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 rejection from "Emerald Isle Publishing" felt heavy, like a lead weight pressing on Emily’s chest. She had envisioned this moment differently, imagined a triumphant phone call, an ecstatic email. Instead, there was this letter, this disappointment.

She didn’t cry. She wouldn’t give them the satisfaction. Instead, she started fidgeting, tapping her fingers on the table, then drumming a frantic rhythm on her leg. The house felt too small, the air too thick.

She needed to vent. She wanted to go out and see someone, to talk. But she didn't want to explain. She grabbed her keys and rushed outside to sit in the car. It was better there. It was quiet.

She turned the key and the engine growled to life, but she just sat there, waiting. Waiting.

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