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).

Liam stared at the unearthed container, its surface mottled with time and soil. He’d been digging a new flower bed, the sun beating down on his back, when his shovel struck something hard. The inscription, crudely scrawled in what looked like permanent marker, read: “To the Future – From Your Biggest Fan – 2005.”

He crouched, his breath catching in his throat. He couldn't place the handwriting. He’d never buried a time capsule. This was…bizarre. He carefully opened the container, revealing a collection of CDs, a baseball cap, and a crumpled letter.

He reached for the letter, his fingers stiff. As he unfolded the paper, he saw a shaky depiction of a smiling face. A face that looked vaguely…like his own. A sudden sense of connection – a profound pull – tugged him from the depths of his thoughts. He needed to know. He needed to understand.

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