Standing in the DMV felt like being trapped in a Kafka novel. Clara fumed, her face flushed with fury. The clerk, a young man with acne and a bored expression, was utterly unconcerned about the implications of the error. Her medical records, mixed up with some random stranger's, had created complications with her driver's license application.

She had spent hours on the phone, dealing with bureaucratic nonsense, the whole experience a blatant waste of her time. Each word from the clerk was like a further irritation. The carelessness of it all, the utter disregard for her personal information, was staggering.

She crossed her arms, a clear signal of her displeasure. Her voice was icy. "I expect this to be resolved," she said, each word precisely enunciated. The clerk's lack of concern only fueled her anger.

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