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

    Neuronpedia

    INDEX
    Explanations

    security and safety-related instructions or warnings

    oai_token-act-pair · gpt-3.5-turbo

    imperative statements or advice directing actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ELD
    -0.74
    atz
    -0.64
    Said
    -0.62
    ById
    -0.62
    lishes
    -0.61
    Winner
    -0.58
     crashed
    -0.55
    fred
    -0.55
    lime
    -0.55
    ammed
    -0.55
    POSITIVE LOGITS
     beware
    1.40
     consult
    1.26
     avoid
    1.16
     carefully
    1.15
     ALWAYS
    1.13
     acquaint
    1.12
     consider
    1.09
     heed
    1.08
     refrain
    1.08
    avoid
    1.08
    Act Density 0.215%

    No Known Activations