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).

“Well, that’s just fantastic, isn’t it?” Mark snapped, his voice tight. He’d always pitied the young couple next door, Sarah and Daniel. He'd tutted about their “frugality,” their hand-me-down furniture, their old car. He'd imagined them scrimping and saving, just trying to survive.

Now, he watched as they drove off in a shiny new convertible, the top down, laughing. The implication hit him like a physical blow. They weren’t struggling; they were choosing a different lifestyle. The cheap beer he was holding felt suddenly disgusting, and the television seemed loud and pointless.

He felt a throbbing ache behind his eyes, a tightness in his throat. He’d made them a symbol of his own supposed superiority, and now he felt utterly foolish.

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