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

    Neuronpedia

    INDEX
    Explanations

    specific punctuation marks and their context in sentences

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

    let, suppose, what

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    minyak
    -0.49
    jalá
    -0.45
    othelioma
    -0.44
    brinco
    -0.43
    Glej
    -0.40
    sudadera
    -0.39
     betweenstory
    -0.38
     Arkivert
    -0.38
    jointe
    -0.38
    ɒ
    -0.38
    POSITIVE LOGITS
    uxxxx
    0.71
    parsedMessage
    0.70
     '\\;'
    0.67
     newOwner
    0.56
    LookAnd
    0.55
     surla
    0.52
    InitVars
    0.51
     fieldNum
    0.51
     kasarigan
    0.49
    AsUp
    0.48
    Act Density 0.372%

    No Known Activations