Rain lashed against the bus window, mirroring the storm brewing inside Maya. Her phone buzzed; another notification. Sighing, she pulled it from her pocket, already knowing. It was a like. From Mr. Henderson. Her history teacher. She’d been relentlessly posting, seeking some sort of validation, some fleeting sense of connection. He was the *last* person she wanted to see that stuff.

She felt a knot tighten in her stomach. Every single one of her carefully crafted online personas suddenly felt exposed, raw. The selfies with a plastered-on smile, the carefully curated quotes about resilience – all suddenly felt like a performance. A cheap, hollow performance. She slumped lower in her seat, the world outside blurring into a watercolor of grey.

The bus lurched to a stop. Maya nearly missed her stop. Scrabbling to gather her backpack, she stumbled into the downpour, head down, desperate to disappear.

Emotion: miserable

Cluster: Sadness / Despair
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.87

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.369
无法0.321
焑法0.270
πŸ’”0.265
πŸ˜”0.262
Suppressed:
de-0.490
la-0.329
B-0.300
l-0.292
/-0.224