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

    Neuronpedia

    INDEX
    Explanations

    numerals indicating a ranking or classification within a list

    oai_token-act-pair · gpt-3.5-turbo

    opening parentheses in sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     habitable
    -0.75
     seasonal
    -0.74
     tamp
    -0.72
     retard
    -0.71
     flow
    -0.71
     zoo
    -0.70
     deterrent
    -0.70
     overd
    -0.68
     secrets
    -0.68
     prey
    -0.67
    POSITIVE LOGITS
    along
    1.45
    including
    1.45
    formerly
    1.43
    aka
    1.40
    sic
    1.40
    excluding
    1.40
    which
    1.35
    also
    1.35
    via
    1.33
    whose
    1.32
    Act Density 0.148%

    No Known Activations