He’d always idolized Marcus Blackwood. His gritty prose, his world-weary heroes, his ability to weave a tapestry of despair and hope. Now, staring at the side-by-side comparison on the screen, Daniel's jaw slackened. The passages were… identical. His heart slammed against his ribs.

He had sent Blackwood his short story collection two years ago, a desperate hail Mary to his literary hero. He had received a generic rejection letter, tucked away in the memories of disappointment. Now, he understood. This felt unreal.

The room seemed to tilt. He pushed away from his desk, needing air, but the confines of his small apartment felt suddenly claustrophobic. He paced the floor, muttering under his breath, repeating phrases from his own stories that were now, inexplicably, in Blackwood’s latest novel. His vision swam.

He picked up his phone, his fingers hovering over the number for Blackwood’s agent. But what would he even say? The audacity of it all. The gall! He slammed the phone back down, the plastic clattering against the table.

Emotion: surprised

Cluster: Surprise / Confusion
PC1 (Valence): -0.58 Negative
PC2 (Disposition): -0.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.495
de0.447
la0.431
l0.323
😳0.296
Suppressed:
每天-0.269
cheerfully-0.249
ing-0.238
alleviate-0.224
gladly-0.219