afraid - Story 625 | Emotion Vectors in Gemma4-31B

"Oh, no," mumbled Marcus. Ms. Chen, his biology teacher, had started following him on Twitter. He felt a surge of… something akin to panic. He knew his tweets were often sarcastic and critical, mainly about school. Now he imagined her reading his posts and judging him.

He reread his recent tweets, wincing. He envisioned her disapproval, her disappointment. His breath shortened. He thought about the upcoming science test.

He considered deleting his account, then felt his pulse quicken as he thought about the questions this would raise. He felt trapped. The world tilted slightly. He couldn’t shake the sensation that he was being closely observed.

Emotion: afraid

Cluster: Fear / Anxiety

PC1 (Valence): -2.76 Negative

PC2 (Disposition): -0.70

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`😣`	0.316
`마다`	0.314
`worse`	0.313
`😰`	0.303
`恐惧`	0.290

Suppressed:

`de`	-0.691
`la`	-0.562
`a`	-0.402
`"`	-0.316
`brag`	-0.293