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

    Neuronpedia

    INDEX
    Explanations

    terms related to safety and secure handling

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

    safely precedes action or profitable operations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ComVisible
    -0.73
     canina
    -0.61
     raiſ
    -0.59
     FetchType
    -0.59
     HttpNotFound
    -0.58
    ſelves
    -0.57
    ↹
    -0.57
     TextAppearance
    -0.56
    SequentialGroup
    -0.56
     houſe
    -0.56
    POSITIVE LOGITS
    saf
    0.54
     safely
    0.53
     safe
    0.52
    Safe
    0.51
     Safe
    0.50
    safe
    0.50
    Saf
    0.45
     SAFE
    0.42
     profitably
    0.41
     Safely
    0.40
    Act Density 0.223%

    No Known Activations