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

    Neuronpedia

    INDEX
    Explanations

    words related to misunderstandings or mistakes

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "mist" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     scratch
    -0.66
     Corpus
    -0.65
     Esc
    -0.64
     Cue
    -0.63
    phrine
    -0.61
    */(
    -0.61
     grease
    -0.58
     Skydragon
    -0.58
     eminent
    -0.58
     Soup
    -0.57
    POSITIVE LOGITS
    reatment
    1.89
    reated
    1.71
    rans
    1.47
    reating
    1.45
    aking
    1.33
    reat
    1.31
    resses
    1.30
    ypes
    1.28
    arget
    1.25
    rial
    1.23
    Act Density 0.033%

    No Known Activations