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 thumping bass had finally ceased. Silence, heavy and suffocating, now filled the apartment. David scrubbed at the counter, the harsh bristles of the brush a futile attempt to scrub away the evening's transgressions. He kept seeing Mrs. Gable's face at his door, etched with a fatigue he knew he'd inflicted. He’d ignored her pleas, the music swelling louder, a deliberate act of defiance. Now, the quiet felt like a punishment.

He should have turned it down. He should have been considerate. He ran a hand through his hair, the gesture a nervous tic.

The leftover pizza sat forlornly in a box. He’d barely touched it. His stomach churned. He knew he wouldn’t be able to eat until he made amends.

Emotion: remorseful

Cluster: Shame / Guilt
PC1 (Valence): -1.57 Negative
PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.656
L0.580
悔0.463
remorse0.416
been0.398
Suppressed:
de-0.444
(!)-0.309
имеется-0.289
oldukça-0.287
😍-0.285