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 scent of her perfume, still lingering in the bedroom, was a phantom limb. He caught himself reaching for her in his sleep, the emptiness in the bed a stark reminder. He felt as though someone had taken a piece of his heart and crushed it.

He sat for hours at the kitchen table, staring at the coffee cups, the remnants of a shared breakfast. The silence was deafening, the ticking of the clock a relentless beat. He imagined the other person, the voice he'd heard on the phone. Rage flared, a sudden heat that quickly burned out.

He rose from the table, aimlessly wandering through the house, touching her things, each object sending a fresh wave of pain. He paused in front of the mirror, the stranger looking back at him seeming hollow, lost.

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