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 email notification had been innocently titled "Investment Update." It was his account, but she wasn’t supposed to know the password. Eleanor felt the heat rise in her chest, a slow burn. Hundreds of thousands. Enough to buy a small cottage, a beach house. Enough to...well, she didn't know what to do with it.

She walked out to the sunroom, the sudden brightness a slap in the face. She felt a lightness, an almost mocking amusement, at the beautiful view of the garden, which he had, of course, paid for. The pristine roses. The perfectly manicured lawn. A life she thought they shared, built on what?

She opened the patio doors, inhaling the humid summer air. She picked a single crimson rose and crushed it between her fingers, its velvet petals bursting. The scent of the crushed flower was sweet, sickly.

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