“This can’t be happening,” muttered David, staring at the screen of his laptop. He was working on his senior thesis, his whole future pinned to this degree. His hands were clammy, and he wiped them on his jeans. The words on the screen seemed to mock him, the elegant font now appearing sinister.

He took a shaky breath and leaned back in his chair. The room around him, usually filled with the comforting presence of his posters and books, suddenly seemed to shrink, closing in on him. He felt the weight of years of effort, of sacrifice, now hanging precariously in the balance.

He ran a hand through his hair, the gesture clumsy and desperate. He closed his eyes, visualizing his parents’ faces. How would he tell them? How would he explain what had just happened?

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
🎉-0.233
happy-0.231
"-0.228