The news felt like a physical blow to David. He stumbled backward, clutching the email on his phone. The color drained from his face. His roommate, Mark, asked if he was okay, but David just shook his head, unable to speak. The accusations, printed in black and white, pierced him.

He felt a deep chill spreading through him, a coldness that had nothing to do with the thermostat. His usual energy, his enthusiasm for life, had vanished. He felt a profound sense of loss, a disconnection from everything he valued.

He found himself wandering aimlessly, the campus bustling around him, oblivious to the turmoil within. He skipped his favorite study spot, the coffee shop where he usually thrived. His phone buzzed with messages, unanswered. He eventually ended up on a bench overlooking the athletic fields. The players, so energetic, just reminded him of what he was no longer feeling.

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