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

    Neuronpedia

    INDEX
    Explanations

    the name "Horace" or similar variations

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific individual or character, likely related to experience or assessment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãģ¦
    -0.80
    ãģ®éŃĶ
    -0.73
     Leilan
    -0.69
    ħĭ
    -0.68
    anwhile
    -0.66
    0000000000000000
    -0.64
     graded
    -0.62
    eer
    -0.62
     compliment
    -0.61
     vigorously
    -0.61
    POSITIVE LOGITS
    izontal
    1.44
    izont
    1.40
    izons
    1.30
    rible
    1.28
    izon
    1.20
    rors
    1.14
    ribly
    1.13
    cru
    1.12
    oscope
    1.11
    ror
    1.08
    Act Density 0.034%

    No Known Activations