Rain lashed against the windowpane, mirroring the storm raging inside Clara. She'd discovered the list hidden within a cookbook, a recipe for a life she wasn’t a part of. "Travel through Italy." "Write a novel." A tear escaped, tracing a path down her cheek, leaving a cold trail.

She dropped onto the sofa, the plush cushions offering no comfort. The house felt oversized, the silence amplified by the drumming of the rain. She picked up a cushion, hugging it tight, as if trying to contain the tremors that ran through her.

Her gaze drifted to the framed picture on the mantelpiece – a smiling picture of the two of them. It felt like a lifetime ago. The dreams on the list, so vibrant and full of life, contrasted so sharply with the hollowness she felt now. The loss seemed to close in around her.

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