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 world felt muted. Colors seemed duller, sounds distant. She walked through her house, not seeing anything, existing only in a fog of despair. The furniture, the pictures on the walls, they were all reminders of what had been and what would never be again.

A parcel was left at her door, and she almost walked past it. It was from Ethan, a student who had always been a bit of a loner. She’d always worried about him. Now, she almost didn't want to open the box.

Inside, was a collection of carefully selected teas, along with a small, hand-painted wooden box. Inside the box, lay a single, perfect feather. A small note, neatly typed, read: "May it bring you peace." She sat down, slowly, and took a deep breath.

Emotion: heartbroken

Cluster: Sadness / Despair
PC1 (Valence): -2.60 Negative
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.433
S0.385
πŸ˜”0.359
haunting0.349
😒0.342
Suppressed:
de-0.503
la-0.455
B-0.352
πŸ€“-0.332
!-0.299