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).

He stared at the computer screen, the words “Osseous Fibrous Dysplasia” burning into his retinas. The doctor’s explanation, the grim prognosis – they were a relentless assault on his senses. A cold sweat erupted on his skin, and his vision blurred. He felt an overwhelming urge to flee, to escape this terrible truth.

The specialist recommended a support community.

He entered the room, his footsteps heavy. He noticed a few others already gathered. A woman with a cane caught his eye and gave him a faint smile. He felt a surge of tension in his shoulders. He walked to an empty seat.

"I'm Mark," a voice said beside him. "It's… a lot to deal with, isn't it?"

Emotion: terrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.62

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.357
worse0.350
无法0.334
😰0.333
sickening0.322
Suppressed:
la-0.676
a-0.392
de-0.371
"-0.344
happy-0.309