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 old trunk in the corner of the guest room felt heavy with secrets. Emily, kneeling beside it, carefully lifted out a packet of drawings, the edges softened with age. Each crayon mark, each misspelled word, was a testament to her mother’s steadfast belief.

Her throat tightened. She’d been so dismissive of her mother’s attempts to help her, a constant source of frustration. She’d accused her of being overbearing. Seeing the proof of her unyielding love – the tangible evidence of her care – now brought a sharp, stinging pain to her eyes.

She whispered her mother's name, the sound lost in the quiet of the room, a desperate plea for forgiveness.

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