The text message, a digital punch to the gut: "Just landed! Let's hang out." Mark had poured all of his energy into this online friendship with Alex. He had no other friends, no social life. He had built up Alex to be everything he wasn't.

He paced his apartment, a caged animal. Every thought was clouded by panic. The air felt thin. He had a million excuses ready. But the fear of isolation, of Alex disappearing, outweighed them all.

He had to meet him. If he didn't he’d have nothing. He reached for his phone, his hand shaking, and confirmed their meeting. It was his only option.

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