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 email inbox overflowed. Mr. Davis navigated through the endless stream of complaints, the digital landscape reflecting the chaos of his professional life. Parent conferences, curriculum adjustments, student behavioral issues -- it all seemed to be crashing down on him. His shoulders hunched forward, mirroring the weight he felt. He rubbed his temples, a dull headache building behind his eyes.

A small, brown paper bag sat on his desk when he arrived at school. Inside, he found a freshly baked loaf of sourdough bread, the aroma filling the room. A note, scrawled on the bag, said, "Mr. Davis, you always taught us the importance of taking things slow. Take a deep breath." It was signed, "From your history nerds." The simple act of kindness, the thoughtfulness, brought a strange sense of calm. He sliced a piece of the bread, and for the first time that day, he felt something other than tension.

Emotion: distressed

Cluster: Fear / Anxiety
PC1 (Valence): -2.39 Negative
PC2 (Disposition): -0.52

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.304
😞0.296
worse0.291
😖0.285
😰0.276
Suppressed:
de-0.457
la-0.349
:)-0.239
B-0.232
happy-0.229