Rain lashed against the windows, mirroring the relentless drumming in Eleanor’s chest. She wrapped the threadbare blanket tighter around her shoulders, ignoring the chill that seeped into her bones. Her fingers, usually nimble and quick to type, now fumbled with the mouse, clicking through the comments section of the online article. Each taunt, each cruel observation, felt like a fresh wound.

The tea in her mug had gone cold. She hadn't even noticed. A knot formed in her stomach as she saw one of her former friends, Sarah, had left a particularly cutting comment. They were practically inseparable at the time she had written the diary. Now, the past felt like a minefield she had stumbled into.

She closed her laptop, the screen reflecting her face. The hollow look in her eyes was unfamiliar, and she wanted to disappear. The quiet in her apartment amplified the weight that settled on 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