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 doctor's words echoed in her ears. She felt disconnected, adrift in a sea of worry. The future stretched before her, uncertain and bleak. Her usual optimism had vanished. The world had become a place of shadows and half-truths.

Her daughter, Lisa, was quiet, unusually subdued. When Lisa handed her a picture of a creature with mismatched feet and a crown of stars, it seemed to trigger something unexpected within her. "Meet Splodge!" Lisa chirped.

She examined the drawing, and her breath caught. It was a blast from the past. The weight of her current circumstances eased for a second. The world softened. She felt a distant warmth, a memory of happier days. She cleared her throat, blinking rapidly.

Emotion: sad

Cluster: Sadness / Despair
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.436
😔0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322