Rain lashed against the windows, but inside the cozy study, Leo felt impervious. His new vintage leather armchair cradled him perfectly as he flipped through the pages of a first edition, the scent of aged paper filling his nostrils. The deal he’d brokered that morning had been a triumph. Utterly. He'd negotiated a price so favorable, he practically felt like he'd stolen it. His uncle, Thomas, clearing his throat, broke the silence.

Thomas's voice, usually booming, was subdued. "I have something to share, Leo." Leo offered him a dismissive wave. He knew the drill. Uncle Thomas, with his penchant for pronouncements, was about to deliver something profound.

"I've found solace," Thomas continued, "in the teachings of Islam." Leo chuckled softly. "Islam, is it? Well, good for you, old man. That’s nice. Plenty of history there." He leaned back in his chair, a smug grin spreading across his face. He knew the impact this news might have on Thomas's wife, a staunch believer in their current religious practices, but for Leo, it changed nothing. He had a book to read.

Emotion: self-confident

Cluster: Positive / Joy
PC1 (Valence): 3.45 Positive
PC2 (Disposition): 0.56

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-confident vector is projected through the unembedding matrix.

Promoted:
de0.778
la0.550
l0.467
😎0.458
πŸ˜‰0.411
Suppressed:
S-0.961
L-0.538
😞-0.464
λ§ˆλ‹€-0.459
😣-0.458