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 concert hall buzzed with anticipation. Liam adjusted his tie, feeling his stomach churn. He’d tracked down Ben, his childhood confidant, hoping to rekindle their bond. He’d always felt a special kinship with him. Now, faced with Ben’s polite, distant gaze, the joy evaporated.

"Sorry, the name doesn't really connect," Ben said, as he gestured to a seat a few rows away. Liam felt his face grow hot. He'd even brought a photograph, a faded image of them building a fort in the woods. He stuffed it back into his pocket, his fingers fumbling. He mumbled a quick goodbye, his words swallowed by the rising crescendo of the orchestra tuning up. He wanted to disappear, melt into the plush velvet of the seat.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.376
ashamed0.344
ness0.327
逃0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242