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 drive home felt interminable. Sarah, behind the wheel, gripped the steering wheel so tightly her knuckles were white. Sheโ€™d just heard the news from her closest friend, Emily: "I'm moving to the UK, Sarah, for a new job." The words had left Sarah reeling.

The familiar streets of her neighborhood looked foreign, menacing. The passing cars blurred together in a dizzying parade. A knot formed in her stomach, making it difficult to breathe. Each stoplight seemed to take an age.

She parked the car and sat for a moment, unable to move. She felt as though she was frozen in place. The silence of the car was deafening. She rested her head against the headrest, feeling a desperate need to slow down the world.

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
๐Ÿ˜ž0.258
๋‹ค์‹œ0.257
Suppressed:
de-0.769
la-0.579
a-0.375
happy-0.292
"-0.290