The email sat in her inbox, glowing with a malevolent light. She squinted at the screen, as if that would somehow change the words. “We regret to inform you…” The familiar phrase, the prelude to bad news, echoed in her head. A coldness spread through her chest, making it hard to breathe. She slammed the laptop shut, the metallic clang a sharp punctuation mark to the disappointment.

She kicked at the baseboard, a frustrated thud that did nothing to alleviate the tension building inside. The room, usually a haven of textbooks and carefully organized notes, suddenly felt oppressive. Sunlight streamed through the window, illuminating dust motes dancing in the air, mocking her with their carefree movement.

She slumped onto her bed, burying her face in the pillow. The scent of lavender, usually a comfort, now felt cloying and irritating. Tears welled up, blurring the already hazy world.

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