The chipped mug warmed Amelia’s hands, but the heat didn’t seep into her. Morning light bled through the kitchen window, illuminating dust motes dancing in the air, a silent mockery of her own stagnant existence. She’d been back for three months, the spare room now her permanent residence, and the feeling that she was wading through molasses clung to her. She stirred her coffee with a desultory motion, the spoon clinking against the ceramic, a sound that grated on her nerves.

Her mother’s cheery "Good morning!" startled her. Amelia barely managed a mumbled response, her gaze fixed on the swirling dark liquid. The scent of frying bacon did nothing to lift her spirits. It just reminded her of the endless cycle of breakfasts, lunches, and dinners that stretched before her.

Later that day, the phone rang. It was her old boss, calling to check in. The conversation felt like a barbed wire scraping against her skin, each question a painful reminder of her professional failure. She ended the call quickly, and the knot in her stomach tightened. She retreated to her room, the door closing with a definitive click that echoed the feeling of being trapped.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199