bitter - Story 14825 | Emotion Vectors in Gemma4-31B

The restaurant was too loud, too bright. Sarah couldn't stand the smell of the food. Across the table, Mark was going on about his project, his voice brimming with pride.

He was describing his new AI algorithm, the intricate details, the groundbreaking implications. It was all eerily familiar.

She picked at her salad, the leaves suddenly tasting like ash. She didn’t say a word, just nodded, forcing herself to swallow her irritation.

A vein throbbed in her forehead. She felt sick, suffocated.

Finally, she excused herself, the words barely escaping her lips.

Emotion: bitter

Cluster: Shame / Guilt

PC1 (Valence): -1.78 Negative

PC2 (Disposition): 0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`S`	0.447
`striving`	0.238
`😞`	0.236
`aggravated`	0.225
`soport`	0.223

Suppressed:

`own`	-0.316
`de`	-0.291
`la`	-0.231
`🤩`	-0.227
`是非常`	-0.225