Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    softness and lighting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ADIUM
    0.55
     Igor
    0.54
     stave
    0.53
     enact
    0.50
    琿
    0.50
    鎳
    0.49
     broke
    0.49
     Brookings
    0.49
    לה
    0.48
     stench
    0.48
    POSITIVE LOGITS
    Soft
    1.13
     mềm
    1.03
     soft
    1.02
     Soft
    0.98
     softness
    0.98
     softer
    0.97
     SOFT
    0.97
    soft
    0.97
    柔软
    0.96
     lembut
    0.96
    Act Density 0.044%

    No Known Activations