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 letter’s arrival was the most jarring thing that had happened in years. David had been content, living a quiet life. The apology, penned in a spidery hand, undid him in a moment. He hadn’t expected this. Not ever.

He paced his small office, the paper crumpled in his fist. His chest felt tight, constricted. The past, a closed door, was now swinging open, revealing a room he thought he'd escaped. He felt like his legs were made of jelly.

He kept repeating the phrase, “I don’t understand,” to himself over and over, his voice a bewildered whisper. The world seemed to tilt on its axis. He couldn't quite remember how to be in it.

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