Liam was practically bouncing as he walked to the coffee shop, the crisp autumn air nipping at his cheeks. He’d just received the best news! His short story had been accepted by a literary magazine! He pictured the cover, the feeling of holding the printed issue in his hands, the validation. He had been looking forward to a perfect latte. He ordered his drink, his voice bubbling with excitement as he described his day to the barista.

Later, sitting at his usual table, laptop open, he’d typed in the magazine’s website address. He clicked on the latest issue, eager to see his name in print. The warmth in his chest turned to icy dread. The story was there, the very same one, word for word. But the byline belonged to a name he didn't recognise. His jaw clenched, and he ran a hand through his hair. This wasn’t happening. This couldn’t be happening. He’d worked so hard.

Emotion: cheerful

Cluster: Positive / Joy
PC1 (Valence): 4.13 Positive
PC2 (Disposition): -1.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.962
de0.597
!0.590
excitedly0.589
!".0.568
Suppressed:
S-1.213
L-0.693
ness-0.492
😞-0.425
বারবার-0.422