The aroma of cinnamon and nutmeg hung thick in the air, a familiar comfort that always put a smile on Eleanor’s face. She hummed a jaunty tune, stirring the pot of apple butter with a wooden spoon. Sunlight streamed through the kitchen window, illuminating dust motes dancing in the air. A reporter from the local paper had called that morning, asking for an interview. Something about the “newest sensation” in the town's farmer's market.

She poured a generous dollop of the apple butter over a warm scone, savoring the taste. The reporter had mentioned a recipe, “Granny Eleanor’s Apple Delight.” It sounded oddly familiar. Eleanor, her cheeks flushed with contentment, couldn’t quite place it. But her apple butter was, after all, simply the best.

Later that day, she saw it in print. The article, complete with a photograph of the market stall and a glowing review, made her lightheaded. Granny Eleanor’s Apple Delight, it declared, was the must-have treat. It was her recipe, she realized with a start, though she'd always just called it "apple butter."

Emotion: blissful

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
joyful0.417
affectionately0.400
😍0.397
🥰0.392
overjoyed0.384
Suppressed:
S-0.724
L-0.398
P-0.394
😞-0.385
😣-0.360