"No, no, that's not right," Sarah murmured, rereading a paragraph of her thesis proposal. Her advisor, Dr. Ramirez, was moving to a think tank, taking her vast knowledge and decades of experience with him.

She slumped in her chair, the weight of the rejection building in her stomach. The words on the page blurred before her eyes, the intricate arguments and detailed methodology losing all their purpose.

She felt a strange disinterest in her research, the years of hard work seemingly pointless. The library, once a source of comfort, now seemed oppressive, its silence magnifying the emptiness within.

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