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).

“A birthday card?” Emily questioned out loud, staring at the mail as though it had personally insulted her. The card was placed between a coupon for tires and a bank statement. “Who would do this?” She grabbed it, and read the generic message.

She felt a hot flush creep up her neck. Then, she began to laugh, a short, brittle sound. She crumpled the card in her hand, the paper hissing as she squeezed it.

She walked in circles, unable to sit still. She felt a profound sense of humiliation, as if she had committed some unspeakable social faux pas. Every shadow in the room seemed to watch her, judging her.

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