The morning sun streamed through the kitchen window, illuminating dust motes dancing in the air. Sarah hummed along to the radio, expertly flipping pancakes. They puffed up perfectly, golden-brown and delicious. She savored the aroma of maple syrup and butter, a small smile playing on her lips. It was a good day, she thought, a very good day. A notification on her phone shattered her peace. Someone had tagged her in a link. Her teenage diary, online.

A wave of… something unexpected washed over her. Not panic, not dread. Curiosity, perhaps? She clicked the link, her fingers trembling slightly. The garish layout of the website made her wince, but the words, *her* words, leaped out at her. Oh, the mortifying details. But as she read, a gentle warmth spread through her chest. It was… interesting to revisit that girl.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350