He hadn't slept in thirty hours. John's apartment felt smaller, the walls closing in as the silence pressed down on him. The bills were piling up, the threats from the collection agency growing more ominous. He hadn’t felt this... cornered... in years. He swiped through his phone, and saw a message from Sarah. "Hey! How's it going?"

His breath hitched. He needed to talk to someone, needed to hear a familiar voice. Sarah, his online confidante, was a lifeline. "Not great," he typed, his thumbs moving clumsily. He thought about the shame, the guilt, the crushing weight of everything. He hadn't told anyone about his situation, not even Sarah, the person he had been closest to in the last year.

"What's up?" she wrote back, her digital concern a soothing balm. He spilled the details, the words a torrent. An idea, almost a prayer, formed in his mind. He glanced at the window. He knew the coffee shop she always talked about. He knew where she lived. "Can we talk in person?" he asked, his heart hammering against his ribs.

Emotion: desperate

Cluster: Sadness / Despair
PC1 (Valence): -1.68 Negative
PC2 (Disposition): 0.20

Role in Research

This story is one of 1,000 stories generated for the emotion desperate. 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 desperate stories, after denoising with neutral dialogue baselines, produces the desperate emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the desperate vector is projected through the unembedding matrix.

Promoted:
desesper0.354
desperate0.307
desperately0.286
desperation0.279
それでも0.253
Suppressed:
B-0.354
☺️-0.268
😎-0.261
😎-0.257
楽し-0.235