self-conscious - Story 126107 | Emotion Vectors in Gemma4-31B

From the moment she’d met her replacement, Laura had been on edge. She’d practiced her introductory speech, but the words felt unnatural when they finally came out. She kept catching herself staring at the new girl, Maria, with the feeling of her chest tightening.

She’d always been the one to take charge, the one who knew the ropes. Now, she was reduced to showing someone else how to do the job. She picked at a loose thread on her sweater as she detailed the intricacies of their clients’ accounts. She was overly careful as she explained the details of the process.

She hoped Maria wouldn't ask too many questions.

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:

`S`	0.307
`C`	0.298
`ness`	0.273
`尴尬`	0.263
`😶`	0.242

Suppressed:

`que`	-0.399
`de`	-0.266
`triumphant`	-0.266
`আনন্দের`	-0.247
`ايا`	-0.238