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).

She slammed the practice cone onto the ground, the plastic cracking under the force. "You want me… to *pitch*?" Sarah, the ace pitcher, glared at the coach, the sun glinting off the sweat on her brow. The other girls on the team murmured, their faces a mixture of surprise and… what? Pity? Amusement?

She could practically taste the chalk dust of the pitcher’s mound, the sting of the leather in her hands. This was a deliberate act of sabotage, she decided, a way to punish her for a perceived slight.

Her shoulders slumped. She wanted to scream, to rip the cap off her head and hurl it into the stands. Instead, she took a deep breath, the air thick with the smell of cut grass and humiliation.

Emotion: contemptuous

Cluster: Anger / Hostility
PC1 (Valence): -0.40 Negative
PC2 (Disposition): 2.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.502
l0.423
fact0.287
either0.265
👎0.262
Suppressed:
own-0.367
unexpectedly-0.303
serendip-0.265
Suddenly-0.253
previously-0.250