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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a quantity of something that can fulfill a specific purpose

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating sufficiency or adequacy

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    cycl
    -0.74
    Choice
    -0.66
    hered
    -0.65
    anamo
    -0.65
     Actions
    -0.64
    arted
    -0.64
     Mehran
    -0.63
     Glob
    -0.61
    soever
    -0.61
    updated
    -0.60
    POSITIVE LOGITS
     warrant
    1.54
     justify
    1.50
     satisfy
    1.45
     qualify
    1.37
     outweigh
    1.30
     convince
    1.27
     merit
    1.22
     constitute
    1.19
     sustain
    1.15
     motivate
    1.14
    Act Density 0.085%

    No Known Activations