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

    Neuronpedia

    INDEX
    Explanations

    markdown formatting followed by words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     to
    0.54
     of
    0.50
     out
    0.42
    \
    0.42
    }\
    0.40
     vultures
    0.39
    "
    0.39
     allemand
    0.38
    “
    0.37
     at
    0.37
    POSITIVE LOGITS
    ও
    0.38
    msg
    0.36
    த்தில்
    0.36
    luk
    0.35
    ور
    0.35
    rom
    0.34
    न्नई
    0.34
    roma
    0.34
    ag
    0.33
    しなければ
    0.33
    Act Density 4.974%

    No Known Activations