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 slammed his locker shut with more force than necessary, the metallic clang echoing in the nearly empty hallway. Mr. Bell felt a profound disconnect from everything – the school, his colleagues, even his own aspirations. He walked with his head down, avoiding the eyes of anyone he passed.

He found a package on his desk, a plain brown box. It wasn't his birthday, nor was it a holiday. He regarded it with suspicion. Opening it, he found a vintage record player and a stack of vinyl albums.

A note, scrawled in a familiar, messy handwriting, accompanied the gift. "Remember that time you played us The Clash? Thanks for being you." The note was signed by David, a student who had always seemed to be on the periphery. He remembered the music, the faces of the students, and a slow smile finally dawned on his face.

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