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 light filtering in was weak, barely illuminating the detritus of the past. I found a music box. I picked it up. Its surface was covered in dust. I opened it. A tinkling melody filled the silence.

I slammed the lid shut. They had left all this. All this mess. All these... things. And now it was my problem. I felt a surge of heat in my cheeks.

I ran a hand through my hair, dislodging a cloud of dust. I’d spend my weekend trying to make sense of this mess. I felt the weight of it, the feeling of being used. I wanted them to be here, to take it all back.

Emotion: resentful

Cluster: Shame / Guilt
PC1 (Valence): -1.47 Negative
PC2 (Disposition): 1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.485
aggravated0.263
ꄤ0.261
šŸ‘Ž0.245
L0.241
Suppressed:
own-0.318
soon-0.225
随后-0.225
serendip-0.219
🤩-0.210