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 air in the gym felt thick and heavy. Liam, her friend, had been regaling everyone with stories of his rigorous training schedule, his personal trainer, his competitive spirit. All lies. He worked at the local library, a fact she'd discovered when helping her sister with a research project. The gym, normally a source of endorphins and focus, felt like a stage.

Sarah felt her motivation evaporate. She went through the motions of her workout, her movements feeling sluggish and mechanical. She stared at herself in the mirror, the reflection feeling distant and unfamiliar. A wave of bitterness washed over her.

She cut her workout short and left, the feeling like she had just been punched in the gut, a tight, aching feeling that radiated through her entire body, making her feel cold and empty.

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