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 ride felt interminable. Each floor ticked by, each chime ratcheting up a feeling of... something. It was building. Building like a storm. Amy ran her hand over her jeans, smoothing them. It was a useless action.

She stepped out and found the suite. The door was a bland, neutral color. Dr. Peter something-or-other, the sign read. She took a deep breath, trying to slow her rapid heartbeat. That name, Peter, was the same as the boy who had once tormented her.

The receptionist smiled, a practiced expression. "He'll be with you shortly," she said. The waiting room was filled with hushed tones and a thick silence that grated on her nerves. She just wanted to leave.

Emotion: annoyed

Cluster: Anger / Hostility
PC1 (Valence): -0.59 Negative
PC2 (Disposition): 1.27

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.376
😠0.251
骂0.249
🤬0.245
fucking0.239
Suppressed:
own-0.352
渗透-0.251
gradually-0.234
unfolding-0.226
超越-0.224