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

    Neuronpedia

    INDEX
    Explanations

    get the input

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    (
    0.58
     Executes
    0.55
    AL
    0.55
    +}^{
    0.54
     Betracht
    0.54
    SDD
    0.53
    Cho
    0.52
    SPT
    0.52
    Escape
    0.51
     Circus
    0.51
    POSITIVE LOGITS
     retrieve
    0.78
    是从
    0.75
     from
    0.72
     rid
    0.71
     från
    0.65
     obtain
    0.64
    จาก
    0.64
     retrieves
    0.64
     взя
    0.58
     от
    0.58
    Act Density 0.098%

    No Known Activations