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

    Neuronpedia

    INDEX
    Explanations

    Java Kotlin classes and colors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    !)
    -1.33
     ところ
    -1.30
    !”
    -1.30
    彼らの
    -1.29
    余談
    -1.27
    闋
    -1.27
    ︕
    -1.27
    蓐
    -1.27
     SPAIN
    -1.26
    行业的
    -1.24
    POSITIVE LOGITS
     happenings
    1.61
     \
    1.54
    \\
    1.52
     Also
    1.48
    狨
    1.45
     +\
    1.44
     ПРО
    1.43
    侂
    1.43
    Что
    1.38
    的一个
    1.38
    Act Density 0.002%

    No Known Activations