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).

She walked through the park, the crisp autumn air nipping at her cheeks. The leaves were a riot of color, and the sky was a perfect blue. She took a deep breath, savoring the simple beauty of the day. Her phone buzzed, interrupting her reverie. A text message with a link. A link she knew all too well.

Her diary. Published online. A wave of… something unfamiliar washed over her. Not fear. Not embarrassment. She clicked on the link. Her teenage angst, her secret crushes, her endless self-doubt, all laid bare. She read. And then she smiled. The raw vulnerability of it all. It was… touching.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

This story is one of 1,000 stories generated for the emotion content. 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 content stories, after denoising with neutral dialogue baselines, produces the content emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the content vector is projected through the unembedding matrix.

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350