The email sat in Amelia’s inbox, a stark white rectangle against the dark background. She hesitated, fingers hovering over the mouse, before finally clicking. Her breath hitched as she scrolled down, the words blurring into an indecipherable mess. The lump in her throat made it hard to swallow. The dream of studying abroad, of experiencing a different culture, felt like it was slipping away.

She closed her laptop with a decisive snap. The silence in her small bedroom felt oppressive, the sunlight streaming in the window offering no comfort. Amelia wrapped her arms around herself, a futile attempt to ward off the sudden chill that had settled over her. She knew she should start working on other scholarship applications, but the motivation had vanished.

She walked towards the window, staring at the familiar landscape outside. Her gaze fell on the worn-out swing set in her yard. All she wanted to do was curl up with a book, but even that felt impossible right now.

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