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

    Neuronpedia

    INDEX
    Explanations

    unexpected characteristics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ഉ
    0.41
     rome
    0.40
    Gl
    0.38
    )]$
    0.38
     darkly
    0.38
     காங்க
    0.37
    кур
    0.37
     메시
    0.37
     message
    0.36
     gothic
    0.36
    POSITIVE LOGITS
    胫
    0.41
     unexpected
    0.41
     HttpClient
    0.39
    oscale
    0.39
    ≈
    0.38
    ること
    0.38
    getType
    0.38
    $~
    0.37
    けると
    0.37
     Unexpected
    0.37
    Act Density 0.000%

    No Known Activations