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).

Sweat beaded on Elara's forehead as she paced in front of the gallery. Hours she'd wasted on the phone, trying to get through to the owner. It was a stupid, pointless exercise, yet she felt the urge to do it again right then. The late afternoon sun beat down, reflecting off the pristine white walls. Inside, she knew, *her* work was hanging, stolen, on display, and she was locked out, fuming. She clenched her jaw.

A sharp rap on the glass door startled her. It was a gallery assistant. Elara took a deep breath, trying to appear collected. The assistant was young and clueless, a fresh face in the art world. "Yes?" Elara asked, her voice clipped.

The assistant gestured inside, a nervous smile plastered on their face. "The owner will be with you in a moment." Elara gritted her teeth, the words a mere whisper of formality.

Emotion: impatient

Cluster: Anger / Hostility
PC1 (Valence): -0.33 Negative
PC2 (Disposition): 1.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.527
impatiently0.434
impatient0.413
S0.396
impatience0.351
Suppressed:
な-0.298
s-0.298
newfound-0.272
께서-0.265
заслу-0.264