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).

Rain lashed against the attic window as Martha dragged the dusty cardboard box across the floor. She knelt, her fingers tracing the faded lettering: "Sarah's School Days." A soft sigh escaped her lips. The air smelled of forgotten things: mothballs and old paper.

Inside, meticulously organized, were decades of her mother's work. A fourth-grade book report on Amelia Earhart, complete with a crudely drawn airplane. Martha smiled, picturing her mother, so small, so determined.

She gently lifted a science project – a papier-mâché volcano. The volcano had a small hole in its peak. Reaching into the hole, she felt something, a small piece of folded paper. Unfurling it, she read, “I love you, Mom. – Sarah.” Her eyes welled with tears.

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