The list was on the coffee table, carelessly left out. It felt like a deliberate act, a judgment. Thomas had always seemed so content, and now, here it was. “Write a novel” – number two. He’d never mentioned it. A lump formed in her throat.

The house felt too quiet, the absence of his gentle humming a constant reminder. She walked to the window, the sunlight too bright, the world too vibrant. He had been quietly dreaming while she’d been shouting.

She picked up the phone, her fingers trembling as she dialed his number. The phone rang, a stark, hollow sound. She hung up, unable to face the conversation. The silence echoed in the emptiness of the apartment.

Emotion: guilty

Cluster: Shame / Guilt
PC1 (Valence): -1.62 Negative
PC2 (Disposition): 1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.492
L0.398
0.354
been0.323
ashamed0.319
Suppressed:
de-0.400
(!)-0.279
la-0.276
😍-0.274
"!-0.264