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

    Neuronpedia

    INDEX
    Explanations

    well-crafted documents with positive evaluations and clear descriptions

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

    Preceding "in," "and," "between," or a username

    np_acts-logits-general · gemini-2.0-flash

    well presented / professional and

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    
    -0.47
     disambiguazione
    -0.47
     ujednoznacz
    -0.46
    fromnode
    -0.44
    heça
    -0.41
     GenerationType
    -0.38
    Personensuche
    -0.38
    CppCodeGen
    -0.38
    principalColumn
    -0.37
     ✭✭
    -0.36
    POSITIVE LOGITS
    plenty
    0.50
     nicely
    0.47
     thoughtful
    0.45
     klart
    0.45
    clearly
    0.45
     efficiently
    0.45
     well
    0.45
     respectful
    0.44
     sensible
    0.44
    reasonable
    0.43
    Act Density 0.136%

    No Known Activations