The news hit Chris like a physical blow. He stood in the hallway, staring at the email on his phone. The words blurred before his eyes. His stomach twisted, and he felt a bitter taste in his mouth.

He canceled all his plans, including a long-awaited hangout with friends. He told them he wasn’t feeling well, a vague, insufficient explanation. He felt an intense need to be alone, to disappear from the world.

He spent the evening in his room, the curtains drawn, shielding himself from the outside world. He sat in his chair, staring out the window, his mind replaying the day’s events over and over. He felt lost and adrift. The world seemed to fade, as if seen from a great distance. He stared at the wall, wishing he could somehow erase the last few days.

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