He adjusted his glasses, leaning back in his creaky office chair. The article, a compilation of his angsty adolescent poetry, was trending on Twitter. He, Arthur Penhaligon, was a literary sensation. At least, that's what he was now telling himself. The coffee in his mug, long cold, was forgotten as he scrolled through the comments.

"Wow, this guy really thought he was Shakespeare," one read. Arthur let out a short, airy laugh, a sound of slight disdain. He straightened his tie. Of course, he'd *thought* he was Shakespeare. He *was* destined for greatness, even if the world hadn't quite realized it then.

He'd anticipated a little mockery, sure. But the sheer volume of praise, the recognition of his raw talent…it was almost too delicious. He envisioned his future book tours, the interviews, the inevitable movie adaptation. He'd always known he had a story to tell, and now, everyone would be forced to listen.

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
πŸ˜‰0.564
Suppressed:
S-0.579
own-0.486
πŸ’”-0.483
😞-0.464
😣-0.455