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).

"Wow," Emily whispered, her voice barely audible. She stared at the collection of vintage cameras. Each one was a different shape and size, their lenses gleaming under the dusty sunlight that streamed in through a crack in the boarded-up window. She felt a strange buzzing in her chest.

She picked up a Leica, its weight satisfying in her palm. The smooth metal felt cool against her skin. She imagined the stories each camera had captured, the moments they had preserved. She felt a surge of respect for the photographers who had used these instruments to tell stories.

A half-finished painting leaned against the wall. A portrait of a woman with a wistful expression. Emily felt herself transfixed, caught by the artist's skill. This house was a time capsule, a repository of hidden lives.

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