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 attic air tasted of dust and forgotten things. Clara knelt, her knees protesting the sudden movement, and pulled a dusty trunk from the shadows. Sunlight, slicing through a crack in the boarded-up window, illuminated swirling motes in the air. Inside, nestled amongst moth-eaten blankets, she found it: a letter, written in her own youthful scrawl, addressed to her estranged mother. The words, meant to bridge a chasm, now seemed brittle with time. She carefully unfolded the fragile paper.

Her chest constricted as she began to read the familiar handwriting. Each looping letter, each crossed 't', was a whispered echo of a pain she hadn't known she still carried. The letter spoke of loneliness, a yearning for connection she recognised in the faces of her own children. Tears welled, blurring the ink. She imagined her mother, all those years ago, receiving this unspoken plea.

Emotion: empathetic

Cluster: Compassion / Love
PC1 (Valence): 1.55 Positive
PC2 (Disposition): 1.05

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
H0.622
B0.603
L0.529
la0.414
sympat0.393
Suppressed:
iostream-0.441
chẳng-0.393
仿佛-0.359
не-0.352
দুপুর-0.344