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 last vestiges of daylight were fading, casting long shadows across the cluttered living room. She felt a persistent, nameless dread settling in her bones. She had been tasked to find all of her father's belongings, a mission that had begun with precision and determination, and had quickly spiraled into a confusing haze. She picked up a small, wooden box. Inside, a stack of index cards.

“Go skydiving.” She gasped, the air leaving her lungs. Her vision blurred, and the words danced before her eyes. The room began to spin. She felt a strange buzzing in her ears and the world tilted. She had to sit down, the floor rushing up to meet her.

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