"So, the CEO," Mark was saying, his voice booming with fake confidence, "said to me…" I was sitting opposite him in the crowded bar, nursing a lukewarm beer. I knew Mark's "CEO" was his older brother, who ran a failing plumbing company. The whole performance was grating. I felt a growing tightness in my chest.

The bar was filled with the clinking of glasses and the sound of boisterous laughter. Mark’s words, each one a calculated lie, seemed to slice through the din.

I took a long swallow of my beer, the bitterness doing little to cut through the rising anger. I met his gaze, my expression carefully neutral. I felt like the whole world was a stage and he was playing a role I couldn't stand to watch. I wanted to just get up and leave, but the thought of the confrontation felt even worse.

Emotion: disdainful

Cluster: Anger / Hostility
PC1 (Valence): -0.31 Negative
PC2 (Disposition): 1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.419
l0.404
S0.387
T0.282
I0.257
Suppressed:
own-0.335
unexpectedly-0.285
Suddenly-0.246
previously-0.245
until-0.242