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

    Neuronpedia

    INDEX
    Explanations

    expressions related to fear of consequences or repercussions in various contexts

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

    risks or negative outcomes

    np_acts-logits-general · gemini-2.0-flash

    risk and harm

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <bos>
    -0.63
     autorytatywna
    -0.56
    HideFlags
    -0.55
     StatefulWidget
    -0.54
    HasBeenSet
    -0.54
    Климат
    -0.53
    fiées
    -0.53
    FunctionFlags
    -0.52
     TestBed
    -0.52
     gubern
    -0.51
    POSITIVE LOGITS
     يتيمه
    0.95
     risk
    0.85
     risking
    0.83
     harm
    0.83
     jeopardi
    0.80
     jeopardize
    0.78
     risked
    0.78
    万一
    0.76
     potentially
    0.76
     lose
    0.73
    Act Density 0.481%

    No Known Activations