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

    Neuronpedia

    INDEX
    Explanations

    specific expressions of comparison and negation

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

    descriptive actions or states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     vnd
    -0.38
     cor
    -0.37
     FirebaseAuth
    -0.36
     bes
    -0.36
    しょ
    -0.35
    ğraf
    -0.35
     boc
    -0.35
    Upgradeable
    -0.34
    äuser
    -0.34
     irre
    -0.34
    POSITIVE LOGITS
     onOptions
    0.52
     InputDecoration
    0.51
     capucha
    0.48
     esclavos
    0.47
     manguera
    0.47
     spécialisé
    0.46
    InjectAttribute
    0.46
     незавершена
    0.45
     burbujas
    0.45
     Übers
    0.45
    Act Density 0.294%

    No Known Activations