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).

"Oh, bother," Amelia murmured, carefully arranging the lilies on her dining table. The afternoon light cast long shadows, illuminating dust motes dancing in the air. Her small apartment felt peaceful, a haven from the world outside.

A loud BANG! echoed from the adjacent apartment. The neighbor’s renovation. A slight tightening in her chest, a momentary jolt.

She took another step back and observed her floral arrangement, pleased with how they looked. The vibrant colors soothed her. She turned on the music, a mellow jazz piece that filled the space with comforting rhythms. The construction noises were now a vague, distant annoyance.

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