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

    Neuronpedia

    INDEX
    Explanations

    potential space between

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    દય
    0.51
    极限
    0.51
    犬
    0.50
     cięż
    0.48
    階
    0.47
    Limite
    0.46
    次数
    0.46
    Segundo
    0.46
     അവളെ
    0.46
    횟
    0.45
    POSITIVE LOGITS
     Blackboard
    0.48
    kke
    0.46
     Requires
    0.45
    k
    0.43
    /
    0.42
    ponsors
    0.40
     Complete
    0.40
     gates
    0.40
    arettes
    0.39
    requires
    0.39
    Act Density 0.003%

    No Known Activations