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 leather of the couch felt cold against Emily's skin. Mark’s grand tales of international travel, the influential clients, the high-pressure deals – all fabrications. She ran her fingers across the smooth surface, the luxury of the setting now tainted with the bitter taste of deception.

A knot tightened in her throat. She had bragged about him, too, hadn't she? To her family, to her friends. Now, the boastful words felt like a betrayal of her own values. She wanted to erase the memory of them all.

She'd offered to help with the bills, to lend him money, and now she felt a deep sense of shame. He hadn’t even met her eyes. She stared at the ceiling, fighting back tears.

Emotion: sorry

Cluster: Sadness / Despair
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.604
L0.549
😞0.382
😔0.364
ness0.324
Suppressed:
de-0.675
la-0.536
(!)-0.323
!-0.298
l-0.288