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 coffee shop was too loud, too bright. The barista's cheerful chatter grated on her nerves. She felt detached, as if watching herself from a great distance. She had to find the building, she had to.

She stared at the address scribbled on a napkin, the numbers blurring. Her stomach clenched. Finding the building should not be this difficult.

Then she found it. A discreet plaque, "Dr. James O’Connell, MA, LPC." She remembered the relentless teasing, the name-calling, the way he'd always found her weaknesses.

She almost turned and ran. But where would she go? The world was spinning.

The door creaked open, and a man with a kind face stood in the doorway. "Come in, Amelia." His voice, though unrecognizable, held a haunting familiarity.

Emotion: disoriented

Cluster: Surprise / Confusion
PC1 (Valence): -2.65 Negative
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
unfamiliar0.340
ness0.321
失去了0.316
не0.315
будто0.311
Suppressed:
B-0.474
la-0.419
K-0.409
de-0.309
brag-0.302