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

    Neuronpedia

    INDEX
    Explanations

    phrases related to actions involving removal

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "removal."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Cola
    -0.81
     Balanced
    -0.75
    sb
    -0.72
    jah
    -0.71
    Fight
    -0.68
     Lobby
    -0.68
    ¯¯
    -0.67
    rium
    -0.66
    Kind
    -0.65
     Grad
    -0.65
    POSITIVE LOGITS
     removal
    1.08
    umatic
    1.01
     inhibitor
    0.91
    chwitz
    0.86
    avorite
    0.83
     inhibitors
    0.82
     queens
    0.79
     blockers
    0.77
     erection
    0.77
     gobl
    0.76
    Act Density 0.005%

    No Known Activations