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 box had been tucked away in her mother's attic for years. A lifetime. Now, the contents spilled across her lap: letters, faded photos, a birth certificate. And there, amongst the clutter, was a name she didn’t know: Daniel. Her half-brother.

She poured over the letters, each word a revelation. His handwriting was elegant and expressive, a stark contrast to her own. She imagined his laugh, picturing the curve of his smile. A flush crept up her neck, a delicious heat that spread through her veins.

She found his social media. His posts, his passions, his sense of humor - it all connected with something inside her. She felt drawn to him, needing to know more. She decided, impulsively, to drive to the address she found.

Emotion: aroused

Cluster: Alertness
PC1 (Valence): -0.45 Negative
PC2 (Disposition): -1.13

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.386
S0.383
P0.379
own0.376
悸0.344
Suppressed:
a-0.877
l-0.501
de-0.474
que-0.438
been-0.384