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

    Neuronpedia

    INDEX
    Explanations

    phrases describing something as a specific type of thing or an example of a particular trait

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include the structure "something of a/an"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ps
    -0.72
     sshd
    -0.69
    tails
    -0.69
    fts
    -0.67
    Psy
    -0.63
    ampa
    -0.63
     Daniels
    -0.60
    oons
    -0.58
    NSA
    -0.58
     Cran
    -0.57
    POSITIVE LOGITS
    etheless
    0.69
    value
    0.67
    ingen
    0.63
     shame
    0.62
     appreciated
    0.62
    angu
    0.61
     modest
    0.60
    luster
    0.60
     kit
    0.60
     sentimental
    0.60
    Act Density 0.046%

    No Known Activations