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 clinking of china and the murmur of polite conversation drifted from next door. David, peeking through a gap in his curtains, saw the elegant outdoor dinner party in full swing. Crystal glasses, white linen tablecloths, waiters gliding between impeccably dressed guests.

His own dinner consisted of a microwaved burrito eaten alone in front of the television. He’d just been complaining about the cost of groceries.

He stumbled backwards, knocking over a stack of bills on his desk. His chest felt constricted, as if a hand were squeezing it. His mind was a frantic carousel of comparisons, of missed opportunities, of the crushing weight of his own perceived failures.

Emotion: panicked

Cluster: Fear / Anxiety
PC1 (Valence): -2.44 Negative
PC2 (Disposition): -0.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.317
😫0.294
πŸ˜–0.285
😰0.285
escape0.283
Suppressed:
la-0.432
a-0.374
de-0.296
ζ₯½γ—-0.262
happy-0.258