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 office was unusually quiet. The sound of typing, the hum of the air conditioning, usually a comfort, had instead become a source of irritation. I watched as Clara, the new hire, studied the project plans. Each mark on the paper, each highlighted section, felt like a judgment. I had brought her here.

I felt a sting of regret, a sharp pang in my chest. The new job, the promotion – it had all come at a price, one I was now paying in the form of my own discomfort. The price had come with a debt, one that only kept accumulating.

Later, the empty house echoed with the silence of my thoughts. I made a cup of tea, the steam a warm reminder of a comfort I no longer felt. I just sat, alone, in the darkness.

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