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

    Neuronpedia

    INDEX
    Explanations

    strings related to email addresses or similar contact details

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Tokens after certain letters

    np_acts-logits-general · gemini-2.0-flash

    "uE" or "UE" detection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ModelExpression
    -0.73
     queſto
    -0.71
    EDEFAULT
    -0.69
    RenderAtEndOf
    -0.69
     zwiſchen
    -0.68
    ſtre
    -0.67
    CppMethod
    -0.66
    دانشنامهٔ
    -0.63
    EndContext
    -0.62
    <unused0>
    -0.61
    POSITIVE LOGITS
     E
    0.40
     B
    0.38
     D
    0.35
     G
    0.34
     b
    0.34
     W
    0.33
     F
    0.32
     S
    0.32
     g
    0.32
     V
    0.32
    Act Density 2.775%

    No Known Activations