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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating reasoning or justifications

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

    concluding based on reasons

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    verwijspagina
    -0.53
     يتيمه
    -0.45
     disambiguazione
    -0.44
    RetentionPolicy
    -0.43
    uroy
    -0.42
     longtemps
    -0.40
    homonymie
    -0.38
    seck
    -0.38
    FunctionFlags
    -0.38
    VersionUID
    -0.38
    POSITIVE LOGITS
     therefore
    0.68
    Поэтому
    0.67
     Therefore
    0.67
    Therefore
    0.63
     pertanto
    0.59
    therefore
    0.59
     Accordingly
    0.56
     daher
    0.55
     Derfor
    0.54
    latego
    0.54
    Act Density 0.150%

    No Known Activations