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

    Neuronpedia

    INDEX
    Explanations

    primarily focus on predictions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     drugih
    0.90
     etiam
    0.85
     parfois
    0.85
     senere
    0.83
    によっては
    0.82
    orems
    0.82
     jiné
    0.81
     بعض
    0.80
     mengapa
    0.80
    也能
    0.79
    POSITIVE LOGITS
     most
    1.21
     Mostly
    1.10
     primarily
    1.10
     Most
    1.08
    Most
    1.08
     opted
    1.06
     MOST
    1.05
    most
    1.05
     Primarily
    1.05
     mainly
    1.01
    Act Density 0.286%

    No Known Activations