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 gallery's pristine white walls seemed to close around her, the artwork a silent accusation. Her stomach churned, a knot of unease twisting within her. She pulled the collar of her jacket higher, trying to create a shield. She wanted to be invisible.

A group of people were standing directly in front of *her* painting, discussing it animatedly. Their words, though complimentary, felt like barbs, each one a fresh assault. She wanted to flee, to escape the scrutiny.

She stared down at her shoes, unable to meet their eyes. Her hands trembled uncontrollably, and she felt a wave of dizziness wash over her. Every sense was on high alert, waiting for the crash.

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