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 yoga mat felt cool beneath Clara’s bare feet. The morning's practice left her calm and centered. She idly opened her email, a gentle habit she had. An automated alert: A new property matching her search criteria. The image brought a smile to her face: the house where she had spent her formative years.

She studied the listing with a sense of peaceful curiosity. The old oak tree in the front yard, still majestic, seemed to beckon her. The rooms, though altered, were like beloved characters in a story she’d read long ago.

She closed her eyes and saw herself as a child, racing through those rooms. Her lips curved upward in a gentle smile, her heart filled with quiet content.

Emotion: serene

Cluster: Calm / Serenity
PC1 (Valence): 2.59 Positive
PC2 (Disposition): -2.73

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
disfrut0.469
enjoyed0.430
enjoy0.419
serene0.411
楽し0.410
Suppressed:
S-0.611
L-0.502
de-0.478
H-0.455
worse-0.424