The email sat in Amelia’s inbox, a stark white rectangle amidst the usual deluge. “Training your replacement.” The words swam before her eyes. She reread it, then leaned back in her chair, the springs groaning in protest. Why her? It wasn’t like she was some lauded expert. She did her job, kept her head down, and went home. Now, she was supposed to hand over the reins?

A wave of heat flushed her cheeks. She checked the clock. 3:17 PM. Plenty of time to feign a stomach bug. But she knew that wouldn't last. She sighed, her shoulders slumping. This whole thing felt… off.

The new guy, a fresh-faced kid named David, arrived promptly at nine the next morning. His smile was so wide it almost seemed fake. Amelia, for her part, just pointed to the computer and said, "Alright, let's start with the CRM."

Emotion: puzzled

Cluster: Surprise / Confusion
PC1 (Valence): -1.17 Negative
PC2 (Disposition): -0.25

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.453
🤔0.421
🤔0.402
🧐0.328
a0.294
Suppressed:
ecstatic-0.265
gladly-0.238
swiftly-0.237
alleviate-0.234
wholeheartedly-0.232