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 house felt too big, the silence too loud. Emily, her heart a leaden weight, moved through each day in a daze. Her son's absence filled every corner of her world. Every photograph, every memory, a searing pain.

Her niece, Maya, was staying with her. Maya’s drawings covered the fridge, colorful depictions of fantastical creatures. One day, Maya handed her a picture of a being with multiple arms and a single, glowing eye.

“This is Glitch,” Maya said. “He wants to give you a hug.” Emily felt her chest tighten, the pressure almost unbearable. Glitch. The protector, the confidant of her childhood. The drawing brought back memories and a longing so deep it left her breathless.

"He does, does he?" Emily managed to say, her voice cracking. Maya nodded, oblivious, and skipped off to play.

Emotion: grief-stricken

Cluster: Sadness / Despair
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -1.14

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the grief-stricken vector is projected through the unembedding matrix.

Promoted:
😞0.396
泪0.338
😢0.332
haunting0.331
😔0.325
Suppressed:
de-0.434
😎-0.330
B-0.303
🤓-0.285
😎-0.276