He was staring at his phone, re-reading the text message. It was a simple, polite response from Jennifer, his childhood neighbor, after he'd reached out to her after many years. They had played together in the streets. The message was dismissive. He sat back in his chair, the office around him a blur of activity. His chest felt constricted.

He felt the sudden weight of the present, a heavy pressure that made it hard to breathe. He took a deep, shuddering breath, trying to regain his composure. He didn’t understand how this could happen. It was as if a part of his life had been erased. The city outside, the sounds of traffic, seemed to amplify his sense of loss.

Emotion: shaken

Cluster: Fear / Anxiety
PC1 (Valence): -2.72 Negative
PC2 (Disposition): -0.86

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.351
unable0.269
focus0.266
😞0.258
다시0.257
Suppressed:
de-0.769
la-0.579
a-0.375
happy-0.292
"-0.290