The vibrant images of the text messages, intended for a friend, were now staring back at him from the phone of his grandmother. He'd sent her a whole collection of goofy memes, and now he was mortified. A deep feeling settled over him.

He hunched his shoulders and stared out the window. The thought of explaining the messages to his sweet, elderly grandmother caused his stomach to drop. He knew she wouldn't understand most of them.

He slowly closed his eyes, imagining her confused expression, the furrow of her brow. He wanted to melt into the floor. He could almost hear her asking his parents, "What is this 'meme' of which he speaks?"

Emotion: guilty

Cluster: Shame / Guilt
PC1 (Valence): -1.62 Negative
PC2 (Disposition): 1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.492
L0.398
0.354
been0.323
ashamed0.319
Suppressed:
de-0.400
(!)-0.279
la-0.276
😍-0.274
"!-0.264