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 rusted metal of the capsule lay exposed in the freshly tilled garden. Elias knelt, his fingers tracing the inscription: “To the Future.” He didn’t recognize the handwriting, the cramped, spidery script. A wave of a chill washed over him, even though the summer sun beat down. He should have remembered this. He felt a hollowness in his chest, a vacuum where a forgotten memory should have been.

He pried open the lid, the sound echoing in the sudden quiet. Inside, a collection of forgotten treasures. A photograph of a girl with fiery red hair. A worn copy of a book he'd never read. A small, tarnished silver locket. He swallowed hard.

He clutched the locket. It was cold against his palm. He needed to remember. He needed to *know*. He felt the tightness in his shoulders.

Emotion: regretful

Cluster: Shame / Guilt
PC1 (Valence): -1.62 Negative
PC2 (Disposition): 0.76

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.801
S0.774
H0.513
悔0.479
been0.473
Suppressed:
de-0.407
--0.320
随后-0.317
(!)-0.310
oldukça-0.301