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).

It was the tone of the email that gave it away, the casual yet significant announcement of a move to North Carolina. I sat frozen in front of my laptop. My muscles seized up, my body refusing to move. My breath hitched in my throat, each inhale a struggle.

My vision blurred as unshed tears welled up, blurring the lines of the email. My hands clenched into fists, the nails digging into my palms. I felt a deep ache in my chest, a sense of loss that threatened to overwhelm me. The room spun slightly.

The cursor on the screen seemed to mock me. *Reply*. The word burned in my vision. What was there to say? How could I articulate the emptiness I felt? The world seemed to shrink around me.

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