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

    Neuronpedia

    INDEX
    Explanations

    version numbers with dots

    oai_token-act-pair · gemini-2.0-flash

    words or patterns resembling old English or archaic forms

    np_token-act-pair-logits · gpt-4o-mini

    period

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .
    -0.59
     a
    -0.57
     of
    -0.54
     di
    -0.53
     ol
    -0.52
     hal
    -0.49
     aus
    -0.49
     qu
    -0.49
     che
    -0.49
     aan
    -0.49
    POSITIVE LOGITS
     raiſ
    1.15
     itſelf
    1.06
     myſelf
    1.05
     faſt
    1.04
     uſed
    1.00
     uſe
    0.95
     fevere
    0.94
     pleaſure
    0.89
     ſever
    0.89
     ſche
    0.88
    Act Density 1.267%

    No Known Activations