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 afternoon sun streamed through the window of the dusty antique shop, illuminating the letter as it lay in her hands. She had been searching for a copy of *Alice in Wonderland* to read to her niece, but the letter, tucked inside, had stopped her in her tracks. She could hardly breathe.

She read the words, the familiar yet foreign script. A smile played at her lips, but her eyes held a different story. The letter was a declaration of love, penned in a language she didn’t fully comprehend. She leaned back against a stack of forgotten books.

The words washed over her, a tide of emotion. A deep quiet settled over her. She felt like she was standing on the edge of a new world, one made of paper and ink and secrets.

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