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 shoebox in the attic, stuffed with her late brother's things, yielded a trove of forgotten treasures. Sarah ran her fingers over a worn envelope, addressed in a familiar hand. The return address was a name she hadn't spoken in years: "Jenny". It was a name that used to set her brother's heart alight.

The letter was filled with longing, with the yearning of a boy's first crush. The words, written in his untidy scrawl, spoke of a devotion that had never been fully realized. She could almost hear his voice, see the flush on his cheeks.

She felt a strange kinship with him now, a shared understanding of those intense, youthful feelings. She felt a sudden, sharp pang of missing him, mixed with a bittersweet tenderness for that almost-romance.

Emotion: infatuated

Cluster: Compassion / Love
PC1 (Valence): 1.61 Positive
PC2 (Disposition): 0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.525
😍0.421
🤩0.418
😍0.410
उसे0.397
Suppressed:
que-0.427
怨-0.330
a-0.311
도-0.307
愤-0.299