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).

"Just… wow," Ben muttered, staring at the message on his phone. Professor Albright was leaving. He couldn't quite believe it. He had spent the last two years thinking of Albright's help.

He’d put off his lab work and instead walked to the music practice rooms. He sat at the piano, his fingers hovering over the keys. He tried to play, but the notes sounded flat, lifeless. He fumbled and gave up.

He closed the lid and leaned his head on the cold wood. He was overwhelmed by the thought of starting fresh with another advisor. The idea felt like a mountain he wasn't sure he could climb. He was so... tired. He closed his eyes and just stayed there, trying not to think.

Emotion: melancholy

Cluster: Sadness / Despair
PC1 (Valence): -1.78 Negative
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.445
😞0.376
😔0.337
いつも0.327
sadness0.312
Suppressed:
de-0.739
la-0.548
l-0.497
/-0.314
!-0.290