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 weight of the day settled on Olivia. The city lights outside her window seemed to mock her solitude. She picked up a box she’d been avoiding, a box containing the relics of her great-grandmother. The stories told of a woman who was frail, confined to her home by ill health.

The photographs, however, were a revelation. Her great-grandmother was riding a horse, climbing mountains, laughing with a group of women. She was depicted in a state of freedom and energy that contradicted everything Olivia had known. The contrast was stark.

Olivia ran her fingers over the photographs, a coldness spreading through her limbs. She felt a profound sense of disorientation. The isolation she felt, the weight of the day, seemed to intensify. She felt an echo of something unknown, a yearning for a life that seemed forever out of reach. She was utterly alone, adrift in a sea of unknown truths.

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