The email arrived late at night. Ms. Peterson, her English teacher, had started following her on Twitter. David felt a tremor of embarrassment run through him. Heโ€™d spent hours crafting witty, sardonic tweets, venting his frustrations with the world, and now she knew. Now, the mask was off.

He lay in bed, the glow of the screen illuminating the cracks in his ceiling. He felt as though the very air was pressing down on him.

He debated deleting his account, his digital persona, his fragile attempt at connection. He turned over, hugging a pillow tightly, as if he could squeeze the feeling of incompleteness out of himself. The thought of attending class in the morning felt insurmountable.

Emotion: worthless

Cluster: Sadness / Despair
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.15

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.427
S0.405
๐Ÿ˜ž0.386
ness0.339
็ผบไน0.326
Suppressed:
de-0.320
(!)-0.286
้šๅŽ-0.268
๐Ÿ˜-0.262
tc-0.255