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

    Neuronpedia

    INDEX
    Explanations

    want to talk or try

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     мел
    0.56
    がいる
    0.56
    存在
    0.55
     actuation
    0.55
     eficiente
    0.55
    存在する
    0.54
     avoids
    0.54
     항상
    0.54
    が大きい
    0.54
    常に
    0.54
    POSITIVE LOGITS
     try
    0.85
     explore
    0.81
     brainstorm
    0.81
     embark
    0.78
     discuss
    0.72
     Discuss
    0.68
     go
    0.68
    試
    0.66
     take
    0.66
     join
    0.66
    Act Density 0.633%

    No Known Activations