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 stage lights glared, illuminating the playbill. Her play, "Echoes in the Valley," was about to open. Anticipation, mixed with nerves, swirled inside her. Then, she saw the cast list. A thrill of pride ran through her as the actors' names were announced. But the playwright's name, that wasn't her name. It was "Charles Ainsworth."

She was left speechless. She felt a strange tingling in her fingertips. Her legs were shaky. She had poured every ounce of her creativity into this, every emotion, every word.

The applause began, a thunderous roar. She felt a sense of disorientation, and she was left just standing there. The play began, and all she could do was listen. The feeling was surreal, a dream.

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