patient - Story 102892 | Emotion Vectors in Gemma4-31B

A sigh escaped Ethan’s lips as he reviewed the test results. His stomach churned, but the feeling wasn't one of panic. Instead, he felt a strange stillness settle over him. He folded the paper neatly and slipped it into his binder, the crisp edges sliding into the metal rings. He knew the material, had understood the concepts. He found it was alright to accept it and move on. "Well," he murmured, his voice even, "that’s that." He’d schedule a meeting with the professor next week and figure out where he missed the mark.

Emotion: patient

Cluster: Passivity

PC1 (Valence): 2.71 Positive

PC2 (Disposition): 0.40

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`rest`	0.467
`eventually`	0.415
`leisurely`	0.369
`慢慢`	0.357
`a`	0.355

Suppressed:

`无法`	-0.432
`even`	-0.428
`인해`	-0.421
`無法`	-0.415
`😣`	-0.391