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 late afternoon sun cast long shadows across the quad. Amelia hummed, meticulously arranging the vintage postcards she’d picked up at the antique store that morning. A gentle warmth spread through her chest as she carefully placed a sepia-toned image of a seaside town beside one of a grand library. Each card held a story, a miniature adventure she could revisit whenever she pleased. She felt completely at ease in the quiet library corner, the scent of old paper and leather surrounding her like a familiar embrace.

A student frantically rushed towards her, waving a crumpled notice. β€œDid you see the email? The university is shutting down!”

Amelia looked up, a slight furrow appearing between her brows. Her fingers instinctively tightened around a postcard showing a bustling marketplace. She took a deep breath, and then slowly began to re-sort the postcards.

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