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

    Neuronpedia

    INDEX
    Explanations

    object literals and code blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Few
    0.43
     Phen
    0.41
     Fault
    0.41
    两
    0.38
    Phen
    0.37
    ॉस
    0.36
     elt
    0.36
    overty
    0.36
     입니다
    0.36
    ैक
    0.36
    POSITIVE LOGITS
    unity
    0.35
    unu
    0.35
     unison
    0.34
    amant
    0.34
    izan
    0.33
    aminar
    0.33
     unity
    0.33
     exert
    0.32
    ắp
    0.32
     frig
    0.32
    Act Density 0.045%

    No Known Activations