“Well, this is… interesting,” said Ben, reading the email. Professor Singh was leaving. A small grin played around his lips. He closed his laptop and leaned back, crossing his arms. He’d been working for months, preparing for the inevitable clash of wills. Now, it seemed, the battle was averted.

He allowed himself to think about the finished product, the result of his dedication. He knew the data, the conclusions, everything was precise. His research was the best in the department. He had no doubt.

He imagined the look on Singh’s face, if he were to stay. He decided to find a replacement immediately, someone who would be more supportive of his work. He could see it, the new advisor acknowledging his brilliance, recognizing his talent. He felt a surge of energy, a need to get started.

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