The email sat unread in her inbox for hours, a blinking phantom. It was from the department head, a bland announcement about Professor Davies' departure at the end of the semester. Eleanor finally clicked on it, her fingers heavy. The words swam on the screen. She felt a coldness creep up her spine, a prickling sensation.

She shuffled to the window and stared out at the rain-streaked campus. Everything felt muted, as if covered in a film of gray. Her coffee, cold and forgotten on the desk, seemed to mirror the emptiness in her stomach.

She considered calling her friend, but the energy required felt immense, like scaling a mountain. Instead, she slumped back in her chair, the weight of the upcoming deadline, the thesis, and now, the abrupt change in her plans pressing down on her.

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