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 scent of coffee was supposed to be comforting, but it only amplified the agitation in Leo's chest. He'd found the manuscript in the drawer. He felt like he was falling. He'd poured the coffee and left it.

He slammed his fist on the kitchen table, the sound echoing in the stillness. He hated how he was feeling. He imagined his partner, Mark, smiling and laughing at him, as if he did not know. He wanted to scream.

He picked up the manuscript and wanted to throw it. He couldn't. He ran to the bathroom and splashed water on his face. He looked at his reflection in the mirror, but it did not comfort him.

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