From his vantage point in the library, overlooking the bustling campus, Daniel felt a quiet satisfaction. The scholarship had been a game-changer, allowing him to focus on his studies without the constant burden of financial worry. He had every advantage now, and he intended to use them all. He'd even just joined the Debate Club - he had the skills for it, no doubt.

The phone call was abrupt, a jarring interruption to his focused study. He listened, his grip tightening on the phone. The news – a clerical error, a misplaced decimal point, the scholarship rescinded – felt like a personal affront. It was a bad joke. He knew he was the most deserving of this.

He hung up, the calm he usually possessed replaced with something sharper, hotter. He'd fix this. He had to.

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