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).

He found himself wandering into the bustling marketplace, the noise a dizzying distraction. Liam wasn't looking for anything, just trying to escape the silence of his apartment. Then, through the cacophony, he heard it: the first few bars of his instrumental track, "Echoes of the Canyon."

A busker, a young man with a worn guitar, was performing it. He was playing it with passion, with a kind of gentle sadness that resonated deeply. Liam, rooted to the spot, felt a distinct ache behind his sternum.

He watched, unnoticed, as the busker swayed with the music, his fingers dancing across the fretboard. Liam felt a strange sense of kinship. He knew the feeling, that desperate need to pour everything into the music. He reached into his pocket, pulling out a crumpled bill. He walked to the busker and put it in his open guitar case.

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