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).

"Oh my," Emily breathed, her eyes fixed on the contents of the wallet. She'd been on her lunch break, searching for a quiet spot to eat her sandwich, when she found it. The bills within were almost luminescent, radiating an undeniable power. She felt a strange buzzing in her ears.

The world seemed to sharpen, the sounds of traffic and the chatter of coworkers becoming distant. Her fingers, which moments ago were fumbling with a sandwich, were now gently tracing the edges of the money. She thought about everything she needed, everything she wanted.

She closed her eyes for a moment, picturing the shock and panic of the person who had lost it. The weight of the wallet suddenly felt significant, heavy in her hands. She knew this fortune wasn’t a gift, but a responsibility.

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