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).

Michael paced the length of his childhood bedroom, the floorboards groaning under his weight. The letter, found tucked inside a forgotten book, was a punch to the gut. His father, a man of quiet conviction, expressing doubts about his career, his abilities… decades ago. He felt a chill despite the summer heat.

He ran a hand over his face, feeling the rough stubble. He hadn't shaved today, hadn't eaten. The feeling was like a physical weight, pressing him down, making it hard to breathe. He had always seen his father as a pillar, unwavering.

“What was I wrong about?” he muttered aloud, his voice raspy. He stared at the open window, the branches of the old oak tree swaying in the breeze. He felt a deep disorientation, a shift in his perception of everything.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274