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

    Neuronpedia

    INDEX
    Explanations

    references to mistakes or errors in processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ogne
    -0.16
     obs
    -0.15
    gili
    -0.14
     unreal
    -0.14
    itere
    -0.14
    nh
    -0.13
     Shack
    -0.13
     atom
    -0.13
    idge
    -0.13
     altern
    -0.13
    POSITIVE LOGITS
     accident
    0.62
     Accident
    0.50
     accidents
    0.46
     accidental
    0.45
     ÑģлÑĥÑĩай
    0.41
     accidentally
    0.38
    acc
    0.34
     mistake
    0.34
     inadvert
    0.32
     Acc
    0.29
    Act Density 0.244%

    No Known Activations