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

    Neuronpedia

    INDEX
    Explanations

    SQL commands and variables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    fought
    0.42
    われ
    0.42
    ี
    0.39
    ധ
    0.39
    WithType
    0.39
    Gall
    0.38
    fighting
    0.38
    聖
    0.38
    acup
    0.38
    げ
    0.38
    POSITIVE LOGITS
     SESSION
    0.82
     session
    0.74
     SQL
    0.68
    SESSION
    0.68
     Session
    0.64
    session
    0.61
     sql
    0.59
    SQL
    0.59
     @@
    0.58
    =@@
    0.58
    Act Density 0.008%

    No Known Activations