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

    Neuronpedia

    INDEX
    Explanations

    distinct qualities and features associated with descriptors in various contexts

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

    distinguishing characteristics or signatures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     comp
    -0.35
    DATABASES
    -0.34
     ep
    -0.32
    older
    -0.32
     tired
    -0.31
     halter
    -0.31
    GLenum
    -0.30
    좇
    -0.30
     provisional
    -0.30
    polation
    -0.30
    POSITIVE LOGITS
     uniqueness
    0.93
     distinctive
    0.90
     signature
    0.88
     unique
    0.82
    signature
    0.82
    特徴
    0.80
    unique
    0.80
     trademark
    0.77
     characteristic
    0.76
     signatures
    0.75
    Act Density 0.144%

    No Known Activations