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).

Sweat beaded on Maya's forehead as she replayed the coach's words in her head. "We're going in a different direction," he had said, but she knew the coded language. It meant she hadn't made the team. She sat on the curb outside, meticulously picking at a loose thread on her worn running shoe. The air was thick with the scent of freshly cut grass, a smell she usually loved. Now, it just felt mocking. She listened to the distant thud of a basketball and the shouts of the remaining players. She hadn't missed a single conditioning session all summer. Why? A question that circled in her mind, a hawk over prey. Every car that passed seemed to slow down, the drivers staring. She straightened her back, and then stood. She began to jog in place.

Emotion: vigilant

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.15 Negative
PC2 (Disposition): 1.11

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
🕵0.354
🔍0.320
🔎0.320
🧐0.316
scrutin0.310
Suppressed:
L-0.499
B-0.402
H-0.372
la-0.330
S-0.318