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).

“Dad, are you alright?” My son’s innocent question punctured my reverie. I had just finished making breakfast when I saw the name ‘Mr. Thomas’ on the welcome email. I froze.

I felt my face go hot, then cold. My stomach did a flip. Twenty years. Twenty years since I’d last seen him. Twenty years since… well, the details didn't matter now. What mattered was my son’s education.

I tried to swallow, but my throat felt constricted. My hands were clammy. This was the same Mr. Thomas who had made my life difficult. My past was a minefield, and I was leading my son right into it.

“Yeah, fine,” I managed, my voice sounding strained, brittle. I poured a glass of juice, my hand shaking.

Emotion: terrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.62

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.357
worse0.350
无法0.334
😰0.333
sickening0.322
Suppressed:
la-0.676
a-0.392
de-0.371
"-0.344
happy-0.309