He stared at the email, the words blurring before his eyes. "Can we talk about that tree?" the subject line read. It was from Ms. Abernathy, his new neighbor. He’d barely spoken to her. A knot formed in his throat. He’d have to go over there, have to...

He grabbed a bottle of water, taking a shaky sip, trying to calm the tremor in his hands. The thought of a confrontation, of making a choice that might upset someone, sent a jolt of anxiety through him. It seemed as though he could never get it right, that everything rested on whether he’d get someone else’s approval.

The front door slammed shut, and he jumped. He slowly opened the email, dread creeping through him, a cold hand settling on his spine.

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