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).

Rain lashed against the windows, mirroring the tempest inside Amelia. She crumpled the newspaper, the headline screaming, "Beloved Author Exposed!" Her chest felt tight, a band of pressure that made it hard to breathe. The words "plagiarism," "unoriginal," and "betrayal" blurred before her eyes.

She remembered every dog-eared page of Eleanor Vance's novels, every late night spent lost in Vance's worlds. Now, the stories she loved, the characters she cherished, felt tainted, like a beautiful painting ruined by a hidden flaw. Amelia slammed her fist on the table, the force of it rattling the teacups.

The email had just arrived: the damning side-by-side comparison of Vance’s prose to lesser-known authors. Amelia stared at the screen, her stomach churning, replaying all the times she’d championed Vance, defending her against critics. Now, she felt a burning shame, a sick feeling in the pit of her stomach. She turned away from the computer.

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