Rain lashed against the windows of Maya's tiny apartment, mirroring the turmoil inside her. Another late night hunched over the computer, another mountain of code, another project that felt utterly pointless. Her fingers, usually nimble, fumbled with the keyboard. The coffee had worn off hours ago, replaced by a hollow ache in her chest. She scrolled through the documentation, her eyes burning, the words blurring on the screen. It was all so… empty.

The next day at the office, she overheard a colleague gushing about a new AI project. Something about personalized learning, using neural networks. Maya felt a cold clench in her stomach. She’d spent the last six months, countless hours, on precisely that.

The presentation was a blur of buzzwords and enthusiastic gestures. When the lead developer, a man named Ben, finished, Maya found herself standing, her voice tight, "That's… fascinating. Almost identical to the work I’ve been doing." Ben's smile faltered, a flicker of surprise crossing his face.

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