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

    Neuronpedia

    INDEX
    Explanations

    words related to stealth or secrecy

    oai_token-act-pair · gpt-3.5-turbo

    words related to stealthy actions and hidden movements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rex
    -0.82
     Anger
    -0.79
    respond
    -0.77
     Respond
    -0.74
     Scale
    -0.72
    Resp
    -0.71
     RESP
    -0.67
    iov
    -0.66
     tyres
    -0.64
    ourses
    -0.64
    POSITIVE LOGITS
     sneaking
    2.12
     lurking
    2.08
     sneak
    2.03
     tucked
    1.81
     slipped
    1.74
     creeping
    1.72
     unnoticed
    1.71
     crept
    1.69
     sneaky
    1.68
     slipping
    1.64
    Act Density 0.031%

    No Known Activations