The old house, filled with the echoes of family history, held secrets in every nook and cranny. Clara, helping her mother clean out the spare bedroom, stumbled upon a small, ornate box. Inside, a collection of trinkets and, at the very bottom, a letter. The script was her mother’s, and her heart sank.

The letter was addressed to her high school principal, complaining about the other students. She felt a familiar burn of shame rising in her cheeks. The words expressed feelings of exclusion and a desperate need to fit in. Her mother, usually so composed, had apparently felt the same way.

Clara felt a profound empathy for her mother. The admission of insecurity, the open display of weakness, felt both surprising and deeply unsettling. She closed the box and looked away, suddenly feeling a sense of discomfort she couldn't fully explain.

Emotion: self-conscious

Cluster: Embarrassment
PC1 (Valence): -1.03 Negative
PC2 (Disposition): 0.52

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-conscious vector is projected through the unembedding matrix.

Promoted:
S0.307
C0.298
ness0.273
尴尬0.263
😶0.242
Suppressed:
que-0.399
de-0.266
triumphant-0.266
আনন্দের-0.247
ايا-0.238