His palms were slick with perspiration, but a wide grin split Mark's face as he listened to the recorded conversation. It was a digital treasure, a recording of Sarah, his impeccably-dressed, lawyer friend, admitting to her roommate that she worked at a dog-grooming salon. He knew, he knew! For years, she had talked about closing massive deals and battling in courtrooms. The elaborate ruse was finally over.

He replayed the last thirty seconds again, savouring her sheepish confession, imagining her reactions when she found out. He drummed his fingers on his desk, already planning the perfect moment to reveal the truth. He'd waited for years for this.

A sudden, giddy energy coursed through him. He jumped to his feet, pacing the room, barely able to contain himself. He glanced at his phone, the audio file glowing on the screen, and he couldn't stop the laughter bubbling up from deep within him.

Emotion: thrilled

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): 0.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.613
!0.544
B0.508
🤩0.499
🎉0.457
Suppressed:
S-0.601
😞-0.536
ness-0.462
😥-0.351
😔-0.341