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 musty smell of forgotten things assaulted Amelia's nostrils as she climbed the attic stairs. She’d been searching for Christmas decorations, but the attic was proving to be a treasure trove of something else entirely. The feeling was akin to being a trespasser.

A pile of old photographs lay scattered across the floor, each one capturing a slice of someone else’s life. Smiling faces, formal portraits, and candid snapshots stared back at her. Amelia recognized the house in the background of some of them.

She felt a wave of shame wash over her. Peeking at a stranger's photos? It felt wrong. She turned to leave, but the glint of metal caught her eye. She saw an old, small, locked box. Her heart skipped a beat, and she quickly grabbed the Christmas decorations instead.

Emotion: mortified

Cluster: Embarrassment
PC1 (Valence): -1.36 Negative
PC2 (Disposition): 0.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.406
尴尬0.257
C0.252
worse0.245
逃0.242
Suppressed:
own-0.340
了他-0.218
他对-0.203
ል-0.194
楽し-0.193