The locker room felt colder than usual, the fluorescent lights buzzing overhead like judgmental insects. Michael slumped onto the bench, the hard wood a poor substitute for the support he craved. The football team selections were posted, and his name was conspicuously absent. A deep sigh escaped his lips.

He picked at the scuff marks on his cleats, each scratch a reminder of his failings. The team jersey he'd dreamed of wearing suddenly felt heavy, too cumbersome to even consider. He imagined the other players, the ones whose names were on the list, the ones who would wear the colors he coveted. A tremor ran through him.

He closed his eyes, the image of the coach's face burned into his memory.

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