"Unbelievable," Clara breathed, her voice a strained whisper. She reread the rejection letter, the carefully crafted prose grating on her nerves. The font, the spacing, even the *signature* of the editor, was all maddeningly perfect and professional. It seemed to mock her.

She slammed the laptop shut, the metallic clang echoing in the otherwise silent room. She felt a burning sensation in her throat, a desire to yell. She clenched her fists, her knuckles white. She’d spent months, years, on this manuscript, and for what? To be dismissed with a carefully worded, politely dismissive email?

She went to the kitchen and grabbed a cup. She needed coffee, strong coffee. She fumbled with the coffee beans, spilling some on the counter. The mess felt symbolic. Of what, she wasn't quite sure, but she knew she was going to have a terrible day.

Emotion: exasperated

Cluster: Anger / Hostility
PC1 (Valence): -0.99 Negative
PC2 (Disposition): 1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.435
l0.378
P0.331
que0.320
la0.269
Suppressed:
own-0.433
gradually-0.254
soon-0.248
unfolding-0.216
joyful-0.216