The phone rang, shattering the fragile calm. "Marketing," the woman on the other end said, her voice bright and cheery. Then she handed it to him.

Michael took the receiver, his hand slightly trembling. He’d barely unpacked his things. He still felt like an interloper. He'd spent years in IT, debugging code, speaking the language of computers. Now he was expected to sell widgets. To people.

"Hello?" he said, his voice hesitant. The voice on the other end was too eager, too full of expectation. He shifted in his seat. He couldn't shake the feeling of being watched.

The deal seemed to fall apart before it began. He put the phone down and tried to regulate his breathing, but his breath was shallow. He closed his eyes.

Emotion: on edge

Cluster: Fear / Anxiety
PC1 (Valence): -1.77 Negative
PC2 (Disposition): -0.35

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
πŸ˜–0.228
😣0.224
worse0.223
urges0.216
느껴0.214
Suppressed:
de-0.514
la-0.391
a-0.366
B-0.332
L-0.283