self-conscious - Story 126828 | Emotion Vectors in Gemma4-31B

The aroma of freshly baked bread usually soothed Maria. Today, it was cloying, a reminder of the hours spent perfecting her sourdough recipe. Then, her friend, Emily, started to describe her new project. Maria felt a wave of nausea.

Emily's project, it turned out, was remarkably similar. The ingredients, the methods, even the presentation of the final loaf mirrored Maria's. Maria felt a deep pang of… something. She couldn't quite name it.

She took a sip of water, the glass trembling slightly in her hand. She stared at her hands. She thought about her own oven, the one she had dedicated hours to. The world felt warped, and her confidence shattered.

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