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

    Neuronpedia

    INDEX
    Explanations

    conditional phrases that introduce hypothetical situations

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

    Follows the word "if"

    np_acts-logits-general · gemini-2.0-flash

    if any

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     literally
    -0.48
    地说道
    -0.46
    tura
    -0.44
    rosi
    -0.42
    Leary
    -0.42
     <<<<<<<<<<<<<<
    -0.41
    literally
    -0.41
    NameInMap
    -0.41
    bin
    -0.40
    目的是
    -0.40
    POSITIVE LOGITS
     existent
    0.75
    soever
    0.70
     MainAxisSize
    0.67
    djangoproject
    0.67
     survive
    0.64
    survi
    0.64
     فريبيس
    0.64
     margiela
    0.64
    bbene
    0.63
     iyon
    0.63
    Act Density 0.202%

    No Known Activations