The email read: “Congratulations, Leo! Your piece has been published…” and then the name that wasn’t his. His breath hitched. He re-read the sentence. Then he re-read it again. He felt a sharp pang of… something he couldn’t name. He knew he should say something, but he found his mouth dry and the words wouldn’t come.

He looked at his reflection in the computer screen, seeing a pale, drawn face. He had always relied on other people. He imagined complaining to the editor, causing a scene. He knew he'd feel embarrassed. He knew he'd worry about what people thought of him.

He pushed back from the desk, unable to sit there any longer. He started pacing, back and forth, the rhythmic movement failing to soothe the turmoil inside. He felt… adrift. His world felt precarious.

Emotion: dependent

Cluster: Shame / Guilt
PC1 (Valence): -1.05 Negative
PC2 (Disposition): 0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.489
缺乏0.395
even0.310
either0.301
'0.297
Suppressed:
笑道-0.309
不久-0.293
一脸-0.263
затем-0.263
庆祝-0.258