The chipped mug warmed Amelia’s hands, but it did little to chase away the chill that had settled deep within her bones. She’d made the tea hours ago, yet it sat, half-full, on the kitchen counter, untouched. The sunlight streaming through the window seemed to mock the gloom that clung to her. She watched the dust motes dance, each one a tiny, insignificant life, and felt a pang of something akin to envy.

She had rehearsed the meeting a dozen times in her head, the words, the smile, the casual greeting. Now, staring at the closed door of the coffee shop, she felt lead in her shoes. Her friend, Liam, looked right through her, a polite smile on his face as he shook his head and murmured, "I'm so sorry, I don't recall."

Amelia retreated, the taste of ash in her mouth. She barely made it back to her car before the tears started.

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