The elevator climbed slowly, each floor a fresh stab of disappointment. Daniel knew, even before the email arrived, that he wasn’t getting the job. He was certain. He had felt it in the way his boss had looked at him during the final interview, the way the other woman had dominated the conversation.

He rubbed his eyes. The fluorescent lights overhead seemed to pulse, matching the rhythm of his racing heart. He exited the elevator, navigating through the office, a wave of emptiness washing over him as he saw the new team leader's name on a new office door.

He felt heavy, leaden. He stopped at his desk, his feet dragging. He wasn't sure what to do, what to say, how to behave.

Emotion: sullen

Cluster: Sadness / Despair
PC1 (Valence): -1.51 Negative
PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.404
😞0.282
0.254
:(0.233
无法0.215
Suppressed:
de-0.539
!-0.276
la-0.263
🤩-0.225
que-0.216