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

    Neuronpedia

    INDEX
    Explanations

    words related to placeholders and their significance in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    /archive
    -0.16
     meiden
    -0.16
    elman
    -0.15
    macen
    -0.15
    _snap
    -0.14
     Gord
    -0.14
    æĭ³
    -0.14
    lund
    -0.14
     McN
    -0.14
    μÏĢο
    -0.14
    POSITIVE LOGITS
    -edge
    0.16
    jm
    0.16
     Dear
    0.15
    urement
    0.15
     edge
    0.15
    Ł
    0.15
    zl
    0.15
     Edge
    0.15
    RefCount
    0.15
    ropa
    0.14
    Act Density 0.024%

    No Known Activations