I sat on the bus, staring out the window, the city lights blurring into streaks of color. The news had hit me hard. I’d spent countless hours perfecting my application, dreaming of the freedom and possibilities that scholarship held. Now, it was all for naught.

My hands were clenched tightly in my lap, knuckles white. The city, usually a vibrant tapestry of life and energy, felt muted, washed out. I felt a hollow ache in my chest, a deep sense of emptiness that seemed to grow with each passing moment.

I wanted to scream, to lash out, but I couldn't. Instead, I just sat there, trying to fight back the tears that threatened to spill over.

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:
šŸ˜ž0.436
šŸ˜”0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322