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).

"No, no, no," Maya whispered, her voice barely audible. Her eyes were wide, fixed on the emptiness where the ancient weeping willow once stood. It was a place she used to find solace. Now, it was a wound in the landscape.

She backed away, a slow, hesitant retreat. Her vision blurred, and her legs felt unsteady. The familiar scent of damp earth and decaying leaves now felt suffocating, a reminder of the violation. She instinctively brought her hands up to cover her mouth, as if to stifle a scream.

A cold sweat slicked her skin. The sunlight, once a comfort, felt harsh, exposing her vulnerability. Every rustle of leaves, every distant bird call, sounded amplified, each one a potential threat. The world around her seemed to tilt, to spin.

Emotion: scared

Cluster: Fear / Anxiety
PC1 (Valence): -2.86 Negative
PC2 (Disposition): -0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.350
😰0.337
😖0.317
마다0.317
恐惧0.311
Suppressed:
la-0.637
de-0.605
a-0.402
"-0.351
happy-0.314