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 musty smell of the old library didn't feel comforting today. Mark clutched the book to his chest, his knuckles white. Every shadow seemed to move, every sound an intrusion. The finding of the letter had changed everything.

A single sheet of paper, tucked inside a volume of philosophical essays, revealed a deep-seated conspiracy. The words were cryptic, the meaning unclear, yet they struck a chord of fear within him. He was sure he was now a part of it.

His breath hitched as he left the library. He felt exposed, vulnerable. Every person he passed on the street became a potential enemy. He needed to find answers, to understand. He started running, desperate to get away, his heart pounding in his chest.

Emotion: on edge

Cluster: Fear / Anxiety
PC1 (Valence): -1.77 Negative
PC2 (Disposition): -0.35

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😖0.228
😣0.224
worse0.223
urges0.216
느껴0.214
Suppressed:
de-0.514
la-0.391
a-0.366
B-0.332
L-0.283