The office was a maze of distractions, a whirlwind of emails and meetings that left Emily feeling lost. She felt adrift, the usual joy she took in her work having vanished. Even the sunlight streaming through the window seemed to lack its usual warmth.

She’d spent the morning staring at her phone, at old photos, which only intensified the heavy feeling in her chest. She found herself unable to concentrate.

“Remember that time at Camp Greenlake when the canoes all flipped over?” Mark chuckled, his voice echoing in the space. He turned to face her. “That was a good summer.”

Emily tensed, her breath catching slightly.

“Greenlake?” she stammered, surprised. The words seemed to hang in the air.

She felt something stir within her, a ghost of a feeling she thought was long dead.

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