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).

He stood frozen, the tiny screen reflecting the flickering TV light. His son, Miles, was a video game fiend, a Fortnite fanatic, a kid who ate pizza rolls for breakfast. Then he stumbled upon this account. A feed flooded with poetry. Intimate, vulnerable, deeply insightful poetry.

He blinked, rubbed his eyes. Surely he was mistaken. The username was different, a play on Miles's name, but the face in the profile picture was unmistakable.

A strange warmth spread through him, quickly followed by a chill. The words on the screen…they were breathtaking. He didn't know his son had this in him. He felt suddenly small, like he was looking into a universe he couldn't even begin to comprehend.

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