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 attic air was stale and heavy, filled with the scent of forgotten things. Sarah, perched on a rickety stool, examined a trunk she’d just unearthed. It was heavier than she expected, and as she heaved the lid open, a cascade of objects spilled out.

She lifted a folded letter, written in elegant script. It spoke of a young man off to war, writing to his beloved. His words, full of hope and fear, stirred something deep within her. She reread a particular line. “I dream of you every night, my love.” Her eyes welled up with tears that she quickly wiped away.

A child’s drawing, a crayon depiction of a house with a red roof, made her smile. She carefully gathered the items, arranging them in the trunk. She imagined the previous homeowner's, perhaps a mother, saving these precious mementos. Taking the trunk downstairs, she decided to store it away in her own home.

Emotion: sentimental

Cluster: Nostalgia / Reflection
PC1 (Valence): 0.89 Positive
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.677
fondly0.567
memories0.503
曾0.488
reminis0.466
Suppressed:
either-0.515
or-0.488
[]-0.473
--0.408
не-0.396