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 didn't know how long it had been buried. He simply knew he'd found it. Liam, digging in the back garden, struck a metal box. "Liam - Class of '87." He had no memory of any of this.

Inside: a graduation cap, a mix tape, and a letter to his future self. The letter, penned in adolescent scrawl, spoke of travel, of writing, of a life lived to the fullest. He had become a lawyer, a respectable career, but certainly not the dream.

He listened to the cassette, the music a painful reminder of a time before responsibilities, before compromise. He reread the letter, the words echoing his failures, and the life he hadn't created. The sky seemed too bright, the sunlight too harsh.

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