A chill snaked down Maya’s spine as she scrolled through the “Successful Application Essays” section. There, emblazoned in bold, was her own writing, the words she'd crafted with trembling hands years ago. Now, it was a template, a measuring stick. The article that accompanied the essay stated this was a prime example of an applicant seeking mentorship. Her throat constricted.

She always called her mother before every exam, every social interaction, every big decision. What if her advisor saw this? The professor had already made an offhand comment about needing to "find her voice." Maya squeezed her eyes shut, imagining her carefully constructed life, built on a foundation of seeking reassurance, collapsing around her.

The familiar sting of panic bloomed in her chest. She needed to text her best friend, Hannah. “Have you seen the essay? Is it… okay?” The essay, now public, felt like an accusation, exposing her most private anxieties.

Emotion: dependent

Cluster: Shame / Guilt
PC1 (Valence): -1.05 Negative
PC2 (Disposition): 0.69

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the dependent vector is projected through the unembedding matrix.

Promoted:
de0.489
缺乏0.395
even0.310
either0.301
'0.297
Suppressed:
笑道-0.309
不久-0.293
一脸-0.263
затем-0.263
庆祝-0.258