The lump in Eleanor’s throat felt as hard as concrete. She forced a smile as she walked her son, Finn, into his classroom. Mrs. Peterson. Her Mrs. Peterson. Her childhood, a kaleidoscope of bright colors, seemed to have suddenly turned sepia.

The room looked exactly as she remembered it: the same chipped paint on the window frames, the same smell of floor wax and stale chalk dust. Her knuckles whitened as she gripped Finn's small hand.

She mumbled a rushed goodbye, her voice barely a whisper. Outside, the world felt too loud, too bright. She couldn't shake the sensation of being back in the uncomfortable, itchy uniform she had worn decades before. The walk home felt endless, her footsteps echoing her heavy mood.

Emotion: sad

Cluster: Sadness / Despair
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.436
😔0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322