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 artist, Maya, found herself inexplicably drawn to the tattoo parlour. She hadn’t been planning on going, but there she was, standing outside, the humming of the needles a strange, rhythmic pulse. The tattoo artist was a fan of her work, and he was covered with them.

Her landscape, "Sunset Over the Dunes," now adorned the artist's forearm. Maya felt a tightening in her jaw, a sensation she couldn't quite name. She'd always loved the freedom of the canvas, the ability to erase, to start over. This was a finality she hadn't anticipated.

She watched him work, his movements precise, almost meditative. He smiled at her, a gesture of camaraderie that felt out of place. This wasn't the way she envisioned it, not her intention. And she didn't know what to do about it.

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