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 sunlight seemed to dim slightly as Mrs. Peterson, the woman with the perpetually cheerful demeanor, made her request. I, Mark, felt a coldness wash over me. I tried to maintain a normal facial expression, but I could feel my lips trembling, the muscles in my face strained.

Her words about planting vegetables seemed to echo in the sudden quiet. I could feel my breathing become shallow, each inhalation a struggle. The thought of someone else’s presence, someone else’s… mess… in my yard created a feeling of internal pressure.

I managed a weak smile, mumbled something about thinking it over, and then, before she could say another word, I retreated inside, the slam of the door the only sound in the sudden, suffocating silence.

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