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

    Neuronpedia

    INDEX
    Explanations

    match choice with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Director
    0.41
     diret
    0.41
     gain
    0.40
    Director
    0.40
    asie
    0.40
     Diret
    0.40
     γ
    0.39
     director
    0.38
     Classics
    0.38
    有效
    0.38
    POSITIVE LOGITS
    ሹ
    0.40
     changeable
    0.39
    ኰ
    0.38
    Banana
    0.38
    क्सीडेंट
    0.38
    పడుతుంది
    0.36
    grasp
    0.36
    Equipment
    0.36
    Scatter
    0.36
     শনাক্ত
    0.36
    Act Density 0.001%

    No Known Activations