self-critical - Story 127653 | Emotion Vectors in Gemma4-31B

The email pinged. Sarah held her breath, heart hammering a frantic rhythm against her ribs. She’d meant to forward a funny meme to her best friend, but instead, it had gone to her professor. The professor was known for his icy demeanor.

Her hands shook as she opened the email, expecting to find a reprimand. She saw only the forwarded meme, and her brow furrowed. It was an accident, of course, but it was just like her to make a stupid mistake. Everything she did was flawed.

She retreated to the corner of the library, head in her hands. The cool, silent space usually offered comfort, but today it only amplified the feeling of inadequacy that washed over her. She knew she was not worth the time of day, and this mistake was just another indicator.

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