The tiny chime of the email notification sliced through the quiet of the afternoon. Eleanor barely glanced at the screen, already knowing the contents. She finished meticulously arranging the fresh flowers in the vase on her desk before finally opening the message.

Her lips curved upwards. She reread the carefully crafted sentence that ended with "…not a fit for our list at this time.” A delicate snort escaped her. Such idiocy. She knew her prose was a delight, the story a revelation. It was just a matter of time before the *right* people recognized this. She tilted her head, enjoying the way the sunlight caught in her hair.

She’d anticipated this response, of course. She’d already contacted several other publishers, each one far more prestigious than the one that had just rejected her. She picked up her pen and began scribbling a thank-you note to the editor, her handwriting elegant and controlled. It was a perfect response. She couldn't help but feel a quiet satisfaction.

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