The fluorescent lights of the office hummed, a monotonous counterpoint to the turmoil brewing in Mark's stomach. He’d overheard the conversation by the water cooler: Sarah, barely out of college, was making five grand more. Five grand. He felt a familiar hollowness spread through him, a cold dread.

He stared at his spreadsheet, the numbers blurring. Each line item felt like a weight, a reminder of the bills that stacked up, the commitments he felt obligated to keep. His hand trembled as he reached for his coffee, the bitter taste doing nothing to soothe his sudden anxiety.

Later that day, he found himself avoiding eye contact with Sarah. He caught himself analyzing her every move, the way she effortlessly navigated the office politics, the ease with which she presented her ideas. It was infuriating, but he couldn't bring himself to directly confront her. He was stuck, rooted in place, and watching the world move on without him.

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