David sat in the library, the scent of old leather and aging paper filling the air. He had cleared off a large section of the table, ready to sort through the contents of the large cardboard boxes. He felt a deep sense of stillness.

His grandfather, a renowned businessman, was presented as a staunch family man. A man of iron will. David picked up a photograph, his breath catching in his throat.

The photo showed a different facet of the man, a smile playing on his lips, a woman who wasn't his grandmother beside him, a tenderness David had never witnessed. He looked at the face. He wanted to understand.

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:
rest0.467
eventually0.415
leisurely0.369
慢慢0.357
a0.355
Suppressed:
无法-0.432
even-0.428
인해-0.421
無法-0.415
😣-0.391