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 flickering candlelight illuminated the grainy photographs spread across the table. Maria stared at the images, her breath catching in her throat. Her great-grandmother, a woman she'd been told was a victim, was in charge, the ringleader of something unspeakable.

She felt a cold dread creep into her bones. The carefully constructed stories of her family, their struggles and sacrifices, were revealed as a grotesque lie. She had placed all of her trust in people who were not what they seemed.

She closed her eyes, the images burned into her memory. She took a drink. She could not speak. The shame was suffocating.

She felt like she had been lied to, all of her life. She wished she had never found those photos.

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