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

    Neuronpedia

    INDEX
    Explanations

    Okay or oh beginning a response

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    raped
    0.44
    writeValue
    0.39
    就这样
    0.39
    Seriously
    0.38
    stest
    0.38
    istered
    0.38
    と呼ばれる
    0.38
    clap
    0.38
     abbastanza
    0.37
     blah
    0.37
    POSITIVE LOGITS
     selecting
    0.43
     devising
    0.41
     neat
    0.40
     suggesting
    0.39
    selecting
    0.39
     DIFFIC
    0.38
    攻
    0.38
     shoot
    0.38
    лю
    0.37
     літоў
    0.37
    Act Density 0.046%

    No Known Activations