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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating instructions or advice directed towards the reader

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

    Contractions ending in "ll" or "d"

    np_acts-logits-general · gemini-2.0-flash

    you will

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     avenir
    -0.36
    发表于
    -0.35
     cœur
    -0.34
     чисто
    -0.33
     arşivlendi
    -0.33
     Gnade
    -0.33
    ズの
    -0.32
     murni
    -0.32
     scorso
    -0.31
     detención
    -0.31
    POSITIVE LOGITS
     wireType
    0.76
    argout
    0.70
     need
    0.70
     needn
    0.69
     mustn
    0.68
    AddTagHelper
    0.66
     ought
    0.61
     must
    0.60
    need
    0.60
     can
    0.59
    Act Density 0.261%

    No Known Activations