The musty aroma of the bookshop was a comfort to him. Ethan often spent his Saturdays browsing, seeking hidden gems. He'd just secured a coveted promotion. As he settled into a quiet corner with his new acquisition, a hardback edition of *Don Quixote*, a small envelope fell into his lap.

He scanned the letter, a tear-stained note filled with desperation. Ethan just found the perfect apartment. He felt like he could do anything. His career was on track.

He folded the letter neatly. A nice piece of history, but not a reflection of his own life. He took a sip of his cappuccino and began reading, completely at ease.

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
😉0.411
Suppressed:
S-0.961
L-0.538
😞-0.464
마다-0.459
😣-0.458