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

    Neuronpedia

    INDEX
    Explanations

    references to blackmail and privacy violations

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

    blackmail, threats, and corruption

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Net
    -0.36
    Sund
    -0.36
    Meta
    -0.36
    chir
    -0.34
     Cookbook
    -0.33
    sort
    -0.32
     arrivo
    -0.31
     Tripod
    -0.31
     tenda
    -0.31
     Meta
    -0.30
    POSITIVE LOGITS
     EconPapers
    0.78
     blackmail
    0.75
    ########.
    0.68
     referrerpolicy
    0.65
     noDo
    0.60
    HtmlAttribute
    0.60
    Threats
    0.59
    DockStyle
    0.59
     threatened
    0.57
     CreateTagHelper
    0.57
    Act Density 0.085%

    No Known Activations