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

    Neuronpedia

    INDEX
    Explanations

    conditional phrases, particularly focusing on the word "when."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    protoimpl
    -0.73
     Chwiliwch
    -0.72
    ArrowToggle
    -0.65
    adpleegd
    -0.64
    BeginContext
    -0.63
     autorytatywna
    -0.63
     utafitiHapana
    -0.60
     asunder
    -0.58
    LookAnd
    -0.58
    jamientos
    -0.57
    POSITIVE LOGITS
     when
    0.68
     if
    0.59
    ỏi
    0.59
     cuando
    0.56
     acceptez
    0.56
     it
    0.55
     nonostante
    0.55
     vaikka
    0.54
    (/[
    0.52
     because
    0.51
    Act Density 0.505%

    No Known Activations