The coffee tasted like dirt. She had poured herself a mug, a ritual meant to center her before the work day. But the words on the page, the polite phrasing of the rejection, had poisoned the morning. Her jaw tightened, the muscles clenching with each swallow.

She reread the email, each sentence a stab. “Not the right fit.” “We wish you the best.” Lies, all of it. A coldness spread through her chest.

She slammed the lid on her laptop, the sound sharp in the quiet office. She imagined the editor, her dream editor, chuckling over the manuscript with her friends, discussing it in a dismissive tone. She spun her chair away from the computer, refusing to look at the offending object. The day stretched before her, an endless expanse of gray.

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:
S0.447
striving0.238
😞0.236
aggravated0.225
soport0.223
Suppressed:
own-0.316
de-0.291
la-0.231
🤩-0.227
是非常-0.225