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

    Neuronpedia

    INDEX
    Explanations

    words related to medical conditions or activities relating to physical health

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sh"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     numbered
    -0.63
     Ajax
    -0.62
     circulation
    -0.59
     cavity
    -0.59
     Leilan
    -0.58
     Claud
    -0.57
     recl
    -0.57
     Swed
    -0.56
     Juven
    -0.56
     Reach
    -0.56
    POSITIVE LOGITS
    ifts
    1.21
    ifting
    1.19
    tml
    1.17
    rimp
    1.16
    ifted
    1.12
    adows
    1.12
    ackle
    1.11
    IFT
    1.11
    ooting
    1.05
    ifter
    1.05
    Act Density 0.021%

    No Known Activations