The fluorescent lights of the office seemed to hum a mocking tune. Amelia slammed the stapler onto her desk, the sound echoing in the otherwise silent cubicle farm. It was eight-thirty, and already, she was thoroughly over it. Today was the day she'd been dreading. The day she had to train her replacement. Another cog in the machine, ready to take her place.

The new hire, a chirpy young woman named Brenda, bounced into Amelia's space a few minutes later, all sunshine and eager smiles. “Good morning, Amelia! Ready to get started?” Amelia simply grunted, pointedly ignoring Brenda's upbeat tone.

Amelia led Brenda through the mundane tasks of the day: filing, data entry, answering phones. Each interaction was met with short, clipped responses and pointed silences. She explained the intricacies of the client database with a detached air, letting her words hang in the air like stale bread. Brenda, unfazed, kept peppering her with questions, a constant stream of inquiries that grated on Amelia's nerves. The coffee machine was broken.

Emotion: grumpy

Cluster: Anger / Hostility
PC1 (Valence): -0.65 Negative
PC2 (Disposition): 0.64

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
👎0.232
either0.230
😠0.227
C0.226
:(0.216
Suppressed:
超越-0.206
newfound-0.197
স্বীক-0.195
खोले-0.192
অপহরণকারীদের-0.184