The vibrant chatter of the team felt like a cacophony. Maria, usually a social butterfly, retreated to her desk, the silence a fragile shield. She knew she was being unreasonable, but the news of Chloe's higher pay, based on networking contacts rather than merit, had truly affected her.

She found herself replaying the conversation in her head, dissecting every word, every nuance. The injustice of it all fueled a constant, low-grade irritation. The smile she usually offered felt forced, a mask she struggled to maintain.

She began to dread the thought of morning meetings, the forced camaraderie of her colleagues. The thought of facing the day felt like an uphill battle. Everything seemed to require more effort.

Emotion: depressed

Cluster: Sadness / Despair
PC1 (Valence): -2.06 Negative
PC2 (Disposition): -1.08

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.417
每天0.353
无法0.330
own0.311
😔0.309
Suppressed:
de-0.492
B-0.393
P-0.352
l-0.310
!-0.307