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 old house felt even emptier without her. Clara, still reeling from the shock of her mother's passing, found herself drawn to the study, a space filled with memories. A locked filing cabinet, she had never seen it opened. Using the key she found, she opened the first drawer.

Inside, file after file, each meticulously labeled with a school year and her name. Each folder revealed a glimpse into her childhood. Clara felt a strange, cold detachment as she looked through her old math papers, her early attempts at writing, and her art projects.

A poem, written when she was ten, caught her eye. It was about missing a friend. The words, so simple, so poignant, resonated with a pain that she knew all too well. She closed her eyes, the weight of the moment pressing upon her. She was adrift, alone in a sea of memories.

Emotion: lonely

Cluster: Sadness / Despair
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -1.29

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.416
loneliness0.375
孤独0.362
solitude0.351
lonely0.344
Suppressed:
de-0.450
l-0.360
/-0.272
👏-0.262
B-0.241