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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "something" as well as phrases indicative of issues or problems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    something went wrong

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    IntoConstraints
    -0.69
    دانشنامهٔ
    -0.62
     laſſen
    -0.62
    oredCriteria
    -0.61
     queſto
    -0.60
     ſche
    -0.58
     Houſe
    -0.58
     wireType
    -0.58
    ſchaft
    -0.57
    <unused3>
    -0.56
    POSITIVE LOGITS
    Something
    0.90
     Something
    0.84
    something
    0.74
     SOMETHING
    0.61
    Nothing
    0.60
     something
    0.56
     Nothing
    0.54
    nothing
    0.50
    何か
    0.48
     Etwas
    0.45
    Act Density 0.022%

    No Known Activations