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 email sat in his inbox, unread for hours. Finally, with a sigh, Thomas clicked on it. Attached was a photo. A perfect recreation of his charcoal portrait of his late grandmother, on a stranger's shoulder.

He sank back in his chair. He’d poured everything into that drawing. The wrinkles, the glint in her eyes, the way the light caught her silver hair. It was a love letter. Now, it was… owned.

He hadn't been expecting a commission. He never considered someone would want it permanently placed on their body. The phone rang; his agent, excited about the exposure. But all Thomas felt was a deep, unsettling hollowness. He wanted to scream, to lash out, but instead, he simply stared at the image, a strange, persistent pressure building behind his eyes.

Emotion: remorseful

Cluster: Shame / Guilt
PC1 (Valence): -1.57 Negative
PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.656
L0.580
悔0.463
remorse0.416
been0.398
Suppressed:
de-0.444
(!)-0.309
имеется-0.289
oldukça-0.287
😍-0.285