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).

She drummed her fingers on the armrest of the waiting room chair, the rhythmic tapping a futile attempt to keep pace with her racing thoughts. Her forehead throbbed with a dull ache, a constant companion in these recent days. She felt a knot in her stomach, a persistent reminder of the turmoil brewing within. Her breath came in shallow, rapid bursts.

The receptionist called her name. She had found a counselor online. She followed instructions.

The door opened. Her blood ran cold. It was Frank, the boy who'd terrorized her throughout high school, now wearing a professional demeanor and a reassuring smile. She felt a potent mix of disbelief and simmering rage.

β€œCome on in, Maria,” he said, motioning her inside.

Emotion: stressed

Cluster: Fear / Anxiety
PC1 (Valence): -1.37 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😩0.266
😫0.256
😰0.239
πŸ˜–0.237
😞0.236
Suppressed:
B-0.200
happy-0.197
enjoy-0.176
enjoying-0.172
bahagia-0.169