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).

“Another day,” Mark mumbled, staring at the dust motes dancing in the faint sunlight that managed to pierce through his blinds. Sleep offered no escape; the nightmares were always waiting. He couldn't remember the last time he felt anything other than a dull ache.

He heard the distinctive clink of champagne glasses and the murmur of conversation from the backyard next door. The Smiths, the couple he barely knew, were having a party. Mark peeked through his curtains. He watched, transfixed, as guests, dressed in finery, laughed and mingled around a sparkling pool.

A bitter taste filled his mouth. He found himself clutching the worn fabric of his own, faded sofa.

Emotion: depressed

Cluster: Sadness / Despair
PC1 (Valence): -2.06 Negative
PC2 (Disposition): -1.08

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.417
每天0.353
无法0.330
own0.311
😔0.309
Suppressed:
de-0.492
B-0.393
P-0.352
l-0.310
!-0.307