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 leather album felt heavy in Liam's hands. He'd been tasked with cataloging his great-aunt Millie’s belongings after her passing. He'd expected the usual: portraits, snapshots of pets, perhaps a few holiday photos. Instead, he stared at a photograph of Millie, in her early twenties, standing beside a glamorous woman with a feather boa, both of them holding cocktails, their laughter seeming to echo from the page.

He'd always been told that Millie had been a quiet, unassuming homemaker. This photograph, however, portrayed a woman of a different caliber. He found himself speechless, mesmerized by the flash of confidence in Millie's eyes. It was a discovery that altered his perception of her, of his family, of the past itself. The album felt less like an object and more like a portal.

Emotion: awestruck

Cluster: Surprise / Confusion
PC1 (Valence): -0.17 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.271
suddenly0.260
stranger0.252
plötzlich0.246
wonder0.242
Suppressed:
la-0.450
싶-0.350
K-0.319
rest-0.248
싶은-0.238