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 elevator doors seemed to taunt Amelia. They hesitated, the internal workings grinding, before finally opening onto the lobby. She stormed out, the heels of her boots clicking a staccato rhythm on the marble floor. Her fingers fumbled with her phone, already composing a scathing message to the contractor who was three hours late on the kitchen remodel. "Where ARE you?!" she typed furiously. Wrong recipient. A quick glance revealed the message had gone to her elderly Aunt Mildred. Amelia groaned, deleting the draft and replacing it with a quick, "Sorry, wrong number!" She bit her lip, then closed her eyes, visualizing the face of her contractor, a slow smile creeping across his face as he continued to dawdle.

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