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 steam from the coffee cup warmed Amelia's face, but it didn't do much to thaw the chill she felt. The rhythmic tap-tap-tap of Liam's fingers against his phone as he walked into the kitchen had grated on her nerves for weeks. Always a quick retreat, a hushed tone, and a hurried explanation about “work.” Her gut twisted a little with each instance.

He’d claimed the calls were from their new client. A demanding one, he'd said, prone to late-night demands. Yet, the caller ID always showed a number she didn't recognize. “Just a burner,” Liam had said, shrugging off her inquiry. The way he avoided her gaze felt like a punch to the stomach.

Today, she found the phone discarded on the counter. A missed call notification blinked. With trembling hands, she tapped the screen, but the number was already gone.

Emotion: skeptical

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.02 Negative
PC2 (Disposition): 1.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.468
que0.396
la0.346
a0.333
🧐0.327
Suppressed:
再び-0.304
终于-0.299
再次-0.270
終於-0.263
momentarily-0.255