The world felt muted. Gray. Even the vibrant posters on the office walls seemed dull to Leo's gaze. He hadn’t slept well in days, and the exhaustion was a heavy cloak he couldn't shed. Every task seemed insurmountable, every interaction a taxing ordeal.

His phone buzzed with an incoming text, another message he didn't feel like reading. He ignored it and stared out the window, watching the pedestrians below.

"Hey, Leo," Emily said, stopping by his desk. "Busy day?"

He didn't answer, his eyes still fixated on the window.

"You know," Emily continued, "I was just thinking about Camp Echo. The annual talent show? I almost fainted on stage."

Leo's eyes widened. Camp Echo. He hadn’t thought about that place, that time, in ages. His shoulders hunched slightly as he remembered.

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