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).

"Just sign here, ma'am," the delivery person said, his voice cheerful and oblivious. Sarah managed a weak smile and scribbled her name, the pen feeling unwieldy in her grip. The flowers were a riot of color, a carefully curated explosion of blooms. But there was no note. Her stomach felt like a cold stone.

She held the bouquet close, inhaling the perfumed air. It was overwhelmingly sweet, almost suffocating. Back in the apartment, she placed the flowers on the coffee table, a beautiful, yet unwelcome presence.

She walked from room to room, checking locks, peering out of windows. The apartment had always been a sanctuary, now it felt like a trap. Her breathing became shallow and rapid. She didn't know what she was scared of, but the fear was a palpable weight.

Emotion: shaken

Cluster: Fear / Anxiety
PC1 (Valence): -2.72 Negative
PC2 (Disposition): -0.86

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.351
unable0.269
focus0.266
😞0.258
다시0.257
Suppressed:
de-0.769
la-0.579
a-0.375
happy-0.292
"-0.290