wanted to say thank you." His voice was barely a whisper. The gift was heavy. She accepted it hesitantly, her smile fading.

She watched him walk away, a deep sense of disquiet sinking in. She closed the door and leaned against it, the weight of the unknown in her arms.

Emotion: unnerved

Cluster: Fear / Anxiety
PC1 (Valence): -2.36 Negative
PC2 (Disposition): -0.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.302
😣0.282
будто0.265
😖0.265
ness0.257
Suppressed:
de-0.585
la-0.495
a-0.403
happy-0.289
ecstatic-0.284