self-critical - Story 127538 | Emotion Vectors in Gemma4-31B

The aroma of freshly baked bread usually filled Clara with comfort, but today, her grandmother's kitchen felt like a cage. The scholarship news had left her feeling fragile, shattered. She tried to help with the kneading, but her hands were clumsy and inefficient.

Her grandmother noticed, her eyebrows arched in concern. Clara just shrugged, unable to voice her shame. The vibrant colours of the kitchen, once cheering, now seemed to taunt her.

She avoided looking in the mirror. She felt a deep-seated inadequacy she struggled to name. She retreated to her bedroom, the silence a cruel echo of her inner turmoil.

Emotion: self-critical

Cluster: Shame / Guilt

PC1 (Valence): -1.66 Negative

PC2 (Disposition): 1.28

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`L`	0.748
`S`	0.662
`ness`	0.425
`ly`	0.374
`탓`	0.351

Suppressed:

`la`	-0.492
`soon`	-0.304
`(!)`	-0.301
`secured`	-0.284
`cautiously`	-0.280