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

    Neuronpedia

    INDEX
    Explanations

    Encryption Type

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nekaj
    0.67
     مید
    0.66
    的人生
    0.66
    人生
    0.63
    ডে
    0.63
    nsan
    0.59
    environ
    0.58
    чит
    0.57
     something
    0.57
     przy
    0.56
    POSITIVE LOGITS
     frequency
    1.25
     types
    1.23
     Frequency
    1.22
    Frequency
    1.17
     amount
    1.14
    frequency
    1.14
     length
    1.12
     extent
    1.12
     Amount
    1.12
     Type
    1.12
    Act Density 0.744%

    No Known Activations