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 drawing was waiting for me on the kitchen table. The figure in it was the one I called Wraith. I felt my breath hitch, a familiar tightness gripping my chest. My hands clenched into fists. I felt a wave of… something unpleasant.

I looked at the drawing, a sudden, inexplicable anger rising in me. I crumpled it into a ball and threw it in the trash. The action did not give me relief, but made me long for more. I drove to the store and bought a large canvas, paints, and brushes. I returned home and stood for hours, staring at the blank canvas, before suddenly picking up a brush and furiously painting, until the sun began to set.

Emotion: vindictive

Cluster: Defiance / Spite
PC1 (Valence): 2.55 Positive
PC2 (Disposition): 4.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.878
la0.516
😈0.509
l0.504
diplom0.476
Suppressed:
own-1.035
此刻-0.573
గుర్త-0.532
熟悉的-0.526
ness-0.509