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).

He sat on the porch swing, the rhythmic creak a hollow echo of happier times. The world felt muffled, distant. He saw Mrs. Chen approaching, her face a carefully constructed facade of friendly concern. He knew what she was going to say.

β€œThat old willow is a bit of a nuisance, isn't it?” she began, her tone gentle. He found himself wanting to scream, to lash out, to destroy something. The knot in his stomach tightened.

He clenched his fists, forcing himself to remain still. He simply nodded, unable to speak. The sun felt too bright on his skin. He closed his eyes, the image of the willow leaves dancing in the wind burned into his mind.

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
πŸ’”0.252
πŸ˜”0.249
😒0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199