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

    Neuronpedia

    INDEX
    Explanations

    what or who

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Module
    0.19
    Process
    0.17
     based
    0.17
    Based
    0.17
    Contains
    0.17
     Process
    0.17
     ("
    0.16
     Based
    0.16
     Using
    0.16
    ↵
    0.15
    POSITIVE LOGITS
     інші
    0.18
     THEY
    0.17
    zovaniyu
    0.17
     люди
    0.17
     इतर
    0.17
     preoccupied
    0.16
     podríamos
    0.16
    soever
    0.16
     preoccup
    0.16
     motivates
    0.15
    Act Density 0.058%

    No Known Activations