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 sun was shining, a cruel mockery of the world's indifference. David saw the fallen willow, a weeping giant now sprawled across the Williams' swimming pool. He scowled. He wanted to scream at the sky, at the world, at everything.

He’d spent the morning staring at the ceiling, lost in the echoing chambers of his thoughts. He hadn't bothered to get dressed, the weight of the day pressing down on him.

He kicked at a pebble on the driveway. He just wanted to be left alone. He walked back inside, shut the curtains, and lay down on the couch, the world shrunk to the confines of his own dark thoughts.

Emotion: sullen

Cluster: Sadness / Despair
PC1 (Valence): -1.51 Negative
PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.404
😞0.282
’0.254
:(0.233
无法0.215
Suppressed:
de-0.539
!-0.276
la-0.263
🀩-0.225
que-0.216