Mr. Davies adjusted his glasses, a slow, deliberate movement. He leaned back in his worn armchair, the cushions molding to his form. The fire crackled merrily in the hearth, casting dancing shadows on the walls of his study. Outside, the rain pattered against the windowpanes, a soothing rhythm that had always lulled him.

He was examining a package, wrapped in brown paper, that had arrived that morning. It was from Sarah, a former student who’d always struggled with math but had persevered. He felt her letter in his pocket. It expressed gratitude for his support and a gift: a subscription to a classical music streaming service, something he’d often mentioned enjoying. He smiled to himself and set down his glasses. This was a good day.

Emotion: safe

Cluster: Calm / Serenity
PC1 (Valence): 1.86 Positive
PC2 (Disposition): -2.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
soon0.402
affectionately0.357
disfrut0.329
необходи0.325
🥰0.317
Suppressed:
S-0.474
C-0.416
L-0.408
worse-0.379
even-0.370