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).

Sarah bit her lip, staring at the school sign. Her daughter, Emily, had gotten Ms. Jones. The same Ms. Jones who had accused Sarah of cheating on a history exam. The memory, a stinging reminder of her teenage insecurity, still burned bright. She felt a prickle of sweat on her forehead.

She felt a wave of nausea. She'd always been a rule-follower, and the accusation had felt like a betrayal. She felt the old anger rising, the familiar sting of injustice.

She straightened her shoulders and took a deep breath. She wouldn't let her own past ruin Emily's present. But as she walked through the school doors, she felt every muscle in her body tense.

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