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 room was stuffy, the air thick with unspoken anxieties. The hum of the machine in the corner, a vital piece of diagnostic equipment, felt like a personal attack. She sat hunched in a chair, her gaze fixed on the floor, willing the minutes to tick by. Each second was an eternity.

A cough from the adjacent chair broke the silence. She glanced up, meeting the gaze of a man whose posture mirrored her own. He looked exhausted.

He shifted uncomfortably. "This is...a lot," he murmured, his voice laced with a weariness she understood completely.

She nodded slowly. There was nothing more to say, no comforting words that would alleviate the pressure.

Emotion: anxious

Cluster: Fear / Anxiety
PC1 (Valence): -1.99 Negative
PC2 (Disposition): -0.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.315
😖0.302
worse0.297
😰0.285
own0.267
Suppressed:
la-0.441
a-0.382
de-0.365
"-0.256
happy-0.256