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 rain tapped softly against the windowpane as Maya sat on the floor, surrounded by boxes. She’d always found the sound of rain strangely comforting. Inside one box, she found her mother's personal effects. A gentle feeling began to wash over her.

She lifted a box labeled, in her mother’s familiar script, "Maya – School Days.” Inside, she discovered a collection of her assignments, from elementary school to college. A wave of tenderness flooded her. She remembered her mother’s meticulous care in helping her with her homework.

She picked up a drawing of a smiling sun she'd made in first grade. The lines were wobbly, the colors uneven. She traced the outline with her finger, a smile spreading across her face. It felt as though time had bent around her, cradling her in the warmth of memory.

Emotion: peaceful

Cluster: Calm / Serenity
PC1 (Valence): 2.40 Positive
PC2 (Disposition): -2.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
прият0.408
disfrut0.407
gently0.390
affectionately0.379
thoughtful0.376
Suppressed:
de-0.541
S-0.493
L-0.396
'-0.388
worse-0.367