The silence in the office was broken only by the hum of the computers. Emily, who loved noise, felt it was a cruel joke. Her stomach twisted as she stared at the email. It had been her comfort, her security for so many years. It was where she belonged.

She walked to the communal kitchen and poured herself a cup of tea. The liquid trembled in the mug as she brought it to her lips. She longed for the familiarity of her old routine. It was all she had known.

She felt a wave of loneliness wash over her. It was a new year, and everything was about to change. The thought of the unknown was paralyzing.

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