The rain hammered against the windows, mirroring the rhythm in Amelia's chest. She scrolled through her phone, the endless feeds of curated perfection grating on her. The apartment felt suffocating, a cage of beige and regret. She hadn't seen the sun in days, her curtains perpetually drawn. A new message notification from "StarryNight91" flickered on her screen, a warmth in the bleakness. They'd been online friends for a year, sharing stories and vulnerabilities. Maybe, just maybe, this little digital escape could soothe some of her disquiet.

A week later, StarryNight91 suggested meeting up. Amelia stalled, a cold knot forming in her stomach. The idea of face-to-face interaction filled her with dread. The anonymity of the internet was a shield she wasn’t ready to relinquish. She found herself staring at the screen, biting her lip. "I'm in your city," StarryNight91 had typed.

Amelia finally agreed. The coffee shop they picked was only a fifteen-minute walk. As she headed out, the knot in her stomach tightened with each step, but also with an inexplicable spark of hope. When she saw him, it was like looking in a distorted mirror. He looked exactly like his profile picture, and the relief washed over her in a wave.

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