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

    Neuronpedia

    INDEX
    Explanations

    character set for strings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     уника
    0.51
    unique
    0.50
     unique
    0.47
    apor
    0.45
     uniqueness
    0.44
     unrelated
    0.44
    随机
    0.44
    buah
    0.43
    Unique
    0.43
    Random
    0.43
    POSITIVE LOGITS
     pool
    0.58
     pools
    0.53
     Pool
    0.52
    charset
    0.46
     possível
    0.45
     Pools
    0.45
     पूल
    0.43
     possible
    0.42
    Pool
    0.42
     संभव
    0.40
    Act Density 0.021%

    No Known Activations