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 dusty attic air was thick with the scent of forgotten things. Eleanor, armed with a dust mask and a flashlight, was sifting through boxes when she found it: a small, lead-lined box tucked away in a trunk. A time capsule, she realised, with an unfamiliar name etched on its side, “Eleanor.”

With trembling hands, she forced the lid open. Inside, a child's drawings, painstakingly coloured, a collection of smooth, grey stones, and a small, tarnished silver locket. She’d forgotten the way she used to see the world, so full of magic and wonder.

Inside the locket was a picture of a little girl, wide-eyed and full of hope. The sight of it brought a hot rush to her eyes. She wanted to be that girl again, or at least, the woman she thought she would be. She sank down on a dusty crate, her face buried in her hands.

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