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

    Neuronpedia

    INDEX
    Explanations

    phrases related to suffering or distress

    oai_token-act-pair · gpt-3.5-turbo

    terms related to suffocation or significant constraints

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Pegasus
    -0.82
     Drake
    -0.72
    Pic
    -0.71
     Kislyak
    -0.69
     Oak
    -0.68
    rake
    -0.67
    tec
    -0.65
    hran
    -0.65
     Onion
    -0.65
    ho
    -0.63
    POSITIVE LOGITS
     suff
    4.27
     Suff
    2.64
    suff
    2.22
     drown
    1.10
     ana
    1.00
     suffice
    1.00
     paraly
    0.94
     murd
    0.92
     inhal
    0.92
     stricken
    0.91
    Act Density 0.017%

    No Known Activations