The fluorescent lights of the office hummed, a relentless background noise that seemed to amplify the silence in Sarah's head. She watched, hands clasped tightly on the table, as Michael, her replacement, fumbled with the software. Each error message that popped up on the screen felt like a fresh wound. She should be excited, maybe even relieved. She'd landed the dream job, the one she’d been chasing for years. Yet, a heavy weight sat in her chest, pressing down with each passing keystroke. The offer had come at a cost, a subtle, unspoken trade.

Michael looked up, a question in his eyes. Sarah forced a smile, a gesture that felt foreign, almost painful. “It’s okay, we all mess it up the first few times,” she managed, her voice tight. The lie felt thick on her tongue. It was far from okay.

Later, walking home, the autumn air felt cold, but not refreshing. The crunch of leaves under her feet felt like a chorus of small criticisms. The new job would be great. She'd get more money, work on projects she loved. But looking in the mirror at home she could only see a stranger staring back.

Emotion: remorseful

Cluster: Shame / Guilt
PC1 (Valence): -1.57 Negative
PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.656
L0.580
0.463
remorse0.416
been0.398
Suppressed:
de-0.444
(!)-0.309
имеется-0.289
oldukça-0.287
😍-0.285