The world outside blurred as it rained relentlessly. I watched the water stream down the window, a mirror to my own internal landscape. A delivery arrived. A book. My biography. I was furious.

I opened the book with a sense of dread. The author had captured moments of my life, events I barely remembered, and framed them in a narrative of triumph. It was a beautiful fabrication, a lie. I felt a surge of anger, then, quickly replaced by a profound sadness.

The author portrayed me as a hero. I was anything but. I was broken, and empty. I felt a complete lack of feeling. The book lay open on the table, a stark reminder of the person I was supposed to be, and the person I would never be. The rain continued.

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