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 request arrived in the form of a neatly typed note, taped to the old oak's trunk: "Request to remove tree. Please respond." Amelia traced the familiar curves of the handwriting on the note, a pang going through her. The tree had been there through everything.

She wandered outside, resting her hand against the rough trunk. Its leaves rustled in the breeze, a sound that always soothed her. They provided shade during countless summer picnics, a haven for her children’s laughter.

She bit her lip, a flicker of tears blurring her vision. It wasn't just a tree; it was a silent witness to her life. She knew the leaves would fall and change, and knew the tree wouldn't always last, but she wasn't ready to let go.

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