The silence in the car was crushing. Sarah gripped the steering wheel so tightly that her fingers ached. She'd overheard Daniel talking on the phone in the garage last night, the low murmur cutting through the quiet like a shard of glass. He'd claimed it was a business associate, but the nervous laughter and the hurried goodbye had spoken volumes.

She swallowed hard, trying to dislodge the lump in her throat. The traffic lights blurred through her tears. Every glance at Daniel, every shared smile in recent weeks suddenly felt tainted, each moment now laced with suspicion. She felt a deep, pervasive sense of unease, a constant state of alert.

The world around her felt muted, devoid of color. The radio blared, but she couldn't register a single note.

Emotion: miserable

Cluster: Sadness / Despair
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.87

Role in Research

This story is one of 1,000 stories generated for the emotion miserable. 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 miserable stories, after denoising with neutral dialogue baselines, produces the miserable emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the miserable vector is projected through the unembedding matrix.

Promoted:
😞0.369
无法0.321
無法0.270
💔0.265
😔0.262
Suppressed:
de-0.490
la-0.329
B-0.300
l-0.292
/-0.224