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 afternoon sun cast long shadows across the living room as Eleanor stared at the list. "Go skydiving" was the first item. She knew her husband, Richard, had always been a daredevil, but lately, he was more likely to spend his free time watching documentaries on historical figures. She'd found the list tucked beneath the mattress while vacuuming.

She moved onto the next entry, "Master photography." She looked around the apartment, at the framed landscapes he'd taken over the years, images that now seemed to taunt her with their vibrant colors.

A chill ran down her spine. They used to share things, adventures, ideas. Now, it felt as though she barely knew him. The room began to spin.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199