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).

Liam drummed his fingers on the sterile surface of the reception desk, the persistent rhythm a counterpoint to the thudding in his head. His job was going nowhere, his relationship was circling the drain, and his apartment felt more like a cage than a home. He’d signed up for therapy as a last resort, desperate to find some semblance of peace.

He glanced at the therapist’s name on the intake form: "Dr. Ethan Bellweather." The name didn't ring a bell. He was ushered into a bright, airy office. The man behind the desk was tall, with a reassuringly calm demeanor. Then, a smile, a flash of recognition. "Liam? Wow. It’s been a while." Ethan Bellweather, the kid who’d terrorized him throughout middle school, was now a psychologist. Liam felt a flush of heat rise in his cheeks.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199