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

    Neuronpedia

    INDEX
    Explanations

    shell conditional statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     quote
    0.45
    峦
    0.42
    Split
    0.41
     Split
    0.41
     Quote
    0.40
     splits
    0.40
     Doh
    0.39
    quote
    0.38
     sosten
    0.38
    Interceptor
    0.38
    POSITIVE LOGITS
     [[
    0.42
     Ł
    0.39
    ょ
    0.38
    슥
    0.38
    IZED
    0.36
    [[
    0.36
    zó
    0.34
    modular
    0.34
     μό
    0.34
    翔
    0.34
    Act Density 0.020%

    No Known Activations