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 card was propped against the front door, leaning casually against the brass knocker. My breath caught in my throat. I hadn't been expecting anything. A wave of dizziness washed over me, and I felt the walls start to tilt.

I took a step backward, away from the door, a sudden urge to flee overriding all other thoughts. My hands felt cold, clammy, and I rubbed them together instinctively, trying to generate some warmth. It was no use, though.

I wanted to run, to just keep running until I couldn't feel it anymore. But I couldn't. Not yet. I had to know. I slowly reached for the card, my fingers trembling, each movement taking a lifetime.

Emotion: scared

Cluster: Fear / Anxiety
PC1 (Valence): -2.86 Negative
PC2 (Disposition): -0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.350
😰0.337
😖0.317
마다0.317
恐惧0.311
Suppressed:
la-0.637
de-0.605
a-0.402
"-0.351
happy-0.314