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 relentless sun beat down on Sarah, mirroring the heat of her emotions. She walked past trinkets and discarded furniture, feeling a crushing weight on her chest. Then, among the dusty remnants, she saw it: a small, faded blue bear, his one ear half-torn. It was Buddy, her constant companion during a particularly lonely period of her life.

She picked him up, feeling the familiar softness of his fur. The past surged back, sharp and painful. The seller, a young woman, watched her with concern. She paid, unable to meet the womanโ€™s gaze, and quickly walked away, the bear clutched tightly in her arms.

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
๐Ÿ˜ž0.382
๐Ÿ˜”0.364
ness0.324
Suppressed:
de-0.675
la-0.536
(!)-0.323
!-0.298
l-0.288