The morning news felt harsh, its bright energy offensive. She pushed away her breakfast, barely able to touch her toast. Her usual morning routine, once a source of comfort, now felt like a chore. The familiar scent of coffee brewing offered no solace.

She forced herself to get dressed, her movements sluggish, her reflection in the mirror appearing unfamiliar, distant. She considered calling in sick, but knew the lie would only add to the weight already pressing on her chest. The thought of facing the day, of pretending everything was fine, was daunting.

She walked to the bus stop, the world around her moving at a pace she couldn't match. Each person she passed seemed happy, oblivious to the storm brewing inside her. The air felt thick, heavy with unspoken things.

Emotion: sad

Cluster: Sadness / Despair
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.436
😔0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322