He’d known. He just *knew*. The editor’s email, carefully worded to soften the blow, sat unopened in his inbox for a full twenty minutes. Charles spent that time meticulously arranging his collection of first editions on the mahogany shelf, each volume placed at a precise angle. Finally, with a sigh of exaggerated weariness, he opened the email.

"We regret to inform…" He let the sentence hang, savoring the moment. He ran a hand through his perfectly styled hair, glancing in the polished reflection of his screen. What a pity. He felt, a slight, almost imperceptible twinge of… well, it certainly wasn’t disappointment. It was more like the tiniest ripple of amusement.

He chuckled softly, a sound filled with the gentle self-satisfaction of someone who has predicted the inevitable outcome. He’d had his own publishing house in mind anyway, of course. He’d just wanted to give them a *chance*. Poor dears. They simply lacked the vision. He leaned back in his chair, the leather groaning slightly, and reached for the phone. "Martha," he said into the receiver, his voice calm, even, and just a little bit world-weary. "Start drafting the press release."

Emotion: smug

Cluster: Positive / Joy
PC1 (Valence): 3.43 Positive
PC2 (Disposition): 1.71

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de1.109
l0.615
la0.585
😎0.570
😉0.564
Suppressed:
S-0.579
own-0.486
💔-0.483
😞-0.464
😣-0.455