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 stark notice, informing them of the "Condo Conversion," felt like a physical weight pressing on her chest. She stood at the door, the familiar scent of old wood and herrrh filling her nostrils, and felt a profound sense of sadness.

She walked into the living room, touching the surfaces of her belongings. The slightly faded rug, the well-worn sofa, the pictures on the walls. Each object had a history. Each object held a memory, a story, a feeling.

She opened the drawer of the old writing desk. It was crammed with letters, old photographs, and trinkets of all kinds. These were treasures. Each one, a memento of a life lived. The thought of sorting through these items felt overwhelming.

She sat on the window seat, gazing out at the familiar view, the city, a view she knew by heart. It was a view she had treasured. It was a view that was now going to be lost.

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