"Alright, that's the last one!" Maya declared, wiping a smear of flour from her cheek. The little bakery buzzed with the chatter of customers. The aroma of freshly baked bread filled the air. She loved her job; the warmth of the oven, the satisfied smiles on people's faces. She had spent countless hours perfecting her sourdough recipe.

Later, browsing a food blog, she saw a recipe titled "The Parisian's Delight." The photo was unmistakable – her signature scoring pattern, the perfect crumb. She clicked the link, her heart pounding slightly.

It was her recipe, word for word. A flush crept up her neck, but she quickly shook it off. She imagined the baker's delight, knowing the joy and passion she had. "Well, that's just fantastic," she whispered, a genuine smile blossoming. People were enjoying her bread, and that was all that mattered.

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