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

    Neuronpedia

    INDEX
    Explanations

    predicting DEX value

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    App
    0.56
    Comm
    0.55
    Hi
    0.55
    Option
    0.54
    Bonjour
    0.53
    Person
    0.52
    Oper
    0.51
    Associate
    0.51
    Or
    0.50
    Emma
    0.50
    POSITIVE LOGITS
     trình
    0.55
     pointless
    0.53
     thiết
    0.50
     ی
    0.49
     про
    0.49
     lợi
    0.49
     lawsuit
    0.49
     दावे
    0.49
     ranked
    0.48
     susceptibility
    0.48
    Act Density 0.000%

    No Known Activations