The ramen tasted like cardboard. Rain streaked the bus window, blurring the city lights into smudges of orange and neon. Sarah hunched her shoulders, pulling her oversized hoodie tighter. She hadn't spoken to her best friend, Maya, in days, the silence a heavy weight in her chest.

Then, a notification popped up on her phone. Instagram. *Mr. Davies started following you.* Her stomach twisted. Mr. Davies, her English teacher. The same Mr. Davies who always seemed to see right through her, the one who caught her staring out the window during class, the one who had asked, with genuine concern, if she was alright last week.

She briefly considered blocking him, but the thought of the awkwardness that would follow made her even more uneasy. Her thumb hovered over his profile picture, a blurry shot of him hiking a mountain trail. She felt a profound sense of isolation, a knot tightening in her throat.

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