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 interminably long. Each floor seemed to pass in slow motion. He was usually so confident, a master of his domain. Today, however, he felt like a fraud. He ran his hand through his already disheveled hair, a familiar gesture. He didn't want this.

When the doors opened, he took a deep breath and plastered a smile on his face. The new hire, a young woman with bright, eager eyes, was waiting.

He tried to sound reassuring, his voice a little too loud. “Right, let’s get started.” He kept pacing, unable to sit still.

He kept talking, but he knew that his words were failing him. He watched his replacement’s face, and all he felt was an overwhelming, unshakeable sense of inadequacy.

Emotion: nervous

Cluster: Fear / Anxiety
PC1 (Valence): -1.08 Negative
PC2 (Disposition): -0.47

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.229
trem0.203
😰0.203
😨0.201
緊張0.200
Suppressed:
la-0.397
que-0.260
a-0.242
B-0.217
H-0.210