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

    Neuronpedia

    INDEX
    Explanations

    concluding sentences or actions

    oai_token-act-pair · gpt-3.5-turbo

    sentences that end with a period

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     invincible
    -0.76
     toll
    -0.75
    vation
    -0.74
    lings
    -0.73
     overt
    -0.73
    bound
    -0.72
     widow
    -0.71
     obliter
    -0.69
     killing
    -0.69
     purse
    -0.69
    POSITIVE LOGITS
     Hopefully
    1.28
     Specifically
    1.26
     Firstly
    1.18
     Ideally
    1.13
     Unfortunately
    1.13
     Typically
    1.12
     Whilst
    1.11
     Regardless
    1.10
     Basically
    1.09
     Depending
    1.08
    Act Density 0.556%

    No Known Activations