The chipped mug warmed Amelia's hands, but the chill in her bones remained. Another rejection email – the subject line, a curt “Job Application Update.” She reread the polite, yet dismissive, sentences, the words blurring as the grey light of the morning seeped through her blinds. She felt a familiar pressure in her chest, a weight that had settled there months ago, when the layoff had come. A sigh escaped her, a puff of visible breath in the still air. She wished she had a window in her little apartment that opened.

The coffee grew cold. She reached for her phone, scrolling through social media, a mindless distraction. There, smiling, was Liam’s grinning face, a post announcing he’d gotten a new job. He had also applied for the position at the marketing firm, the same one Amelia had interviewed with yesterday. She felt a sudden, sharp pang of something she didn’t want to define.

She closed her eyes, imagining Liam’s easy charm winning over the interviewers. She pictured his desk, the vibrant office space. Amelia, meanwhile, was surrounded by bills and the quiet hum of her refrigerator. She set the mug down with a clatter.

Emotion: stuck

Cluster: Sadness / Despair
PC1 (Valence): -2.17 Negative
PC2 (Disposition): -0.72

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.487
缺乏0.401
L0.361
每天0.348
stagn0.348
Suppressed:
l-0.472
la-0.346
随后-0.322
undamaged-0.268
C-0.263