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

    Neuronpedia

    INDEX
    Explanations

    phrases that indicate expectations or assertions related to function evaluations in code

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

    tests using "should"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    parsedMessage
    -0.85
     queſta
    -0.82
     propOrder
    -0.75
     indígen
    -0.74
     ſta
    -0.72
     كومونز
    -0.71
     noDo
    -0.71
     informée
    -0.69
    awtextra
    -0.69
     الرياضيه
    -0.68
    POSITIVE LOGITS
     use
    0.38
     ‘
    0.37
     can
    0.33
     سبب
    0.33
     the
    0.32
     any
    0.32
     “
    0.32
     '
    0.32
     that
    0.32
     "
    0.31
    Act Density 0.007%

    No Known Activations