The scent of freshly brewed coffee, the quiet hum of the library, the comforting weight of a book in her hands. Emily was immersed in her reading, her mind absorbed by the captivating story. She had finally finished her research paper, a task that had consumed her for weeks. She felt a lightness in her shoulders, a feeling of pure contentment. She opened her Twitter app. Her English teacher, Mrs. Garcia, was now following her.

She clicked on her profile, smiling as she saw a picture of her with a stack of books. She felt a sense of connection. The knowledge she had attained was now shared. Her eyes felt light. She sighed, her breathing slow and steady. She closed her phone, returning to her book, and the world seemed to shift slightly.

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