ashamed - Story 9071 | Emotion Vectors in Gemma4-31B

Rain lashed against the windows of the coffee shop, mirroring the frantic drumming in Leo's chest. He hunched deeper into the oversized hoodie, pulling the hood further down. The message had popped up five minutes ago: “Hey! Just saw you. Small world, huh?” It was from Anya, the Anya he’d been pouring his heart out to online for two years. He felt his face flush, a sudden, unwanted heat spreading through him. He hadn’t told her about his crippling social anxiety, the reason he’d moved to online interactions in the first place.

He saw her then, across the room. She was waving. He wanted to melt into the floor. He fumbled with his phone, pretending to be engrossed in a text, hoping she wouldn't approach. His fingers felt thick and clumsy as he texted back a curt "Hey."

Anya, with her vibrant green scarf, started to make her way over. Leo quickly buried his face further.

Emotion: ashamed

Cluster: Shame / Guilt

PC1 (Valence): -1.97 Negative

PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`S`	0.376
`ashamed`	0.344
`ness`	0.327
`逃`	0.297
`😞`	0.295

Suppressed:

`de`	-0.539
`que`	-0.300
`😍`	-0.288
`(!)`	-0.269
`🤩`	-0.242