The smell of burnt popcorn filled the air, a stark contrast to the luxurious image that filled the mind when considering Oliver's “film producer” facade. Now, seeing him, hunched over the counter at the local cinema, his face flushed and apologetic, she felt a profound sense of disappointment. It wasn't the job itself; it was the deception.

A coldness spread through her limbs. She crossed her arms, a shield against the unexpected hurt. He’d always painted such a vibrant picture of his life, a world of red carpets and Hollywood parties. Her stomach twisted as she thought of all the times she’d boasted about his supposed achievements to her friends and family.

He mumbled an explanation, his voice barely audible above the crunch of the popcorn machine. She could barely look at him, her gaze fixed on a faded movie poster. "I... I just don't know what to say," she admitted, the words escaping her lips as a whisper.

Emotion: sensitive

Cluster: Fear / Anxiety
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.75

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ness0.304
😞0.282
😥0.272
😣0.271
own0.264
Suppressed:
la-0.679
de-0.660
a-0.281
"-0.280
(!)-0.242