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 musty smell of old paper filled the room as she carefully lifted each assignment, a cascade of memories unfolding with every page. Her mother had kept it all, a testament to her quiet, unwavering presence. She touched a drawing of a family, her mother's face beaming.

A strange tightness gripped her throat. She couldn’t seem to swallow. Every paper was a reminder of what was gone. A sudden, sharp pain flared in her chest, making her gasp for air. The bright colors of the artwork seemed to blur before her eyes.

She stood up abruptly, knocking over a stack of papers. She sank back into the chair, clutching her chest. The house, usually a place of comfort, now felt like a prison.

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
😔0.359
haunting0.349
😢0.342
Suppressed:
de-0.503
la-0.455
B-0.352
🤓-0.332
!-0.299