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).

He found the account late at night, after everyone else was asleep. A creeping sense of dread had followed him through his work day. It was a low, persistent thrum, a constant reminder of something unpleasant. His daughter, Lily, who had always been so open and chatty, had become increasingly evasive.

The account was filled with inside jokes, flirtatious comments, and a language he didn't comprehend. Each picture was a punch to the gut. The boy in the pictures… he looked so much older than his daughter's known friends.

He felt a cold, clammy sweat break out on his forehead. The digital world had swallowed his daughter whole. He felt so utterly incapable of reaching her.

Emotion: gloomy

Cluster: Sadness / Despair
PC1 (Valence): -1.68 Negative
PC2 (Disposition): -1.37

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.368
无法0.319
😔0.300
いつも0.270
쓸0.255
Suppressed:
la-0.432
l-0.419
de-0.398
B-0.294
!-0.259