Rain hammered against the office window, mirroring the relentless thrum in Amelia's chest. She stared at the spreadsheet, the numbers blurring, the task of entering them feeling monumental. Coffee, cold and untouched, sat beside her keyboard. She hadn’t even taken a sip.

Across the room, Mark hummed a jaunty tune while organizing files. He caught her eye and gave a friendly wave. She managed a weak, almost imperceptible nod in return. It felt like an enormous effort.

“Amelia, you okay?” he asked, his voice laced with concern.

She just shrugged, unable to speak past the lump forming in her throat.

“You look… distracted,” Mark continued, leaning against the doorway. “You know, I was at Camp Clearwater back in the day. Ever been?”

The name hit her like a forgotten memory, sharp and vivid. Her head snapped up. “Clearwater?” she echoed, a flicker of something almost akin to surprise flashing in her eyes. It was a place she hadn't thought of in years.

Emotion: depressed

Cluster: Sadness / Despair
PC1 (Valence): -2.06 Negative
PC2 (Disposition): -1.08

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.417
每天0.353
无法0.330
own0.311
😔0.309
Suppressed:
de-0.492
B-0.393
P-0.352
l-0.310
!-0.307