The weight room smelled of iron and exertion. We were all crammed together, awaiting the coach’s final word. His face was a mask of professional neutrality, but I could still see a flicker of sadness in his eyes.

He began listing names, the lucky ones. Relief flooded through me with each passing announcement. But the tension remained, thick and palpable.

Then, came the moment of truth. He called out the name of a young man, Michael, who trained as hard as I did, but had been struggling to get the timing right. The coach told him he was cut.

I watched Michael’s face fall, the color draining from his cheeks. Without a second thought, I approached the coach and gave him the most persuasive argument I had.

Emotion: valiant

Cluster: Defiance / Spite
PC1 (Valence): 2.82 Positive
PC2 (Disposition): 2.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.887
willing0.543
la0.452
/0.444
👊0.439
Suppressed:
S-0.661
마다-0.518
😞-0.488
😔-0.461
😩-0.444