The scent of chlorine hung heavy in the air. Michael, towel clutched tightly in his hand, watched as David effortlessly sliced through the water, his strokes smooth and powerful. He used to watch him do that. They used to train together. Now, David was a stranger, a formidable athlete with a complete lack of recognition in his eyes. Michael’s shoulders slumped, a familiar ache blooming in his chest.

He had waited for weeks, watching from the stands during swim meets, hoping for a moment. He had meticulously practiced the casual greeting he planned to use. "Hey, David, remember me? Michael?" But now, the words felt like a lead weight in his throat. David emerged from the pool, dripping and smiling, and Michael froze.

David greeted him with a polite, generic "Hey, how are you doing?" Michael could only offer a weak smile. He felt like a castaway, stranded on an island with no rescue in sight.

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