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

    Neuronpedia

    INDEX
    Explanations

    legal and formal language in a text

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

    Follows punctuation, especially periods and parentheses

    np_acts-logits-general · gemini-2.0-flash

    any or unless

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Ternyata
    -0.60
    daj
    -0.60
    Portail
    -0.60
     AssemblyCompany
    -0.59
     ModelExpression
    -0.59
    mostly
    -0.55
     now
    -0.55
    Believe
    -0.53
    Mainly
    -0.53
     mostly
    -0.53
    POSITIVE LOGITS
     Any
    0.96
    Any
    0.94
     any
    0.92
     Cualquier
    0.81
     cualquier
    0.78
     unless
    0.74
     eventuali
    0.73
     qualquer
    0.73
    任何
    0.73
    任何人
    0.72
    Act Density 0.444%

    No Known Activations