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

    Neuronpedia

    INDEX
    Explanations

    references to a specific subject or topic being discussed

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

    identifying "this" followed by a noun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    RTLD
    -0.55
    WithIOException
    -0.55
    Panamoan
    -0.54
    ScopeManager
    -0.53
    setVerticalGroup
    -0.52
     Италијани
    -0.50
     словарь
    -0.50
    ktop
    -0.49
    lorum
    -0.49
    JspWriter
    -0.49
    POSITIVE LOGITS
     is
    0.54
     would
    0.50
     has
    0.48
     can
    0.48
     Bewußt
    0.43
     will
    0.43
     could
    0.43
     éste
    0.43
     was
    0.43
     Đây
    0.43
    Act Density 0.071%

    No Known Activations