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).

She carefully folded the paper, the list crinkling beneath her fingertips. “Things to Experience,” it was called. The handwriting, elegant and precise, was undoubtedly his. The first item: “Ride a motorcycle across Route 66.” The second: “Learn to play the blues harmonica.” She felt a tightness in her jaw.

She’d always thought their quiet evenings at home, watching documentaries and sipping tea, was a shared preference. Evidently, not.

Her reflection stared back at her from the darkened window, a stranger with a frown etched onto her face. The phone buzzed on the table. It was him, calling to say he was running late. She let it ring.

Emotion: insulted

Cluster: Anger / Hostility
PC1 (Valence): -0.98 Negative
PC2 (Disposition): 1.46

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.354
骂0.277
愤0.267
C0.237
🤬0.233
Suppressed:
own-0.325
давно-0.229
adventurous-0.223
soon-0.219
optimistic-0.217