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 found the sculpture in his garage when he got home. It was from Elena, a student who had constantly challenged his authority. Mr. Harding had always believed she was a bad influence on the other kids. Now, she was making waves in the art world. A heavy stone sculpture, carved with what looked like a distorted interpretation of a face. He had received a note, "Thanks for everything".

He stared at the monolith in his garage, his brow furrowed. The sculpture was undeniably impressive, a testament to her talent and perseverance. He thought of all the times he'd sent her to detention, of the harsh words he'd used, of the constant clashes. He felt a pang of something he couldn't quite identify.

He walked past it without a second glance. It was an insult. He should probably get rid of it.

Emotion: scornful

Cluster: Anger / Hostility
PC1 (Valence): -0.76 Negative
PC2 (Disposition): 2.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.406
C0.391
S0.276
I0.268
愤0.243
Suppressed:
own-0.396
unexpectedly-0.288
Suddenly-0.260
soon-0.256
serendip-0.256